ACTA GEOGRAPHICA GEOGRAFSKI ZBORNIK SLOVENICA 2020 60 1 ACTA GEOGRAPHICA SLOVENICA GEOGRAFSKIZBORNIK 60-1 2020 2 ZNANSTVENORAZISKOVALNI CENTER SLOVENSKE AKADEMIJE ZNANOSTI IN UMETNOSTI GEOGRAFSKI INŠTITUT ANTONA MELIKA • RESEARCH CENTRE OF THE SLOVENIAN ACADEMY OF SCIENCES AND ARTS ANTON MELIK GEOGRAPHICAL INSTITUTE ACTA GEOGRAPHICA SLOVENICA GEOGRAFSKIZBORNIK 60-1 2020 LJUBLJANA 2020 ACTA GEOGRAPHICA SLOVENICA 2020 ISSN: 1581-6613 UDC: 91 © 2020, ZRC SAZU, Geografski inštitut Antona Melika Internationaleditorialboard/mednarodniuredniškiodbor:ZoltánBátori(Hungary),David Bole(Slovenia),MarcoBontje(theNetherlands), Mateja Breg Valjavec (Slovenia), Michael Bründl (Switzerland), Rok Ciglič (Slovenia), Lóránt Dénes Dávid (Hungary), Mateja Ferk (Slovenia), Matej Gabrovec (Slovenia), Matjaž Geršič (Slovenia), Maruša Goluža (Slovenia), Mauro Hrvatin (Slovenia), Ioan Ianos (Romania), Peter Jordan (Austria), Drago Kladnik (Slovenia), Blaž Komac (Slovenia), Jani Kozina (Slovenia), Andrej Kranjc (Slovenia), Matej Lipar (Slovenia), Dénes Lóczy (Hungary), Simon McCarthy (United Kingdom), Slobodan B. Marković (Serbia), Janez Nared (Slovenia), Cecilia Pasquinelli (Italy), Drago Perko (Slovenia), Florentina Popescu (Romania), Garri Raagmaa (Estonia), Ivan Radevski (North Macedonia), Marjan Ravbar (Slovenia), Nika Razpotnik Visković (Slovenia), Aleš Smrekar (Slovenia), Vanya Stamenova(Bulgaria), Annett Steinführer (Germany), Mateja Šmid Hribar (Slovenia), Jure Tičar (Slovenia), Jernej Tiran (Slovenia), Radislav Tošić (Bosnia and Herzegovina), Mimi Urbanc (Slovenia), Matija Zorn (Slovenia), Zbigniew Zwolinski (Poland) Editors-in-Chief/glavna urednika: Rok Ciglič; rok.ciglic@zrc-sazu.si, Blaž Komac; blaz.komac@zrc-sazu.si Executive editor/odgovorni urednik: Drago Perko; drago.perko@zrc-sazu.si Chief editors for physical geography/glavni uredniki za fizično geografijo: Mateja Ferk; mateja.ferk@zrc-sazu.si, Matej Lipar; matej.lipar@zrc-sazu.si, Matija Zorn; matija.zorn@zrc-sazu.si Chief editors for human geography/glavni uredniki za humano geografijo: Jani Kozina; jani.kozina@zrc-sazu.si,Mateja Šmid Hribar; mateja.smid@zrc-sazu.si, Mimi Urbanc; mimi.urbanc@zrc-sazu.si Chief editors for regional geography/glavni uredniki za regionalno geografijo: Matej Gabrovec; matej.gabrovec@zrc-sazu.si, Matjaž Geršič; matjaz.gersic@zrc-sazu.si, Mauro Hrvatin; mauro.hrvatin@zrc-sazu.si Chief editors for regional planning/glavni uredniki za regionalno planiranje: David Bole; david.bole@zrc-sazu.si, Janez Nared; janez.nared@zrc-sazu.si, Nika Razpotnik Visković; nika.razpotnik@zrc-sazu.si Chief editors for environmental protection/glavni uredniki za varstvo okolja: Mateja Breg Valjavec; mateja.breg@zrc-sazu.si, Jernej Tiran; jernej.tiran@zrc-sazu.si, Aleš Smrekar; ales.smrekar@zrc.sazu.si Editorial assistant/uredniška pomočnica: Maruša Goluža; marusa.goluza@zrc-sazu.si Journal editorial system manager/upravnik uredniškega sistema revije: Jure Tičar; jure.ticar@zrc.sazu.si Issued by/izdajatelj: Geografski inštitut Antona Melika ZRC SAZU Published by/založnik: Založba ZRC Co-published by/sozaložnik: Slovenska akademija znanosti in umetnosti Address/naslov: Geografski inštitut Antona Melika ZRC SAZU, Gosposka ulica 13, SI – 1000 Ljubljana, Slovenija The papers are available on-line/prispevki so dostopni na medmrežju: http://ags.zrc-sazu.si (ISSN: 1581–8314) Ordering/naročanje: Založba ZRC, Novi trg 2, p. p. 306, SI – 1001 Ljubljana, Slovenija; zalozba@zrc-sazu.si Annual subscription/letna naročnina: 20 € for individuals/za posameznike, 28 € for institutions/za ustanove. Single issue/cena posamezne številke: 12,50 € for individuals/za posameznike, 16 € for institutions/za ustanove. Cartography/kartografija: Geografski inštitut Antona Melika ZRC SAZU Translations/prevodi: DEKS, d. o. o. DTP/prelom: SYNCOMP, d. o. o. Printed by/tiskarna: Present, d. o. o. Print run/naklada: 400 copies/izvodov The journal is subsidized by the Slovenian Research Agency and is issued in the framework of the Geography of Slovenia core research pro-gramme (P6-0101)/Revija izhaja s podporo Javne agencije za raziskovalno dejavnost Republike Slovenije in nastaja v okviru raziskovalnega programa Geografija Slovenije (P6-0101). The journal is indexed also in/Revija je vključena tudi v: SCIE – Science Citation Index Expanded, Scopus, JCR – Journal Citation Report/Science Edition, ERIH PLUS, GEOBASE Journals, Current geographical publications, EBSCOhost, Geoscience e-Journals, Georef, FRANCIS, SJR (SCImago Journal & Country Rank), OCLC WorldCat, Google scholar, and CrossRef. Design by/Oblikovanje: Matjaž Vipotnik. Front cover photography: In Debeli Rtič Landscape Park, coastal flysch cliffs and wave-cut platform intertwine (photograph: Jure Tičar). Fotografija na naslovnici: V Krajinskem parku Debeli rtič se prepletajo obalni flišni klifi in abrazijske police (fotografija: Jure Tičar). ISSN: 1581-6613 UDC: 91 Number: 60-1 Year: 2020 Contents Mojca POKLAR Comparison of the sonar recording method and the aerial photographymethod for mapping seagrass meadows 7 Vanja PAVLUKOVIĆ, Uglješa STANKOV, Daniela ARSENOVIĆ Social impacts of music festivals: A comparative study of Sziget (Hungary)and Exit (Serbia) 21 Péter János KISS, Csaba TÖLGYESI, Imola BÓNI, László ERDŐS,András VOJTKÓ, István Elek MAÁK, Zoltán BÁTORI The effects of intensive logging on the capacity of karst dolines to providepotential microrefugia for cool-adapted plants 37 Radu SĂGEATĂ Commercial services and urban space reconversion in Romania (1990–2017) 49 Kristina IVANČIČ, Jernej JEŽ, Blaž MILANIČ, Špela KUMELJ, Andrej ŠMUC Application of a mass movement susceptibility model in the heterogeneous Mioceneclastic successions of the Slovenj Gradec Basin, northeast Slovenia 61 Andrej GOSAR Measurements of tectonic micro-displacements within the Idrija fault zone in theUčja valley (W Slovenia) 79 Piotr RAŹNIAK, Sławomir DOROCKI, Anna WINIARCZYK-RAŹNIAK Economic resilience of the command and control function of cities in Central and Eastern Europe 95 Mateja FERK, Rok CIGLIČ, Blaž KOMAC, Dénes LÓCZY Management of small retention ponds and their impact on flood hazard prevention in the Slovenske Gorice Hills 107 5 Gregor KOVAČIČSediment production in flysch badlands: A case study from Slovenian Istria 127 Vesna LUKIĆ, Aleksandar TOMAŠEVIĆ Immigrant integration regimes in Europe: Incorporating the WesternBalkan countries 143 Mitja DURNIKCommunity development: Local Immigration Partnerships in Canadaand implications for Slovenia 155 6 COMPARISON OF THE SONAR RECORDING METHOD AND THE AERIAL PHOTOGRAPHY METHOD FOR MAPPING SEAGRASS MEADOWS Mojca Poklar Underwater image of a meadow in Semedela Bay. DOI: https://doi.org/10.3986/AGS.5161 UDC: 911.2:582.533.1(497.4Semedelski zaliv) 681.883:582.533.1(497.4Semedelski zaliv) 528.7:582.533.1(497.4Semedelski zaliv) COBISS: 1.01 Mojca Poklar1 Comparison of the sonar recording method and the aerial photography method for mapping seagrass meadows ABSTRACT: This article presents a new perspective on the study of the spatial distribution of seagrass meadows,which–duetotheirsensitivitytocoastalhydrodynamics,sedimenttransport,changesinnutri­entcontent,anddisruptionsduetohumaninterventionintheirenvironment–areagoodindirectindicator of the properties of seawater. Monitoring their extent and characteristics is essential for determining the properties of seawater, but this requires developing a precise methodology that involves acquiring data on the occurrence of seagrass meadows and mapping them. The base data for the survey presented are sonar recording and aerial photography data, which were utilized to create a seabed classification using geographicinformationsystems(GIS).Thisprovidedinformationontheextentandcharacteristicsofthe seagrass meadows. Spatial analysis offers a new look at the coastal belt and reveals some new features. KEYWORDS:geography,SemedelaBay,seagrassmeadows,multibeamsonardata,aerialphotography,GIS, line transect method, coastal area Primerjava metode sonarskega snemanja in metode zračne fotografije za namen kartiranja morskih travnikov POVZETEK: Prispevek prikazujenov pogled na preučevanje prostorske porazdelitve morskih travnikov, ki so zaradi njihove občutljivosti na obalno hidrodinamiko, transport sedimentov, spremembe vsebnos­tihranilinmotnjezaradičlovekovegaposeganjavnjihovookolje,doberposrednipokazateljlastnostimorske vode.Spremljanjenjihovegaobsegainlastnostijenamrečbistvenopriugotavljanjulastnostimorskevode, zahteva pa natančno izdelano metodologijo, ki vključuje pridobivanje podatkov o razširjenosti morskih travnikov in njihovo kartiranje. Izhodišče za izvedeno raziskavo so bili podatki sonarskega snemanja in zračnefotografije,nakaterihsmozuporabogeografskoinformacijskihsistemovizvedlitipizacijomorskega dna, kjer je bila posebna pozornost posvečena morskim travnikom. S tem smo dobili podatke o obsegu inlastnostimorskihtravnikov.Prostorskeanalizesoomogočilenovpoglednaobalnipasinrazkrilenekatere nove značilnosti. KLJUČNE BESEDE: geografija, Semedelski zaliv, morski travniki, podatki večsnopnega sonarja, zračna fotografija, GIS, metoda linijskih presekov, obalno območje The article was submitted for publication on July 5th, 2017. Uredništvo je prejelo prispevek 5. julija 2017. 1 University of Primorska, Faculty of Humanities, Koper, Slovenia mojca.poklar@fhs.upr.si 8 1 Introduction Seagrass meadows are one of the most important marine ecosystems in the world in terms of the goods they produce and ecosystem services they provide (Telesca et al. 2015). Due to their characteristics and sensitivitytocoastalhydrodynamics,sedimenttransport,changesinnutrientcontent,anddisruptiondue to human intervention in their environment, seagrasses are important species in determining the quali­tyofacoastalecosystem(Krause-Jensenetal.2004;Ralphetal.2007;McMahonetal.2013;Peterlin2013; Vacchi et al. 2014). Despite their importance, they are constantly threatened by numerous human activ­ities that eventually lead to their degradation and rapid loss (Duarte 2002), estimated at a rate of 110km2/year since 1980 (Waycott et al. 2009). Consequently, seagrass meadows are regularly included in monitoringprograms,bothfortheirprotectionandfortheirvalueasabioindicator(McMahonetal.2013). To understand the dynamic nature of seagrass meadows and to predict their response to future environ-mentalchanges(Unsworthetal.2014),itisnecessarytosynopticallymonitorthechangesinthecomposition of a meadow, its spatial distribution or cover, andits biomass. Therefore, developing an effective method­ology for monitoring meadows is a very topical issue (Comas Gonzales 2015). Thisstudyfocusesonmappingseagrassmeadowsandthusexaminingtheirspatialdistributionasone of the parameters of monitoring seagrass meadows (Hossain et al. 2014). Because the changes in spatial distribution occur on small (<1km2) and large (>100km2) spatial scales, traditional field surveys (diver observations, sampling using rakes or scrapers, and other methods) are often inconvenient for mapping large areas (McKenzie 2003; Hossain et al. 2014). With the development of geographic information sys­tems (GIS) and technical improvements in remote sensing techniques (Robbins 1997), indirect methods have become more popular. Due to the ability of remote sensing to detect changes in the spatial distrib­ution of seagrass meadows on larger spatial and time scales, it is among the most important tools in the managementofseagrassmeadowsbecauseoftimeefficiency,speedofuse,largecoverage,andreproducibility of observations (Hossain et al. 2014). On a global scale, mapping seagrass meadows using remote sensing techniques is already well known (Hossain et al. 2014), whereas in Slovenia mapping seagrass meadows is still carried out using only field surveys (Turk et al. 2002; Lipej et al. 2007). Despite the fact that there have been some individual attempts to map the seabed using remote sensing techniques (Berden Zrimec, Poklar and Moškon 2015; Berden Zrimec et al. 2015; Moškon et al. 2015), the traditional approach is still predominant. In order to reduce theconstraintsimposedbythisapproach,thisstudycomparedthesonarrecordingmethodandaerialpho­tography method, and it verified this with the already established line transect method for determining the spatial distribution of seagrass meadows. The aim of this research was to evaluate the selected meth­odsbasedontheobtaineddataqualityandtodeterminetheirsuitabilityandopportunitiesforuseinfurther research.Ofparticularinterestwastheaccuracyofbothmethods,especiallyaerialphotography,forwhich it was assumed that the significant water turbidity typical for Slovenian waters and for a large part of the northern Adriatic Sea would be a limiting factor. 2 Methods 2.1 Research area TheresearchareacoveredSemedelaBayasthesoutheasternmostpartofKoperBaybetweenŽusternaand theold town of Koper. This is a shallow bay with an average depth of 6m (Harpha sea 2013), and, despite its strong anthropogenic transformation, its coastline has the characteristics of a depositional coast. Due to its erodible flysch hinterland (Zorn 2009), the Badaševica River carries sediments that are deposited in the sea (Malačič 1994; Orožen Adamič 2002). The area is a uniquehabitatbecause it differs from the cen­tralpartofKoperBayinitsnaturalcharacteristics.Themixingofseawaterandfreshwatervariesconsiderably over the course of the year (Poklar 2016). Due to this variability, the area is suitable for researching the impact of changing water properties on seagrass coverage. Two types of seagrass are found in Semedela Bay: little Neptune grass (Cymodocea nodosa) and common eelgrass (Zostera marina; Lipej et al. 2006). 2.2 Definition of a seagrass meadow Because the perimeter of a seagrass meadow, which is the basis of determining its entire area, cannot be absolutely determined, problems may arise in defining it. In measuring phenomena that are not directly measurable, the need for an operational definition arises. This ensures that the understanding of phenomena and the data collection method are unified and repeatable (Adanza 1995). Therefore, for the purpose of this survey, the operational definition of a seagrass meadow and thereby the minimum mapping unit of 0.01 ha were defined. Even though very sparse seagrass may indicate that seagrass appears in a certain area, such areas were excluded from the operational definition of the seagrass meadow. There are several rea­sons for this: very sparse seagrasses have very little ecological value and also visually do not correspond to the idea of a meadow. In addition, monitoring very sparse seagrass and tracking its changes is very dif­ficult (Virnstein et al. 2000). 2.3 Mapping seagrass meadows using the sonar recording method Sonar data, which are essential for this survey, were obtained from bathymetric measurements with a Reson SeaBat 8125 multibeam echosounder. Measurements were conducted within seven working days (August 28th, September 26th, October 15th, 17th, 23rd, and 25th, and November 5th, 2013) in the morning in clear to cloudy weather with precipitation with winds from 0.0m/s (smooth sea level) up to 5.9m/s (small waves, peaks already breaking; Internet 1; Internet 2). The measurements provided a georeferenced point cloud, which was manually examined in order to avoid incorrect data that occasionally arise due to disturbances in measurements. From processed and systematically organized data, a bathymetric model with a reso­lution of 0.5×0.5m was created, which served as a basis for mapping seagrass meadows. Based on this mapping, a spatial seabed slope analysis was made. Seagrass meadows are higher than the seabed and it was Legend depths (m) 0–1 1–2 2–3 3–4 4–5 5–6 6–7 0 100 200 m Scale: Content by: Mojca Poklar Map by: Mojca Poklar Source: DOF 2014, TK 50 © 2013, Harpha Sea, Koper ´ Figure 1: Depths of the Semedela Bay research area (Source: Podatki snemanja morskega dna z večsnopnim sonarjem 2013). expectedthattheslopesatthetransitionsbetweensiltandmeadowwouldbequitehigh.Theresultinglayer was examined in detail. Because the area of seagrass meadow occurrence was previously recognized from an orthophoto (Digitalni ortofoto 2012), it was known in advance where they could be expected. In these areas, an attempt was made to identify key patterns or edges of seagrass meadows. A vector layer of sea-grass meadows was acquired from the seabed slope raster by exporting all contours of slopes greater than 40°, which was completed and verified with raw sonar data at the end. 2.4 Mapping seagrass meadows using the aerial photography method Aerial photography was used to obtain aerial photos, which were used to digitize seagrass meadows. This wascarriedoutwithaprofessionalcamerawithautomatictriggeringintermsofaircraftheightandveloc­ity, providing 60% overlap of the photos in the forward direction of the flight. The exact location of the aerial photos was ensured by monitoring the position and orientation of the camera on the aircraft using a GNSS receiver and a gyroscope. Aerial photography was carried out in one working day (September 6th, 2013), in the morning during clear weather. Prior to the digitization of seagrass meadows, pre-processing of aerial photos was carried out, which included geometric and lighting corrections. Aerial photos were then merged into a unique photo of the entire research area, which was orthorectified and georeferenced, and its contrast was improved. Dataonthespatialdistributionandthereforeedgesofseagrassmeadowswereobtainedthroughasuper-visedimageclassificationoftheRGBlayersoftheaerialphoto,incombinationwithitsvisualinterpretation. Intheprocessofasupervisedimageclassification,trainingsampleswerefirstcreated;theseareareaswith aknown type of seabed, on which the spectral signature of the seabed type was calculated. Training sam­ples were marked interactively using the training sample drawing tools and were determined by manual limitation. Twelve training samples were determined for various seabed types and in various situations (shadows,seagrassmeadowdensity,etc.).Fortheclassification,themaximumlikelihoodclassificationmethod was used because it is the most accurate, although it is a very demanding computing process (Oštir 2006). The quality of the classification was improvedby visual interpretation of the entire photograph, for which the edges of seagrass meadows were manually corrected by evaluating the basic elements of visual photo interpretation(tone,shape,size,pattern,texture,shadows,etc.);thisisthemostsubjectivepartofthemethod. 2.5 Verification of both methods by comparison with the line transect method and the final map of seagrass meadows Because the sonar recording and aerial photography methods are indirect remote sensing methods, after their implementation they always require ground truth observations toverify the results already obtained (Komatsu et al. 2003). They are helpful in interpreting the distinctive characteristics of seagrasses from sonar data or aerial photos, where they also serve as a reference point for verifying the interpretation of photos; for example, tocheckthatnomacroalgaeorshellsweremisidentifiedasseagrassmeadows(Krause-Jensenetal.2004). Accordingly, to verify the spatial distribution of seagrass meadows obtained by sonar data or aerial photography,andtoevaluatetheaccuracyofselectedmethods,afieldsurveywascarriedout,whichinvolved seabed recordings with an underwater camera. Underwater recordings were made directly from a vessel on predetermined line transects and meadow centroids (Figure 2). Because the line transects were plot­ted by a computer, the precise geographical position and the angles of the recordings were verified with a GNSS receiver and a gyroscope simultaneously with the recordings. The recorded videos of line tran­sects were then processed and converted into underwater photos or raster data. The raster data obtained represented the reference state for the verification of sonar and aerial pho­tographydata.Thefirstphasecomparedthemappededgesofseagrassmeadowsandmeasureddeviations fromthereferencestate.Alongfivelinetransects,forty-sixcontrolpointswererandomlyselected,onwhich the seabed type was determined (the analysis was limited to two types: silt and seagrass meadows) and then compared with sonar and aerial photography data. A comparison also included four points that rep-resentedthecentroidsoftheseagrassmeadows.Basedonthecomparison,theaccuracyofseagrassmeadows mapped with each method was assessed using a confusion matrix (Mumby and Green 2000). Based on the evaluation of the accuracy of two remote sensing methods, the polygon layer of seagrass meadowswasestablished.Wheretheedgesofseagrassmeadowsweredetectedbybothmethods,theedge with the greater accuracy was considered. Where the edges were detected by only one method, the avail­able ones were considered. When checking raster data from the field survey, it did not occur that an edge was not be detected by any method. The result was then mapped using the tools for analysis and spatial display of measured data. 3 Results The multibeam sonar data and the seabed slope analysis showed that seagrass meadows’ edges are clear­lyvisibleinmostcasesbecausetheslopeatthetransitionbetweenthesiltandthemeadowcanrangefrom 0°to80°.Largeranddenserseagrassmeadowsarewellvisible(Figure3),whereastheareaswherethemead­ows are sparse are not. Such areas are difficult to separate from the silt, and so accurate mapping requires a review of raw data (the distance between points was about 10cm) or the combination of sonar data with data from another method. The seagrass meadow edges in Semedela Bay mapped from sonar data are shown in Figure 4. Not all the edges are connected. The western edge, which lies in the eastern part of the bay, was not completely visible because of a gradual transition between seagrass and silt. In this area, the seagrass is sparse and lower in growth, making it difficult to determine its edge. The same applies to seagrass meadows around themouthoftheBadaševicaRiver.Thesonarrecordingmethodmadeitpossibletodrawtheseagrassmead­ow edges with a total length of 5,310.10m. From aerial photography data (i.e., classified aerial photos), it was determined that the edges between theseagrassmeadowsandsiltweremostlyvisible.However,therewereareaswherethephotodidnotmake Figure 2: Selected line transects and sampling points (centroids) for verifying seagrass meadow occurrence. it possible to recognize whether there is a meadow or not. This is especially true for deeper areas, where the lower edge of the meadow is often difficult to determine due to the smaller proportion of light that can penetrate to the seabed or to the meadow. Problems also occurred in areas where the meadow edges were less visible due to reflection of light from the sea surface. The problem was solved by changing the direction of the flight. For the research area, it turned out that the reflection of sunlight from the sea sur­faceislessvisibleintheaerialphotos,whichweretakenbyflyinginanorth–southdirection.Nevertheless, it was not possible to completely solve these problems, and so in the previously described areas the mead­ow edges were difficult to determine. Figure 5 shows edges of seagrass meadows mapped using the aerial photography method and with a total length of 5,727.30m. Anoverviewoftheunderwaterphotosofthefieldsurveyshowedthepresenceofbothtypesofseabed as predicted with the sonar recording method and aerial photography method (silt and seagrass mead­ows). In addition, other species were also found in underwater photos; specifically, various macroalgae that appeared closer to the mouth of the Badaševica River and the noble pen shell or fan mussel (Pinna nobilis), found on the outer part of the seagrass meadow along the harbor at Koper. Figure 3: Example of a seagrass meadow on a seabed slope raster. Figure 4: Seagrass meadow edges in Semedela Bay obtained using the sonar recording method. Figure 5: Seagrass meadow edges in Semedela Bay obtained using the aerial photography method. 14 Acomparisonoftheseagrassmeadowedgesinunderwaterphotoswithmeadowedgesobtainedusing thesonarrecordingmethodshoweddeviationsofupto1m,whereasatthemeadowedgesobtainedthrough aerial photography deviations of up to 3m occurred. Considering the position errors – which were esti­mated between 0.2 and 0.3m (sonar data), between 0 and 1m (aerial photography), and between 0.2 and 0.3m(linetransectdata)–deviationsoccurforvariousreasons.Inthecaseofsonardata,deviationsoccur in areas of gradual transition between seagrass and silt, whereas at the sharp edges of seagrass meadows the contours are completely coincident (centimeter-level accuracy). Major deviations in aerial photogra­phydatacanbeattributedtoerrorsingeoreferencingofaerialphotos,aswellasthepoorvisibilityofseagrass meadow edges from the aerial photo in areas of greater depth and in areas where light reflected from the sea surface during the shooting. To assess the classification accuracy of selected methods (Lillesand and Kiefer 1994), two confusion matriceswereproduced,comparingthepredicteddataofthesonarrecordingoraerialphotographymethod with ground truth (reference) data of the field survey. Figure 6: Example of verifying the sonar recording and aerial photography method by using the line transect method, recorded with an underwater camera on line transect T4. Table 1: Confusion matrix of the a) sonar recording method and b) aerial photography method. a) Reference data Seagrass Silt User accuracy Sonar recording method Seagrass Silt 30 1 3 26 90.9% 96.3% Total number of sampling points Producer accuracy 31 96.8% 29 89.7% Overall accuracy=93.3% b) Reference data Seagrass Silt User accuracy Aerial photography method Seagrass Silt 24 7 13 16 64.9% 69.6% Total number of sampling points Producer accuracy 31 77.4% 29 55.2% Overall accuracy=66.6% Figure7:SeagrassmeadowedgesinSemedelaBaybasedonthemappingmethodused,andthefinalmapofthemeadowsinSemedelaBayinautumn2013. 16 Table1showsthatsamplingpointslabelingseagrassweremorecorrectlyclassifiedthanthosethatlabeled silt. The difference is small (7.1%) for the sonar recording method, whereas for the aerial photography method it is considerably larger and amounts to 22.2%. In contrast, user accuracy, which serves as a guide to the results’ reliability as a prediction tool, shows that in both the sonar recording and aerial photogra­phy methods the silt class is more correctly classified (96.3% by sonar recording and 69.6% by aerial photography).Nevertheless,themostnoticeableinformationinTable1isthedifferencebetweentheover­all accuracy of the methods by which the seagrass meadows were detected. For the sonar recording the overall accuracy was 93.3%, and for the aerial photography it was 66.6%. Based on the accuracy of both methods, a spatial data layer of seagrass meadows was created using complete sonar data (because in this case it is more accurate than aerial photography data) supplement­ed with aerial photography data. Based on the mapping method, a map of seagrass meadows edges was created (Figure 7). Most of the meadow edges (52%) were plotted using the sonar recording method, and theaerialphotographymethodwasusefulfor24%oftheplottededges.Someedgesweredetectedbyboth methods, which is shown as an independent category in Figure 7 (24%). Figure 7, which also shows the spatial distribution of seagrass meadows in Semedela Bay, shows that seagrass meadows are distributed along the coast and in the inner part of the bay. There are two major seagrass meadows as well asa number of minor ones, constituting »islands,« separated from major mead­ows. Seagrass meadows were not detected at depths exceeding 5m, where light conditions do not allow the growth of seagrass, and directly along the coast, especially at the mouth of the Badaševica River. In the past, the Badaševica deposited contaminated and nutrient-rich water in the bay, which contributed to the extremely depleted vegetation at its mouth. Because both Cymodocea nodosa and Zostera marina are sensitive to elevated levels of nutrients in the water column (Lipej et al. 2006; Orfanidis et al. 2007), this could be the main reason for the lower coverage of the seabed with seagrass in the area. Lower coverage of the seabed with seagrass directly along the Semedela promenade in the eastern part of the bay can be attributed to the renovation of the promenade in 2010. The renovation works also consisted of deepen­ing the seabed, which led to physical damage to the seabed and associated vegetation. 4 Discussion Measurements of the spatial distribution of seagrass meadows with the methods presented for Semedela Bayprovidedsomekeyfindingsregardingtheircharacteristics.Thefirstrelatestothetimeframeformak­ingthemeasurements.Theaerialphotographymethodisfasterincomparisontothesonarrecordingmethod because photographing the entire research area was carried out in one day, whereas sonar measurements lasted several days. Another characteristic investigated was the spatial and temporal dependence of the method. It was determined that the sonar recording method, in contrast to the aerial photography method, is a spatial-lyandtime-independentmethodbecausethedatacapturewithamultibeamsonarisindependentofwater transparencyandsunlight,andwithaccurateGNSSandINSreceiversitispossibletoperformqualitymea­surements in the undulating sea. In contrast, the use of the aerial photography method in the Slovenian seais limiteddue tohigh water turbidity. It turned out that the greatest problems arise inthebays (the sea currents are not so strong, the influence of waves is greater than in the open sea, and siltation is promi­nent), where the largest share of seagrass meadows is located. In addition to increased water turbidity, the problem also lies in the refraction and reflection of light on sea surface, which makes it necessary to cap­ture photos at the best time of the day and under the best environmental conditions. The greatest weakness of the aerial photography method is certainly its subjectivity in determining thedistributionofseagrassmeadows.Inordertodetermineseagrassmeadowsfromaerialphotos,animage classification was made, partially also with a manual capture of the edges of seagrass meadows, where it was necessary tovisually evaluate the basic elements of photointerpretation (such as tone, color, contrast, texture, shadows, etc.), which each individual can recognize differently. Because the multibeam sonar spreads beams at ± 60° steering angles (Fridl, Kolega and Žerjal 2008), the method is useful for flat and for more morphologically diverse seabeds, and in addition it is also pos­sibletomeasuretheheightofseagrassanditsbiomassabovetheseabed.However,ifoneisonlyinterested in information on the occurrence of a meadow in a certain area or if the required precision of the mapped meadows is low, the aerial photography method is more appropriate from a user perspective. This is espe­ciallytruewhenanalyzingalreadyexistingaerialorsatelliteimages(inthiscase,onemusttakeintoaccount thelowerresolutionandthusthelowerqualityofsuchimages),which,incontrasttothemethodsdescribed, are more accessible. Inadditiontothesecharacteristicsofbothmethods,theiraccuracywasofprimaryinterest.Considering positionerrors–whichwereestimatedbetween0.2and0.3m(sonardata),0and1m(aerialphotography), and0.2and0.3m(linetransectdata)–theoverallaccuracyofthesonardatawas93.3%,whereastheover-all accuracy of the aerial photos was only 63.3%. Considering that accuracy of classification over 90% is good,andthatover80%issatisfactory(Oštir2006),thesonarrecordingmethodwasgoodinthisresearch case, whereas the aerial photography method did not yield the most accurate results. In this study, sonar recording is a more reliable method of data acquisition on the spatial distribution of seagrass meadows. Ofcourse, this does not apply to less turbid waters and thus to more accurate visibility of seagrass mead-ows,wheretheaerialphotographymethodcanachievethesameaccuracyasthesonarrecordingmethod. In the case at hand, this was noticeable in determining the edges of seagrass meadows in the area from the mouth of the Badaševica River to the inner part of the bay. In that area, seagrass meadows were poor­lyvisibleonsonardataduetotheaforementionedgradualtransitionbetweensparseseagrassandsilt,whereas they were clearly visible in aerial photos due to shallow water (0 to 2m in depth) and thus increased light penetration through the water column to the seabed. 5 Conclusion The purpose of this study was to show that the use of modern remote sensing technologies and GIS tech­niques makes it possible to obtain new and more useful results in mapping seagrass meadows, yielding betterknowledgeoftheecologicalstatusoftheseaandabetterunderstandingoftheprocessesinitscoastal area.Comparisonofthesonarrecordingandaerialphotographymethodswiththealreadyestablishedline transect method for determining the spatial distribution of seagrass meadows in part of the Slovenian sea showed that both methods allow efficient mapping of seagrass meadows, but they differ significantly in certain characteristics. The sonar recording method proved to be more accurate, more objective, and, in contrasttotheaerialphotographymethod,spatiallyandtemporallyindependent,whichisaconsequence ofthehigherwaterturbiditytypicaloftheSloveniansea.Intermsofaffordability,bothmethodsareexpen­sive because high-precision data require high-priced equipment. Iftherequiredprecisionofthemappedseagrassmeadowsisloworoneisonlyinterestedintheoccur­renceofseagrassmeadowsinacertainarea,theaerialphotographymethodismoreappropriatefromauser perspective because already existing aerial or satellite images are much easier to access. However,thechoiceofmethodologyprimarilydependsonthepurposeofresearch/mapping,andthen ontheenvironmentalconditions,wherethewatertransparencyorturbidity,bathymetryandmorphology ofthebay,weather conditions,andavailableresources mustbetaken intoaccount. Itshouldalso benoted thattheremotesensingmethodsusedalonearenotenough.Regardlessofthechoiceofthemappingmethod, after applying it ground truth observations are always required in order to verify the results obtained and to explain the characteristics of seagrass recognized by remote sensing images or sonar measurements. 6 References Adanza, E. G. 1995: Research methods, principles and applications. Manila. BerdenZrimec,M.,Poklar,M.,Moškon,S.2015:Underwaterhabitatmappingusingremotesensingtech­niques. Proceedings GEOSUB – Underwater geology. Trieste. Berden Zrimec, M., Poklar, M., Mozetič, D., Vranac, D. 2015: Seabed habitat mapping with underwater photogrammetry. Abstract book, 52nd European Biology Symposium. Piran. Comas Gonzales, R. 2015: Mapping seagrass meadows, using low altitude aerial images. Ph.D. thesis, Universidade de Lisboa, Faculdade de Ciencias. Lisboa. Internet: http://repositorio.ul.pt/bitstream/ 10451/23040/1/ulfc117350_tm_Robert_Comas_Gonzalez.pdf. Digitalniortofotoposnetki,območjeMestneobčineKoper.LjubljanaGeodetskaupravaRepublikeSlovenije. Ljubljana, 2012. Digitalni ortofoto posnetki, območje Mestne občine Koper. Geodetska uprava Republike Slovenije. Ljubljana, 2014. Duarte,C.M.2002:Thefutureofseagrassmeadows.EnvironmentalConservation29-2.DOI:https://doi.org/ 10.1017/S0376892902000127Fridl, J., Kolega, N., Žerjal, A. 2008: Pomen digitalnega batimetričnega modela za trajnostni razvoj morja. Geodetski vestnik 52-4. Hidrografija 2014: Internet: http://www.hidrografija.si/p1/4-1-5.php (26.11.2014). Hossain, M. S., Bujang, J. S., Zakaria, M. H., Hashim, M. 2014: The application of remote sensing to sea-grassecosystems:anoverviewandfutureresearchprospects.InternationalJournalofRemoteSensing 36-1. DOI: https://doi.org/10.1080/01431161.2014.990649 Komatsu, T.,Mikami, A.,Sultana, S., Ishida,K., Hiraishi,T., Tatsukawa,K. 2003:Hydro-acoustic methods as a practical tool for cartography of seagrass beds. Otsuchi marine science 28. Krause-Jensen, D., Queresma, A. L., Cunha, A. H., Greve, T. M. 2004: European seagrasses: an introduc­tion to monitoring and management. Internet: http://www.seagrasses.org/handbook/european_ seagrasses_high.pdf (5.7.2017). Lillesand, T. M., Kiefer, R. W. 1994: Remote sensing and image interpretation. New York. Lipej, L., Dobrajc, Ž., Forte, J. Mavrič, B., Orlando Bonaca, M., Šiško, M. 2007: Kartiranje habitatnih tipov inpopisvrstnamorskihzavarovanihobmočjihNSDebelirtič,NRStrunjaninNSRtMadona.Nacionalni inštitut za biologijo, Morska biološka postaja. Piran. Lipej, L., Turk, R., Makovec, T. 2006: Ogrožene vrste in habitatni tipi v slovenskem morju. Ljubljana. Malačič, V. 1994: Razvojni projekt občine Koper 2020: varstvo morja in priobalnega pasu. Raziskovalno poročilo. Nacionalni inštitut za biologijo, Morska biološka postaja. Piran. McKenzie, L. 2003: Guidelines for the rapid assessment and mapping of tropical seagrass habitats. Cairns. McMahon, K., Collier, C. J., Lavery, P. S. 2013: Identifying robust bioindicators of light stress in seagrasses: A meta-analysis. Ecological Indicators 30. DOI: https://doi.org/10.1016/j.ecolind.2013.01.030Moškon, S., Žibert, J., Kavšek, B. 2015: Kartiranje morskih travnikov s podatkih mnogosnopnega sonarja. Geografski vestnik 87-1. DOI: https://doi.org/10.3986/GV87106 Mumby, P. J., Green, E. P. 2000: Field survey: building the link between image and reality. Remote Sensing Handbook for Tropical Coastal Management. Paris. OpazovaniinmerjenimeteorološkipodatkipoSloveniji2014.Internet:http://meteo.arso.gov.si/met/sl/archive (18.11.2014). Orfanidis, S., Papathanasiou V., Gounaris, S. 2007: Body size descriptor of Cymodocea nodosa indicates anthropogenicstressincoastalecosystem.TransitionalWatersBulletin2.DOI:https://doi.org/10.1285/i1825229Xv1n2p1 OroženAdamič,M.2002:GeomorfološkeznačilnostiTržaškegazalivainobrobja.Geografijainnjeneaplika­tivne možnosti. Ljubljana. Oštir, K. 2006: Daljinsko zaznavanje. Ljubljana. Peterlin, M. 2013: Načrt upravljanja morskega okolja. Začetna presoja morskih voda v pristojnosti Republike Slovenije. Bistvene lastnosti in značilnosti morskih voda. Ljubljana. Podatki snemanja morskega dna z večsnopnim sonarjem, ožji morski pas. Harpha sea d.o.o. Koper, 2013. Poklar,M.2016:Metodologijageografskegaraziskovanjastikasladkovodnegainmorskegaokoljanaprimeru Semedelskegazaliva.Ph.D.thesis,FakultetazahumanističneštudijeUniverzenaPrimorskem.Koper. Prostorsko-podatkovni sloj meje obale morja. Agencija Republike Slovenije za okolje. Ljubljana, 2010. Ralph, P. J., Durako, M. J., Enriquez, S., Collier, C. J., Doblin, M. A. 2007: Impact of light limitation on seagrasses.JournalofExperimentalMarineBiologyandEcology350,1-2.DOI:https://doi.org/10.1016/ j.jembe.2007.06.017 Robbins, B. D. 1997: Quantifying temporal change in seagrass areal coverage: The use of GIS and low res­olutionaerialphotography.AquaticBotany58-3-4.DOI:https://doi.org/10.1016/S0304-3770(97)00039-9 Telesca, L., Belluscio, A., Criscoli, A., Ardizzone, G., Apostolaki, E. T., Fraschetti, S., Gristina, M., Knittweis, L., Martin, C. S., Pergent. G., Alagna, A., Badalamenti, F., Garofalo, G., Gerakaris, V., Pace, M. L., Pergent-Martini, C., Salomidi, M. 2015: Seagrass meadows (Posidonia oceanica) distribution and trajectories of change. Scientific Reports 5. DOI: https://doi.org/10.1038/srep12505 Turk, R., Orlando Bonaca, M., Makovec, T., Vukovič, A., Lipej, L. 2002: A topographical survey of habitat types in the area characterized by seagrass meadow of Posidonia oceanica in the southern part of the Gulf of Trieste (Northern Adriatic). Annales, Series historia naturalis 12-2. Unsworth, R. K. F., Keulen, M., Coles, R. G. 2014: Seagrass meadows in a globally changing environment. Marine Pollution Bulletin 83-2. DOI: https://doi.org/10.1016/j.marpolbul.2014.02.026 Vacchi, M., Montefalcone, M., Schiaf.no, C. F., Parravicini, V., Bianchi, C. N., Morri, C., Ferrari,M. 2014: TowardsapredictivemodeltoassessthenaturalpositionofthePosidoniaoceanicaseagrassmeadows upper limit. Marine Pollution Bulletin 83-2. DOI: https://doi.org/10.1016/j.marpolbul.2013.09.038 Virnstein, R., Avery, W. M., Johansson, R. O. 2000: On defining the »edge« of a seagrass bed. Reports 145. Waycott, M., Duarte, C. M., Carruthers, T. J. B., Orth, R. J., Dennison, W. C., Olyarnik, S., Calladine, A., Fourqurean, J. W., Heck, K. L., Hughes, A. R., Kendrick, G. A., Kenworthy, W. J., Short, F. T., Williams, S. L. 2009: Accelerating loss of seagrasses across the globe threatens coastal ecosystem Proceedings of the National Academy of Sciences 106-30. DOI: https://doi.org/10.1073/pnas.0905620106 Zorn,M. 2009: ErosionprocessesinSlovene Istria – part1:Soilerosion. ActageographicaSlovenica 49-1. DOI: https://doi.org/10.3986/AGS49102 SOCIAL IMPACTS OF MUSIC FESTIVALS: A COMPARATIVE STUDY OF SZIGET (HUNGARY)AND EXIT (SERBIA) Vanja Pavluković, Uglješa Stankov, Daniela Arsenović The main stage of the Exit festival, 2017. DOI: https://doi.org/10.3986/AGS.6514 UDC: 911.3:78.079(497.11+439) COBISS: 1.01 Vanja Pavluković1, Uglješa Stankov1, Daniela Arsenović1 Social impacts of music festivals: A comparative study of Sziget (Hungary) and Exit (Serbia) ABSTRACT: Music festivals are often seen as a key driver of the city’s economies. Therefore, there is an increasing interest in the impacts associated with them. The aim of this research is to examine residents’ perceptions of the social impacts of two European music festivals, Sziget (Budapest, Hungary) and Exit (Novi Sad, Serbia), applying the modified Festival Social Impact Attitude Scale and to compare the results using Importance–Performance Analysis. Similarities and differences in perceptions of social impacts of two festivals are discussed with the proposition of priorities for destinations. The findings contribute to a deeper understanding of residents’ attitudes toward the impacts of festivals and can be utilized by local authorities to increase the positive and reduce the negative impacts of the festival. KEYWORDS: social impact, music festival, residents’ perception, Serbia, Hungary Družbeni vplivi glasbenih festivalov: Primerjalna študija festivalov Sziget na Madžarskem in Exit v Srbiji POVZETEK:Glasbenifestivalipogostoveljajozaglavnagonilamestnegagospodarstva,zatoseraziskovalci čedaljeboljzanimajoza vplive,povezanez njimi.Ciljteraziskavejena podlagiprilagojenerazličiceocen­jevalne lestvice družbenih vplivov festivalov (angl. Festival Social Impact Attitude Scale) preučiti stališča prebivalcev glede družbenih vplivov dveh evropskih glasbenih festivalov – Szigeta v Budimpešti in Exita v Novem Sadu – ter primerjati rezultate z uporabo analize pomembnosti in uspešnosti (angl. importance-performance analysis). Avtorji predstavijo podobnosti in razlike v stališčih glede družbenih vplivov obeh festivalovterpredlagajoprednostnenalogezaobemesti.Izsledkiraziskaveomogočajoboljšerazumevan­jeodnosovprebivalcevdovplivovfestivalov,lokalneoblastipajihlahkouporabijozaizboljšanjepozitivnih in zmanjšanje negativnih vplivov festivalov. KLJUČNE BESEDE: družbeni vpliv, glasbeni festival, stališča prebivalcev, Srbija, Madžarska The paper was submitted for publication on 23rd November, 2017. Uredništvo je prejelo prispevek 23. novembra 2017. 1 University of Novi Sad, Faculty of Sciences, Department of Geography, Tourismand HotelManagement, Novi Sad, Serbia vanja.dragicevic@dgt.uns.ac.rs, ugljesa.stankov@dgt.uns.ac.rs, daniela.arsenovic@dgt.uns.ac.rs 1 Introduction Theorganizationoffestivalsisoneofthefastest-growingsegmentsofthetourismindustry(Getz2010;Lashua, SpracklenandLong2014;Kim,DuncanandChung2015;BagiranandKurgun2016;GetzandPage 2016). Consequently,thereisanincreasinginterestinresearchonthethemeoffestivals,specificallyintermsofmea­suringtheimpactsoffestivalsonhostcommunities.GetzandPage(2016)highlightthateventtourism,and festivalsaspartofit,areprimarilydrivenbytheeconomicbenefits(Dwyer,ForsythandSpurr2006;Herrero et al. 2006; Dwyer, Jago and Forsyth 2016), and so much research has been devoted to economic impacts, while other outcomes were neglected for many years. In addition, local authorities and festival organizers focus on the economic benefits of the event, and there is no doubt that they are important, but the social impacts may have an even more profound effect on the local community (Delamere 2001). Recently a fair amount of research (Delamere 2001; Delamere, Wankel and Hintch 2001; Fredline, Jago and Deery 2003; SmallandEdwards2003;Small,EdwardsandSheridan2005;Small2007;RollinsandDelamere2007;Woosnam, VanWinkleandAn2013;Dragićevićetal.2015;BagiranandKurgan2016;Woosnametal.2016;Pavluković, ArmenskiandAlcantara-Pilar2017)hasbeenconductedconcerningthesocialimpactsoffestivals.However, Pavluković,Armenski,andAlcantara-Pilar(2017)highlightedthattherewasalackofstudiescomparingthe socialimpactsofsimilareventsbetweenhostdestinations.Moreover,inmanycountries,especiallyinemerg­ingones,suchinthecaseofSerbiaandHungary,socialimpactsofevents,andspecificallyfestivals,areempirically stillunderresearched (except the work of Pavluković, Armenskiand Alcantara-Pilar 2017). Oneofthefirstscalesusedtomeasureresidents’perceptionofsocialimpactsoffestivalswastheFestival SocialImpactAttitudeScale(FSIAS),developedbyDelamere(2001)andDelamere,WankelandHintch(2001). While FSIAS was firstly used on small community festivals, there was a need to further validate it by test­ing it in other community types and with different types of festivals (Delamere 2001). Therefore, the objective of this study is to assess the local residents’ perceptions of the social impacts of two large-scale and worldwide popular music festivals, Sziget (Budapest, Hungary) and Exit (Novi Sad, Serbia), using the modified FSIAS. Exploratory factor analysis was carried out in order to determine the underlying factor structure of modified FSIAS. In addition, Importance-Performance Analysis (IPA) was employed to compare the results and to examine the implications of the findings for both festivals’ man­agementanddestinationsthatcanassistthemtodevelopafocusedactionagendatoachieveandmaintain festivals’ sustainability and community support. As these two festivals take place annually in summer months, attract similar music performers and visitors,theyareoftenseenascompetitors.Moreover,thefestivalsareheldinEurope,inneighboringcoun­triesthathavesomesimilarpoliticalandeconomicsettings–post-communistcountriesthatexperienced a transition to a market-based economy (Stankov and Dragićević 2015). In this context, the comparative study is of importance. 2 Literature review 2.1 Social impact of festivals »Festivalsareemergingasgrowingandvibrantsectorofthetourismandleisureindustriesandareseentohave significant economic, sociocultural, and political impacts on the destination area and host groups« (Arcodia and Whitford 2007, 1). According to Getz (2008) festivals produce various outcomes and managers can-notconcentrateonlyoneventprofitabilityasameasureofsuccess.Instead,socialandenvironmentalimpacts of an eventshouldbe equally considered(Smalletal. 2005; Wood2005; Reid2007; Kimand Petrick2005; Delamereetal.2001;Fredline,JagoandDeery2003).However,therearedifficultiesinvolvedindistinguishing between social, cultural, environmental and economic impacts of the events all of which can have politi­cal repercussions (Getz and Page 2016) and affect the quality of life of the host community. Park (2007) definessocialimpactsaspositiveornegativechangesinsocialandculturalconditionsdirectlyorindirectly resulting from an activity, project,or program. Forthepurpose ofthispaper, socialimpactsare definedas anyimpactsthatpotentiallyaffectthequalityoflifeforlocalresidents(Fredline,JagoandDeery2003).Similarly, Sharpley and Stone (2012) pointed out that the social impacts of events refer to effects on people’s life. There have been significant efforts to create measurement scales in order to assess how residents per-ceivethesocialimpactsofevents. Delamere,WankelandHintch(2001),Fredline,JagoandDeery(2003), andSmallandEdwards(2003)developedthemostcommonlyusedscales.Fredline,JagoandDeery(2003) developedascalesimilartoFSIAStomeasurethesocialimpactsofavarietyofmediumtolarge-scaleevents.Small and Edwards (2003) created the Social Impact Perception (SIP) scale with 35 items across six fac­torswhicharecomparablewithfactorspresentedbyDelamere(2001)andFredline,JagoandDeery(2003). Delamere,WankelandHintch(2001)developedFSIAStobeusedforthemeasurementandinterpretationof residents’ perceptions of the social impacts of community-based festivals. Through exploratory factor analysis and removal of cross loading and low loading items (from 70 to 47 items), two main factors ofthe scale were determined: social benefits (comprising 21 items – positive impacts) and social costs (26 items – negative impacts) of festivals. Delamere (2001) further refined FSIAS and verified the two-factor scale across 25 items. The social benefits factor comprised items relating to community image, identity,wellbeing, experiencing new things, opportunities to develop new skills. The second factor (social costs) explained a range of items related to overcrowding, traffic, litter, noise and disruption and intrusion intothe lives of local residents. Recently, there has been an increase in the utilization of FSIAS in different community settings and typesoffestivals.BagiranandKurgan(2016)appliedoriginalFSIASintheirresearchontheresidents’per­ceptionsoftheFocaRockFestivalinIzmir,Turkey.Theyconfirmedthetwo-dimensionalnatureofFSIAS (social benefits and social costs) with 35 items in total. Similarly, Woosnam, Van Winkle and An (2013)confirmedthefactorstructureoftheFSIASutilizingthecontextofaculturalheritagefestivalinruralTexas.Theyfoundthatresidents’perceptionsofimpactsdifferedacrossthelengthofresidencyandannualhouse­hold income. Dragićević et al. (2015) assessed Maribor residents’ attitudes towards the social impacts ofthe European capital of culture, using modified FSIAS. They found that residents perceived more posi­tive than negative impacts of the event. Woosnam et al. (2016) examined perceived social impacts of the annual harvest festival in the ruraltown of Morden, Manitoba (Canada) on the community as well as the underlying structures of motiva­tions to attend the festival among residentsand visitors through modified FSIAS and a newly developedfestival–attending motivation scale. Their modified FSIAS had a four-factor structure: social costs (com­prisedofthesameitemsasinoriginalFSIAS),communitybenefits,individualbenefits(thesameasfactorsocial benefits in original FSIAS) and new factor labeled new opportunities. In addition, this was the firststudy to consider motivations as a predictor of perceived impacts. Pavluković, Armenski and Alcantara-Pilar(2017)utilizedmodifiedFSIAStoidentifytheunderlyingdimensionsofsocialimpactsoftwolarge-scalemusicfestivals,Exit (Serbia)andSziget (Hungary),andtoexploremoderationeffect ofHofstede’s nation-alculturaldimensionsonresidents’perceptionsofimpactsofthesefestivalsontheircommunities.Results reveal the six-factor substructure of FSIAS that represents two main, positive and negative, dimensionsof social impacts of large–scale music festivals. In addition, they found that national culture significant­ly influenced residents’ perception of the impacts of the festival on their community. All the above–mentioned studies call for further testing and modification of FSIAS in order to con­tribute to the academic literature on the social impacts of the events and on a practical basis to improvethe management of festivals. Therefore, this research is of importance. 2.2 Importance–Performance Analysis Importance-Performance Analysis (IPA) introduced by Martilla and James (1977) is a widely used tech­nique for developing management strategies. IPA identifies attributes for which, given their importance, a product or service underperforms or over-performs. It combines measures of attribute importance andperformanceintoatwo–dimensionalgridinordertofacilitatedatainterpretationandattainpracticalrec­ommendations (Dwyer et al. 2016). Figure1illustratestheIPAgrid.TheY-axisshowstheperceivedimportanceofspecificattributeswhile the X-axis reflects the performance when compared with these attributes. The four quadrants are labeled asfollows:Concentratehere,Keepupthegoodwork,LowpriorityandPossibleoverkill.IntheConcentrateherequadrant,attributesareconsideredtobeveryimportant,butoflowperformance,meaningthatimprove­ment efforts should be concentrated here. In the quadrant labeled Keep up the good work very important attributes with high levels of performance in relation to these activities are concentrated. In the Low pri­ority quadrant, attributes have both low importance and low performance and therefore should not be of managementconcern.Possibleoverkillquadrantgathersattributesoflowimportanceandofrelativelyhigh performance meaning managers should consider their efforts on these attributes as being overexploited Figure 1: Importance–performance grid (Dwyer et al. 2016). (Chu and Choi 2000). The standard four quadrants matrix helps stakeholders/managers to identify the areas for improvement and actions for minimizing the gap between importance and performance. Importance-Performance Analysis has increasingly been applied in tourism contexts, including the hotelindustry(BeldonaandCobanoglu2007;ChuandChoi2000),travelmarkets(MurdyandPike2012), leisure and recreation (Deng 2007; Chen 2014), tourism destination competitiveness (Dwyer et al. 2013; Dwyeretal.2015;Dwyeretal.2016),festivalqualityandattractiveness(Kim,AhnandWick2014;Choi2015), transportation (Huang, Wu and Hsu 2006). 3 Methods 3.1 Study site Exit festival is a summer music festival held annually since 2000 in the city of Novi Sad, Serbia. It started as a student movement fighting for democracy in Serbia, but already in 2001, it became one of the most important music festivals in Europe. The four-day festival takes place on the first weekend of July. It was rankedtop10bestmajorfestivalsatEuropeanFestivalAwardsfrom2009to2013,oneofthe10BestOverseas Festivals at UK Festival Award 2014 and Best Summer Music Festival in Europe for 2016 by travel portal »EuropeanBestDestinations«incooperationwiththeEuropeanCommission.Morethan2.5millionpeo­plefromover60countriesaroundtheworldhavevisitedthefestivalsofar(Internet1;Pavluković,Armenski and Alcantara-Pilar 2017). From a student event in 1993 Sziget Festival has become one of the largest summer music events in Europe, held every August in Budapest, Hungary. This seven days festival hashad more than 500,000 vis-itorsfromover100countriesin2016.ItwasrankedoneofthefivebestfestivalsinEuropebyTheIndependent in 2011 and is a two–time winner at the European Festivals Awards in the category Best Major European Festival in 2012 and 2015 (Internet 2). 3.2 Research instrument, sampling and data collection To achieve the objective of the study, a two-step procedure was conducted. First, in order to adopt origi­nal FSIAS to the research setting of large–scale music festivals, five academics from Serbia and Hungary who have research experience in tourism and event management were asked to discuss on FSIAS. Based on this discussion, the original scale was refined to suit specific cases of Exit and Sziget festivals. Namely, 20originalitemsfromFSIASwerekeptand11newitemswereproposed.Thefinalscaleconsistsof31items with good internal consistency (.=0.88). Second, by using the refined scale, the main survey was conducted among residents of Novi Sad and Budapest, host destinations of Exit and Sziget, in summer months June–September 2014, when festivals areheld.Thequestionnaireusedwascomposedofthreesections.Thefirstpartincludedbackgroundinfor­mation of participants. The second part consisted of 31 items – social impacts, for which local residents of Novi Sad/Budapest were asked to evaluate the perceived importance of the impacts when organizing any festival in their community on a five-point Likert scale ranging from 1 (least important) to 5 (most important).Inthethirdpartresidentsevaluatedthesame31itemsinrespectofactualfestivalperformance using a five-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). The questionnaire was prepared in two languages: Hungarian (for respondents from Budapest) and Serbian(for respondents from Novi Sad). It was created using Google Docs and distributed electronically. In total 505 usable ques­tionnaires were obtained (301 from Novi Sad and 204 from Budapest). Some authors believe that the assessmentwillbegoodonlyifthesamplecontainsaminimumof51units(Bagozzi1981;BarrettandKline 1981),whileothersthinkthat150–300observationswillbesufficient(Pallant2011).AccordingtoMacCallum etal.(2001),asamplesizebetween100and200isacceptable.Therefore,thesamplesizeusedinthisstudy is adequate for the analyses conducted. The data was processed with the statistical package SPSS 2.0. 4 Results 4.1 Respondents’ profile TherespondentscharacteristicsareshownindetailinTable1.Inbothsubsamples,females,youngerresidents (lessthan31)andthosewhoattendedthefestivalevidentlyshowhigherinteresttotakepartintheresearch. Table 1: Respondents’ characteristics. City/Festival Novi Sad/EXIT Budapest/SZIGET Characteristics Absolute frequencies (%) Absolute frequencies (%) Female 200 66.4 120 58.8 Male 101 33.6 84 41.2 Less than 31 167 55.5 111 54.4 Between 31–41 92 30.5 56 27.5 More than 41 42 14.0 37 18.1 Less than 10 years Between 10 and 20 years More than 20 years 76 78 147 25.3 25.9 48.8 68 42 94 33.3 20.6 46.1 Level of Education High school 70 23.2 50 24.5 2–years higher education 31 10.3 59 28.9 Graduate studies 133 44.2 54 26.5 Post graduate studies 67 22.3 41 20.1 Yes 216 71.8 147 72.1 No 85 28.2 57 27.9 Never 85 28.2 57 27.9 Once 34 11.3 27 13.3 Two times 27 9.0 32 15.7 Three times 25 8.3 19 9.3 More than three times 130 43.2 69 33.8 4.2 Exploratory factor analysis To explore dimensions of modified FSIAS exploratory factor analysis was carried out, using the principal componentmethodandObliminrotation.TheKaiser–Meyer–Olkin(KMO)overallmeasureofsampling adequacy (KMO=0.927) and Barlett’s test of sphericity (p=0.000) suggested that the data were suitable for factor analysis. In this study, all factors with an eigenvalue greater than 1 and with factor loadings of morethan0.5wereretained.Theresultsofthefactoranalysissuggestedatwo-factorsolution,whichexplained 46.35% of the total variance. The results produced a clean factor structure with relatively higher loadings on the appropriate factors. Most variables were loaded heavily on one factor and this reflected that there was minimal overlap among factors and that all factors were independently structured. Cronbach’s . val­ues for each factor were greater than recommended 0.7 (DeVellis 2003), suggesting that the scale used in the survey has considerable reliability (Nunnally 1978). Table 2: Results of exploratory factor analysis for modified FSIAS. Factors Variance explained Eigen value Parallel analysis 95 percentile of random Eigenvalues Cronbach’s alfa Number of items F1 Social benefits – Positive impacts 32.35 10.027 1.538 0.937 22 F2 Social costs – Negative impacts 14 4.342 1.472 0.863 9 The factors are labeled as in the original FSIAS scale (Delamere 2001): first factor »social benefits«, as it involves 22 items referring to positive impacts of a music festival on the local community, and sec­ond factor »social costs« consists of nine items – negative impacts of a festival on the local community. Inthefollowingsections,wewilldiscusseachoftheitems–socialimpactsoftwofestivalsacrosstwodimen­sions of scale, positive and negative. 4.3 IPA results Table 3 shows the mean values of social impacts of music festivals on local communities in relation to importance and performance. For almost each of the impacts, the respondents rated importance rela­tivelyhighlyandconsistentlyhigherthanperformance.Inaddition,t-testofpairedsampleswasemployed in ordertotestthedifference betweeneach festivalimpactperformanceandimportance mean. Almost allimportanceandperformancemeansforbothfestivalswerefoundtobesignificantlydifferent(p<0.05) (seeTable 3). Interestingly,forbothgroupsofrespondents,thereisnostatisticallysignificantdifference in evaluating the importance and performance of impact labeled P3. In addition, this social impact of both festivals demonstrates strong performance. Further, the data were transferred to the IPA grid pre­sentation in order to provide easier interpretation and discussion of the results and comparison of the festivals. Figure2andFigure3areconstructedusingdatameans(themeanscoresofimportanceandperformance across all 31 social impacts) as theintersection point of the x (performance) and y (importance) axes. For easier interpretation, positive impacts are labeled P1–P22 and negative impacts N1–N9 (see Table 3). Almostallofthenegative impacts(except two)of Exitfestivalare located in Lowpriorityor Possible overkill quadrants, while most of the positive impacts are in Keep up good work and Concentrate here area, meaning that respondents from Novi Sad are more focused on positive impacts of the festival. This is consistent with the literature in general, which emphasizes that local communities are more aware of festival benefits to the community. In the case of the Sziget festival, the benefits and costs are more dispersed among quadrants, and we assume a higher level of tourism development and event industry. Table 3: Descriptive statistics of the importance and performance of social impacts of festivals and t-test paired samples. Impacts EXIT SZIGET I P t p* I P t p* P1 Festival enhances image of the community 4.59 4.40 –3.828 0.000 3.64 3.57 –0.729 0.467 P2 Community identity is enhanced through festival 4.47 3.89 –8.825 0.000 3.41 2.87 –6.512 0.000 P3 Hosting festival improves promotion of the city internationally 4.66 4.73 1.803 0.072 4.42 4.44 0.291 0.771 P4 Festival enables local community to present itself to others (visitors) as special and unique 4.38 4.09 –4.430 0.000 3.78 3.64 –1.601 0.111 P5 Festival acts as a showcase for new ideas for the locals 4.46 3.76 –11.116 0.000 4.00 3.36 –7.246 0.000 P6 Festival contributes to sense of community well-being 4.24 2.89 –19.271 0.000 4.00 3.11 –6.166 0.000 P7 Community feels a sense of pride due to hosting festival 4.37 3.75 –10.224 0.000 3.70 3.50 –7.367 0.000 P8 Festival has ongoing positive cultural impact on community 4.41 3.81 –11.594 0.000 4.18 3.51 –6.462 0.000 P9 Festival improves the quality of life in community 4.57 3.03 –16.970 0.000 4.13 2.57 –11.810 0.000 P10 Festival provides residents with opportunity to learn new things 4.33 3.58 –9.936 0.000 3.64 3.05 –6.498 0.000 P11 The local community has a chance to meet festival performers 4.30 3.04 –8.325 0.000 3.75 3.11 –3.911 0.000 P12 Local community is exposed to a variety of cultural experiences through festival 3.69 3.92 –6.576 0.000 3.50 3.65 –3.136 0.002 P13 Local community gains positive recognition as a result of festival 4.35 3.70 –7.847 0.000 3.96 3.26 –9.043 0.000 P14 Festival provides new job opportunities for residents 4.45 3.63 –11.702 0.000 4.41 3.72 –7.725 0.000 P15 Due to hosting the festival, residents have the opportunity for additional income 4.53 4.31 –4.230 0.000 4.52 3.75 –9.022 0.000 P16 Festival is of great importance for exploring the local culture by visitors 4.60 3.95 –10.365 0.000 4.13 3.13 –9.593 0.000 P17 There is high security level of festival visitors 4.81 3.60 –18.475 0.000 3.36 3.58 2.214 0.028 P18 There is high security level of residents during the festival 4.82 3.55 –19.135 0.000 4.48 3.38 –12.415 0.000 P19 Festival visitors behave properly 4.70 3.28 –20.716 0.000 4.49 3.15 –15.091 0.000 P20 Festival program is rich and diverse 4.71 3.83 –13.407 0.000 4.63 3.79 –10.347 0.000 P21 Local community is involved in the organisation of the festival 4.32 3.21 –15.327 0.000 3.46 2.79 –7.153 0.000 P22 Local community attitudes toward organi­ zation of the festival are acknowledged 4.28 2.79 –18.731 0.000 3.98 2.78 –12.354 0.000 N1 Festival leads to disruption in normal routine of residents 3.95 3.18 –7.911 0.000 3.85 3.27 –5.586 0.000 N2 Community facilities are overused 3.38 3.45 0.720 0.472 3.57 3.39 –1.796 0.074 N3 The influx of festival visitors reduced privacy in community 3.56 2.61 –11.142 0.000 3.99 3.18 –9.479 0.000 N4 Community is overcrowded during festival 3.08 3.68 6.266 0.000 3.34 3.11 –2.131 0.034 N5 Local traffic increases to unacceptable levels 3.52 3.01 –5.865 0.000 3.48 2.90 –5.837 0.000 N6 Noise levels increase to an unacceptable level 3.33 3.71 4.599 0.000 3.77 2.84 –9.136 0.000 N7 Litter increases to unacceptable levels 4.52 3.44 –12.915 0.000 4.10 2.99 –11.935 0.000 N8 Prices of products and services increases 4.16 3.69 –5.404 0.000 4.14 3.11 –9.644 0.000 N9 Crime in community increases 4.78 2.59 –28.928 0.000 4.59 2.25 –24.708 0.000 Mean value of whole scale (for all items) 3.55 Note: I –mean value of Importance, P –mean value of Performance, *p<0.05 5 45. 4 35. Concentrate here Performance N9 N3 P22 P6 P9 P11 N5 Keep up good work P18 P19 P20 P3 P17 P16 P1 N7 P5 P15 P14 P2 P21 P10 P13 P7 P8 P4 N8 N1 P12 N2 N6 3 N4 25. 2 15. Low priority Importance Possible overkill 1 1 15. 2 25. 3 35. 4 45. 5 Figure 2: IPA grid for the Exit festival. 5 Concentrate here Keep up good work 45. 4 35. Performance N9 P9 P20 P19 P18 P15 P14 N8 N7 P16 P8 P22 P6 N3 P13 P5 N1 N6 P11 P4 P10 P7 P1 N5 N2 P21 P12 P2 N4 P17 P3 3 25. 2 15. Low priority Importance Possible overkill 1 1 15. 2 25. 3 35. 4 45. 5 Figure 3: IPA grid for Sziget festival. 5 Discussion 5.1 Social impacts of Exit and Sziget in common Inthissection,thesocialimpactsofExitandSziget,whicharelocatedinthesameIPAquadrants(seeTable4), will be discussed in order to highlight festivals’ similarities. Both sets of respondents from Novi Sad and Budapest nominated eight positive impacts of festivals as combining relatively high importance and per­formance. Table 4: Social impacts of Exit and Sziget located in the same IPA quadrants. IPA Quadrant Impact of festival Keep up the good work P3 Hosting festival improves the promotion of the city internationally P5 Festival acts as a showcase for new ideas for the local community P8 Festival has an ongoing positive cultural impact on the community P13 Local community gains positive recognition as a result of the festival P14 Festival provides new job opportunities for residents P15 Due to hosting the festival, residents have the opportunity for additional income P18 There is the high-security level of residents during the festival P20 Festival program is rich and diverse Concentrate here P9 Festival improves the quality of life in the community P19 Festival visitors behave properly P22 Local community attitudes toward the organization of the festival are acknowledged N7 Litter increases to unacceptable levels during festival N9 Crime in community increases during the festival Low priority N5 Local traffic increases to unacceptable levels Possible overkill P12 Local community is exposed to a variety of cultural experiences through the festival That festival program is rich and diverse and adds entertainment opportunities for the community is consistentwithrecentawardsbothfestivalsgotanincreasednumberofvisitors.Inthisfield,festivalorga­nizersshouldkeepupthegoodwork.Respondentsinbothdestinationsaffirmedtheimportanceofresidents’ safety during the festival, which was expected as safety risks are associated with outdoor music festivals. Research suggests that safety and security are fundamental requirements for tourism destination com-petitiveness(DwyerandKim2003)aswellasforfestivals’sustainability.ItseemsthatsofarExitandSziget have performed well with respect to residents’ safety and should continue in the same manner. Affirmation by both sets of respondents that Exit and Sziget festivals provide new jobs and addition­alincomeopportunitiesisconsistentwithresearchliteratureemphasizingthecapacityoffestivalstogenerate increased revenues and job opportunities for locals (Dwyer, Forsyth and Spurr 2006). This is specifically important for countries like Serbia and Hungary where average wages are among the lowest in Europe (Stankov and Dragićević 2015; OECD 2016). That hosting festival improves the promotion of the city internationally was another benefit of festi­vals. Both festivals have been attracting an increasing number of people from different parts of Europe as well as from other continents in the past decade (Nagy and Nagy 2013). Namely, the number of foreign visitorsinboth,NoviSadandBudapestincreasedinJulyandAugust,monthswhenfestivalsareheld.Both festivalsemergedfromsmalllocaleventstomulticulturalEuropeanmusicfestivals.Inaddition,theimpact of festivals on the promotion of cities and consequently tourism development is well documented in the literature on festivals (Getz 2008; Montgomery 2007). Findings that local communities gain positive recognition as a result of festivals and that both festi­vals have an ongoing positive cultural impact on host communities are consistent with research literature (Bowdin et al. 2006; Gursoy, Kim and Uysal 2004). FiveimpactsoffestivalsfallintoConcentrateherecategoryforbothExitandSziget.Itseemsthatboth festivals management together with local authorities should focus on decreasing crime rates during fes­tivals,solvingenvironmentalissues(increasedgarbageinpublicareas)andcontrollingthebehavioroffestival visitors.Theliteratureemphasizesthatthedelinquentbehavioroffestivalvisitorsinthelong-termimpacts residents’ support for the festivaland the image of destination (Deeryand Jago 2010), as well as thequality of life in the community. Residents’ attitudes toward festivals are of great importance as the local com­munity is directly involved in creating an experience for visitors and furthermore, their support for the eventwillprobablyaffectthefestivalsustainability(GursoyandKendall2006).Festivals’managementand local authorities should acknowledge that there could not be a successful festival unless the community is involved in it (Mason 2015). Bothsetsofrespondentsaccordedlow-prioritytotrafficjamsduringthefestival.SinceExitandSziget last a few days, the residents of both cities may not think that festivals can create any major traffic prob­lems in a few days period. Local residents are willing to accept short-term irritation (such as traffic and parking problems) as they are aware of numerous benefits that festivals generate (Deery and Jago 2010). 5.2 Differences in social impacts priorities of Exit and Sziget There are a large number of benefits and costs of Exit and Sziget festivals valued differently by the two sets of respondents. We herein focus on Keep up the good work and Concentrate here categories since these have important implications for the action agenda of festival management and tourism leaders in each destination. RespondentsfromNoviSadconsiderthatExitisperformingwellinrespectofseveralimpactsofrel­atively high importance. They identified high performance in the enhancement of community image, which is consistent with literature that emphasizes the role of the festival in improving the place’s image (Getz2008;VandenBerg2012).Incontrast,respondentsfromBudapestplacedthisimpactinaPossible overkill quadrant, which was expected as Budapest has already created an image as a travel and event destination. Inaddition,respondentsfromNoviSadallocatedseveralbenefitsofthefestival(enhancementofcom­munityidentity,senseofpride,opportunitytolearnnewthings,thehigh-securityleveloffestivalvisitors) toKeep up good work quadrant while respondents from Budapest placed them into Possible overkill and Low priority quadrant. Thesedifferencescouldbeduetoslightlydifferenttourismdevelopmentstagesofhostingdestinations. Budapest is the most populous city in Hungary, with about 1.7 million inhabitants and more than 3 mil­lion tourists in 2017, while Novi Sad is a smaller community with about 300,000 residents and 130,000 touristsin2014. AccordingtoPizam(1978),tourismconcentrationonadestinationarealeadstonegative attitudestowardtouristsandtourismingeneral.Thedifferencesinpopulationandinthenumberoftourists in Budapest and Novi Sad could affect both communities’ attitudes towards benefits of festivals, meaning residents of Novi Sad perceived more benefits of festivals than residents of Budapest. Respondents from Budapest think more efforts should be made towards presenting the local culture to the festival visitors. Respondents from Novi Sad indicated two impacts of the festival that deserve priority: the local com­munity should be involved in the organization of the festival and should have a chance to meet festival performers. However, these are low priorities for respondents from Budapest. In the case of Serbia, these results are not surprising as younger respondents with less than 31 years old make more than half of the sample in this research, and they are usually willing to socialize and meet new people, specifically festi­valperformers,whichbringexcitementtotheireverydaylifeandcreatepositivememoriesrelatedtofestival. Interestingly,therespondentsfromNoviSadaffirmedthatExitfestivalmanagementandlocalauthor-itiesshouldconcentrateoninvolvingthelocalcommunityinthefestivalorganizationandshouldacknowledge community’s attitudes, which is consistent with general opinion in the research literature (Arcodia and Whitford 2007; Gursoy and Kendall 2006; Mason 2015). Both festivals’ management should concentrate on collaboration with community and consultation before, during and after the festival, in order to pro­vide community well-being and its support for the festival in long terms (Arcodia and Whitford 2007). RespondentsfromBudapestemphasizedaspriorityareathefestival’scontributiontothesenseofcom-munitywell-being.However,respondentsfromNoviSadregarditasalowpriority,whichisnotinaccordance with the positive impacts of the Exit festival located in Keep up good work quadrant that altogether con-tributetocommunitywell–being.Inaddition,itisinconsistentwiththegrowingevidenceintheliterature that festivals have public good aspect beyond generating revenue (Getz and Page 2016). RespondentsfromHungaryindicatedtwonegativeimpactsofthefestival,whichshouldbeinthefocus of local authorities. The research literature indicates that like any other type of tourism development, fes­tivals generate an increase in prices of goods and services and overcrowding in streets and public places, whichaffectsresidents’privacy(Gursoy,JurowskiandUysal2002;Tosun2002).FortheExitfestival,these impacts are located in Low priority and Possible overkill category. A number of researchers who exam­inedthelinkbetweentheperceptionofnegativeimpactsandthesupportfortourismdevelopmentreported thattherewasanegativerelationshipbetweennegativeimpactsandresidents’perceptionsoftourismdevel­opment (Gursoy, Jurowski and Uysal 2002; Tosun 2002). 6 Conclusion This study has presented and discussed the social impacts of two famous European music festivals Exit and Sziget on their communities using modified FSIAS and IPA. So far, there was a lack of studies com­paring the social impacts of similar events between host destinations with an exception to the work of Pavluković,ArmenskiandAlcantara-Pilar(2017).Moreover,thisstudywasconductedinresponsetorecent calls for further testing and modification of FSIAS in different settings. The paper has theoretical and practical contributions. First, it expands the body of knowledge on the social impacts of festivals. Then, it validates the usage of FSIAS in a novel context of large-scale festivals in urban communities. The scale can still produce reliable results even if slightly modified and utilized in differentenvironments.Theresultsofexploratoryfactoranalysisshowtheacceptableandmeaningfultwo-factorstructureofmodifiedFSIAS,asinDelamereetal.(2001)originalFSIAS.Inaddition,thisstudyseeks to encourage the use of IPA in geography, specifically urban and social, as this technique is widely used in a tourism context. Also, within urban social geography,considerable effort has been directed to assess-ingthequalityoflifeandurbanenvironmentalquality(Pacione2003).Giventheimportanceofsocialimpacts offestivalsonthequalityofevery-daylifeforlocalresidentsandontheurbanspace,thispapercontributes to the body of knowledge in urban social geography field. The results have practical implications for festival/urban destination management in each community. Namely,thepaperdiscussedsimilaritiesanddifferencesbetweentwofestivalsproposingtheprioritiesfor eachfestival/destinationmanagementimpliedbyIPAresults.Strongareasofbothfestivalsweredetermined as well as those that should be improved by festivals’ management and local authorities in order to max­imize benefits and minimize the costs of hosting thefestival. In addition, low priority or possible overkill pointsaredefinedandshouldnotbeignoredastheycanlargelyinfluencebothcommunitywellbeingand the sustainability of the festival itself. It is the social impact in Concentrate here quadrant that deserves special emphasis by festival organizers. As with any research, limitations exist that need to be discussed. Although two countries havea com­mon history of being communist countries, differences exist in their level of economic and tourism development, and in the transition to a market-based economy, which affects the respondents’ percep­tions of festival impacts. Cultural differences also exist and may well impact on the study results (see Pavluković, Armenski and Alcantara-Pilar 2017). Although it was not the purpose of this study to examine whether sociodemographic variables and previous attendance at the festival affect residents’ perceptions of social impacts, a description of survey samples from Novi Sad and Budapest may potentially shed light on the findings. For instance, more than 50% of the respondents in both samples are females and younger (under the age of 31), and more than 70%oftherespondentsinbothsamplesattendedthefestival.Perhapssucha highleveloftherespondents who attended the festival, and specifically younger respondents who are usually in favor of music festi­vals,affecttheperceptionofpositiveandnegativeimpactsofthefestival.Therefore,futureresearchwould need to examine this speculation. In addition, Small (2007) and Woosnam, Van Winkle and An (2013) note that within any given community residents will perceive the same impact in different ways based on theirpersonalbackgroundwhetheritisfrompreviousfestivalparticipation,sociodemographicandsocio-economic classification. Sincemostoftheresearchersusethequantitativeapproachinexaminingperceptionsofsocialimpacts of festivals, further research orientations could be towards a qualitative approach (e.g. interviews with the representative stakeholders and/or focus groups). As perceptions of the impacts of the festival are not static and the festival itself changes, it would be interestingtoseehowperceptionsoffestivalimpactschangeovertimeinalongitudinalstudy.Totheauthors’ knowledge, this has rarely been done by researchers and festival organizers, although it can provide valu­able information for the festival and destination management. Despitetheabovementionedlimitationsofthework,findingsfromthisstudysupporttheuseofFSIAS andmakeasignificantcontributiontounderstandingresidents’attitudestowardthesocialimpactsoflarge- scale music festivals. ACKNOWLEDGEMENT: This research is part of the project Naselja I stanovništvo Vojvodine, financed by the Matica Srpska, Serbia. 7 References Arcodia,C.,Whitford,M.2007:Festivalattendanceandthedevelopmentofsocialcapital.JournalofConvention and Event Tourism 8-2. DOI: https://doi.org/10.1300/J452v08n02_01 Bagiran, D., Kurgun, H. 2016: A research on social impacts of the Foça Rock Festival: the validity of the Festival Social Impact Attitude Scale. Current Issues in Tourism 19-9. DOI: https://doi.org/10.1080/ 13683500.2013.800028 Bagozzi,R.P.1981:EEvaluatingstructuralequationmodelswithunobservablevariablesandmeasurement error: A comment. Journal of Marketing Research 18-3. DOI: https://doi.org/10.2307/3150979 Barrett,P.T.,Kline,P.1981:Theobservationtovariableratioinfactoranalysis.PersonalityStudyandGroup Behavior 1-1. Beldona, S., Cobanoglu, C. 2007: Importance–performance analysis of guest technologies in the lodging industry. Cornell Hotel and Restaurant Administration Quarterly 48-3. DOI: https://doi.org/10.1177/ 0010880407304023 Bowdin, G., Allen, J., O’Toole, W., Harris, R., McDonnell, I. 2006: Events managements. Oxford. Chen, K.-Y. 2014: Improving importance-performance analysis: The role of the zone of tolerance and competitor performance. The case of Taiwan’s hot spring hotels. Tourism Management 40-1. DOI: https://doi.org/10.1016/j.tourman.2013.06.009 Choi, S. S. 2015: Importance–Performance Analysis of attractiveness assessment for festival: a case of SobaeksanRoyalAzaleaFestival.IndianJournalofScienceandTechnology8-18.DOI:https://doi.org/ 10.17485/ijst/2015/v8i18/83947 Chu, R. K. S., Choi, T. 2000: An importance-performance analysis of hotel selection factors in the Hong Konghotelindustry:Acomparisonofbusinessandleisuretravellers.TourismManagement21-4.DOI: https://doi.org/10.1016/S0261-5177(99)00070-9 De Vellis, R. F. 2003: Scale development: Theory and application. Thousand Oaks. Deery,..,..g.,L.2010:Socialimpactsofeventsandtheroleofanti-socialbehaviour.InternationalJournal of Event and Festival Management 1-1. DOI: https://doi.org/10.1108/17852951011029289 Delamere, T. A., Wankel, L. M., Hinch, T. D. 2001: Development of a scale to measure resident attitudes towardthesocialimpactsofcommunityfestivals,PartI:Itemgenerationandpurificationofthemeasure. Event Management 7-1. DOI: https://doi.org/10.3727/152599501108751443 Delamere, T. .. 2001: Development of a scale to measure resident attitudes toward the social impacts of communityfestivals,PartII:verificationofthescale.Eventmanagement7-1.DOI:https://doi.org/10.3727/ 152599501108751452 Deng,W.2007:Usingarevisedimportance-performanceanalysisapproach:thecaseofTaiwanesehotsprings tourism. Tourism Management 28-5. DOI: https://doi.org/10.1016/j.tourman.2006.07.010 Dragićević,V.,Bole,D.,Bučić,A.,Prodanović,A. 2015:EuropeanCapitalofCulture:residentsperception of social benefits and costs – Maribor 2012 case study. Acta geographica Slovenica 55-2. DOI: https://doi.org/10.3986/AGS.747 Dwyer, L., Armenski, T., KneževićCvelbar, L., Dragićević, V., Mihalic, T. 2015: Modified Importance– Performance Analysis for evaluating tourism businesses strategies: comparison of Slovenia and Serbia. International Journal of Tourism Research 18-4. DOI: https://doi.org/10.1002/jtr.2052 Dwyer, L., Dragićević, V., Armenski, T., Knežević Cvelbar, L., Mihalič, T. 2016: Achieving destination competitiveness: an Importance–Performance Analysis of Serbia. Current Issues in Tourism 19-13. DOI: https://doi.org/10.1080/13683500.2014.944487 Dwyer, L., Dragićević, V., Armenski, T., Mihalič, T., Knežević Cvelbar, L. 2016: Achieving destination competitiveness: an Importance–Performance Analysis of Serbia. Current Issues in Tourism 19-13. DOI: https://doi.org/10.1080/13683500.2014.944487 Dwyer,L.,Forsyth,P.,Spurr,R.2006:Assessingtheeconomicimpactsofevents:acomputablegeneralequi­librium approach. Journal of Travel Research 45-1. DOI: https://doi.org/10.1177/0047287506288907 Dwyer,L.,Jago,D.,Forsyth,P.2016:Economicevaluationofspecialevents:Reconcilingeconomicimpact andcost-benefitanalysis.ScandinavianJournalofHospitalityandTourism16-2.DOI:https://doi.org/ 10.1080/15022250.2015.1116404 Dwyer, L., Kim, C. W. 2003: Destination competitiveness: Determinants and indicators. Current Issues in Tourism 6-5. DOI: https://doi.org/10.1080/13683500308667962 Dwyer,L.,KneževićCvelbar,L.,Edwards,D.,Mihalič,T.2013:Tourismfirms’ strategic flexibility:thecase of Slovenia. International Journal of Tourism Research 16-4. DOI: https://doi.org/10.1002/jtr.1931 Fredline, L., Jago, L., Deery, M. 2003: The development of a generic scale to measure the social impacts of events. Event Management 8-1. DOI: https://doi.org/10.3727/152599503108751676 Getz, D. 2008: Event tourism: Definition, evolution and research. Tourism Management 29-3. DOI: https://doi.org/10.1016/j.tourman.2007.07.017 Getz, D. 2010: The nature and scope of festival studies. International Journal of Event Management Research 5-1. Getz, D., Page, S. J. 2016: Progress and prospects for event tourism research. Tourism Management 52. DOI: https://doi.org/10.1016/j.tourman.2015.03.007 Gursoy, D., Jurowski, C., Uysal, M. 2002: Residents attitudes: A structural modeling approach. Annals of Tourism Research 29-1. DOI: https://doi.org/10.1016/S0160-7383(01)00028-7 Gursoy,D.,Kendall,K.2006:Hostingmegaevents:modellinglocals’support.AnnalsofTourismResearch 33-3. DOI: https://doi.org/10.1016/j.annals.2006.01.005 Gursoy,D.,Kim,K.,Uysal,M.2004:Perceivedimpactsoffestivalsandspecialeventsbyorganizers:Anexten­sionandvalidation.TourismManagement25-2.DOI:https://doi.org/10.1016/S0261-5177(03)00092-X Herrero, L. C., Sanz, J. Á., Devesa, M., Bedate, A., del Barrio, M. J. 2006: The economic impact of cultural events: a case-study of Salamanca 2002, European Capital of Culture. European Urban and Regional Studies 13-1. DOI: https://doi.org/10.1177/0969776406058946 Huang, Y. C., Wu, C. H., Hsu, C. J. 2006: Using Importance-Performance Analysis in evaluating Taiwan medium and long distance national highway passenger transportation service quality. The Journal of American Academy of Business 8. Internet 1: www.exitfest.org (10.03.2016). Internet 2: www.szigetfestival.com (10.03.2016). Kim, N., Ahn, Y. J., Wick, B. E. 2014: Local festival quality and the application of a revised Importance– Performance Analysis: The Case of the JirisanCheon–Wang Festival. Event Management 18-2. DOI: https://doi.org/10.3727/152599514X13947236947266 Kim, S. S., Petrick, J. F. 2005: Residents’ perceptions on impacts of the FIFA 2002 World Cup: the case of Seoul as a host city. Tourism Management 26-1. DOI: https://doi.org/10.1016/j.tourman.2003.09.013 Kim, Y. H., Duncan, J., Chung, B. W. 2015: Involvement, satisfaction, perceived value, and revisit inten­tion:Acasestudyofafoodfestival.Journalofculinaryscience&technology13-2.DOI:https://doi.org/ 10.1080/15428052.2014.952482 Lashua,B.,Spracklen,K.,Long,P.2014:Introductiontothespecialissue:Musicandtourism.TouristStudies 14-1. DOI: https://doi.org/10.1177/1468797613511682 MacCallum, R. C., Widaman, K. F., Preacher, K. J., Hong, S. 2001: Sample size in factor analysis: The role ofmodelerror.MultivariateBehavioralResearch36-4.DOI:https://doi.org/10.1207/S15327906MBR3604 Martilla, J. A., James, J. C. 1977: Importance-performance analysis. Journal of Marketing 41-1. DOI: https://doi.org/10.2307/1250495 Mason,P.2015:TheCommercializationofmusicfestivalsandtheriseofsuperconcerts.Internet:www.huff­ ingtonpost.com/pete-mason/the-commercialization-of-_1_b_7873236.html (10.12.2015). Montgomery, J. 2007: The new wealth of cities: city dynamics and the fifth wave. Aldershot. Murdy, S., Pike, S. 2012: Perceptions of visitor relationship marketing opportunities by destination mar­ keters:Animportance-performanceanalysis.TourismManagement33-5.DOI:https://doi.org/10.1016/ j.tourman.2011.11.024 Nagy,A.,Nagy,H.2013:TheimportanceoffestivaltourismintheeconomicdevelopmentofHungary.Visegrad JournalonBioeconomyandSustainableDevelopment2-2.DOI:https://doi.org/10.2478/vjbsd-2013-0011 Nunnally, J. C. 1978: Psychometric theory. New York. OECD 2016. OECD better life index Hungary. Internet: www.oecdbetterlifeindex.org/countries/hungary (1.3.2016). Pacione,M.2003:Urbanenvironmentalqualityandhumanwellbeing–asocialgeographicalperspective. Landscape and Urban Planning 65, 1-2. DOI: https://doi.org/10.1016/S0169-2046(02)00234-7 Pallant,J.2011:SPSSsurvivalmanual:AstepbystepguidetodataanalysisusingtheSPSSprogram.Berkshire. Park, C. 2007: A dictionaryof environment and conservation. Oxford University Press. Oxford Reference Online. Goteborg University Library (5. 3. 2016). Pavluković, V., Armenski, T., Alcantara-Pilar, J. M. 2017: Social impacts of music festivals: Does culture impactlocals’attitudetowardeventsinSerbiaandHungary?TourismManagement63.DOI:https://doi.org/ 10.1016/j.tourman.2017.06.006 Pizam, A. 1978: Tourism’s impacts: the social costs to the destination community as perceived by its res­idents. Journal of Travel Research 16-4. DOI: https://doi.org/10.1177%2F004728757801600402 Reid, S. 2007: Identifying social consequences of rural events. Event Management 11, 1-2. DOI: https://doi.org/10.3727/152599508783943192 Rollins, R., Delamere, T. 2007: Measuring the social impact of festivals. Annalsof TourismResearch 34-3. DOI: https://doi.org/10.1016/j.annals.2007.01.004 Sharpley,R.,Stone,P.2012:Socio-culturalimpactsofevents:meanings,authorisedtransgressionandsocial capital. The Routledge handbook of events. London. Small, K. 2007: Social dimensions of community festivals: An application of factor analysis in the devel­opment of the social impact perception (SIP) Scale. Event Management 11-1. Small,K.,Edwards,D.2003:Evaluatingthesocioculturalimpactsofafestivalonahostcommunity:Acase studyoftheAustralianFestivaloftheBook.Proceedingsofthe9th AnnualConferenceoftheAsiaPacific Tourism Association. Sydney. Small, K., Edwards, D., Sheridan, L. 2005: A flexible framework for socio-cultural impact evaluation of a festival. International Journal of Event Management Research 1-1. Internet: http://hdl.handle.net/ 10453/10173 (14. 12. 2018). Stankov,U.,Dragićević,V.2015:Changesinthespatialpatternofnetearnings:evidencefromSerbia.Acta Oeconomica 65-3. DOI: https://doi.org/10.1556/032.65.2015.3.1 Tosun, C. 2002: Host perceptions of impacts: A comparative tourism study. Annals of Tourism Research 29-1. DOI: https://doi.org/10.1016/S0160-7383(01)00039-1 VandenBerg,M.2012:FemininityasacitymarketingstrategygenderbendingRotterdam.UrbanStudies 49-1. DOI: https://doi.org/10.1177/0042098010396240 Wood,E.H.2005:Measuringtheeconomicandsocialimpactsoflocalauthorityevents.InternationalJournal of Public Sector Management 18-1. DOI: https://doi.org/10.1108/09513550510576143 Woosnam, K. M., Jiang, J., Van Winkle, C. M., Kim, H., Maruyama, N. 2016: Explaining festival impacts onahostingcommunitythroughmotivationstoattend.EventManagement20-1.DOI:https://doi.org/ 10.3727/152599516X14538326024919 Woosnam, K., Van Winkle, C., An, S. 2013: Confirming the festival social impact attitude scale in the contextofaruralTexasculturalfestival.EventManagement17-3.DOI:https://doi.org/10.3727/152599513 X13708863377917 THE EFFECTS OF INTENSIVE LOGGING ON THE CAPACITY OF kARST DOLINES TO PROVIDE POTENTIAL MICROREFUGIA FOR COOL-ADAPTED PLANTS Péter János Kiss, Csaba Tölgyesi, Imola Bóni, László Erdős, András Vojtkó, István Elek Maák, Zoltán Bátori Snow cover in the bottom of a large doline in the Mecsek Mountains, in the spring of 2018 (11th March, 2018). DOI: https://doi.org/10.3986/AGS.6817 UDC: 911.2:581.9(439) 581.9:551.448(439) COBISS: 1.01 PéterJánosKiss1,CsabaTölgyesi2,ImolaBóni2,LászlóErdős3,AndrásVojtkó4,IstvánElekMaák5,Zoltán Bátori2 The effects of intensive logging on the capacity of karst dolines to provide potential microrefugia for cool-adapted plants ABSTRACT: Dolines are local depressions of karst surfaces. They can be considered potential microrefu­gia for various species. We investigated the plant species composition and vegetation pattern of two medium-sized dolines in Hungary before and 10 years after logging, and analysed how different species groups (oak forest species, beech and ravine forest species and disturbance-tolerant species) were affect­ed.Thecoverandnumberofoakforestspeciesanddisturbance-tolerantspeciesincreased,whilethecover and number of beech and ravine forest species decreased within dolines due to logging. Therefore, their speciescompositionandvegetationpatternhavechangedsubstantially,anddolineshavepartiallylosttheir capacity to act as safe havens for plant species adapted to cooler conditions. KEY WORDS: biology, geography, climate change, logging, cool-adapted plants, karst area, vegetation pattern, Hungary Vpliv intenzivne sečnje na sposobnost kraških vrtač za zagotavljanje potencialnih mikrozatočišč za hladnoljubne rastline POVZETEK:Vrtačesokraškekotanjenakraškempovršju,kilahkozagotavljajomikrozatočiščarazličnim vrstam. Preučili smo vrstno sestavo in razporeditev vegetacije v dveh srednje velikih vrtačah na Madžar­skem pred sečnjo gozda in deset let po njej ter analizirali vpliv sečnje na različne skupine rastlinskih vrst (vrstev hrastovih gozdovih,vrste v bukovih gozdovihin vrstev gozdovih plemenitih listavcev ter vrste, ki soodpornenarazličnemotnje).Pokrovnostinštevilovrstvhrastovihgozdovihtervrst,odpornihnamot­nje se je povečalo, pokrovnost in število vrst v bukovih gozdovih ter v gozdovih plemenitih listavcev pa se je zaradi sečnje lesavvrtačah zmanjšalo. Zatosta se močno spremenili vrstnasestava in razporeditev veg-etacijevtehgozdovih,vrtačepasodelnoizgubilefunkcijovarnihzatočiščzahladnoljubnerastlinskevrste. KLJUČNEBESEDE:biologija,geografija,podnebnespremembe,sečnja,hladnoljubnerastline,kraškisvet, razporeditev vegetacije, Madžarska The paper was submitted for publication on June 24th, 2018. Uredništvo je prejelo prispevek 24. junija 2018. 1 UniversityofSzeged,DepartmentofEcology,DoctoralSchoolofEnvironmentalSciences,Szeged,Hungary kisspeterjanos003@gmail.com 2 University of Szeged, Department of Ecology, Szeged, Hungary festuca7@yahoo.com, imolaboni@gmail.com, zbatory@gmail.com 3 MTA Centre for Ecological Research, Department of Terrestrial Ecology, Vácrátót, Hungary erdos.laszlo@okologia.mta.hu 4 Eszterházy Károly University of Applied Sciences, Department of Botany, Eger, Hungary vojtkoa@gmail.com 5 University of Szeged, Department of Ecology, Szeged, Hungary, and Polish Academy of Sciences, Museum and Institute of Zoology, Warsaw, Poland bikmakk@gmail.com 1 Introduction Stable habitats buffered from climate changes are known as refugia (Ashcroft 2010; Keppel et al. 2012). Microrefugia are small sites with locally favourable environmental conditions (e.g., microclimate and soil moisture)amidstunfavourableregionalenvironments(Rull2009;Gentilietal.2015).Therefore,theyfacil­itate the in situpersistence of species. Convergent environments (e.g., basins, ravines and valleys) may act asmicrorefugia(Dobrowski2010;Bátorietal.2014a),sincetheirtopographiccomplexityprovidesadiver­sity of microclimates (Whiteman et al. 2004), allowing species to track suitable conditions with minimal movement.InEurope,dolines(localdepressionsofkarstsurfaces)mayalsoconstitutemicrorefugia(Bátori et al. 2009), as cold-air pooling occurs within them and north-facing slopes receive less insolation than surroundingareas(Bárány-Kevei1999).Forinstance,dolinesinCentralandSoutheasternEuropeareknown tomaintainrelictpopulationsofborealandhigh-mountainplantspecies(RitterBeckvonMannagetta1906; Horvat 1953; Kobal et al. 2015). Therefore, they have the capacity to harbour many plant and animal taxa that are rare or absent in the surrounding areas (Modrić Surina and Surina 2010; Kemencei et al. 2014; Raschmanová et al. 2015; Růžička et al. 2016; Bátori et al. 2019a). This capacity depends on small- and large-scaleenvironmentalfactors(Keppeletal.2015),suchasmacroclimate,depth/diameterratio,slopeaspect andvegetationcover(Bátorietal.2012;2017).Althoughdolinesactaskeyhabitatsforvarioustaxa,theeffects of human activity on their species-holding capacity have rarely been assessed and raise further questions. Expansiveeconomicgrowthandurbanizationposeaseriousthreattokarstlandforms(BregValjavec, Zorn and Čarni 2018; Bátori et al. 2019b). For instance, many dolines have been filled up with construc­tionwasteandrockdebrisinSloveniaduringtheexpansionofsettlementsandhighwayconstructions(Breg 2007;KovačičandRavbar2013).Consequently,theshapeanddepthofthesedolineshavechangedmarked­ly(Cernatič-GregoričandZega2010;BregValjavec2014;BregValjavec,RibeiroandČarni2017)andmany of them have lost their capacity to act as safe havens for biodiversity in changing climate. In Crete, road construction,overgrazingandnutrientadditionbysheeparethemainthreatstotheendemicplantspecies of dolines (Egli 1991; Iatroú and Fournaraki 2006). Human-induced changes in forest cover also have the potential to negatively influence the species diversity of karst depressions (Bárány-Kevei 2011). Deforestation is a serious threat to the biological diversity of karst surfaces (Calo and Parise 2006). Species respond to changes in light, nutrient, soil moisture and temperature individualistically (Stewart 2010),however,whenchangesarestrongandfast,functionallysimilarspeciesmayrespondsimilarly(Birks and Ammann 2000). Microhabitat changes induced by logging are especially pronounced in dolines (Lehmann 1970), where environmental conditions vary considerably from slope to slope and from top to bottom (Whiteman et al. 2004; Bátori et al. 2011). Previous studies revealed that the diurnal fluctuations in temperature, humidity and wind speed are less extreme in forested dolines than in non-forested ones (Lehmann1970).Tobetterunderstandtheimpactofloggingonthespecies-holdingcapacityofkarstdolines, we need to document the changes in species composition and vegetation pattern over time. ThewesternpartoftheMecsekMountains(Hungary)hastypicalkarstlandformfeatures,suchasdolines, withfunnel-shaped geometry and unique microclimate (Bátori et al. 2011). Our previous studies showed that larger dolines have the capacity to enable the persistence of cool-adapted plant species and diverse foresttypesthatareabsentfromthesurroundingplateau(Bátorietal.2012;2014a;2014b;2017).Intensive loggingbeganabout10yearsagointhisregion,andmatureforestcoverhasbeenreducedtoapproximately 30–40 percent of its original extent in some dolines. In the present study, we investigated the changes in the species-holding capacity of karst dolines induced by logging. Specifically, we studied the plant species compositionandvegetationpatternintwodolinesintheMecsekMountainsbeforeandafterlogging,and compared the focal species groups to each other. 2 Methods 2.1 Study area Thestudyareaisalimestonekarstlandscapeofabout30km2inthewesternpartoftheMecsekMountains (Figure 1), located at an altitude between 250–500m. The climate is continental with sub-Mediterranean influences: moderately warm (mean annual temperature is 9.5°C) and moderately humid (mean annual precipitation is 740mm) (Dövényi 2010). The number of dolines is about 2200 (Hoyk 1999). Beech and oak-hornbeamforestscovertheslopesofdolinesandthemajorpartoftheplateausbetweendolines,while ravine forests cover the bottom of larger dolines (Bátori et al. 2012). 2.2 Vegetation resurvey Twomedium-sizeddolineswereselected(doline1:WGS46.13359N,18.16589E;doline2:WGS46.13131N, 18.17164 E) with a funnel-shaped geometry. Dolines were about 70m in diameter and about 14m deep. Both dolines were sampled before logging (about 110 years old forests, in 2007) and 10 years after log­ging (in 2017), using the same method. The cover of mature trees within dolines has been reduced to approximately 30–40 percent of its original extent (Figure 1). Because the greatest differences in species compositionwereexpectedbetweenthenorth-andsouth-facingslopes(Bátorietal.2012),weestablished atransectwithnorth-to-southorientationacrossthedolines,traversingthedeepestpoints.Transectsbegan and ended on doline rims. Each transect consisted of 1m×1m plots spaced at 2-m intervals. We record-edthecoverofallherbs,shrubsandtreesaplingsinallplots.Atotalof114plotsweresurveyed.Nomenclature follows The Plant List (2018). 2.3 Species grouping Weclassifiedallplantspeciesaccordingtotheirhabitatpreference(Horváthetal.1995).Threelargerfunc­tional groups (cf. Troiani et al. 2016) were established: • »oakforestspecies«(i.e.typicalspeciesofturkeyoak–sessileoakforestsanddrieroak-hornbeamforests), • »beech and ravine forest species« and • »disturbance-tolerant species« (i.e. ruderals). All three functional groups were analysed using both the cover and presence/absence data of species. Since beech and ravine forest species are the best indicators of cool microhabitats in our study area, this functionalgroupwasconsideredtobeparticularlyrelevanttodetectchangesinthespecies-holdingcapac­ity of dolines. 2.4 Data analyses We used permutational multivariate analysis of variance (PERMANOVA) to test the effect of logging on plant assemblages. We used the raw cover data of species for each sampling plot in the source matrices. We applied the Bray–Curtis index, and performed 5,000 permutations. PERMANOVAs were calculated inRstatisticalenvironment(RCoreTeam2018)usingtheadonisfunctionofthe»vegan«package(Oksanen et al. 2018). We prepared non-metric multidimensional scaling (NMDS) ordinations (cover values and Bray–Curtis index) to visually illustrate differences in vegetation pattern. The diagnostic species of the dolines before and after logging were determined by calculating the phi (.) coefficient of association between species and habitat (Chytrý et al. 2002). Species with .>0.1 were considereddiagnostic.Non-diagnosticspecieswereexcludedwithFisher’sexacttest(p<0.05).Calculations were done with the JUICE 7.0.25 program (Tichý 2002). Weusedlinearmixed-effectmodels(LMMs)withGaussianerrortermforthecomparisonofthecover of functional groups, and generalized linear mixed-effects models (GLMMs) with Poisson error term for thecomparisonofthespeciesnumbersoffunctionalgroups.AllanalyseswerecarriedoutinR(RCoreTeam 2018).Inthemodels,themanagementtypes(unloggedandlogged)wereincludedasfixedfactors,thecover andnumberofspeciesasdependentvariables,andthelocation(i.e.doline1and2)asrandomfactor.LMMs wereperformedusingthelmefunctionfromthe»nlme«package(Pinheiroetal.2018),whileGLMMswere performed using the glmer function from the»lme4« package (Bates, Maechler and Bolker2013). Figure 1: Location of the study site (a) and studied dolines (b) in the Mecsek Mountains (Hungary). Dolines were sampled before logging (c) and 10 years after logging (d). Red lines indicate the position of transects. p 3 Results Atotalof72vascularplantspecieswererecordedintheplots.Fivediagnosticspecies(oneoakforestspecies andfourbeechandravineforestspecies)wereidentifiedbeforeloggingand15(sixoakforestspecies,three beech and ravine forest species, and six disturbance-tolerant species) after logging (Table 1). NMDS ordi­nations (stress factor:0.24 and0.22,respectively) showedthatthecompositional patternof thevegetation changed significantly (PERMANOVA: F=6.8 and 7.8, respectively, p<0.001) after logging (Figure 2). The cover and number of oak forest species (e.g., common bugle (Ajuga reptans), fragrant hellebore (Helleborusodorus)andbutcher’sbroom(Ruscusaculeatus))anddisturbance-tolerantspecies(e.g.,dead­lynightshade(Atropabelladonna),Americanburnweed(Erechtiteshieracifolia)andJapanesehedgeparsley (Torilisjaponica))werehigherafterlogging(p<0.001andp<0.05,respectively),whilethecoverandnum­ber of beech and ravine forest species (e.g., enchanter’s nightshade (Circaea lutetiana), common lungwort (Pulmonaria officinalis) and spineless butcher’s broom (Ruscus hypoglossum)) were higher before logging (p<0.001 and p=0.06, respectively) (Figure 3). 4 Discussion Karstdepressionsprovidesuitablemicrohabitatsforavarietyofspecies,buthuman-activitymayinfluence their species-holding capacity. To our knowledge, this is the first study to report evidence of a decrease in species-holding capacity within forested dolines confirmed with repeated sampling. We have shown that 10 years after logging (30–40 percent of the original forests remained intact) the vegetation pattern and speciescompositionchangedsubstantiallywithindolines,andtheyhavepartiallylosttheircapacitytoact as safe havens for a number of plant species adapted to cooler conditions. Aspreviousstudieshaveshown,coolandhumiddolinesinEurasiahavethecapacitytosupportanimal and plant populations outside their main distribution ranges (Raschmanová etal. 2015; Bátori et al. 2017; Su et al. 2017), and microclimate is a significant predictor of the species diversity within them (Růžička etal.2016).Althoughmicroclimaticdifferencesbetweentheslopesofforesteddolinesarelesspronounced Table 1: Diagnostic species of the dolines before and after logging (p<0.05). Species are listed by decreasing values of the phi (.) coefficient of association between species and habitat. Letters in parenthesis indicate functional groups: (a) oak forest species, (b) beech and ravine forest species and (c) disturbance-tolerant species. Ivy – Hedera helix (a) 0.47 Yellow archangel – Lamium galeobdolon s.l. (b) 0.37 Wood speedwell – Veronica montana (b) 0.33 Woodruff – Galium odoratum (b) 0.28 Dog’s mercury – Mercurialis perennis (b) 0.28 Hairy blackberry – Rubus hirtus agg. (c) 0.42 Stinging nettle – Urtica dioica (c) 0.42 European hornbeam – Carpinus betulus (b) 0.35 Wood small-reed – Calamagrostis epigejos (c) 0.34 Wood sedge – Carex sylvatica (b) 0.27 Annual fleabane – Stenactis annua (c) 0.24 Hairy St Johns-wort – Hypericum hirsutum (c) 0.23 Turkey oak – Quercus cerris (a) 0.23 Lady fern – Athyrium filix-femina (b) 0.21 Wall lettuce – Lactuca muralis (a) 0.21 Goat willow – Salix caprea (c) 0.21 Silver linden – Tilia tomentosa (a) 0.21 Germander speedwell – Veronica chamaedrys (a) 0.21 Wood melick – Melica uniflora (a) 0.20 Sessile oak – Quercus petraea (a) 0.19 Figure 2:Non-metricmultidimensionalscaling(NMDS)ordination(covervaluesandBray–Curtisindex)diagramsofdoline1(a)anddoline 2(b)before and after logging (stress factor: 0.24 and 0.22, respectively). Figure 3: Cover and species number of the different functional groups (a: oak forest species; b: beech and ravine forest species and c: disturbance-tolerant species) in the dolines of the Mecsek Mountains (Hungary) before (white boxes) and after (green boxes) logging. Significant differences are indicated by asterisks (*p <0.05; ***p <0.001). (Bátori et al. 2011; 2014b), dolines in the Mecsek Mountains introduce great variation in species compo­sitionandmanyplantspeciescanbefoundwithinthemthatarerareortotallyabsentfromthesurrounding karst plateaus (Bátori et al. 2012). In general, changes in vegetation cover should affect the surface albedo (i.e. the amount of reflected solarradiation),windpattern,soilhydrology,soilnutrientcontent,near-surfacetemperatureandhumid-ity,whichinturn,shouldaffectthevegetation(Saikh,VaradachariandGhosh1998;GuariguataandOstertag 2001; Berbet and Costa 2003; Lukić et al. 2017; Stančič and Repe 2018). Our results show that intensive loggingcancausesignificantchangesinlocalspeciesabundanceandcomposition,andhasanegativeeffect on the species-holding capacity of dolines (Figures 2 and 3). Dolines contained fewer beech and ravine forest species (e.g., narrow buckler-fern (Dryopteris carthusiana), herb Paris (Paris quadrifolia) and wood speedwell (Veronica montana)) 10 years after logging, and their cover was much lower than before log-ging.Incontrast,thenumberandcoverofoakforest(e.g.,fieldmaple(Acercampestre),woodmelick(Melica uniflora)andwhiteviolet(Violaalba))anddisturbance-tolerantspecies(e.g.,woodsmall-reed(Calamagrostis epigejos), hairy blackberry (Rubus hirtus agg.) and stinging nettle (Urtica dioica)) increased significantly due to logging. The increase in the number and cover of oak forest species likely resulted from a decrease in albedo and a corresponding increase in absorbed solar radiation, while greater post-disturbance light and nutrient availability, changes in species interactions (such as competition) could increase the num­ber and cover of disturbance-tolerant species. Since different species may recover at different rates during forestregeneration(Dunn2004),itisunclearhowspeciescompositionwillchangeinthesedolinemicrorefu­gia in the future. ExtremeheateventsareincreasinginfrequencyinCentralEurope,andwillcontinuetodosothrough-out the next few decades (Bartholy et al. 2008). Climate warming affects the local abundance, phenology and distribution of species and alters the interactions between populations (Hegland et al. 2009). The dis­tributions of many plant species in Europe are likely to shift upwards (Geßler et al. 2007; Czúcz, Gálhidy andMátyás2011;Hlásnyetal.2011),orspeciescanpersistinenvironmentallystablehabitats(Willis,Rudner and Sümegi 2000; McLaughlin et al. 2017), such as dolines (Bátori et al. 2017). A number of studies indi­cate that forest cover has the potential to mitigate climate warming at local scales and to maintain suitable microclimate, enhancing species persistence (Frey et al. 2016; Keppel et al. 2017). The combined effects of forest cover and topographic complexity may contribute to the long-term persistence of forest species adapted to cooler conditions (beech and ravine forest species) (Figure 3) in the karst region of the Mecsek Mountains,providedthatthecurrentmanagementintensitydecreases.Ifthismanagementpracticecontinues, thespecies-holdingcapacityofthesedolineswouldfurtherdecrease,andmanycurrentlywidespreadplant species may become rare or extinct during forest regeneration in a warming climate. 5 Conclusion We conclude that intensive logging poses a serious threat on the species-holding capacity of karst depres­sions. To meet the demands of preservation of cool-adapted species in dolines, it would be necessary to establish more forest reserves on the karst surfaces where no forestrytakes place. Alternatively, the main­tenanceofcontinuousforestcoverinatleastlargerdolinesandtheirsurroundingswouldbeindispensable to facilitate the recolonization or the in situ persistence of cool-adapted species under global warming. ACKNOWLEDGEMENT: This research was founded by the NKFIH K 124796 grant. We are thankful to Izabella Benczur and Petra Vass for their help in field works. 6 References Ashcroft, M. B. 2010: Identifying refugia from climate change. Journal of Biogeography 37-8. DOI: https://doi.org/10.1111/j.1365-2699.2010.02300.x Bárány-Kevei,I.1999:Microclimateofkarsticdolines.ActaClimatologicaUniversitatisSzegediensis32/33. Bárány-Kevei, I. 2011: Changes in the vegetation of dolines in Aggtelek and Bükk Mountains. Acta Climatologica et Chorologica 44/45. Bartholy,J.,Pongrácz,R.,Gelybó,G.,Szabó,P.2008:AnalysisofexpectedclimatechangeintheCarpathian Basin using the PRUDENCE results. Időjárás 112, 3-4. Bates, D., Maechler, M., Bolker, B. 2013: lme4: linear mixedeffects models using S4 classes. R package ver­sion 0.999999-2. Internet: http://cran.r-project.org/package=lme4 (26. 4. 2018). Bátori, Z., Csiky, J., Erdős, L., Morschhauser, T., Török, P., Körmöczi, L. 2009: Vegetation of the dolines in Mecsek Mountains (South Hungary) in relation to the local plant communities. Acta Carsologica 38, 2-3. DOI: https://doi.org/10.3986/ac.v38i2-3.125 Bátori,Z.,Csiky,J.,Farkas,T.,Vojtkó,E.A.,Erdős,L.,Kovács,D.,Wirth,T.,Körmöczi,L.,Vojtkó,A.2014a: TheconservationvalueofkarstdolinesforvascularplantsinwoodlandhabitatsofHungary:refugiaand climatechange.InternationalJournalofSpeleology43-1.DOI:https://doi.org/10.5038/1827-806X.43.1.2 Bátori,Z.,Lengyel,A.,Maróti,M.,Körmöczi,L.,Tölgyesi,C.,Bíró,A.,Tóth,M.,Kincses,Z.,Cseh,V.,Erdős, L. 2014b: Microclimate-vegetation relationships in natural habitat islands: species preservation and conservation perspectives. Időjárás 118-3. Bátori, Z., Gallé, R., Erdős, L., Körmöczi, L. 2011: Ecological conditions, flora and vegetation of a large doline in the Mecsek Mountains (South Hungary). Acta Botanica Croatica 70-2. DOI: https://doi.org/ 10.2478/v10184-010-0018-1 Bátori, Z., Körmöczi, L., Erdős, L., Zalatnai, M., Csiky, J. 2012: Importance of karst sinkholes in preserv­ing relict, mountain, and wet-woodland plant species under sub-Mediterranean climate: A case study from southern Hungary. Journal of Cave and Karst Studies 74-1. DOI: https://doi.org/10.4311/ 2011LSC0216 Bátori, Z., Vojtkó, A., Farkas, T., Szabó, A., Havadtői, K., Vojtkó, E. A., Tölgyesi, C., Cseh, V., Erdős, L., Maák, I., Keppel G. 2017: Large- and small-scale environmental factors drive distributions of cool-adaptedplantsinkarsticmicrorefugia.AnnalsofBotany119-2.DOI:https://doi.org/10.1093/aob/mcw233 Bátori, Z., Vojtkó, A., Maák, I. E., Lőrinczi, G., Farkas, T., Kántor, N., Tanács, E., Kiss, P. J., Juhász, O., Módra,G.,Tölgyesi,C.,Erdős,L.,Aguilon,D.J.,Keppel,G.2019a:Karstdolinesprovidediversemicro­habitatsfordifferentfunctionalgroupsinmultiplephyla.ScientificReports9.DOI:https://doi:10.1038/ s41598-019-43603-x Bátori,Z.,Vojtkó,A.,Keppel,G.,Tölgyesi,C.,Čarni,A.,Zorn,M.,Farkas,T.,Erdős,L.,Kiss,P.J.,Módra,G., Breg Valjavec, M. 2019b: Anthropogenic disturbances alter the conservation value of karst dolines. Biodiversity and Conservation. DOI: https://doi.org/10.1007/s10531-019-01896-4 Berbet,M.L.C.,Costa,M.H.2003:Climatechangeaftertropicaldeforestation:seasonalvariabilityofsur­facealbedoanditseffectsonprecipitationchange.JournalofClimate16.DOI:https://doi.org/10.1175/ 1520-0442(2003)016<2099:CCATDS>2.0.CO;2 Birks,H.H.,Ammann,B.2000:Twoterrestrialrecordsofrapidclimaticchangeduringtheglacial–Holocene transition (14,000–9,000 calendar years B.P.) from Europe. PNAS 97-4. DOI: https://doi.org/10.1073/ pnas.97.4.1390 Breg Valjavec, M. 2014: Study of filled dolines by using 3D stereo image processing and electrical resis­tivityimaging.InternationalJournalofSpeleology43-1.DOI:https://doi.org/10.5038/1827-806X.43.1.6 Breg Valjavec, M., Ribeiro, D., Čarni, A. 2017: Vegetation as the bioindicator of human-induced degra­dationinkarstlandscape:casestudyofwaste-filleddolines.ActaCarsologica46-1.DOI:https://doi.org/ 10.3986/ac.v46i1.4712 BregValjavec,M.,Zorn,M.,Čarni,A.2018:Bioindicationofhuman-inducedsoildegradationinenclosed karst depressions (dolines) using Ellenberg indicator values (Classical Karst, Slovenia). Science of the Total Environment 640/641. DOI: https://doi.org/10.1016/j.scitotenv.2018.05.294 Breg,M.2007:DegradationofdolinesonLogaškoPolje(Slovenia).ActaCarsologica36-2.DOI:https://doi.org/ 10.3986/ac.v36i2.191 Calo, F., Parise, M. 2006: Evaluating the human disturbance to karst environments in Southern Italy. Acta Carsologica 35, 2-3. DOI: https://doi.org/10.3986/ac.v35i2-3.227 Cernatič-Gregorič,A.,Zega,M.2010:Theimpactofhumanactivitiesondolines(Sinkholes)–typicalgeo­morphologicfeaturesonkarst(Slovenia)andpossibilitiesoftheirpreservation.GeographicaPannonica 14-4. DOI: https://doi.org/10.5937/GeoPan1004109C Chytrý,M.,Tichý,L.,Holt,J.,Botta-Dukát,Z.2002:Determinationofdiagnosticspecieswithstatisticalfidelity measures.JournalofVegetationScience13-1.DOI:https://doi.org/10.1111/j.1654-1103.2002.tb02025.x Czúcz, B., Gálhidy, L., Mátyás, C. 2011: Present and forecasted xeric climatic limits of beech and sessile oakdistributionatlowaltitudesinCentralEurope.AnnalsofForestScience68-1.DOI:https://doi.org/ 10.1007/s13595-011-0011-4 Dobrowski,S.Z.2010:Aclimaticbasisformicrorefugia:theinfluenceofterrainonclimate.GlobalChange Biology 17-2. DOI: https://doi.org/10.1111/j.1365-2486.2010.02263.x Dövényi, Z., Ambrózy, P., Juhász, Á., Marosi, S., Mezosi, G., Michalkó, G., Somogy, S., Szalai, Z., Tiner, T. 2010: Magyarország kistájainak katasztere (Inventory of microregions in Hungary). Budapest. Dunn, R. R. 2004: Recovery of faunal communities during tropical regeneration. Conservation Biology 18-2. DOI: https://doi.org/10.1111/j.1523-1739.2004.00151.x Egli, B. R. 1991: The special flora, ecological and edaphic conditions of dolines in the mountain of Crete. Botanika Chronika 10. Frey, S. J. K., Hadley, A. S., Johnson, S. L., Schulze, M., Jones, J. A., Betts, M. G. 2016: Spatial models reveal the microclimatic buffering capacity of old-growth forests. Science Advances 2. DOI: https://doi.org/ 10.1126/sciadv.1501392 Gentili, R., Baroni, C., Caccianiga, M., Armiraglio, S., Ghiani, A., Citterio, S. 2015: Potential warm-stage microrefugia for alpine plants: Feedback between geomorphological and biological processes. Ecological Complexity 21. DOI: https://doi.org/10.1016/j.ecocom.2014.11.006 Geßler, A., Keitel, C., Kreuzwieser, J., Matyssek, R., Seiler, W., Rennenberg, H. 2007: Potential risks for European beech (Fagus sylvatica L.) in a changing climate. Trees 21-1. DOI: https://doi.org/10.1007/ s00468-006-0107-x Guariguata, M. R., Ostertag, R. 2001: Neotropical secondary forest succession: changes in structural and functionalcharacteristics.ForestEcologyandManagement148,1-3.DOI:https://doi.org/10.1016/S0378­1127(00)00535-1 Hegland, S. J., Nielsen, A., Lázaro, A., Bjerknes, A-L., Totland, O. 2009: How does climate warming affect plant-pollinatorinteractions?EcologyLetters12-2.DOI:https://doi.org/10.1111/j.1461-0248.2008.01269.x Hlásny,T.,Barcza,Z.,FabrikaM.,Balázs,B.,Churkina,G.,Pajtík,J.,Sedmák,R.,Turčáni,M.2011:Climate change impacts on growth and carbon balance of forests in Central Europe. Climate Research 47-3. DOI: https://doi.org/10.3354/cr01024 Horvat, I. 1953: Vegetacija ponikava. Geografski Glasnik 14/15. Horváth, F., Dobolyi, Z. K., Morschhauser, T., Lőkös, L., Karas, L., Szerdahelyi, T. 1995: FLORA database. 1.2. Vácrátót. Hoyk,E.1999:GeoecologicalstudiesonthekarsticsurfacesoftheplannedprotectedareainWesternMecsek, South Hungary. Acta Carsologica 28-2. DOI: https://doi.org/10.3986/ac.v28i2.485 Iatroú, G., Fournaraki, C. 2006: Horstrissea dolinicola. The IUCN Red List of Threatened Species 2006. DOI: http://doi.org/10.2305/IUCN.UK.2006.RLTS.T61613A12522933.en Kemencei,Z.,Farkas,R.,Páll-Gergely,B.,Vilisics,F.,Nagy,A.,Hornung,E.,Sólymos,P.2014:Microhabitat associations of land snails in forested dolinas: implications for coarse filter conservation. Community Ecology 15-2. DOI: https://doi.org/10.1556/ComEc.15.2014.2.6 Keppel,G.,Anderson,S.,Williams,C.,Kleindorfer,S.,O’Connell,C.2017:Microhabitatsandcanopycover moderatehighsummertemperaturesinafragmentedMediterraneanlandscape.PLoSONE12-8.DOI: https://doi.org/10.1371/journal.pone.0183106 Keppel, G., Mokany, K., Wardell-Johnson, G. W., Phillips, B. L., Welbergen, J. A., Reside, A. E. 2015: The capacity of refugia for conservation planning under climate change. Frontiers in Ecology and the Environment 13-2. DOI: https://doi.org/10.1890/140055 Keppel,G.,VanNiel,K.P.,Wardell-Johnson,G.W.,Yates,C.J.,Byrne,M.,Mucina,L.,Schut,A.G.T.,Hopper, S.D.,Franklin,S.E.2012:Refugia:identifyingandunderstandingsafehavensforbiodiversityundercli­matechange.GlobalEcologyandBiogeography21-4.DOI:https://doi.org/10.1111/j.1466-8238.2011.00686.x Kobal, M., Bertoncelj, I., Pirotti, F., Dakskobler, I., Kutnar, L. 2015: Using lidar data to analyse sinkhole characteristics relevant for understory vegetation under forest cover – case study of a high karst area in the Dinaric Mountains. PLoS ONE 10-3. DOI: https://doi.org/10.1371/journal.pone.0122070 Kovačič,G.,Ravbar,N.2013:Analysisofhumaninducedchangesinakarstlandscape–thefillingofdolines in the Kras plateau, Slovenia. Science of the Total Environment 447. DOI: https://doi.org/10.1016/ j.scitotenv.2013.01.002 Lehmann,A.1970:Tarvágásáltalokozottökológiaiváltozásokazabaligetikarszton.PécsiMűszakiSzemle25. Lukić, T., Marić, P., Hrnjak, I., B. Gavrilov, M., Mladjan, D., Zorn, M., Komac, B., Milošević, Z., Marković, B. S., Sakulski, D., Jordaan, A., Đorđević, J., Pavić, D., Stojsavljević, R. 2017: Forest fire analysis and classification based on a Serbian case study. Acta geographica Slovenica 57-1. DOI: https://doi.org/ 10.3986/AGS.918 McLaughlin, B. C., Ackerly, D. D., Klos, P. Z., Natali, J., Dawson, T. E., Thompson, S. E. 2017: Hydrologic refugia,plants,andclimatechange.GlobalChangeBiology23-8.DOI:https://doi.org/10.1111/gcb.13629Modrić Surina, Ž., Surina, B. 2010: Snowbed vegetation in Croatia: Phytosociology, ecology and conser­vation status. Plant Biosystems 144-4. DOI: https://doi.org/10.1080/11263504.2010.502716 Oksanen, J., Blanchet, F. G., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Minchin P. R., O’Hara, R. B.,Simpson,G.L.,Solymos,P.,StevensM.H.H.,Szoecs,E.,Wagner,H.2018:Vegan:CommunityEcology Package. Internet: http://CRAN.R-project.org/package=vegan (26.4.2018). Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., and R Core Team 2018: nlme: linear and nonlinear mixed effects models. R package version 3.1-137. Internet: https://cran.r-project.org/web/packages/nlme/ index.html (26.4.2018). R Core Team, 2018: R: A language and environment for statistical computing. Vienna. Internet: http://www.R-project.org/ (6.5.2018). Raschmanová, N., Miklisová, D., Kováč, L’., Šustr, V. 2015: Community composition and cold tolerance of soil Collembola in a collapse karst doline with strong microclimate inversion. Biologia 70-6. DOI: https://doi.org/10.1515/biolog-2015-0095 Ritter Beck von Mannagetta, G. 1906: Die Umkehrung der Pflanzenregionen in den Dolinen des Karstes. Deutsche Universität Prag, Prag. Rull,V.2009:Microrefugia.JournalofBiogeography36-3.DOI:https://doi.org/10.1111/j.1365-2699.2008.02023.xRůžička,V.,Mlejnek,R.,Juřičková,L.,Tajovský,K.,Šmilauer,P.,Zajíček,P.2016:InvertebratesoftheMacocha Abyss(MoravianKarst,CzechRepublic).ActaCarsologica45-1.DOI:https://doi.org/10.3986/ac.v45i1.896 Saikh,H.,Varadachari,C.,Ghosh,K.1998:Changesincarbon,nitrogenandphosphoruslevelsduetodefor­estation and cultivation: A case study in Simlipal National Park, India. Plant and Soil 198-2. DOI: https://doi.org/10.1023/A:1004391615003 Stančič,L.,Repe,B.2018:Post-firesuccession:SelectedexamplesfromtheKarstregion,southwestSlovenia. Acta geographica Slovenica 58-1. DOI: https://doi.org/10.3986/AGS.1942 Stewart,J.R.,Lister,A.M.,Barnes,I.Dalén,L.2010:Refugiarevisited:individualisticresponsesofspecies inspaceandtime.ProceedingsoftheRoyalSocietyB277-1682.DOI:https://doi.org/10.1098/rspb.2009.1272 Su,Y.,Tang, Q.,Mo, F., Xue, Y. 2017: Karsttiankengs asrefugia for indigenous tree flora amidsta degraded landscapeinsouthwesternChina.ScientificReports7.DOI:https://doi.org/10.1038/s41598-017-04592-x The plant list, 2018. Internet: http://www.theplantlist.org (6.5.2018). Tichý, L. 2002: JUICE, software for vegetation classification. Journal of Vegetation Science 13-3. DOI: https://doi.org/10.1111/j.1654-1103.2002.tb02069.x Troiani, N., Tardella, F. M., Malatesta, L., Corazza, M., Ferrari, C., Catorci, A. 2016: Long-term cropland abandonmentdoesnotlead persetothe recoveryofsemi-naturalherbcommunitiesdeemed habitats of community interest. Acta Botanica Croatica 75-2. Whiteman, C. D., Haiden, T., Pospichal, B., Eisenbach, S., Steinacker, R. 2004: Minimum temperatures, diurnal temperature ranges, and temperature inversion in limestone sinkholes of different sizes and shapes. Journal of Applied Meteorology 43. DOI: https://doi.org/10.1175/1520-0450(2004)043<1224: MTDTRA>2.0.CO;2 Willis, K. J., Rudner, E., Sümegi, P. 2000: The full-glacial forests of Central and southeastern Europe. Quaternary Research 53-2. DOI: https://doi.org/10.1006/qres.1999.2119 COMMERCIAL SERVICES AND URBAN SPACE RECONVERSION IN ROMANIA (1990–2017) Radu Săgeată »Afi Cotroceni«, the biggest mall in Romania (over 90,000 sqm, commercial surface), was open on Oct. 29, 2009. The structure was built in a former industrial area (Electrical Equipment Plant, Bucharest). DOI: https://doi.org/10.3986/AGS.6995 UDC: 911.375.12:339.378.2(498) COBISS: 1.01 Radu Săgeată1 Commercial services and urban space reconversion in Romania (1990–2017) ABSTRACT:Concentrating incomes in large cities has encouraged the development of specialist services andtheopeningofbigcommercialunits.ThedownfallofideologicalbarrierseastoftheformerIronCurtain made global culture combine with endemic sub-cultures, influenced by the living standard. The only lim­itation of this process appears to be social segregation which restricts demand and creates preferential segments of users. In Romania, financial segregation is directly reflected in the commercial investment made in Bucharest and in the large cities, mostly in the centre and western part of the country. The paper analysesthecorrelationoffinancialandcommercialservices,aswellastheirlocationanddispersionstrate­gies at the level of the Romanian urban system. KEY WORDS: globalization, urban spatial reconversion, services, commercial investments, Romania Trgovske storitve inponovnapreobrazbamestnega prostora vRomuniji(1990–2017) POVZETEK: Zgoščanje dohodka v velikih mestih je spodbudilo razvoj specializiranih storitev in odpi­ranjevelikihtrgovskihsredišč.Zaradipadcaideološkihovirvzhodnoodnekdanježeleznezavesesejezačela globalna kultura mešati z lokalnimi subkulturami, kolikor jim je to dopuščal življenjski standard. Edina slabosttegaprocesajesocialnasegregacija,kiomejujepovpraševanjeinustvarjaprednostnesegmenteuporab­nikov.VRomunijisefinančnasegregacijaneposrednokaževnaložbahvgradnjotrgovskihsrediščvBukarešti in večjih mestih, zlasti v osrednjem in zahodnem delu države. V članku avtor analizira povezavo med finančnimi in trgovskimi storitvami ter strategije, povezane z njihovim umeščanjem in razpršenostjo na ravni romunskega urbanega sistema. KLJUČNE BESEDE: globalizacija, ponovna preobrazba mestnega prostora, storitve, naložbe v trgovska središča, Romunija The paper was submitted for publication on April 26th, 2018. Uredništvo je prejelo prispevek 26. aprila 2018. 1 Romanian Academy, Institute of Geography, Bucharest, Romania rsageata@gmail.com 50 1 Introduction Withthecollapseofideologicalbarriersattheendofthe1980s,deep-goingeconomicmutationstookplace in the former communist space. In Romania, transition from a central-based economy to a free-compe-titionsystem,destructuredtheeconomy,aneweconomicstructuresbeingputinstead.However,corruption andfailedprivatizationsmadeespeciallyindustrialunitsbankrupt,whicheventuallywereamongthefirst to disappear. A new space-dependency between production and consumption and new spatial poles that concentrated the population’s incomes would develop (Ianoş and Heller 2006). The main aim of the present study is to highlight changes in the urban space organization after the decline of industry and the development of commercial services which have re-balanced the territorial systemsdisturbedbyoversizedindustrializationundercommunism.Asaresult,largerural-urbanmigra­tionflowsgotmomentum,associatedwithafast-growinghousing-stockinthecity(GoodmanandMarshall 2013). This process, which could be labelled »quantitative urbanization«, evolved faster than »qualitative urbanization«,thatis,thedevelopmentofatechnical-buildingandservicesinfrastructure(Novotný2016). The issues discussed in this paper highlight the relationships between territorial disparities, purchas­ingpowerandthelocationoffinancial-bankingandcommercialinvestmentsinRomania.Aftertheindustrial sector had collapsed, a services sector developed, these processes being characteristic of the former cen­tral-basedeconomiesinCentralandEasternEuropeafter1989(Brülhart,CarrereandRobert-Nicoud2018). Also, the logic behind big commercial investments and the social impact of promoting the services sec­tor through big commercial investments is discussed herein. Deindustrializationmeansreducingtheshareofindustryinacountry’seconomy,orinahumancom­munity. This is specific to the periods of transition, or of economic crises and affects both the industrial productionandpeople’slivesasrealwagesandthelivingstandardaredecliningandunemploymentisris­ing (Rita Sedita, De Noni and Pilotti 2017). Transition from an excessively industrialized economy to a free-competition system is marked by globalization, deindustrialization preceding the development of the services sector. The complex relations between deindustrialization and the emergence of the services sector, as well as their impact on urban space organization within the general context of globalization, is a topical issue of major interest for the Central- and East-European countries (Jacoby and Korkut 2015). The fall of the communist system and the elimination of ideological barriers created the premises for a closer connection among the urban systems in neighbouring countries (McFarlane 2006). Cities acted as polarization nuclei, redirecting human flows in the territory. And yet, urban development could not makeupforindustrialdeclineanditsdemographic,butespeciallysocialconsequences(Crescenzi,Pietrobelli and Rabellotti 2016). After four decades of forced industrialization associated with hypertrophic devel­opment, the Romanian urban system experienced a radical change, basically de-urbanization of the city population and city-life quality, the functional characteristics of cities undergoing essential modifications (Holmén 1997). Industry, which had been the main factor of urbanization, would be overcome by the services sector, a phenomenon specific mainly to the big cities. Thus, Romania’s capital-city and the regional metropo­lisesstrengthenedtheirpositionsofco-ordinationintheterritory(Ianoş,PumainandRacine2000;Bürkner and Totelecan 2018). Bigcitiestendedtoassumetheroleofcosmopolitecitiesduetoanincreasingethnicaldiversity,aswell as to the development of specialist services and the large-scale assimilation of consumption products of theglobalculturethatgobeyondculturalfrontiers(Stead,deVriesandTasan-Kok2015;Wikström,Jönsson andL’EspoirDecosta2016).Foodandfashionitemshavethegreatestimpactonthepopulation.Immigration toRomaniaofsomeArab,Turkish,andChinesecitizensdealingintrade,orhavingsmallindustrialenter­prises,thepresenceofforeigncitizensstudyinginthiscountry,thestaffofdiplomaticmissions,ofmultinational companies,orofnon-governmentorganizationskeptdiversifyingtheservicessector.So,Chinese,Lebanese, Italian,orGreekrestaurantswouldopen,asdidFrenchorGermanbakeries,African,Indian,Latin-American, or other artisanal shops (van Ham and Tammaru 2016). The ever-greater concentration of incomes in the country’s capital-city, or in large cities, stimulated the development of products and specialist services, profile commercial units cropping up, usually in big commercial centres of the mall and supermarket type. This would explain the importance of these units forspreadingtheitemsoftheglobalconsumeristcultureatlocallevel(BezinandMoizeau2017).Thelim-its of this type of localization lie presumably in the growing social segregation through the ever-wider gap between the population’s income and demand, so that access is restricted to some particular segments of users (Picard and Zenou 2018). Hence, new spatial polarizations, directly proportional to social and cul­turalsegregation,anddependentonthecapacityoftheRomanianurbansystemtotakeinglobalizingflows. Thus,thegrowingfinancialsegregationbetweentheurbanpopulation,asaresultofrestructuredeconomic activities, opens up the way to globalizing flows (Demetry, Thurk and Fine 2015). Even if the items of the global consumerist culture are easily penetrating at local level, the population’s access to them is still fair­ly limited. Global culture tends to combine with local culture grafted on poverty and lack of education. In view of it, the quality of urban life is degrading through the development of urban subcultures and the proliferation of organized crime (Cohen 2016). Studying the distribution of banking investments could be a starting-point for looking at the spatial distribution of other types of services, commercial ones, in particular. 2 Data collection and methods 2.1 Data-set Thepresentresearchreliesonfieldinvestigation,bibliographicalsourcesandquantitativeanalyses.Inthis way, correlations could be established between the analyzed elements and the statistical prognoses. Field investigations focused on current land-use practices in certain urban and peri-urban areas and their situation in the 1988-1989 period, the peak of socialist industrialization in Romania (Dumitrescu 2008). The differences found show the changes occurred in the organization of the urban and peri-urban area, as well as in the functional zoning of cities in the post-communist period of deindustrialization. Preferences in the distribution of commercial investments reveal the extent of socio-economic avail­ability both at country level and in each city. Assessing the situation meant estimating the distribution of financial-banking investments. To this end, the internet sites of the 38 commercial banks in our study-sample were accessed. Investment distribution was compared with the territorial disparities highlighted by the poverty-level indicators reported in official statistics. Collecting data on the territorial distribution of malls and supermarkets was made by consulting the internetsitesofthemainbigcommercialcentres.Acorrelationwasestablishedbetweentheterritorialdis­tributionofthesecentersandthefinancial-bankinginvestments,correspondingtothepopulation’spoverty level. Our analyses covered all mall-type commercial investments of rank 0, I and II cities, as well as the stores of the 25 chains of big European retailers located in these cities. 2.2 Research methods In order to get an insight into the changes of the urban space once industrial zones were turned into com­mercial and services zones, one should proceed by looking at the impact of the population’s financial segregation when commercial investments are made (Mermet 2017). So, the analysis of the spatial distri­butionofacharacteristicsampleof38commercialbanks,whichprovedtoberelativelystableintheRomanian capitalmarket,hasbeenmadebyabi-dimensionallinearregressionmodel,conceivedasacorrelationbetween the logic of financial-banking investment distribution area and commercial investments. What underlies the dependency of placing investments in the services sector is the perpetuation, in time, of path dependency (Isaksen and Jakobsen 2017). Once the logic of locating the main big investments was established, a correlation was made between thelocationofnewcommercialsitesandwhathadpreviouslyexistedinthoseareas.Inthisway,afewtypolo­giesofurbanandperi-urbanspatialreconversionwereoutlinedintermsofthedevelopmentofcommercial services. The impact of ethnical minorities on the physiognomy and functional particularities of commercial zones was analyzed as part of the process of large Romanian cities acquiring a cosmopolitan character. This phenomenon is ever more often met in Central and Eastern Europe. There is a close correlation between the territorial distribution of poverty (i.e. minimum guaranteed income, labour employment rate, purchasing power/inhabitant) (Paugam 1998), and the distribution of financial-banking and commercial investments. This distribution can be expressed by a bi-dimensional regression equation between the number of banking units and of commercial centres: y=a + bx, where x=number of banking units, y=number of commercial units and a = a point on the axis of com-mercialunitswhenthenumberofbankingunits=0andb=regressionslopespecifictotherelationbetween the two variables. BucharestisahypertrophiccityintheRomanianurbansystemandthecorrelationbetweenbothvari- ablesgivesawrongimageonterritorialdistribution.Thisexplainswhyitisnecessarytoanalyzetheurban system with and without Bucharest. Synthetically, the main methodological steps which define our approach were as follows: a) Highlighting the main territorial disparities in Romania; b) Correlating territorial disparities with the purchasing power and the placement of financial-banking and commercial investments; c) Analyzingthecorrelationbetweentheterritorialdistributionoffinancial-bankingandcommercialinvest­ments using the bilinear regression (with and without Bucharest); d) Defining the changes occurred within the urban space organization through deindustrialization and tertiarization,thedevelopmentofcommercialservicesandtheestablishmentofatypologytothiseffect; e) Evaluating the impact of big commercial investments on Romania’s cities. 3 Results and discussion 3.1 Territorial disparities in Romania: General remarks InordertobetterunderstandthelogicbehindlocatingnewcommercialinvestmentsinRomania,oneshould firstlook atthemaincharacteristicsoftheurbansystemandthemajorterritorialdisparities inthiscountry. The Romanian urban system (Figure 1) contains 320 towns (Mitrică, Grigorescu and Urucu 2016). Afour-rank hierarchy of towns had in view the economic and demographic polarization grade in the ter­ritory and the social-urbanism level. The first three ranks (0, I and II) include 93 urban centres (29% of all of Romania’s towns) with 8.6 million inhabitants (68.5%) of the urban population and ca. 43% of Romania’s total population. These are best representative for the urban system in this country. The poverty rate of the population is decisive for the purchasing power and the location of financial-banking and commercial investments. Territorial disparities show two poverty polarization areas: 1) in the east (Moldavia), extreme poverty (lack of means of daily subsistence) in the counties of Vaslui and Botoşani; 2) in the south, extreme poverty in Ialomiţa, Giurgiu and Teleorman counties. According to statistical data (2014), most poverty-prone people, running the risk of social exclusion, liveintheSouth-Eastdevelopmentregion(53.4%/totalpopulation),North-East(48.9%)andSouth-West (45.6%).Theseterritorialdisparitiesareconfirmedbythefollowinganalyzedindicators:percentageofpop­ulationreceivingaminimumguaranteedincome,labouremploymentlevel,andpurchasingpower/inhabitant In Romania, 1.1% of the population benefits from a guaranteed minimum (31.85 Euro/person) from the State budget; on average, most socially-assisted people live in the east (Vaslui County – 2.7%) and south (Teleorman and Mehedinţi counties – 2.4% each; Buzău and Dolj counties – 2.2% each). Labour employ­ment level in Romania: 59.8% of the total active population (15–64-year olds) in 2016, minimum values being recorded again in the eastern and southern counties. Disparities are also shown by the purchasing power index: the poorest counties lying in the east and south (Vaslui: 3,054 Euro/inhabitant) (Figure 2). Figure 1:The urban system in Romania. p p. 54 Figure 2: Purchasing power and banking unit density in Romania (2016). p p. 55 Legend: Development regions: BUCHAREST Capital city – – I.North EastU k r a i n e Dorohoi Municipalites– II.South East – IAŞI county seats, – (regional metropolises) III.South11 Rădăuţi Boto ani ş Maramureş SATU IV. Bucharest–Ilfov V.South West – VI.West VII.Centre VIII. North West – Carei BAIA MARE BOTO ANI Ş MARE Câmpulung SUCEAVA Satu Mare Municipalites– Moldovenesc Fălticeni Suceava SUCEAVA county seats – Vatra Dornei VIII. Iaşi Bistri a­ ţ Marghita Paşcani S laj ă N ăăs ud Dej BISTRI A ZAL UŢ PIATRA-IAŞI Ă ORADEA NEAMŢ Roman Gherla Bihor Vaslui– County Salonta Topliţa I. Câmpina Municipalites Towns Reghin CLUJ-Cluj Neamţ Rank 0 Mureş BeiuşHuşi SIBIU Orăştie Codlea Lugoj 6 GHEORGHE Gala iţ Gheorgheni Rank I NAPOCA 10 TÂRGU MUREŞ I VIII – DevelopmentTurda – Harghita MoineştiVASLUI regions BAC UVaslui MIERCUREAĂ Rank II Târnăveni CIUC Bacău Aiud Arad Bârlad Oneşti Blaj 8 ARAD Rank III Alba 9 ALBAMediaş Brad IULIA VII. Adjud 7 DEVACovasnaTecuci Republic of Sebeş Sibiu F g raş ăă Timiş SFÂNTU VranceaMoldova TIMI OARA Ş ş BRA OVŞ Bra ov FOC ANI Ş Hunedoara GALAŢI Râmnicu Ukraine Săcele VI. Caransebeş Lupeni 5 Sărat 4 CâmpulungII. Buzău ŞŢA Ă Cara ş -1 Argeş Vâlcea Argeş Moreni Prahova BUZ UĂ Brăila Tulcea RE ICurtea deCâmpina BR ILATULCEA Severin 2 PITE TI GorjŞţ PLOIE TI Ş Motru Drăgăşani Dâmbovi a Orşova TÂRGOVI TEUrziceni Ş Ilfov 3 Ialomi a ţ SLOBOZIA Mehedin iţ Feteşti IV. Black V. III.BUCHAREST Sea RÂMNICU VÂLCEA, TÂRGU JIU, 1. 2. SLATINA C l raiş 3. DROBETA - TURNU SEVERIN, 4. Vulcan, Dolj ăă C L RAI ĂĂ Ş Olt 3 Petro ani, 5. ş 6. Hunedoara, 7. Târgu Secuiesc,CRAIOVA Medgidia CONSTANŢA Teleorman Olteniţa Giurgiu 8. Odorheiu Secuiesc, 9. ş 10. Câmpia Turzii,Caracal RoşioriConstan a Sighi oara, ţ de VedeGIURGIU Băileşti 11. Sighetu Marma iei. ţ Calafat ALEXANDRIA Mangalia Turnu Măgurele Content and map by: Radu S geatăă , 0 20 40 60 kmB u l g a r i a © 2018 Romanian Academy, Institute of Geography 3.2 Purchasing power, financial-banking and commercial investments The logic of making financial-banking and commercial investments lies at the base of the territorial dis­parities of the population’s purchasing power. An analysis of the 38 commercial banks reveals their clear-cut concentration in the urban area, with one exception (CEC Bank). Another characteristic feature of the Romanian banking system is the pyra­midal distribution of banking units, most of them located in Bucharest (19% of all) and in rank I and II cities,regionalandcountypolarizationnuclei,whichalsohavethebestservicescoverage/inhabitant.Small, especiallyone-industrytowns,ortownssituatedinhigh-povertyareasandoccupationaldependency,with little financial-banking investments, make the population dependent on the services offered only by cer-tainbanks,whichimplicitlyhavetheregionalmonopole.Thebestcoverageofbankingservicesunits(positive externalities) is in Bucharest, as well as in some large cities from the west of Romania, and in Constanţa (aBlackSeaportandtouristcity).Here,thereisagoodcorrelationbetweenincomedistributionandpeo­ple’s purchasing power. At the other endofthe spectrum (negative externalities) stand the poor countries intheeast(especiallyVasluiandBotoşani)andinthesouth-east(Călăraşi,Ialomiţa,GiurgiuandTeleorman) (Mitrică et al. 2017). Thereisanobviouspath-dependencybetweentheterritorialdistributionofthepurchasingpowerand ofbankingservices,visibleinthepopulationdensity/bankingunit.Positiveanomaliesrecordedsomecities inwhichtheindustrialdeclinecouldbecompensatedforbytourism(e.g.Constanţa,PiatraNeamţ),enabling the employment of the industry-released labour surplus, or by some cities that have a good geographical position versus the big communication axes (e.g. Suceava). At the other end of the spectrum one finds negativeanomalieswhichlargelyaffecttheiron-and-steelindustrycentres(Galaţi,Hunedoara,Reşiţa,Călăraşi and Târgovişte) and the mining ones (Petroşani and Motru), in which the process of economic reconver­sion, generated unemployment and a low purchasing power, or towns located in the neighbourhood of some regional metropolis (Săcele), where the labour force is taken in by the city (Braşov) (Figure 2). Thesamepyramidaldistributionoffinancialinvestmentsholdsalsoforcommercialunits,theextentof investments being directly proportional to town-size. Big commercial units are frequently opened in large regionalpolarizationcitieswithapositiveeconomicdynamic,offeringafairlycompetitivefinancialmarket thatensuresprofitability.Commercialinvestmentsmadeintownssituatedatthelowerendoftheurbanhier­archy(under30,000inhabitatnts)aremostlyofthesmallsupermarkettype,becausemakingbigcommercial investments insuch townsisnot profitable. IntheterritoryofRomania, the east/west financialsegregation isthedirectreflectionofcommercialservices,too;bigcommercialinvestmentsareattractedmainlybyBucharest, Constanţa and the large cities in Banat and Transylvania (Timişoara, Arad, Oradea and Cluj-Napoca). Analyzing this regression equation for the 93 urban-rank centres 0, I and II, yielded a 0.85 correla­tion coefficient which confirms the close correlation among these variables. This strong correlation is the result of the high hypertrophic character of the Capital-city (H index is 5.8) within the Romanian urban system. If Bucharest were excluded, this coefficient would be equal to 0.67, which shows a weaker, yet sig­nificant, correlation (R2=0.45). 3.3 The spatial impact of commercial investments on urban areas The first malls in Romania were opened by changing the destination of some buildings designed (in the years of the centralized economy) to host big food stores. In the second stage, the policy of centralized industrializationfocussedonthebuildingoflargeagro-foodandpublicfoodunitswithinindustrialareas, or on their boundary with dwelling quarters to enable the workers’ rapid access to these services and thus reduce meals-and-shopping breaks. In 1990, these constructions were abandoned; later on, they would be taken over by foreign investors who turned them into mall-type centres (all in Bucharest). Since urban developmentpolicieshadinviewdevelopingindustrialandresidentialareas(totakeinindustry-employed migration flows), locating big commercial and services units inside these areas to make up for a shortage of commercial and services facilities. Investments proved profitable and contributed to diminishing peo­ple’s migration from the city-centre to the periphery (Shertzer, Twinam and Walsh 2018). Figure 3: Typology of localizing commercial investments in Romania. p Legend: Inhabitants Count i es w here no (cens. 2011) A 1 new ma l was b ui l t 2 000 000 2 btwe n 1990 a nd 2016 350 000 300 000 U k r a i n e 200 000 3 B ALBA – County nameDorohoi100 000 50 000 430 000 Satu Mare Arad –County-seat nameRd ă u iăţ Sighetu Marma iei ţ Boto ş ani 4.1 Motru –City/Town name Suceava SATU MARE MARAMURE Ş BTO ANI Ş Carei 4.2CâmpulungMoldovenesc Ba ia Mare SUCEA VA F lticeni ă 5 Bis tria ţ IA ŞI S Ă LAJ Vatra Dornei Pa cani ş Commer cil Pa rks NEAM Ţ Dej Orad ea BISTRI Ţ A­ I ai ş Zal u ă Ma l ls Gherla ţ N Ă S Ă UD Topli a Piatra- Roman CLUJ Su perma rkets Neam ţ BIHOR Reghin C l u j- Acta geographica Slovenica, 60-1, 2020 BAC U Ă Bârlad F ocani ş Hu iş MURE Ş Npo c a Gheorgheni Vaslui Turda Târgu Bac ău HARGHITA Mure ş V A S L UI Odorheiu Mi rcure a ARAD Secuiesc Moine ti ş AiudTârn ă veni Ci uc ARAD ALBA Blaj Ar ad Al b a One ti ş Brad Sighişoara Iu li 57 aDe Mediaş COVASNA Târgu va BRA OV Ş Secuiesc Sebeş SIBIU T imi o ş ra Sfntuâ Tecuci VRANCEA Orăştie Gorghe F ă g ă raş Lugoj GALA ŢI Sibiu TIMI Ş Bra ş ov U k r a i n e S cele ă Hunedoara Gla i ţ HUNEDOARA R mnicu â S ă rat Petro ani ş CâmpulungCaransebeş BUZ ĂU PRAHOVA Br ă ila R mni c uâ Curtea de Reia ş ţ T rgu Ji u â Argeş V l c ea â T ulc e a CARA Ş -SEVERIN ARGE Ş B RILA Ă uz u ă Moreni Câmpina ÂLCEA V GORJ Motru Orşova Dr TULCEA Pl oet ş i ă g ăş ani P et i ş Trgo vi t e â ş ILFOV Urziceni Slob ozi a Dr ob e ta- IALOMI ŢA Fete ti T urn u Sv e rin ş B l a c k MEHEDIN ŢI S e a Slatina Bu cha r est Medgidia C ăă l ra ş i 3 CLRA I Ă Ş OLT GIUR Olteni a ţ DOLJ Ro iori de Vede ş CONS TAN A Ţ GIU Caracal C rai ova Constan ţa TELEORMAN Giurgiu Al Calafat e xandria Mangalia Turnu M gurele ă Content and map by: Radu S geatăă 0 20 40 60 km , © 2018 Romanian Academy, Institute of Geography B u l g a r i a Other big commercial investments were placed on farmland on the outskirts of cities, or on the ter­rainsofsomeformerindustrialunits(thathadbeendemolished),usingtheavailablerail-and-roadfacilities (Mirea2011).Herearesomeexamplesofformerindustrialzonesturnedcommercial:inBucharest(Mioriţa milk factory, Electric Machine Plant, Semănătoarea), Craiova (Electroputere), Timişoara (former slaugh­terhouse), Braşov (former rubber items and agricultural equipment factories), etc. There are many other situations of big investments localized on the territory of some communes that lieintheproximityoflargecities,wherelandpriceislower.Theadvantagesofferedbyeasyaccessaccount for the development of peri-urban commercial parks: in the proximity of Bucharest, on the motorways to Ploieşti, to Piteşti and to Urziceni; near Sibiu, on the motorway to Bucharest; close to Ploieşti, on the highway to Braşov; Constanţa (to Mangalia); Braşov (to Ploieşti and Bucharest); Galaţi (to Brăila); Piteşti (on the motorway to Bucharest), etc. Another tendency is to update the big commercial units, built before 1989 in county-seats, and turn them into malls. For example, the big universal store Unirea in Bucharest, updated and turned into mall-type centre. In some 1st and 2nd-rank cities, they built commercial units in the central parts of the city, or of big residential areas (Figure 3). Some commercial complexes are located in the proximity of large student camps. In Bucharest, a typ­icalexampleisCarrefourOrhideeasitedintheclosevicinityoftheRegie-Grozăveştistudentcamp;similar examples in Cluj-Napoca and Iaşi. The same location pattern holds for entertainment services comple­mentary to the shopping ones. 3.4 The social impact of commercial investments. The Romanian city from traditional to cosmopolitan Whetherlocatedontheoutskirtsoflargecities,orwithintheircentralsections,ofgreatdemographiccon-centration,theimpactofbigcommercialinvestmentsisworthconsidering.The»mall«concepttranscends a strictly commercial function, it being associated with luxury and fashionable; one goes to the mall to seenewtrends,tosocialize,tobecomeacquaintedwithacertainsegmentofthepopulation(Skivko2016). So, financial segregation goes hand-in-hand with social segregation, both being embodied in commercial segregation, and the best way of expressing it is going to the mall. No wonder, therefore, that in some big commercial centers, commercial facilities are associated with social and business facilities (in Timişoara, Bucharest, etc.). Apartfromfashiongarments,cosmetics,orelectricitems,mallshavefast-foodrestaurants,coffee-bars, cinemas, casinos, entertainment areas, destined to disseminate global consumer products within the Romanian urban society. Looking at the number of visitors going to these places, the social function of malls is quite obvious. At the same time, the global character of this commercial type is reflected in the countriestheseproductsoriginatefrom,theybeingsoldbyover90%ofthecommercialnetworkinRomania. Incomes are concentrated in Bucharest and in the large cities; this determined specialist goods and servicestobeconcentratedinbigcommercialcomplexes.Theirlocationappearstobeconnectedwithever-greater social segregation, tending to delimit the areas where such goods and services can be distributed, that is, areas in which a certain type of users can be found. Thus, a new kind of urban-rural polarization emerged, directly proportional to social and cultural segregation, which depends on the Romanian urban system’s capacity to take in globalizing flows. The deep-going financial disparities among the urban population, caused by industrial restructuring, filter the penetration of globalizing flows. On the other hand, the global consumerist culture, combined with endemic poverty-induced lack of culture, is a fertile ground for the resurgence of urban subcultures andorganizedcrime.ThisphenomenonisincreasinglyaffectingRomania’stowns,andespeciallythelarge cities (Ban 2014). 4 Conclusion Developing the commercial services sector has become one of the main characteristics of the Romanian economy after 1989, in line with evolution trends in the global economy. Changes in the urban functional zone have been made as commercial and services areas kept replacing the industrial ones. The density of modern commercial areas goes hand-in-hand with the financial-banking investments, and is directly proportional to the population’s state of well-being. In Romania, the commercial area out­side Capital-city is 2,43 million m2(density: 184 m2/1,000 inh.) and of 1.19 million m2 in Bucharest City (651m2/1,000 inh.) (C&W Echinox 2018). Bigcommercialinvestmentshavebeenmadeonthesiteofformerindustrialunitsthathadbeendemol­ished (Cercleux, Peptenatu and Merciu 2015), or on some former production spaces, former farm lands located on the outskirts of large cities, or in peri-urban zones (using existing road access), in some resi­dential areas short of services facilities, or in the proximity of some student camps. TheglobaleconomicandfinancialcrisishadaseriousimpactoncommercialinvestmentsinRomania, theinvestment-marketbecomingsaturatedaspeople’spurchasingpowerdropped.Someprojectswereaban­doned for lack of investors (e.g. in Craiova, Buzău, Constanţa and Bucharest), others were turned into office-rooms, or residential areas. Globalization, through big investments in commercial and agreement areas has materialized in that over90%areheldbyinternationalstorechainsinRomania’scommercialstructure.Thiscountryisamong the European states with the fastest economic dynamics in this field. 5 References Ban, C. 2014: Beyond anticomunism. The fragility of class analysis in Romania. East European Politics and Societies 29-3. DOI: https://doi.org/10.1177/0888325415599197 Bezin, E., Moizeau, F. 2017: Cultural dynamics, social mobility and urban segregation. Journal of Urban Economics 99. DOI: https://doi.org/10.1016/j.jue.2017.02.004 Brülhart,M.,Carrere,C.,Robert-Nicoud,F. 2018:Tradeandtowns:Heterogeneousadjustmenttoaborder shock. Journal of Urban Economics 105. DOI: https://doi.org/10.1016/j.jue.2017.09.005 Bürkner,H-J.,Totelecan,S.G.2018:Assemblagesofurbanleisurecultureininner-cityBucharest.International Journal of Urban and Regional Research 42-5. DOI: https://doi.org/10.1111/1468-2427.12630 Cercleux, A-L., Peptenatu, D., Merciu, F-C. 2015: Structural dynamics of tertiary activities in industrial parksinBucharest,Romania.ActaGeographicaSlovenica55-2.DOI:http://doi.org/10.3986/AGS.753 Cohen,J-N.2016:ThemythofAmerica’s»cultureofconsumerism«:PolicymayhelpdriveAmericanhouse­hold’sfrayingfinances.JournalofConsumerCulture16-2.DOI:https://doi.org/10.1177/1469540514528196 Crescenzi, R., Pietrobelli, C., Rabellotti, R. 2016: Regional strategic assets and the location strategies of emerging countries’ multinationals in Europe. European Planning Studies 24-4. DOI: https://doi.org/ 10.1080/09654313.2015.1129395 C&W Echinox 2018. Internet: https://www.agendaconstructiilor.ro/files/oportunitati-comunicate/Cresc livrările de spaţii comerciale până la sfârşitul anului 2018 (16. 1. 2019) Demetry,D.,Thurk,J.,Fine,G.A.2015:Strategicpoverty:Howsocialandculturalcapitalshapeslow-income life. Journal of Consumer Culture 15-1. DOI: https://doi.org/10.1177/1469540513493205 Dumitrescu,B.2008:OraşelemonoindustrialedinRomâniaîntreindustrializareforţatăşideclineconomic. Bucharest. Goodman, J., Marshall, J. P. 2013: Crisis, movement and management in contemporary globalizations. Globalizations 10-3. DOI: https://doi.org/10.1080/14747731.2013.793896 Holmén,H.1997:Limitstoglobalization.EuropeanReview5-1.DOI:https://doi.org/10.1002/(SICI)1234­ 981X(199701)5:1%3C75::AID-EURO151%3E3.0.CO;2-W Ianoş, I., Heller, W. 2006: Spaţiu, economie şi sisteme de aşezări. Bucharest. Ianoş, I., Pumain, D., Racine, J. B. (eds.) 2000: Integrated urban systems and sustainability of urban life. Bucharest. Isaksen,A.,Jakobsen,S-E.2017:Newpathdevelopmentbetweeninnovationsystemsandindividualactors. European Planning Studies 25-3. DOI: https://doi.org/10.1080/09654313.2016.1268570 Jakoby,W.,Korkut,W.2015:Vulnerabilityandeconomicre-orientation.RhetoricandrealityinHungary’s »Chinese Opening«. East-European Politics and Societies 30-3. DOI: http://doi.org/10.1177/ 0888325415604355 McFarlane,C.2006:Transnationaldevelopmentnetworks:bringingdevelopmentandpostcolonialapproaches intodialogue.TheGeographicalJournal172-1.DOI:https://doi.org/10.1111/j.1475-4959.2006.00178.x Mermet, A-C. 2017: Global retail capital and the city: towards an intensification of gentrification. Urban Geography 38-8. DOI: https://doi.org/10.1080/02723638.2016.1200328 Mirea, D-A. 2011: Industrial landscape – a landscape in transition in the Municipality Area of Bucharest. Geographical Phorum – Geographical Studies and Environment Protection Research X-2. DOI: https://doi.org/10.5775/fg.2067-4635.2011.014.d Mitrică, B., Grigorescu, I., Urucu, V 2016: Dezvoltarea urbană şi ariile metropolitane. România – natură şi societate, Bucharest. Mitrică, B., Mocanu, I., Dumitraşcu, M., Grigorescu, I. 2017: Socio-economic disparities in the devel­opment of Romania’s border areas. Social Indicators Research 134-3. DOI: https://doi.org/10.1007/ s11205-016-1462-7 Novotný, L. 2016: Urban development and migration processes in the urban region of Bratislava from the post-socialist transformation until the global economic crisis. Urban Geography 37-7. DOI: https://doi.org/10.1080/02723638.2016.1139413 Paugam,S.1998:Lesformescontemporainesdelapauvretéetdel’exclusion.Lepointdevuesociologique. Geneses. Femme, Famille, Individu 31. DOI: https://doi.org/10.3406/genes.1998.1516 Picard, P. M., Zenou, Y. 2018: Urban spatial structure, employment and social ties. Journal of Urban Economics 104. DOI: https://doi.org/10.1016/j.jue.2018.01.004 Rita Sedita, S., De Noni, I., Pilotti, L. 2017: Out of the crisis: an empirical investigation of place-specific determinants of economic resilience. European Planning Studies 25-2. DOI: https://doi.org/10.1080/ 09654313.2016.1261804 Shertzer,A.,Twinam,T.,Walsh,R.P.2018:Zoningandtheeconomicgeographyofcities.JournalofUrban Economics 10. DOI: https://doi.org/10.1016/j.jue.2018.01.006 Skivko, M. 2016: Touring the fashion: Branding the city. Journal of Consumer Culture 16-2. DOI: https://doi.org/10.1177/1469540516635806 Stead, D., de Vries, J., Tasan-Kok, T. 2015: Planning cultures and histories: Influences on the evolution of planning system and spatial development patterns. European Planning Studies 23-11. DOI: https://doi.org/10.1080/09654313.2015.1016402 vanHam,M.,Tammaru,T.2016:Newperspectivesonethnicsegregationovertimeandspace.Adomains approach. Urban Geography 37-7. DOI: https://doi.org/10.1080/02723638.2016.1142152 Wikström,S.,Jönsson, H.,L’EspoirDecosta,P.2016:Aclashofmodernities:Developinganewvalue-based framework to understand the mismatch between production and consumption. Journal of Consumer Culture 16-3. DOI: https://doi.org/10.1177/1469540514528197 APPLICATION OF A MASS MOVEMENT SUSCEPTIBILITY MODEL IN THE HETEROGENEOUS MIOCENE CLASTIC SUCCESSIONS OF THE SLOVENJ GRADECBASIN,NORTHEASTSLOVENIA Kristina Ivančič, Jernej Jež, Blaž Milanič, Špela Kumelj, Andrej Šmuc The hilly landscape between Slovenj Gradec and Velenje consists of soft sedimentary rock, which is susceptible to landslide formation. DOI: https://doi.org/10.3986/AGS.7040 UDC: 551.435.62:528.94(497.413) COBISS: 1.01 Kristina Ivančič1, Jernej Jež1, Blaž Milanič1, Špela Kumelj1, Andrej Šmuc2 Application of a mass movement susceptibility model in the heterogeneous Miocene clastic successions of the Slovenj Gradec Basin, northeast Slovenia ABSTRACT: In Slovenia, mass movements are not only a threat to the population, but also a major envi­ronmentalandsocialsciencechallenge.Lithologicallyheterogeneousareashavebeenfoundtobeproblematic, and the Miocene Slovenj Gradec basin (in northeast Slovenia) is one such area. For this area, we devel­opedlandslideandrockfallsusceptibilitymapsbasedondetailedgeologicalresearchcombinedwithstatistical modeling schemes. Crucial factors include lithological composition, land use, geological structural ele­ments, slope curvature, aspect and inclination, and bed dipping. The approach taken in the development of mass movement susceptibility maps presented here is transferable to other areas defined by heteroge­neouslithology.Suchmapscouldproveusefulspatialplanning,forestry,environmentalprotection,landscape architecture, and other fields. KEY WORDS: Miocene, landslides, rockfalls, mass movement process modeling, heterogeneous lithology, Slovenia Model nevarnosti za pobočne procese na primeru heterogenih miocenskih zaporedij klastičnih kamnin v slovenjgraški kotlini v severovzhodni Sloveniji POVZETEK: V Sloveniji pobočni procesi ogrožajo prebivalstvo, hkrati pa so tudi velik okoljski in druž­boslovni izziv. S tega vidika so se za problematična izkazala litološko heterogena območja, med katerimi jetudimiocenskaslovenjgraškakotlinavseverovzhodniSloveniji.Zatoobmočjesoavtorjinapodlagipodrob­nihgeološkihraziskavvkombinacijisshemamistatističnegamodeliranjaoblikovalizemljevidenevarnosti za zemeljske plazove in skalne podore. Glavni dejavniki, ki vplivajo na oblikovanje tovrstnih procesov, so litološkasestava,rabatal,geološkestrukturneprvine,ukrivljenostpobočja,ekspozicija,nakloninusmer­jenost kamninskih plasti. Predstavljeni pristop k oblikovanju zemljevidov nevarnosti za pobočne procese jeprenosljivnadrugalitološkoheterogenaobmočja.Tovrstnizemljevidibibililahkouporabnizapodročja, kot so prostorsko načrtovanje, gozdarstvo, okoljevarstvo in krajinska arhitektura. KLJUČNE BESEDE: miocen, zemeljski plazovi, skalni podori, modeliranje pobočnih procesov, litološka heterogenost, Slovenija The paper was submitted for publication on October 9th, 2018. Uredništvo je prejelo prispevek 9. oktobra 2018. 1 Geological Survey of Slovenia, Ljubljana, Slovenia kristina.ivancic@geo-zs.si, jernej.jez@geo-zs.si, blaz.milanic@geo-zs.si, spela.kumelj@geo-zs.si 2 University of Ljubljana, Faculty of Natural Sciences and Engineering, Department of Geology, Ljubljana, Slovenia andrej.smuc@geo.ntf.uni-lj.si 1 Introduction MassmovementsareaverycommongeologicalphenomenoninSlovenia(Mikoš,BrillyandRibičič2004; Zornand Komac2008;Mikoš and Majes 2010; Zorn, Komacand Kumelj2012). Theiroccurrenceis related to diverse geological and tectonic structures, relief, and land use. They have caused considerable damage and occasionally threatened or even claimed human lives (Mikoš 2000/2001). In recent decades, geolog­icalphenomenaincludinglandslides,debrisflows,androckfallshavebeenintensivelyinvestigatedinSlovenia (Mikošetal.2006;Mikoš,FazarincandMajes2007;ZornandKomac2002,2007,2008,2011;JemecAuflič et al. 2017). Preventive measures have been identified as an important basis for avoiding damage and loss caused by mass movements and helping solve socioeconomic challenges. Among these measures, land­slide susceptibility maps are a useful tool for minimizing potential hazard (Komac and Jež 2018) in terms of appropriate spatial planning based on the results of the investigations. Many susceptibility models for landslide zonation (Carrara et al. 1991, 1995; Guzzetti et al. 1999, 2006; Zorn and Komac 2007; Rossi et al.2010;Petschkoetal.2014)androckfallsusceptibilityzonation(Pannatieretal.2009;Shirzadietal.2012; Böhme, Derron and Jaboyedoff 2014) have been used around the world. The methodology used in our research was developed as part of research projects at the Geological Survey of Slovenia (Bavec, Budkovič and Komac 2005; Komac and Jež 2018). It was developed using a linear model of weighted spatial factors testedbyaunivariatechi-square(.2)statisticalmethod,whichhasalreadybeenutilizedbyseveralauthors (Stančič and Veljanovski 1998, 2000a, 2000b; Veljanovski 1999; Komac 2005b). Initially, the methodolo­gy was used at smaller scales at the national level (Komac and Ribičič 2006), and later it was also applied tolargerscales;thatis,atthemunicipalandlocallevels.Inordertoverifythetransferabilityofthemethod­ology to other environments (also outside of Slovenia), it was tested on the example of the Municipality of Zvornik (Republika Srpska, Bosnia and Herzegovina), where it proved successful (Kumelj et al. 2014). The methodwaspreviouslyusedforlandslideandrockfallsusceptibilityzonationinthewiderstudyareaatascale of1:25,000.Itturnedoutthatapproximately70%oftheareaofthemunicipalitiesofSlovenjGradecandVelenje are exposed toa medium, high, andvery high occurrence of mass movement (Bavec et al. 2012a; 2012b). Inpreparingthegeologicalmodelinginputdata,themostproblematicareaswerefoundtobeareaswith lithologicallyheterogeneoussequences.Insuchareas,rockswithdifferentgeomechanicalpropertiesalter­natefrequentlywithinveryshortdistances.Suchrocks includeMioceneandPermian–Carboniferousclastic sedimentarysuccessions,whicharemostcommonlyfoundincentralandnortheasternSlovenia(Figure1A). According to the Landslide Susceptibility Map of Slovenia, these are the most exposed hazardous land­slide areas (Figure 1B). The problem of common occurrences of landslides in Neogene rocks in the wider Pannonian Basin was also addressed by Tošić et al. (2014). This paper describes a possible approach in geologically diverse areas where detailed geological map­pinganddetailedgeologicalprofileloggingconstitutethemainanalyticaltools.Theyprovidemoreprecise geological input data required for the model. The main purpose is to improve maps so that they are pre­cise for large scales (e.g., 1:5,000), on which the susceptibility zonation fora specific location can be seen. The area between Podgorje (in the Municipality of Slovenj Gradec) and Gaberke was appropriate for our research because it is lithologically and morphologically heterogeneous with different land-use charac­teristics. The approach can be transferred to other areas with similar rock successions. 2 Study area The investigated area is located between Podgorje (in the Municipality of Slovenj Gradec) and Gaberke innortheasternSlovenia,anditcovers11km2(Figures1A,2).TheareastudiedispartoftheAlpinemacrore­gion and corresponds to the Eastern Karawanks, the Velenje and Konjice Hills, and the Strojna, Kozjak, andPohorjemountains(Perko1998).Paleogeographically,theinvestigatedareaispartoftheSlovenjGradec Basin and consists of alternating Miocene clastic sedimentary rocks. The terrain in the study area is morphologically diverse. Only 3.6% of the study area has a slope less than 5°. In areas with steeper slopes, landslides could occur. A slope steeper than 38° is found in 17.3% of the area and rockfalls can occur (Komac 2005a). Rarely, conglomerate and sandstone beds can even form overhanging walls. The highest point of the area reaches 825m, and the lowest is at 425m. The morpho­logical diversity of the area is conditioned by the geological diversity there and, consequently, a branched A B Scale: 1:2,000,000 Content and map by: Kristina Ivančič Source: Buser 2009 (A); Komac 2005 (B) © 2018, Geological Survey of Slovenia 0 12.5 25 50 km Miocene rocks Carboniferous and Permian rocks Study area Landslide susceptibility map [A - Distribution of a landslide susceptibility rate area for Miocene, Carboniferous and Permian rocks] None [A: 10.9%] Very low [A: 2.0%] Low [A: 7.7%] Moderate [A: 11.7%] High [A: 41.5%] Very high [A: 26.2%] Figure 1: A) Areas with Miocene and Permian–Carboniferous clastic rocks, B) landslide susceptibility map of Slovenia (Komac and Ribičič 2006). 64 Figure 2: Topographic map with the location of the investigated area (red). river system of the torrent type. The main valleys and ridges formed in a northwest–southeast direction and are partly conditioned by the tectonic structure present there. The investigated area is faulted and folded. Faults and fault zones are more frequent in the southern partofthearea.TheybelongtothePeriadriaticfaultzone,whichseparatestonalitefromclasticsedimentary rocks. Smaller-scale faults are present in other parts of the study area. In the entire investigated area, the rocks are fractured. The most prominent fractures are in the southern part of the area, occurring mainly in conglomerate (Figure 3A) and sandstone, and they are subordinate in siltstone. In the area there are also two synclines and an anticline with a northwest–southeast axis orientation. The folding has resulted in variable dip of the beds, which ranges from 310–30/20–60 to 190–225/30–50. 3 Methods Landslide and rockfall susceptibility models were developed for the area between Podgorje and Gaberke. The study is based on two primary types of data collection and processing schemes: a) geological map­ping and sedimentological analyses, and b) the preparation of input data and statistical modelling. 3.1 Geological methods Detailedgeologicalmappingofrockoutcropswasperformedatascaleof1:5,000inordertoobtainadetailed lithological map with geomechanical properties. In addition, tectonically fractured rocks, the location of faults and fault zones, and bed dip were evidenced, and a special focus was placed on existing landslides. Seven lithological sections present in different parts of the Slovenj Gradec Basin succession were record­ed at a scale of 1:100 in order to precisely determine the type and properties of rocks (Figure 4A). Thelithologicalunitswerereclassifiedintosixsusceptibilitycategoriesaccordingtolandslideandrock­fallsusceptibility(Bavec,BudkovičandKomac2005;KomacandJež2018),whereCategory1showsareas withnooccurrence(areaunder5°)andCategory6showsareasthatareverypronetomassmovementevent occurrence. 3.2 Input data and statistical modeling Twoexistingconceptualmodels(amethodologyforestimationofgeohazardinducedbymassmovements; Komac 2003a, 2005a; Bavec, Budkovič and Komac 2005) were used to determine landslide and rockfall susceptibilityrates.ThemethodisexplainedinKomac(2003a,2003b,2005a)andKomacandRibičič(2006). It was developed using a linear model of weighted spatial factors. Univariate statistical methods (the chi-square method) were used to test the influences of individual spatial factors on landsliding, and multivariate statistical methods were used to test the importance of individual factors in landslide occur­rence. Komac (2005a) developed 3,142 models for landslide susceptibility and 7,674 for rockfall susceptibility using the Monte Carlo method. For the landslide model, lithological data are combined lin­early with the synchronicity of the bed dips and the slope aspect, applying a weighting ratio of 0.8 for the lithologicalcompositionand0.2forthedipofthebeds.Inaddition,land-usefactors,distancesfromstruc­turalelementsandfaultzones,slopeaspect,andcurvaturewereallincluded.Preconditionspatial-temporal factors for landslide occurrence are lithology 0.3, slope inclination 0.25, landcover type 0.25, slope cur­vature 0.1, distance to structural elements 0.05, and slope aspect 0.05 (Figure 5A). For the rockfall model, lithologicalcompositionandfaultzoneparameterswereweightedasfollows:0.5forslopeinclination,0.35 synchronicity of dip of beds, and 0.15 for slope aspect weight (Figure 5B). In general, the methodology of both models at a scale of 1:25,000 is based on four consecutive phas­es (Bavec, Budkovič and Komac 2005; Komac 2005a; Komac and Jež 2018): 1) synthesis of the existing cartographic archival geological data and verification fieldwork examinations in the phase map of geo­logicalhazardsduetoslopemassmovements;2)theproductionofaprobabilitystatisticalmodelofgeological Figure 4: A) Lithological map of the investigated area with structural elements (faults). The red circle marks the location of the Gaberke section; B) the Gaberke section, where frequent alternation of different rock types with different geomechanical properties can be observed. p Figure 5: A) Input parameters for the landslide susceptibility model, B) input parameters for the rockfall susceptibility model. hazards due to slope mass movements; 3) production of maps of geological hazards due to processes of slope mass movement based on the synthesis of Phase Map 1 and Probability Model 2 by exclusion of the mostproblematicareas,and4)detailedmappingofmoreproblematicareasatascaleof1:5,000or1:10,000 and production of a detailed geological hazards map due to slope mass movements for these areas. In this paper, the modeling is based exclusively on new geological data, and therefore Phase 1 related to archival geological data is not included. The methodology starts with the process of Phase 4. Data on slope inclination, aspect, and curvature were derived from the digital relief model produced onthebasisofLiDARdataataspatialresolutionof1m(MOP2018).Areasunder5°forlandslidesandareas under 38° for rockfall are considered areas where the probability of landslide or rockfall is essentially zero, and they were therefore not included in the model (cf. Komac 2005a). The values of the calculated models were classified into six landslide and rockfall susceptibility cate­goriesandwerearrangedaccordingtochangesinthedistributionoffrequencyoccurrence(NaturalBreaks– Jenks; natural boundary method). Class breaks are identified that best group similar values and that max-imizethedifferencesbetweenclasses.Thefeaturesaredividedintoclasseswhoseboundariesaresetwhere there are relatively large differences in the data values (Esri 2006). All landslides in the area studied from the national database (Komac and Hribernik 2015) were ver­ified in the field, and new landslides identified during fieldwork were added to the database. This served as a basis for model validation. If the input parameters were identified as incorrect in the field (e.g., lithol­ogy or land use), we modified them accordingly in the model and ran a new one. 4 Results 4.1 Lithological map The investigated area is composed of tonalite and alternation of conglomerate, sandstone, siltstone,marl­stone,andclaystone(Figure4A).Tonaliteisnotpartoftheclasticsedimentarysuccession,anditistherefore not involved in our research. Conglomerate and sandstone predominate and were found throughout the entire investigated area. They frequently occur together, meaning that layers of sandstone alternate with conglomerate layers(Figure3B). Ingeneral,morecoarse-grainedlithologiespredominate inthe northern part of the investigated area, whereas fine-grained intercalations and packages are more common in its southern part. 4.2 Detailed description of lithological units Basedondetailedfieldwork(Ivančičetal.2018),alllithologicalunitsweredefinedandtheirgeomechanical properties were evaluated. The frequent alternation of different lithological units could be well observed in the Gaberke section, where conglomerates, sandstone, and silty marlstone occur (Figure 4B). Fine- to coarse-grained conglomerate is grain-supported, thin- to thick-bedded, or massive in places. The conglomerate is very well to poorly lithified, and sometimes fractured. It is present throughout the entire investigated area, and it is common in hilltops and ridges. Fine-tocoarse-grainedsandstoneiswelltopoorlylithified.Thegrainsarebondedwithcarbonateand quartz cement, with a carbonate matrix. Sandstone was found throughout the investigated area, mostly in combination with conglomerate layers. Siltstone occurs locally throughout the entire mapped area. It is laminated and bedded, and contains plant remains (Figure 3C). Although silty layers are usually present only in thin layersbetween sandstone and conglomerate beds, locally successions can be up to 32m thick. Marlstoneandsiltymarlstonearefrequentlyfoundonthesouthernpartofthemap,occurringincom­bination with conglomerate and sandstone. Marlstone is laminated and usually shaly, which contributes to its rapid weathering and poor geomechanical characteristics. Claystoneismostlyfoundinthenorthernpartoftheinvestigatedarea,occurringasthinlayersincom­bination with conglomerate and sandstone. Retained water on the surface suggests impermeability of the unit (Figure 3D). 4.3 Model The values of the final model (between 0 and 1) were classified into six landslide or rockfall susceptibili­ty classes. The method determines the boundaries between groups of data that exhibit relatively large differences between pairs of adjacent values. The classes are not equally distributed (Figures 8 and 9). 4.3.1 Landslide susceptibility model The landslide susceptibility model indicatesthat19.9% of the investigated area exhibits a very high prob-abilityoflandslideoccurrence.Amoderatelyhighprobabilityisexhibitedby22.7%ofthearea,and44.3% has a relatively low to a very low probability of landslide occurrence. The greatest landslide hazard areas are Plešivec (Figure 6) and Vodriž (Figure 7). The areas coincide with the real state of nature. The very highlandslidesusceptibilityclassthatcharacterizesthePlešivecareacoincideswithalreadyidentifiedland­slidehazardareas.Theserelativelyvastunstableareasaremainlylimitedtomeadowsandpastures,whereas inforeststheyarenotcommon.Generallyspeaking,thosemoresusceptibleareascoincidewithfine-grained sedimentary rocks (Figure 4A), such as siltstone, marlstone, and claystone. These locationsare presented in Figure 8. 4.3.2 Rockfall susceptibility model The rockfall susceptibility model indicates that 22.1% of the investigated area belongs to the class of very high susceptibility, and 28.6% of the area to moderate susceptibility. Locations where rockfall occurrence has been documented are rare. The highest possibility for their occurrence is on the northwest part, the central part, and the southwest part of the investigated area (Figure 9). The area is mostly composed of a conglomerate, subordinate sandstone. It is bound to vertical walls in places. The result presented by the modelcoincideswiththerealsituationinnature.ThelocationsofthesusceptibleareasaremarkedinFigure9. 5 Discussion With this approach, the mass movement susceptibility models of the rockfall and landslides provedto be very precise and gave a very good approximation of the natural state. This was proved by verification. The areas of highest landslides and rockfall susceptibility coincide with the actual state of nature even in the study area between Podogrje and Gaberke, where the lithology frequently alternates. Very high and high landslidesusceptibilityinthestudyareaturnsouttobeonthemeadowsandpastureswithsiltstone,marl-stone, or claystone as bedrock. Landslides from the national database and new landslides are located in the high and very high susceptibility areas. Moreover, new landslides were determined during the verifi­cation process. They are located in areas of high and very high probability for landslide occurrence. The model did not specify the correct landslide susceptibility zonation in the case of locally changed land use or in the case of newly changed land use that is not included in the modeling process. One specific land­slide occurs in the area of deforestation (Figure 7C). In this case, the bedrock is composed of alternating sandstone and conglomerate. Rockfalls in the study area are subjected to the lithology of conglomerate with rare sandstone layers. Manydifferentmodelswereusedinthepast(Carraraetal.1991,1995;Guzzettietal.1999,2006;Zorn and Komac 2004, 2007; Rossi et al. 2010; Reichenbach et al. 2018) to produce mass movement suscepti­bility zonation. Such models usually do not precisely define susceptibility to slope mass movements in lithological heterogeneous areas. The problem of heterogeneous lithology has already been discussed Figure 6: Examples of larger labile areas in the Plešivec region. A) The area below the Koližnik farm, B) the area near the Grah farmhouse, C) the area at the Grabnar farm. The locations of areas 6A to 6C are also indicated in Figure 8. p p. 71 Figure 7: Examples of individual landslides in the investigated area. A) Landslide at the Petelanšek farm, B) cleanup of the landslide along the road toward the Pelc farm, C) newly formed landslide on the freshly deforested section along Vodriž Creek, with marked main scarp. The locations of areas 7A to 7C are also indicated in Figure 8. p p. 72 Figure8:Landslidesusceptibilitymapofthe investigatedarea.Points6A,6B,6C,7A,7B,and7Cmarkthelocationsofthephotos in Figures6and7. p p.73 Figure 9: Rockfall susceptibility map of the investigated area. p p. 74 73 74 (Lee et al. 2008; Zorn and Komac 2009; Blahut, Van Westen and Sterlacchini 2010; Petschko et al. 2014). Thegreatestdifferenceinourmodelisinitsinputfactors,whicharemorepreciselydefinedforslope(cur­vature,inclination,andaspect)andstrata(dippingandaspect).Moreover,detailedmappingimprovesthe input data of lithology and tectonic elements, and therefore the final landslide and rockfall susceptibility zonationismoredetailedandusableatlargerscales(e.g.,1:5,000).Themodelprovidesaccuracywithrespect to the worst-scale input parameter, but nevertheless more accurate data (such as LiDAR) show charac­teristic features that must be taken into account when interpreting the results. Geological maps at smaller scales (e.g., 1:100,000 or 1:25,000) usually do not sufficiently separate or differentiate lithologically het­erogeneousunits.Consequently,suchmapsareoflimiteduseformassmovementsusceptibilitymodeling. The quality of the final predictions of the formation of slope mass movements primarily depends on the quality of the input geological data, and therefore it is necessary to combine the existing methodolo­gy of the modeling (Komac, Kumelj and Krajnik 2012; Bavec et al. 2012a, 2012b) with classic techniques of geological research (e.g., geological mapping and detailed recording of lithological sections). Only in this way can we obtain quality data, which, in combination with the geomorphological parameters of the terrain, make a significant contribution to the production of a useful final model of susceptibility to slope mass movements. Todaypubliclyavailablereliefdata(i.e.,LiDARdatasusceptibility)aresignificantlymoreaccuratethan that used by existing lithological maps, and this therefore contributes to more detailed geological field­work. In addition, current land-use data are also very important in the modeling process. Statistical data indicate that shallow landslides frequently occur in areas used as meadows, pastures, orchards, and vine-yards(Komac2005a).Certainfactorsrelatedtochangesinlanduse,suchasdeforestationandthecultivation ofpasturesandorchards,playamajorroleinthedestabilizationoflabileareas(Fidejetal.2018).Consequently, agricultural areas located in hazardous areas are often affected. Examples of such phenomena were also observed and documented in the investigated area. Determining the weighting ratio of the impact of the individual land use class (forest, vineyard, etc.) proves a particular challenge within the basic input fac­tor: land use. In addition to activities related to agriculture and forestry, the data in the model are indispensable in allspatialplanningprocessesintheenvironment.Regionalandlocalspatialplanningandlanduseshould beadjustedaccordingtosuchmodels.Itisworthmentioningthatthemodelsindicatesourceareasofpoten­tial mass movements, not their transport paths or deposition areas. As a rule, landslide deposition areas aregenerallynotfarfromthelandslidesource,withtheexceptionofcasesinwhichlargequantitiesofwater are present, and the material can be converted into a mass (debris) flow. In the case of rockfalls, the mate­rial (blocks) may be deposited far from the source site (Zorn 2002). 6 Conclusion Existing geological and pedological maps are not sufficiently accurate to produce quality detailed mod-els,makingitnecessarytoincludeprecisefieldmappingandotherbasicgeologicalresearchintheprocess. This is absolutely suggested in lithologically heterogeneous areas. A good example of this is thestudy area between Podgorje and Gaberke, where more than 40% of the area exhibits a very high, high, or moder­ate probability of landslide occurrence. Alongside lithology, land use as model input factor (e.g., meadows and pastures, and deforestation) wasfoundtobeveryimportantinthemodelingprocess.Inaddition,changesinlandusemayplayamajor role in the destabilization of labile areas. The susceptibility models presented here are sufficiently accurate and have been verified by checking known slope movement events in the investigated area. The approach used is transferable and compara­ble at all susceptibility levels, and it can be used in lithologically heterogeneous areas for large-scale maps (1:5,000). Theresultscanbeusedforspatialplanningintheenvironment.Byconsideringsuchmodels,themost hazard-prone areas can be avoided or can be dealt with in a geotechnically professional manner. ACKNOWLEDGEMENT:ThisstudywasfundedbytheSlovenianResearchAgency(ARRS)intheframe­workoftheYoungResearchersprogram,theGroundwatersandGeochemistryresearchprogram(P1-0020), andaspartofresearchcorefundingno.P1-0011(regionalgeology),whichiscarriedoutattheGeologicalSurveyofSlovenia.WewouldliketothankMladenŠtumergarforpreparationofsamplesforpetrographic analysis. The authors are also grateful to Dragomir Skaberne and Matevž Novak for their generous help and support. 7 References Bavec,M.,Budkovič,T.,Komac,M.2005:Geohazard–geološkopogojenanevarnostzaradiprocesovpobočnega premikanja. Primer občine Bovec. Geologija 48-2. DOI: https://doi.org/10.5474/geologija.2005.025 Bavec, M., Čarman, M., Krivic, M., Kumelj, Š., Požar, M., Komac, M., Šinigoj, J., Rižnar, I., Jurkovšek, B.,Trajanova,M.,Poljak,M.,Celarc,B.,Demšar,M.,Milanič,B.,Mahne,M.,Otrin,J.,Čertalič,S.,Štih, J., Hrvatin, M. 2012a: Izdelava prostorske baze podatkov in spletnega informacijskega sistema geološko pogojenih nevarnosti zaradi procesov pobočnega premikanja, erozijskih kart ter kart snežnih plazov, Občina Velenje. Geološki zavod Slovenije. Ljubljana. Bavec, M., Čarman, M., Krivic, M., Kumelj, Š., Požar, M., Komac, M., Šinigoj,J., Rižnar, I., Jurkovšek, B.,Trajanova,M.,Poljak,M.,Celarc,B.,Demšar,M.,Milanič,B.,Mahne,M.,Otrin,J.,Čertalič,S.,Štih, J., Hrvatin, M. 2012b: Izdelava prostorske baze podatkov in spletnega informacijskega sistema geološko pogojenih nevarnosti zaradi procesov pobočnega premikanja, erozijskih kart ter kart snežnih plazov, Občina Slovenj Gradec. Geološki zavod Slovenije. Ljubljana. Blahut, J., Van Westen, C. J., Sterlacchini, S. 2010: Analysis of landslide inventories for accurate prediction ofdebris-flowsourceareas.Geomorphology119,1-2.DOI:https://doi.org/10.1016/j.geomorph.2010.02.017 Böhme,M.,Derron,M.-H.,Jaboyedoff,M.2014:Quantitativespatialanalysisofrockfallsfromroadinven­tories:acombinedstatisticalandphysicalsusceptibilitymodel.NaturalHazardsandEarthSystemSciences Discussions 2. DOI: https://doi.org/10.5194/nhessd-2-81-2014 Buser, S. 2009: Geološka karta Slovenije 1 : 250.000. Geološki zavod Slovenije. Ljubljana. Carrara, A., Cardinali, M., Detti, R., Guzzetti, F., Pasqui, V., Reichenbach, P. 1991: GIS Techniques and statistical models in evaluating landslide hazard. Earth Surface Processes and Landform 16-5. DOI: https://doi.org/10.1002/esp.3290160505 Carrara,A.,Cardinali,M.,Guzzetti,F.,Reichenbach,P. 1995:GIStechnologyinmapping landslidehazard. Geographical Information Systems in Assessing Natural Hazards. Dordrecht. DOI: https://doi.org/ 10.1007/978-94-015-8404-3_8 Esri. ArcGIS Desktop Help. ESRI, 2006.Fidej, G., Mikoš, M., Jež, J., Kumelj, Š., Diaci, J. 2018: Presoja varovalnega učinka gozda pred drobirskimi tokovi. Gozdarski vestnik 76-4. Guzzetti,F.,Carrara,A.,Cardinali,M.,Reichenbach,P.1999:Landslidehazardevaluation:areviewofcur­renttechniquesandtheirapplicationinamulti-scalestudy,CentralItaly.Geomorphology31,1-4.DOI: https://doi.org/10.1016/S0169-555X(99)00078-1 Guzzetti F., Reichenbach, P., Ardizzone, F., Cardinali, M., Galliet, M. 2006: Estimating the quality of land­slidesusceptibilitymodels.Geomorphology81,1-2.DOI:https://doi.org/10.1016/j.geomorph.2006.04.007 Ivančič, K., Trajanova, M., Skaberne, D., Šmuc A. 2018: Provenance of the Miocene Slovenj Gradec Basin sedimentaryfill,WesternCentralParatethys.SedimentaryGeology375.DOI:https://doi.org/10.1016/ j.sedgeo.2017.11.002 Jemec Auflič, M., Jež, J., Popit, T., Košir, A., Maček, M., Logar, J., Petkovšek, A., Mikoš, M., Calligaris, C., Boccali, C., Zini, L., Reitner, J., Verbovšek, T. 2017: The variety of landslide forms in Slovenia and its immediate NW surroundings. Landslides 14-4. DOI: https://doi.org/10.1007/s10346-017-0848-1 Komac, M. 2003a: Geohazard map of the central Slovenia – the mathematical approach to landslide pre­diction. Geologija 46-2. Komac,M.2003b:Landslidehazardpredictionwithanalysis ofsatelliteimagesandotherspatialdata.Ph.D. Thesis, University of Ljubljana. Ljubljana. Komac,M.2005a:Verjetnostnimodelnapovedinevarnihobmočijgledenapremikepobočnihmas–primer občine Bovec. Geologija 48-2. DOI: https://doi.org/10.5474/geologija.2005.022 Komac, M. 2005b: Landslide occurrence probability prediction with analysis of satellite images and other spatial data. Geološki zavod Slovenije. Ljubljana. Komac, M., Ribičič, M. 2006: Landslide susceptibility map of Slovenia at scale 1 : 250,000. Geologija 49-2. DOI: https://doi.org/10.5474/geologija.2006.022 Komac, M., Hribernik, K. 2015: Slovenian national landslide database as a basis for statistical assessment of landslidephenomenainSlovenia.Geomorphology249.DOI:https://doi.org/10.1016/j.geomorph.2015.02.005 Komac, M., Jež, J. 2018: TXT-tool 1-20 Landslide susceptibility assessment method. Landslide dynamics: ISDR-ICLlandslideinteractiveteachingtools1:Fundamentals,mappingandmonitoring.Cham.DOI: https://doi.org/10.1007/978-3-319-57774-6_9 Komac, M., Kumelj, Š., Krajnik, M. 2012: Izdelava lokalnega/regionalnega načrta preprečevanja tveganj prednaravnimiindrugiminesrečamizaSevernoPrimorsko(Goriškastatističnaregija)zaprimerizbranih nesreč: (zemeljski) plazovi, poplave, potres, požar v naravi in ujma / burja. Geološki zavod Slovenije. Ljubljana. Kumelj, Š., Jež, J., Milanič, B., Bavec, M. 2014: Zvornik municipality (Republika Srpska) – Landslide sus­ceptibility map. Internal report on methodology, Geološki zavod Slovenije. Ljubljana. Lee, C.-T., Huang, C.-C., Lee, J.-F., Pan, K.-L., Lin, M.-L., Dong, J.-J. 2008: Statistical approach to earth­quake induced landslide susceptibility. Engineering Geology 100, 1-2. DOI: https://doi.org/10.1016/ j.enggeo.2008.03.004 Mikoš, M. 2000/2001: Značilnosti drobirskih tokov. Ujma 14-15. Mikoš, M., Brilly, M., Ribičič, M. 2004: Landslides and floods in Slovenia. Acta hydrotechnica 22-37. Mikoš,M.,Brilly,M.,Fazarinc,R.,Ribičič,M.2006:StruglandslideinWSlovenia:Acomplexmulti-process phenomenon. Engineering Geology 83, 1-3. DOI: https://doi.org/10.1016/j.enggeo.2005.06.037 Mikoš,M.,Fazarinc,R.,Majes,B.2007:DelineationofriskareainLogpodMangartomduetodebrisflows from the Stože landslide. Acta geographica Slovenica 47-2. DOI: https://doi.org/10.3986/AGS47202 Mikoš, M., Majes, B. 2010: Mitigation of large landslides and debris flows in Slovenia, Europe.Landslides: Causes, types and effects. New York. MOP – Ministrstvo za okolje in prostor: Geodetska uprava Republike Slovenije, Topografija: Javne infor­macije Slovenije. Ljubljana, 2018. Pannatier,A.,Oppikofer,T.,Jaboyedoff,M.,Stock,G.M.2009:RockfallsusceptibilitymappingofYosemite Valley (USA) using a high-resolution digital elevation model. Geophysical Research Abstracts 11. Perko, D. 1998: The Regionalization of Slovenia. Geografski zbornik 38. Petschko, H.,Brenning,A.,Bell, R.,Goetz,J.,Glade,T. 2014:Assessingthequalityof landslidesusceptibility maps–casestudyLowerAustria.NaturalHazardsandEarthSystemSciences14.DOI:https://doi.org/ 10.5194/nhess-14-95-2014 Reichenbach,P.,Rossi,M.,Malamud,B.D.,Mihir,M.Guzzetti,F.2018:Areviewofstatistically-basedland­slidesusceptibilitymodels.Earth-ScienceReviews180.DOI:https://doi.org/10.1016/j.earscirev.2018.03.001 Rossi, M., Guzzetti, F., Reichenbach, P., Cesare, A., Peruccacci, M. S. 2010: Optimal landslide susceptibility zonation based on multiple forecasts. Geomorphology 114-3. DOI: https://doi.org/10.1016/ j.geomorph.2009.06.020 Shirzadi,A.,Saro,L.,Joo,O.H.,Chapi,K.2012:AGIS-basedlogisticregressionmodelinrock-fallsusceptibility mapping along a mountainous road: Salavat Abad case study, Kurdistan, Iran. Natural Hazards 64-2. DOI: https://doi.org/10.1007/s11069-012-0321-3 Stančič, Z.,Veljanovski, T. 1998: Arheološki napovedovalni modeli in GIS. Geografski informacijski sistmi v Sloveniji. Ljubljana. Stančič,Z.,Veljanovski,T.2000a:UnderstandingRomansettlementpatternsthroughmultivariatestatistics and predictive modelling. Beyond the map. Washington. Stančič,Z.,Veljanovski,T.2000b:UnderstandingRomansettlementpatternsthroughmultivariatestatistics and predictive modelling. Geoarchaeology of the landscapes of Classical Antiquity. Leiden. Tošić, R., Dragićević, S., Zorn, M., Lovrić, N. 2014: Landslide susceptibility zonation: a case study of the municipality of Banja Luka (Bosnia and Herzegovina). Acta geographica Slovenica 54-1. DOI: http://doi.org/10.3986/AGS54307 Veljanovski, T. 1999: Prostorsko modeliranje in napovedovanje lokacij arheoloških najdišč. B.Sc. Thesis, University of Ljubljana. Ljubljana. Zorn, M. 2002: Rockfalls in Slovene Alps. Geografski zbornik 42. Zorn,M.,Komac,B.2002:PobočniprocesiindrobirskitokvLogupodMangartom.Geografskivestnik74. Zorn,M.,Komac,B.2004:Deterministicmodelingoflandslideandrockfallrisk.ActageographicaSlovenica 44-2. DOI: https://doi.org/10.3986/AGS44203 Zorn, M., Komac, B. 2007: Probability modelling of landslide hazard. Acta geographica Slovenica 47-2. DOI: https://doi.org/10.3986/AGS47201 Zorn, M., Komac, B. 2008: Zemeljski plazovi v Sloveniji. Georitem 8. Ljubljana. Zorn, M., Komac, B. 2009: The importance of landsliding in a flysch geomorphic system: The example of theGoriškabrdahills(WSlovenia).ZeitschriftfürGeomorphologie53-2.DOI:https://doi.org/10.1127/ 0372-8854/2009/0053S3-0057 Zorn,M.,Komac,B.2011:Damagecausedby natural disasters inSlovenia andgloballybetween1995and 2010. Acta geographica Slovenica 51-1. DOI: https://doi.org/10.3986/AGS51101Zorn, M., Komac, B., Kumelj, Š. 2012: Mass movement susceptibility maps in Slovenia: The current state. Geografski vestnik 84-1. MEASUREMENTS OF TECTONIC MICRO-DISPLACEMENTS WITHIN THE IDRIJA FAULT ZONE IN THE UčJA VALLEY (W SLOVENIA) Andrej Gosar Introductory figure caption: Left: A crack in the Idrija fault inner fault zone in the Učja valley on which the TM 71 extensometer is installed. Right: Outcrop of another fault plane located 50m to the east with clear striations indicating subhorizontal movements. DOI: https://doi.org/10.3986/AGS.7444 UDC: 551.242:551.34(497.47) COBISS: 1.01 Andrej Gosar1,2 Measurements of tectonic micro-displacements within the Idrija fault zone in the Učja valley (W Slovenia) ABSTRACT:Arecentslip-rateofanactivefaultisaveryimportantseismotectonicparameter,butnoteasy todetermine.Idrijafault,120kmlong,isaprominentgeomorphologicfeaturewithlargeseismogenicpoten­tial,stillneededtoberesearched.Measurementsoftectonicmicro-displacementscanprovideinsightinto its recent activity. The Učja valley extends transversally to the Idrija fault and was therefore selected for the installation of TM 71 extensometer. Measurements on the crack within its inner fault zone are con-ductedfromtheyear2004.In14yearsofobservationsasystematichorizontaldisplacementswithaverage rate of 0.21mm/year and subordinate vertical displacements of 0.06mm/year were established, proving the activity of this fault. An overview of methods of displacement measurements related to active faults and of newer interdisciplinary investigations of the Idrija fault is given. Displacement rates are beside for geodynamicinterpretationsimportantforimprovementofseismotectonicmodelsandthusforbetterseis­mic hazard assessment. KEY WORDS: tectonics, geodynamics, fault, extensometer, Idrija fault, Učja, Slovenia Meritve tektonskih mikro-premikov v prelomni coni Idrijskega preloma v dolini Učje (Z Slovenija) POVZETEK: Recentna hitrost premikov ob aktivnem prelomu je zelo pomemben seizmotektonski para­meter, ki pa ga ni lahko določiti. Idrijski prelom je z dolžino 120km pomembna geomorfološka struktura z velikim seizmogenim potencialom, ki pa ga je treba še raziskati. Meritve mikro-premikov lahko nudijo vpogledvrecentnoaktivnostpreloma.DolinaUčjepotekaprečnonaIdrijskiprelom,zaradičesarjebilaizbrana za namestitev ekstenziometra TM 71. Meritve na razpoki v notranji prelomni coni potekajo od leta 2004. V14letihopazovanjjebilougotovljenosistematičnovodoravnozmikanjespovprečnohitrostjo0,21mm/leto inpodrejenivertikalnipremikishitrostjo0,06mm/leto,kardokazujeaktivnosttegapreloma.Podanjepregled različnihmetodraziskovanjarecentnihpremikovobaktivnihprelomihternovejšihinterdisciplinarnihraziskav Idrijskega preloma. Ocene hitrosti premikov so poleg geodinamskih interpretacij pomembne predvsem za izboljšanje seizmotektonskih modelov in s tem boljše ocenjevanje potresne nevarnosti. KLJUČNE BESEDE: tektonika, geodinamika, prelom, ekstenziometer, Idrijski prelom, Učja, Slovenija The paper was submitted for publication on 5th June, 2019. Uredništvo je prejelo prispevek 5. junija 2019. 1 Slovenian Environment Agency, Seismology and Geology Office, Ljubljana, Slovenia 2 University of Ljubljana, Faculty of Natural Sciences and Engineering andrej.gosar@gov.si 80 1 Introduction A recent slip-rate of an active fault is among the most important seismotectonic parameters needed for realisticearthquakehazardassessment.However,itisnoteasytodetermineitandinterdisciplinaryapproach isneededforareliableestimate.ThemethodsappliedsofarontheIdrijafaultoritsvicinitycanbegrouped into a) geodetic (Kogoj9 2000; Rižnar, Koler and Bavec 2007; Weber et al. 2010), geological (Čar 2010), b) tectonic geomorphology (Moulin et al. 2014), c) paleoseismology (Bavec et al. 2013), d) outcrops dat­ing methods (Moulin et al. 2016) and e) seismological (Bajc et al. 2001; Živčić et al. 2011). Measurements of tectonic micro-displacements using extensometer represent an additional method (Stemberk, Košťak andVilimek2003),whichcancontributetothesolutionoftheproblem.Suchmeasurementswereenabled through invention of a crack-gauge instrument (Košťak 1991), which is installed on the crack which sep­arates two tectonic blocks. In Slovenia seven TM 71 extensometers were installed between 2004 and 2010 at or near the active faults,threeofthemonthesurface(Idrija,KnežaandRašafaults)andfourinsidekarstcaves(twoinPostojna cave, in Polog cave and in Kostanjevica cave) (Gosar et al. 2007, Gosar et al. 2011). Until 2011 the highest rates with stable sense of movements were observed on the Idrija fault (average horizontal displacement rateof0.24mm/year).Theobserveddisplacementrates(mainlyhorizontal)atallotherlocationsweremuch smaller(from0.006to0.05mm/year).OnlyontheRašafaultmoresignificantshort-termverticaldisplacement rateof0.16mm/yearwasobserved,atallotherlocationstherateswereanorderofmagnitudesmaller(Gosar et al. 2011). The fact that the monitoring on the Idrija fault showed with the leap the highest displacement ratesinSlovenia,additionaleightyearsofmonitoringtill2019andsignificanceoftheIdrijafaultforearth­quakehazardassessmentandtectonicgeomorphology,motivatedastudydedicatedonlytothemeasuring site in the Učja valley which is presented in this paper. Idrija fault is because of its length and prominent expression in topography one of the most impor­tant faults in Slovenia (Fig. 1) and it is supposed that it has a large seismogenic potential (Atanackov et al. 2014; 2016). However, its recent tectonic activity is not definitely proved yet by seismological monitoring or geodetic observations, but there are several indications for its activity from geological mapping, tec­tonic geomorphology, dating of outcrops and paleoseismological studies. TogetadditionalinsightintotheactivityoftheIdrijafault,weinstalledin2004theTM71extensometer on a crack within the inner fault cone which is exposed in the Učja valley (Figures 2 and 3). This location was selected as the best, because elsewhere along its trace there are no suitable outcrops. After 14 years of measurements,weprovidetheresultsofobservedtrendsanddisplacementrates.First,anoverviewofinter-disciplinary investigations of the Idrija fault is given. It is followed by description of different methods for displacement measurements and tectonic slip-rate estimates. The fault zone exposed in the Učja valley is describedtogetherwiththeselectionofthelocationfortheinstallationofextensometer.Observedmicro-tectonic displacements are evaluated and some possible interpretations indicated. 2 Idrija fault and measurements of tectonic displacements 2.1 Geology and geomorphology of the Idrija fault The Idrija fault is geomorphologically the most prominently expressed fault in W Slovenia and is clearly visible on satellite and aerial images or digital elevation models (Figure 1). Active tectonic movements are the only geological force which can influence the landscape in such a way. They can be studied by tec­tonic geomorphology methods or direct measurements along faults. The Idrija fault can be traced in the length of 120km from Friuli (Italy) on the NW (Figure 1) to Gorski Kotar in Croatia on the SE (Buser 1986).Itsaveragestrikedirectionis310o,andthedip85o(Atanackovetal.2014;2016).Inthesouthernpart there are several karst poljes of the Ljubljanica river system distributed along the fault. In the northern partthefaultextendsalongtheIdrijca,KanomljicaandSočarivervalleys(Figure1).Thehistoryofitsinves­tigationsisverylong(ČarandGosar2011),alsoduetotheimportanceoftheIdrijaHgoredeposits,because in the geological history part of the ore body (Ljubevč) was displaced along the fault for approximately 2.5km to its present position. Based on this fact and considering the supposed age of the fault of 10–12 million years, Placer (1971) estimated that the average slip-rate through the whole period was from 0.25 Figure 1: Mapofactivetectonicfaults inNWSlovenia(afterAtanackov et al.2014;2016) onshadedLiDAR 1mresolutiondigitalelevation model (ARSO 2015) with the location of TM 71 extensometer on the Idrija fault in the Učja valley. The rectangle indicates a detailed map shown in Figure 2. to0.16mm/year,andtheslipwasobliquewithverticaldisplacementof480m(Placer1982).Newerinves­tigationsalsoshownobliquedisplacement,whichistheresultoftwoseparatedtectonicphases(Čar2010). In middle Miocene, 12 million years ago, the area was under strong extensional forces and normal faults weredevelopedinNW–SEdirectiondippingtowardstheNE.AlongtheIdrijafaulttheNEblockwaslow­ered for up to 480m (Čar and Gosar 2011). Later the stress regime has changed to compression in approx. N–Sdirectionandthefaultwasreactivatedasadextralstrike-slip(Čar2010).Relatedtothischangesome new fault traces were developed. Due to this fact the whole fault zone is rather complex. In a newer study Placer, Vrabec and Celarc (2010) made an estimate that the apparent displacement along the Idrija fault in Tolmin area is around 10km. Because the Idrija fault runs mostly along river valleys and karst fields, there are very few outcrops suitable for detailed tectonic analyses. The best exposure of the whole fault zone is in the Učja valley at the far NW end of the fault (Figures 1 and 2), because it extends in a transverse direction to the fault and at the location of the fault zone it forms a small canyon (Čar and Pišljar 1993). In this canyon the structures of outer and inner fault zones are very good visible (Figure 2). The direct evidence of Quaternary to recent deformation in this area was described by Vrabec (2012). He found an outcrop of poorly sorted Quaternary breccia, which is dissect­ed by several NW–SE oriented subvertical faults and associated fractures. Activity of the fault is further indicated by dextral offsets of the Učja river coinciding with the strands of the Idrija fault zone. Using the tectonic geomorphology (e.g. Žibret and Žibret 2014) and detailed LiDAR digital elevation model, recent kinematics of the Idrija fault was studied by Moulin et al. (2014). Later they applied also thedatingmethodsbasedontheexposureofoutcropstocosmicrays(isotope36Cl)andestimatedtheaver-ageslipratealongthefaulton1.15mm/yearforlatePleistoceneonwards(Moulinetal.2016).Intheframe ofseismotectonicparameterisationofactivefaultsinSlovenia,Atanackovetal.(2016)estimatedtherecent slip rate on 1mm/year. It is assumed that the Idrija fault is seismically active, although the number of earthquakes detectedso far in its vicinity is rather low (Živčić et al. 2011; Vičič et al. 2019). However, the fact that seismic sta­tionswereinthepastlocatedatrelatively largedistances,contributestothelargeextenttotheuncertainty of determinations. For the Idrija earthquake in 1511 with estimated magnitude of 6.8, it is supposed that it occurred in the wider area of the Idrija fault. However, the exact location of this event is still subject of investigations (Fitzko et al. 2005). Recently, the most important investigations of this subject are paleo­seismological(Bavecetal.2013).Inthe20th centuryinawiderareaoftheIdrijafaulttwostrongearthquakeshappened, the Cerknica one in 1926 and the 1998 in Krn Mountains (Živčić et al. 2011). 2.2 Measurements of tectonic displacements Estimatesonrecentslip-ratesareveryimportanttounderstandactivetectonicsandtectonicgeomorphol­ogy,aswellasforearthquakehazardassessments.Intheregionswithmoderateratesofdeformations,including Slovenia,tomeasurerecentslip-ratesisnotaneasytask,especiallyifstrongfaultsarecharacterisedbyrather wide fault zones. To accomplish the task the methods of terrestrial and satellite geodesy are used. In the past several repeated levelling measurements were conducted along various infrastructure. They usually measured only vertical deformations, but horizontal displacements can be measured as well. In the west­ern Slovenia vertical deformations were measured along the profile from Sečovlje to Bled (Rižnar, Koler andBavec2007) indicating fasterupliftoftheJulian Alpsnorth of the Idrija faultwith respecttothe terri­tory south of it. For detailed analysis of the activity of individual faults, geodetic measurements should be performed at locations where structural geological setting is well known and the fault zone is preferable nottoowide.OntheIdrijafaultitwasappraisedthatforterrestrialmeasurementsasuitabletransectisbetween DolenjaTrebušaandKanomeljskoRazpotje(ČarandGosar2011).Forthispurposeanetworkoffourpoints wasdeployedintheKanomljicavalleyalreadyin1985,butonlyinitialmeasurementswereconducted(Kogoj 2000). ForgeodeticmonitoringofactivefaultzonesPlacerandKoler(2007)proposeddeploymentofgeo­detic networks of points in both blocks outside the fault zones and long-term measurements. Some investigations which include repeated GNSS measurements were also conducted (Weber et al. 2010; Serpelloni et al. 2016) in W Slovenia. They provided velocity vectors for wider area indicating gen­eral movements in the north direction for 2–3mm/years. However, the density of measuring points was ingeneraltoolowtoenableassessmentofdisplacementsalongindividualfaults.BesidesGNSSmeasurements, fordetectionofverticalmovementsofthesurfaceSyntheticApertureRadarmethodofpersistentscatterers (InSAR PS) is also used. This method was applied also in the Julian Alps, but not evaluated for tectonicdeformations (Žibret, Komac and Jemec Auflič 2012). With respect to the scale of measurement, on the other end, compared to satellite geodesy, are micro-displacement measurements on individual fault planes or cracks inside the fault zones (Stemberk, Košťak andVilimek2003),whicharedescribedinthispaper.Bytheirapplicationwecanprovetheactivityofafault planeand thus activity of the fault, but we cannot assess the total deformation across the whole fault zone. 3 The Idrija fault zone in the Učja valley In the Učja valley the Idrija fault zone is approximately 750m wide (Figure 2). In the frame of COST pro­ject 3D monitoring of active tectonic structures it was geologically mapped by Igor Rižnar. The wider area isbuiltmainlyofupperTriassicDachsteinlimestone.Slightlytothenorth,thereisthemainthrustofKanin Mountains,alongwhichtheDachsteinlimestoneisoverthrustedtothesouthonCretaceousflysch(Buser 1986). To the SW of the main fault plane in the Učja valley there is a smaller area of Jurassic limestone, and in the canyon floor in some places Cretaceous flysch outcrops. The inner fault zone is approximate-ly260mwideandcomprisesoftwoborderfaults,themainandsidefaultplanes(Figure2).Alongthemain fault plane we didn’t find any suitable crack for micro-displacement measurements. On the other hand, thesidefaultplanewhichruns70mtotheeastacross50mhighcanyonwall,representsaprominentcrack (Figure 3a), which was found as the most appropriate for the installation of extensometer. Down the river to the east, there is another well exposed fault plane with striations which indicate subhorizontal move­ments (Figure 3b). However, the access to this narrow part of the canyon with equipment needed for the installation was not possible. Figure2:DetailedmapofouterandinnerfaultzonesoftheIdrijafaultintheUčjavalleyonshadedLiDAR1mresolutiondigitalelevationmodel(ARSO 2015) with the location of TM 71 extensometer. 4 Methods 4.1 Extensometer TM 71 TM 71 is a mechanic extensometer (Figure 4) aimed for installations on cracks to monitor relative micro-displacements of two tectonic blocks separated by the crack. It operates on the principle of Moire optical effect. Displacements are measured through interference pattern (Košťak 1977; Košťak 1991), formed by two optical grids engraved in two glass plates which undergo a relative shift. The instrument measures deformations in three dimensions as displacement vectors in two perpendicular planes (horizontal and vertical) and angular deviation (rotation). The accuracy of the instrument is 0.05–0.0125mm for dis­placements and greater than 3.2x10-4 rad (0.018o) for angular deviation (Stemberk, Košťak and Vilimek 2003; Stemberk, Košťak and Cacon et al. 2010). The main advantage of TM 71 is that it has no electric componentsandisthusveryrobustforoperationindifficultoutdoorconditions.Therefore,itisverysuit­ableforlong-termmonitoring.Adisadvantageisthatitrequiresmanualreadingsatregularintervals.However, this problem was recently solved for temperature stable and protected environments (for instance karst Figure 3: a) A crack in the Idrija fault inner fault zone in the Učja valley. Arrow indicates the location of the TM 71 extensometer. b) Outcrop of another fault plane located 50m to the east with clear striations indicating subhorizontal movements (after Gosar 2007). caves) by automated shooting images of interference pattern in selected intervals (Briestensky et al. 2010; Šebelaetal.2009).Instrumentreadingsarealwayscorrectedfortemperaturevariations.Todayalmost300 instruments are installed in the whole world. Beside measurements of tectonic displacements, it is used also in engineering geology for monitoring stability of rock blocks or landsliding. 4.2 Measurements of tectonic micro-displacements in the Učja valley Afterdetailedsurveyofthefaultzoneweselected,asthemostsuitableforinstallationoftheTM71,apromi­nentcrackintheinitialpartofthecanyon,whichcut50mhighsouthorientedface(Figure3a).Thecrack extends across the whole face, and the installation was realised at the foot of it, above a large scree cone (Figures 3a and 5). In this area seismically triggered rockfalls occurred during the 1976 Friuli earthquake (Čar and Pišljar 1993) and also during the 1998 earthquake in the Krn Mountains (Gosar 2012; 2019c).InstallationofTM71wasperformedinNovember2004(Šebelaetal.2005;Gosaretal.2007).Quiteachal­lengewasalreadyatransportofheavydrillingequipmentintothecanyonandacrosstheriver.Takinginto account configuration of the crack, a still rod is anchored in the western block in two points and in the eastern block in one point (Figure 5). Toguarantee a sustainable installation without exposureto damage, it was important that the instrument is located under the small overhang in the wall (Figure 3a) which protects it from falling rocks and ice. The instrument is protected with a metal case (Figure 5a). 5 Results The extensometer TM 71 records displacements for already 14 years. This is long enough period that it is possible to deduce on representativity of long-term displacements. The results of the first six years of measurements(endof2004–2010)werealreadypublishedinGosaretal.2011),butadditionaleightyears Figure 4: The TM 71 extensometer which measures tectonic micro-displacements in three directions. (2011–2018) of measurements allows much more firm interpretations and conclusions. The results (Figure6)showsespeciallysystematictrendofhorizontaldisplacements(y-axis)andrelativelysmallerdis-placements in vertical direction (z-axis). This observation is in agreement with the fact that Idrija fault is today mainly a strike-slip fault with minor vertical component of slip (Čar and Gosar 2011). This was so far known from geological observations only. Other source of information on the sense of displacements couldbeseismologicaldata.HoweverduringthelastdecadesofdetailedseismologicalmonitoringinSlovenia andFriuli,therewerenoreallystrongearthquakesontheIdrijafaultwhichwillallowcomputationoffocalmechanisms or detailed analysis of a coseismic slip along the faultplane (Živčić et al. 2011; Gosar 2019b). Suchanalyseswereperformedforthe1998and2004earthquakesintheKrnMountains,wherefocalmech­anisms show on almost pure dextral slip along the Ravne fault with only minor vertical component (Bajc et al. 2001; Živčić et al. 2011; Gosar 2019a). Since Ravne and Idrija faults are parallel and both exposed to thesamestressregime(compressioninN–Sdirection),wecandeducethatalsoalongtheIdrijafaultrecent displacements are predominantly dextral strike-slip. However, this is not necessarily valid for all cracks within the wide fault zone (e.g. Twiss and Moores 1992). In fact measurements with TM 71 on the crack which is parallel to the main fault plane shows predominantly left-lateral strike-slip with minor vertical component. Although sucha result is a surprise, a thorough discussion on possible causes at this moment is not possible. This would perhaps become possible, if detailed geodetic measurements in a dense net of points would be performed for several years. Such a survey should include a measuring points that are definitely located in a stabile blocks far outside from the fault zone, as well as within the blocks separat­edbyindividualfaultplaneswithintheouterandinnerfaultzones.Localpermutationsofthestressdirection withincomplexfaultzonesarealwayspossible,andtheycanresultinunequalsenseofmovementsofindi­vidual blocks also due to rotation of blocks (Twiss and Moores 1992). Therefore, in spite the whole Idrija fault clearly expresses right-lateral strike-slip movements, a displacement on a crack inside the fault zone composed of several blocks and fault planes can show opposite sense. Large scale neotectonic rotations of rigid blocks in the Adria-Eurasia collision strike-slip zones in W Slovenia were recently revealed from paleomagnetic data of Pliocene-Quaternary cave sediments (Vrabec et al. 2018). However, this large scale observation cannot be directly transferred to the local scale situation within the complex fault zone. Average displacement rate of left-lateral slip in the first ten years (2004–2014) is 0.21mm/years (Figure 6). Inthefirsttenmonthsafterinstallationthevelocitywasevenmuchhigher(y=+0.54mm/year) (Gosar et al. 2011). It was followed with anomalous reading at the beginning of 2006. Although there are nodirect proofs,this outlieris most probably a result ofanoutermechanical influenceontheinstrument, which can be a falling rock or ice. This explanation is likely, because outlying reading was later completely recovered. In the first 2.5 years the average displacement rate was 0.31mm/year, and in the first six years 0.24mm/year (Gosar et al. 2011). This can be an indication of gradually diminishing of displacement rate 2.8 2.6 2.4 y 2.2 2.0 1.8 Displacement (mm) +0.21 mm/yr +0.18 mm/yr 1.6 1.4 1.2 +0.24 mm/yr 1.0 0.8 +0.31 mm/yr 0.6 +0.54 z 0.4 +0.06 mm/yr 0.2 mm/yr +0.03 mm/yr 0.0 x –0.2 –0.4 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Year Figure 6: Displacements observed with TM 71 on the Idrija fault; +x indicates closing of the crack, +y indicates left-lateral displacement, +z indicates lowering of the SW block. 35 0.3 Temperature +x – crack opening 0.2 30 0.1 0 25 –0.1 Temperature (°C) 20 15 10 5 0 –0.2 –0.3 –0.4 –0.5 –0.6 –0.7 –0.8 –0.9 Displacement (mm) 12.11.2004 6.01.200510.03.200515.04.200512.06.2005 5.08.200524.09.200511.11.200520.01.200611.03.2006 2.04.200612.05.200624.06.200626.07.2006 9.10.200614.11.200617.02.200713.05.200718.08.200725.11.2007 2.03.200814.05.200829.06.200810.11.2008 7.05.200928.08.200923.11.2009 1.09.201014.10.201019.04.201118.11.201122.11.201223.07.201323.07.201324.07.201419.07.201619.10.201622.03.2017 5.10.201723.04.201824.07.201826.09.2018 Date Figure 7: Correlation between temperature variations and opening (-x extension) or closing (+x compression) of the crack on the Idrija fault. 88 withtime.Possiblereasonsforthisarenotknown.However,on12July,2004astrongearthquakewithmoment magnitudeof5.2occurredintheKrnMountains(Gosar2019b)atthedistanceof10–12kmfromUčja,only fivemonthsbeforeinstallationofTM71. Itiswellknownthatlargeearthquakescauseastaticstresschange on neighbouring faults. Therefore also the Krn Mountains earthquake definitely had an influence also on theIdrijafaultwhichrunsonly8kmtotheSW.Coulombstaticstresschangefor1998and2004earthquakes wascomputedbyGanas,GosarandDrakatos(2008)andshowsslightunloadinginthepartoftheIdrijafaultnear Žaga. Although this can be only a hypothesis, the stress change can have an influence on the slip-rate withdiminisheswithtimeaftertheearthquake.Moreover,aftertheyear2004uptonowtherewerenoother strong earthquakes in the area which exceed magnitude 3.5 (Gosar 2019b). In the last years a smaller dis­placement rate of 0.08mm/year was measured with TM 71 from mid-2014 up to the end of 2018. Taking intoaccountthewhole14-yearsofobservation,theaveragehorizontaldisplacementratewas0.18mm/year. Ontheverticalaxis(z-axis)duringfirstsixyearsofobservationsverystablepositivedisplacementrates wereestablished+0.06mm/year(Gosaretal.2011),whichmeansrelativesubsidenceoftheSWblockwith respect to the NE block (Figure 6). After 2010 this rate has diminished and for the whole 14-years period it is only +0.03mm/year. For the Idrija fault prevailing strike-slip tectonics, subordinated vertical com­ponent was expected also from geological data. The horizontal x-axis extends in transverse direction to the crack and reflects opening or closure of the crack. As expected it shows only seasonal variations, which are in general well correlated with mea­suredtemperaturesatthetimeoftakingreadings(Figures6and7).Positivevalues(compression)corresponds to higher temperatures in summer months and negative values (extension) to lower temperatures in win­ter months. Observed displacements do not exceed 0.2mm. Angular deformations (rotations) in both measuring planes are small (Figure 8). In the xz plane they reach maximum of +0.4 ./200 and in the xy plane maximum of -0.2 ./200. The largest deformations in the xz plane were in years 2007, 2009 in 2011, but they were later mostly recovered and there is no sys-tematictrendoverthewholeperiodofobservation.Onthexyplanethereisonlyaslighttrendtonegative values over the whole period. 0.5 0.4 0.3 0.2 gama xz 0.1 0.0 gama xy 0.1 0.2 0.3 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 Angular deviation ( /200) . Year Figure 8: Angular deviations observed with TM 71 on the Idrija fault in two planes. 6 Discussion of results and conclusions Displacement measurements with TM 71 extensometer on the crack in the inner fault zone of the Idrija fault proved that this is an active fault. This is a very important result, because no direct geodetic (Kogoj2000) or seismological (Živčić et al. 2011) proofs of its recent activity are available so far, although it was deduced from the geologic (Čar 2010), tectonic geomorphological (Cunningham et al. 2006; Moulin et al. 2016) and paleoseismological (Bavec et al. 2013) investigations that Idrija fault is an active fault. AmongallsupposedactivefaultsmonitoredwithTM71extensometersinSlovenia(Gosaretal.2011) the observed horizontal displacement rates on the Idrija fault are with the leap the highest (average rate of 0.21mm/year) and with the most stable and consistent trend. The measured displacement rates on the Raša, Kneža, Predjama-Avče (Figure 1) and Brežice faults (from 0.006 to 0.05mm/year) are for one order of magnitude smaller. Only measuring location on the Raša fault showed for 2.5 years a higher vertical displacement rate (0.16mm/years), but later the sense of the movements has changed and the long-term averagedisplacementtrendis0.06mm/year(Gosaretal.2011),whichisthesameasmeasuredinthewhole 14-years period on the Idrija fault, but the trend on the later was stable and the sense of movements con­sistent in the whole period. DuringthewholeperiodthedisplacementtrendsontheIdrijafaultwereconsistent,howevertheslip-rate diminishes with time. This can be only hypothetically related to the static stress changes on the Idrija fault after the strong 2004 earthquake (Ganas, Gosar and Drakatos 2008). Prevailing are left-lateral hor­izontaldisplacementswithaveragerateof0.21mm/year,andsubordinatedverticaldisplacementswiththe rate of 0.06mm/year. For observed left-lateral displacements within dextral strike-slip fault we can only guess that they are caused by local permutation of the stress field within a complex and very wide fault zone or by rotations of the blocks. Measured displacement rates can be compared with long-term dis­placementsestimatedfromthegeologicaldata,whicharefrom0.25to0.16mm/year(Placer1971).Based on tectonic geomorphology and cosmic rays exposure dating, the slip-rate of the Idrija fault for post late Pleistocene period is assessed on 1.15mm/year (Moulin et al. 2016). In the frame of seismotectonic para­meterisation of active faults Atanackov et al. (2014) estimated the recent slip-rate on 1mm/year. GNSS measurements performed in W Slovenia revealed general movements of the territory in the north direc­tion with velocity of 2–3mm/year(Weber et al. 2010; Serpelloni et al. 2016). Deformations related to this movementsaredistributedovernumerousfaultsofDinaricdirectionwhichprevailintheWSlovenia(Moulin et al. 2016). In a recent study on the present-day kinematic behaviour of active faults in the Eastern Alps in Austria using TM 71 extensometers (Baroň et al. 2019), it was found that annual displacement rates of the monitored faults were mostly about an order of magnitude smaller than the rates of the entire crustal wedges revealed from GNSS measurements. This is consistent with our observations on the Idrija fault. Comparison of deformation rates on various faults is very important for earthquake hazard assess-ment,especiallyintheWSlovenia,whereactiveDinaricstrike-slipfaultsprevail.Themostintriguingabove all is the Idrija fault, due to rather low recent seismic activity in the last decades of instrumentally seis­mological monitoring (Živčić et al. 2011). This fact rises a question which of Dinaric faults could be a seismogenic sources of the strongest known historical or paleoseismological earthquakes in the region, for instance the Idrija 1511 earthquake (Fitzko et al. 2005), because these earthquakes to large extent con­trol the seismic hazard in W Slovenia. Results of TM 71 measurements on the Idrija and Raša faults and in vicinity of the Predjama-Avče fault in the Postojna cave revealed that observed deformations are much higher on the Idrija fault, although the seismic activity in the last decades was higher in the vicinity of the Raša and Predjama-Avče faults especially with several earthquake swarms or aftershock sequences in the Ilirska Bistrica, Pivka-Knežak and Vipava valley areas (Vičič et al. 2019). AlthoughinvestigationsontheactivetectonicsoftheDinaricfaultsystemhavebeenmoreintensivein thelastdecadeandcomprisestectonicgeomorphology,paleoseismology,satellitegeodesy,seismology,micro-displacement measurements etc., they still do not provide sufficient data for reliable estimates on recent slip-ratesoftheIdrijafault.Therefore,itwouldbeveryimportanttosupplementthemwithsystematicand long-term geodetic measurements in a well designed and dense network of measuring points installed on stableoutcropsonbothtectonicblocksseparatedbythefault,outsideofitshighlyfracturedfaultzone(Placer andKoler2007).RealisticdataonrecentratesoftectonicdeformationsalongtheIdrijafaultandotherDinaric faults in the W Slovenia are beside geodynamic interpretations very important for improvement of seis­motectonic models and thus for better seismic hazard assessment in this earthquakes prone area. ACKNOWLEDGEMENT: Installation of extensometer TM 71 was realised in the frame of the EU action COST6253DmonitoringofactivetectonicstructuresleadedbytheInstituteofRockStructureandMechanics of the Czech Academy of Sciences in Prague. The author is in debt to Josef Stemberk, Blahoslav Košťakand Stanka Šebela (ZRC SAZU, Karst Research Institute) for their effort in setting up micro-deformation monitoringinSloveniaandtoMilošBriestenskyformaintenanceofinstrumentsinthelastyears.Thestudy wasrealizedwiththesupportoftheresearchprogramP1-0011financedbytheSlovenianResearchAgency. 7 References ARSO2015:DigitalelevationmodelofSlovenia–1mresolution.SlovenianEnvironmentAgency.Ljubljana. Atanackov,J.,Bavec,M.,Celarc,B.,JamšekRupnik,P.,Jež,J.,Novak,M.,Milanič,B.2014:Seizmotektonska parametrizacija aktivnih prelomov Slovenije. 1. del. Geološki zavod Slovenije, Ljubljana. Atanackov, J., Jamšek Rupnik, P., Jež, J., Milanič, B., Novak, M., Celarc, B., Bavec, M. 2016: Database of activefaultsinSlovenia.Procedingsofthe7th InternationalINQUAMeetingonPaleoseismology,Active Tectonics and Arheoseismology. Crestone. Bajc, J., Aoudia, A., Sarao, A., Suhadolc, P. 2001: The 1998 Bovec-Krn mountain (Slovenia) earthquake sequence. Geophysical Research Letters 28-9. DOI: https://doi.org/10.1029/2000GL011973 Baroň,I.,Plan,L.,Sokol,L.,Grasemann,B.,Melichar,R.,Mitrovic,I.,Stemberk,J.2019:Present-daykine­maticbehaviourofactivefaultsintheEasternAlps.Tectonophysics752.DOI:https://doi.org/10.1016/ j.tecto.2018.12.024 Bavec, M., Atanackov, J., Celarc, B., Hajdas, I., Jamšek Rupnik, P., Jež, J., Kastelic, V., Milanič, B., Novak,M.,Skaberne,D.,Žibret,G.2013:EvidenceofIdrijafaultseismogenicactivityduringtheLateHolocene including the 1511 Mm 6.8 earthquake. Seismic hazard, critical facilities and slow active faults: pro­ceedings of the 4th International INQUA Meeting on Paleoseismology, Active Tectonics and Archeoseismology. Aachen. Briestenský,M.,Košťák,B.,Stemberk,J.,PetroL.,Vozár,J.,Fojtíková,L.2010:Activetectonicfaultmicrodis-placement analyses: a comparison of results from surface and underground monitoring in western Slovakia. Acta Geodynamica et Geomaterialia 7-4. Buser, S. 1986: Osnovna geološka karta SFRJ 1 : 100.000, lista Tolmin in Videm. Zvezni geološki zavod, Beograd. Cunningham, D., Grebby, S., Tansey, K., Gosar, A., Kastelic, V. 2006: Application of airborne LiDAR to mappingseismogenicfaultsinforestedmountainousterrain,SEAlps,Slovenia.GeophysicalResearch Letters 33. DOI: https://doi.org/10.1029/2006GL027014 Čar,J.2010:Geološkazgradbaidrijsko-cerkljanskegahribovja.TolmačhGeološkikartiidrijsko-cerkljanskega hribovja med Stopnikom in Rovtami v merilu 1 : 25 000. Geološki zavod Slovenije, Ljubljana. Čar, J., Gosar, A. 2011: Idrijski prelom in premiki ob njem. Idrijski razgledi 56-1. Čar,J.,Pišljar,M.1993:PresekIdrijskegaprelomainpotekdolineUčjegledenaprelomnestrukture.Rudarsko­ metalurški zbornik 40, 1-2. Fitzko,F.,Suhadolc,P.,Aoudia,A.,Panza,G.F.2005:Constraintsonthelocationandmechanismofthe1511 Western-Sloveniaearthquakefromactivetectonicsandmodelingofmacroseismicdata.Tectonophysics 404, 1-2. DOI: https://doi.org/10.1016/j.tecto.2005.05.003 Ganas, A., Gosar, A., Drakatos, G. 2008: Static stress changes due to the 1998 and 2004 Krn Mountain (Slovenia)earthquakesandimplicationsforfutureseismicity.NaturalHazardsandEarthSystemScience 8-1. DOI: https://doi.org/10.5194/nhess-8-59-2008 GeoZS, 2016: Seizmotektonska parametrizacija aktivnih prelomov Slovenije. Geološki zavod Slovenije, Ljubljana. Gosar, A. 2007: Monitoring of micro-deformations along Idrija and Raša faults in W Slovenia. Geologija 50-1. DOI: https://doi.org/10.5474/geologija.2007.004 Gosar, A. 2012: Application of Environmental Seismic Intensity scale (ESI 2007) to Krn Mountains 1998 Mw=5.6 earthquake (NW Slovenia) with emphasis on rockfalls. Natural Hazards and Earth System Science 12-5. DOI: https://doi.org/10.5194/nhess-12-1659-2012 Gosar, A. 2019a: Review of geological and seismotectonic investigations related to 1998 Mw5.6 and 2004 Mw5.2earthquakesinKrnMountains.Geologija62-1.DOI:https://doi.org/10.5474/geologija.2019.002 Gosar,A.2019b:Reviewofseismologicalinvestigationsrelatedto1998Mw5.6and2004Mw5.2earthquakes in Krn Mountains. Geologija 62-1. DOI: https://doi.org/10.5474/geologija.2019.003 Gosar, A. 2019c: The size of the area affected by earthquake induced rockfalls: Comparison of the 1998 KrnMountains(NWSlovenia)earthquake(Mw5.6)withworldwidedata.ActageographicaSlovenica, 59-1. DOI: https://doi.org/10.3986/AGS.4845 Gosar,A.,Šebela,S.,Košťák,B.,Stemberk,J.2007:Micro-deformationmonitoringofactivetectonicstruc­tures in W Slovenia. Acta Geodynamica et Geomaterialia 4-1. Gosar, A., Šebela, S., Koštak, B., Stemberk, J. 2009: Surface versus underground measurements of active tectonic displacements detected with TM 71 extensometers in western Slovenia. Acta Carsologica 38, 2-3. DOI: https://doi.org/10.3986/ac.v38i2-3.123 Gosar,A.,Šebela,S.,Košťák,B.,Stemberk,J.2011:OnthestateoftheTM71extensometermonitoringinSlovenia: Sevenyearsofmicro-tectonicdisplacement measurements. Acta Geodynamica et Geomaterialia 8-4. Kogoj,D.2000:GeodetskemeritvestabilnostitalobtektonskihprelomihnaobmočjuSlovenije.Geodetski vestnik 44, 1-2. Košťak, B. 1977: Terčové meřidlo TM-71 a jeho užití pro meření velmi pomalých pohybů na poruchách a trhlinách. Inž. Stavby 25-5. Košťak, B. 1991: Combined indicator using Moire technique. Proceedings of the 3rd international sym­posium on field measurements in geomechanics. Oslo. Moulin,A.,Benedetti,L.,Gosar,A.,JamšekRupnik,P.,Rizza,M.,Bourles,D.,Ritz,J.-F.2014:Determining thepresent-daykinematicsoftheIdrijafault(Slovenia)fromairborneLiDARtopography.Tectonophysics 628. DOI: https://doi.org/10.1016/j.tecto.2014.04.043 Moulin,A.,Benedetti,L.,Rizza,M.,JamšekRupnik,P.,Gosar,A.,Bourles,D.,Keddadouche,K.,Aumaitre, G., Arnold, M., Guillou, V., Ritz, J.-F. 2016: The Dinaric fault system: Large-scale structure, rates of slip,andPlio-PleistoceneevolutionofthetranspressivenortheasternboundaryoftheAdriamicroplate. Tectonics 35-10. DOI: https://doi.org/10.1002/2016TC004188 Placer, L. 1971: Nekaj osnovnih podatkov o idrijskem prelomu. Idrijski razgledi 16-1. Placer, L. 1982: Tektonski razvoj idrijskega rudišča. Geologija 25-1. Placer, L., Koler, B. 2007: Predlog geodetske spremljave aktivnih prelomnih con. Geologija 50-2. Placer,L.,Vrabec,M.,Celarc,B.2010:ThebasesforunderstandingoftheNWDinaridesandIstriaPeninsula tectonics. Geologija 53-1. DOI: https://doi.org/10.5474/geologija.2010.005 Rižnar, I., Koler, B., Bavec, M. 2007: Recent activity of the regional geologic structures in western Slovenia. Geologija 50-1. DOI: https://doi.org/10.5474/geologija.2007.009 Serpelloni,E.,VannucciG.,Anderlini,L.,Bennett,R.A.2016:Kinematics,seismotectonicsandseismicpoten­tialoftheeasternsectoroftheEuropeanAlpsfromGPSandseismicdeformationdata.Tectonophysics 688. DOI: https://doi.org/10.1016/j.tecto.2016.09.026 Stemberk,J.,Košťak,B.,Vilimek,V.2003:3Dmonitoringofactivetectonicstructures.JournalofGeodynamics 36, 1-2. DOI: https://doi.org/10.1016/S0264-3707(03)00042-5 Stemberk, J., Košťak, B., Cacoń, S. 2010: A tectonic pressure pulse and geodynamic activity recorded from long-term monitoring of faults in Europe. Tectonophysics 487. DOI: https://doi.org/10.1016/ j.tecto.2010.03.001 Šebela, S., Gosar, A., Košťak, B., Stemberk, J. 2005: Active tectonic structures in the W part of Slovenia – setting of micro-deformation monitoring net. Acta Geodynamica et Geomaterialia 2-1. Šebela,S.,Turk,J.,Mulec,J.,Košťak,B.,Stemberk,J.2009:Statisticalevaluationofthe3Dmonitoringofdis­ placementsofDinaricFaultZone inPostojna Cave, Slovenia. ActaGeodynamica etGeomaterialia6-2. Twiss, R. J., Moores, E. M. 1992: Structural geology. New York. Vičič, B., Aoudia, A., Javed, F., Foroutan, M., Costa, G. 2019: Geometry and mechanics of the active fault system in western Slovenia. Geophysical Journal International 217-3. DOI: https://doi.org/10.1093/ gji/ggz118 Vrabec,M.2012:EvidenceofQuaternaryfaultingintheIdrijafaultzone,Učjacanyon,NWSlovenia.RMZ– Materials and Geoenvironment 59, 2-3. Vrabec, M., Preuner, P., Zupan Hajna, N., Mihevc, A., Bosak, P. 2018: Neotectonic vertical-axis rotations intheAdria-Eurasiacollision zonereviledfrom Paleomagnetic dataofPliocene-Quaternary cavesedi-ments(Slovenia).INQUASEQSQuaternaryStratigraphyinKarstandCaveSediments,Program,Abstract and Guide Book. Postojna. Weber, J., Vrabec, M., Pavlovčič-Prešeren, P., Dixon, T., Jiang, Y., Stopar, B. 2010: GPS-derived motion of theAdriaticmicroplatefromIstriaPeninsulaandPoplain,andgeodynamicimplications.Tectonophysics 483, 3-4. DOI: https://doi.org/10.1016/j.tecto.2009.09.001 Žibret, G., Komac, M., Jemec Auflič, M. 2012: PSInSAR displacements related to soil creep and rainfall intensities in the Alpine forelandof western Slovenia. Geomorphology 175-176. DOI: https://doi.org/ 10.1016/j.geomorph.2012.07.002 Žibret,L.,Žibret, G. 2014: Useof geomorphologicalindicatorsfor the detectionof activefaults insouthern partofLjubljanamoor,Slovenia.ActageographicaSlovenica54-2.DOI:https://doi.org/10.3986/AGS54203Živčić, M., Čarman, M., Gosar, A., Jesenko, T., Zupančič, P. 2011: Potresi ob Idrijskem prelomu. Idrijski razgledi 56-1. ECONOMIC RESILIENCE OF THE COMMAND AND CONTROL FUNCTION OF CITIES IN CENTRAL AND EASTERN EUROPE Piotr Raźniak, Sławomir Dorocki, Anna Winiarczyk-Raźniak Warsaw CBD. DOI: https://doi.org/10.3986/AGS.7416 UDC: 911.375:338.1(4) COBISS: 1.01 Piotr Raźniak1, Sławomir Dorocki1, Anna Winiarczyk-Raźniak1 Economic resilience of the command and control function of cities in Central and Eastern Europe ABSTRACT: The authors propose a new approach to the analysis ofcities in a time of potential major cri­sis in a dominant sector consisting of the largest firms generating the command and control function of acity.ThispurposeisservedbythecreationoftheCentralandEasternEuropeanEconomicCentreIndex (CEEECI), which reflects the potential of each studied city and its development and/or fields of economic specialisation of its largest companies capable of generating regional command and control (C&C) func­tionsofcities.ResearchhasshownthattheC&CfunctionsofcitiessuchasWarsaw,Prague,andBudapest are the most resistant to economic crisis of the dominant sector. More than half of the analysed cities are economically dominated by the consumer business and transportation and manufacturing sectors. KEYWORDS:cities,economicresilience,headquarters,regionalcommandandcontrolfunction,Central and Eastern Europe Ekonomskaodpornostfunkcijevodenja inupravljanja srednje-in vzhodnoevropskih mest POVZETEK: Avtorji v članku predlagajo nov pristop k analizi mest med morebitno veliko gospodarsko krizo vodilnega sektorja, v katerem največje gospodarske družbe opravljajo funkcijo vodenja in uprav­ljanjamesta. V ta namen avtorji oblikujejoindeks srednje- in vzhodnoevropskih gospodarskih središč, ki izraža potencial posameznega proučevanega mesta in njegov razvoj in/ali področja gospodarske speciali­zacijenjegovihnajvečjihgospodarskihdružb,kilahkoopravljajoregionalnofunkcijovodenjainupravljanja mest.Raziskavajepokazala,dasofunkcijevodenjainupravljanjamest,kotsoVaršava,PragainBudimpešta, najbolj odporne na gospodarsko krizo vodilnega sektorja. Glavni gospodarski sektorji v več kot polovici analiziranih mest so sektor izdelkov za široko rabo, prevozništvo in proizvodni sektor. KLJUČNEBESEDE:mesta,ekonomskaodpornost,sedež,regionalnafunkcijavodenjainupravljanja,Srednja in Vzhodna Evropa The paper was submitted for publication on May 16th, 2019. Uredništvo je prejelo prispevek 16. maja 2019. 1 Pedagogical University of Cracow, Institute of Geography, Poland, Cracow, Poland piotr.razniak@up.krakow.pl, slawomir.dorocki@up.krakow.pl, anna.winiarczyk-razniak@up.krakow.pl 1 Introduction The concept of a command and control function appears in many research publications and descriptions of the economic strength of cities and their international connectivities. One seminal paper on this sub­ject is that of Hall (1966) who created the world city concept. Friedmann (1986, 70–77) argues that one measure of a world city is its control function in the global economy: »…The global control functions of worldcitiesaredirectlyreflectedinthestructureanddynamicsoftheirproductionsectorsandemployment…«. He also notes that world citiesinvite foreign capital (Friedmann 1986, 73) by functioning as open systems, thus attracting investment and yielding additional command and control functions in the world econo­my.Sassen(1991)describesNewYork,London,andTokyoasthemostcriticalcitiesintheworldeconomy due to their accumulation of world economic control functions. Beaverstock, Smith and Taylor (1999) analysedtheinternationalconnectivitiesofcitiesviathelargestcompaniesintheservicessector,account­ingsector,advertisingsector,bankingsector,andthefieldoflaw,butconnectivitiesmayalsobeexamined for other sectors including global media firms, maritime-producer services, and non-governmental organisations (Derudder and Taylor 2018). It appears at this point in time that New York and London are the most highly linked cities via what is known as the NY-LON city-dyad concept (Taylor et al. 2014). Command and Control Function (C&C) is determined on the basis of location of the headquarters of the main multinational companies. The C&C functions in city research are strongly linked to the notion of citiesascommandandcontrolcentresoftheworldeconomy(Csomós2013;CsomósandDerudder2014). The command and control (C&C) function may also contribute to the prestige of a city (Alderson and Beckfield 2004). It seems that research on the corporate headquarters of the largest firms shows a partic­ular strength of cities in terms of their command and control role in the global economy; however, this is not the only measure of the global rank of a city (Taylor 2004). Other studies on the number of corporate headquarters per city and corporate financial results as factors in the economic strength of cities as well as their impact on other parts of the world include the following:Heenan(1977),FriedmannandWolff(1982),Tayloretal.(2009),Huang,LeungandShen(2013), Raźniak, Dorocki and Winiarczyk-Raźniak (2018b), Csomós (2017), Derudder et al. (2018), Śleszyński (2018),Raźniak,DorockiandWiniarczyk-Raźniak(2019).Sincethe1970stransnationalcorporationshave become increasingly important in the world economy and many have relocated their principal office or corporate headquarters to new locations in Asia (Csomós and Derudder 2014). This may be due to the fact that geographical diversification of investments is one of the key issues considered by investors eager toreducetheirlevelofrisk(Bacsosz2019).ItisnoteworthythatcompanieslocatedinCentralandEastern Europearestartingtoplayanincreasinglysignificantroleintheworldeconomy(RaźniakandWiniarczyk­Raźniak 2015). Existing conceptualisations of the city, e.g. world city (Internet 3), global city (Sassen 1991), city com-mandandcontrolfunction(Csomós2013)donotassesstheresilienceofthesefunctionstoperiodiccrises in the world economy. While cities may be very important in the world, they may not be ready to counter economicrecessioneventsthatmaystronglyaffectagivenkeyfunctionoftheireconomy.Itmaybeargued thatacitycharacterisedbyastrongcommandandcontrolfunctionisnotnecessarilypreparedforamajor crisisevent.Thecommandandcontrolfunctionofcitiesdominatedbyasinglesectororcompanyisprone toeconomiccollapseintheeventthatthedominantsectororcompanyfacesproblems.Ontheotherhand, the C&C function of cities basing their growth on multiple successful companies or sectors is better able tooffsetlossesatselectedcompaniesorstructuralproblemsinsomesectorsbythesuccessfulperformance ofothercompaniesorentiresectorsofthecommandandcontrolfunction(Raźniak,DorockiandWiniarczyk­Raźniak 2017). Citiesarecomplex,adaptivesystemsofpeople,economics,andthenaturalenvironmentdrivenbykey processes sustaining them and determining their abilityto resist crisis events (Hooling 2001). This means thatacity’sresistabilitymaybedefinedintermsoftheflexibilityofitscommunitiesandeconomies,which areabletopredict,preparefor,andrespondtodisruptionsintheirfunctioning(Barnett2001;Foster2007). We argue that a given city’s command and control function stability can be measured in terms of its abilitytoresisteconomiccrisisofthedominantsectorrepresentedbyleadingcorporations(Raźniak,Dorocki and Winiarczyk-Raźniak 2017). This paper assesses this level of resistance to crisis for key economic sec­torsandkeycorporateemployersinthestudiedcities.Crisisisdefinedbytheauthorsinthiscaseasadecline in the financial performance of firms in a given sector, which causes them to become excluded from the list of the 500 largest firms in Central and Eastern Europe. This type of decline forces a sector to lose its dominant economic position and effectively its ability to play a regional command and control role in its parent city. In current world economy the increasingly important question is the following: »How will these cities change when/if these principal functions are lost (Raźniak, Dorocki and Winiarczyk-Raźniak 2017)?« It is not only important for a city to rank high in a good economic situation, but also in an economic crisis thatmayaffectitsmainfunctions.Thispromptsthefollowingquestion:»Whatwillhappentothecitycom­mandandcontrolfunctionsincaseofeconomicrecession?«Theauthorsusetheconceptofthecitycommand and control function to determine the resilience of the cities to possible future economic crises that can affect a city’s overall rank in the world economy. 2 Research data and methods ThestudyanalysesthelargestcompaniesinEasternEuropeandcomparestheirperformanceovertheperi­od 2008–2015 based on data obtained from the Deloitte company (Internet 1), which has been compiling sales revenue data on the top 500 companies headquartered in Central and Eastern Europe since 2008. According to Anh, Thuy and Khanh (2018) an economic crisis may be triggered by a budget deficit, international currency reserves or increases in national debt. In a purely financial sense, a crisis may be defined by a depreciation of local currency against the U.S. dollar of at least 30%, and this decline is high­er than 10 percentage points relative to previous years (Delbianco, Fioriti and Tohmé 2019). Corporate performancesuffersoverthecourseofafinancialcrisis,asdoescorporatemanagementquality,withexec­utives attempting to maximise their own compensation packages at the expense of company financial performance (Cornett et. al. 2009). In this research crisis is defined as a decline in financial performance among companies in the most profitable sector present in a city, which results in a substantial loss of market status of the sector’s com­paniesthatgeneratethecommandandcontrolfunctionofacity.Thistypeofanalysisisservedbythecreation of an World Economic Center Index that illustrates the rank of a city in terms of the value of the largest corporateheadquartersbysectoraswellasthestabilityofthisrankuponsuddenremovalofthemostimpor­tantsector.Thisindexiscalculatedusingbasicstatisticsincludingstandardisationviathestandarddeviation value and the weighted average. These methods represent the basis for descriptive statistics and are used in the calculation of most statistical indices employed in socioeconomic studies (Benos, Karagiannis amd Karkalakos 2015). This method is designed to produce information on the command and control poten­tial of a city, but also on its path of economic development or economic specialisation of the companies creating this function. In addition, it may be assumed that strong company resistance to economic crisis is linked with a city’s high degree of international connectivity, as financial losses generated by a compa­nyatonelocationmaybeoffsetbyprofitsgeneratedananotherlocationintheglobalisedworld(Raźniak, Dorocki and Winiarczyk-Raźniak 2017). Deloitte’s Central Europe Top 500 List + 50 banking and 50 insurance companies (banking and insurance companies are in the same reports but not in the main Top 500 list) (Raźniak, Dorocki and Winiarczyk-Raźniak 2018a) is not representative of the studied region, as it could be affected by a poten­tialoverrepresentationofthebankingandinsurancesectors.Inaddition,salesrevenueisnottheonlyfactor that describes the rank of a company. In light of this issue, Deloitte Central Europe Top 500 Reports were usedtocreateacustomisedlistofthemostimportant500companiesinCentralandEasternEurope.Deloitte Central Europe Top 500 Report (Internet 2) is based on standardised financial data, and it was used to create a New Eastern Europe Top 500 list of major companies. Sales revenue was standardised along with net profit for all 600 Deloitte companies by calculating the share for each individual company where the maximum relative value is 100. Next, an average value (Equation 1) was calculated for this set of revenue and profit values (xtop) (Shiller 1991). @ ä A (1), @ a 6 where x is the sales revenue and y is the net profit. The xtop value was used to assign a rank to each company on the list. The first 500 companies were then selected for further analysis. Further analysis focused on selecting only cities with corporate head­quarters representing three or more sectors of the economy in any given year. ThestudiedcompaniesweredividedintoninesectorsusedbyDeloitteinitsresearchwork(Internet 1): banking, consumer business and transportation, energy and resources, insurance, life sciences and health care, manufacturing, the public sector, real estate, technology media and telecommunications. The geo­graphic location of each company was determined based on the location of its headquarters. Companies were assigned to specific major metropolitan areas. BasedonRaźniak,DorockiandWiniarczyk-Raźniak(2018a)itisassumedthatregionalaswellaslocal commandandcontrolfunctionsaregeneratedinEasternEuropeancitiesbycompaniesontheNewEastern Europe Top 500 List created by the authors. The benefits of the presence of these companies presumably gototheirparentcities.BasedonNewEasternEuropeTop500ListCentralandEasternEuropeanEconomic Center Index(CEEECI) was created, which assesses cities in terms of the value of corporate headquarters by sector and stability in cases where one or more sectors are excluded. This approach yields information not only on the potential of a given city, but also on its economic evolution and/or specialisation. Cities with at least three economic sectors were included in the analysis. Cities featuring only two sectors were excluded in light of the substantial decline of the command and control function in the event of crisis in one of the sectors. The rank of the studied cities was compared by calculating standardised values based on mean nor-malised revenue and net profit (x) for each studied sector of the economy (z) in the calculation of the comprehensive potential index (CPI) for each studied city (Equations 2 and 3): Ô L . . (2) . (3) 5 Ô where x=revenue plus net profit for each sector of the economy, s – is the sectors of the economy and N – is the number of cities per sector of the economy. One goal of the research was to verify which sector of the economy impacts a city’s economic poten­tial the most. This purpose was served by subtracting the value of individual sectors from the total standardised value calculated for 2009, 2012 and 2015. The objective was to observe how the sum of stan­dardised values (z’) changes for a given city,assumingthat the initial value constitutes 100% (Equation 4). Citiescharacterisedbyahighrangefollowingsectorsubtractiontendtospecialiseinoneareaoftheecon­omy and tend to remain underdeveloped in other areas of the economy. On the other hand, cities with low range tend to have both, advanced and multifaceted economies. The paper analyses cities with three or more sectors. . (4) Â Then a stability index (SI; Equation 5) was constructed using standardised values CPI and variances invaluesresultingfromthesubtractionoftheselectedindividualsectorsoftheeconomy(z’)fromallstud-  iedsectors.Thevalueoftheindexwasthendividedbythestandarddeviation(SD)ofvaluesresultingfrom the subtraction of selected sectors (z’) [4]. This index also illustrates how financial crisis in a city’s main economic sector impacts the city’s overall economy. (5) a  The final step consisted of the construction of a Comprehensive Development Index for cities based on sectors (CEEECI). The index includes the degree of stability, number of sectors, and the number of corporateheadquarters(Equation6).TheweightsofthevariablesusedtoconstructtheCEEECIwerecal­ culatedusingprincipalcomponentanalysis(PCA),whichincludedthestabilityindex(SI),numberofsectors  of the economy (S) and the number of headquarters (HQ). Data for 2008, 2012 and 2015 were analysed separately and the value of the CEEECI was calculated only based on the first principal component. ; (6) e whereHQisthenumberofheadquarters,SisthenumberofsectorsoftheeconomyandSIisthestabilityindex. Calculationscannotbeperformedforameansectorvalueofzero.Normalisationwasperformedonly inrelationtoanormaldistributionwithoutcheckingfordataasymmetryandassumingthatanormaldis­tribution will suffice. This must be considered in data analysis and normalisation must not be used for data that is strongly skewed. Similarmethodsofconstructingeconomicindicatorswiththeuseofstandardizedvalues,weightedaver­ages,principalcomponentanalysis,andmixedmodelswerediscussedbyMorrison(1967),andMarinoand Tebala(2016).However,themethodusedinthispaperwascreatedbyitsauthors,anditisusedalsobythe authorsofotherpublications(Dorocki,RaźniakandWiniarczyk-Raźniak2018and2019;Raźniaketal.2019). 3 Ranking of Eastern European cities based on the CEEECI A total of 500 companies had their headquarters in 125 cities in 2008, 124 in 2012, and 156 in 2015. The number of cities declined by one (–0.8%) during the global economic crisis in 2008. A significant diver-sificationofheadquarterslocationswasnotedintheperiod2012–2015,andthenumberofcitiesincreased 25.8%.Anoppositetrendwasobservedinthecaseofthelocationofthelargestworldcorporations(Internet2). Figure 1: CEEECI values for 2008, 2012 and 2015 based on Deloitte’s Central Europe Top 500 Reports. Thenumberofcitieswiththelargestcorporationsincreased7.9%intheperiod2008–2012andthendeclined 5.4% in the period 2012–2015. Hence, a spatial deconcentration is observed for the largest corporations in Central and Eastern Europe, while a spatial concentration is observed on a world scale. The highest CEEECI values in 2008 were noted for Warsaw (4.7 points), Budapest (2.5 points), and Prague (2.4 points; Figure 1). Warsaw had the highest CEEECI in subsequent years. Resistance to crisis in the main sector of Budapest’s economy declined in 2012 and 2015 leading to a decline to third rank, behind Prague. Other capitals were also characterised by rather high CEEECI values in 2015: Bratislava (1.99),Bucharest(1.94),Ljubljana(1.46).Ontheotherhand,Kiev’sresistancetocrisissignificantlydeclined. The reason for this may be the current state of war between Russia and Ukraine, influencing the econo­my (Charap and Colton 2017; Feklyunina and Romanova 2017). Its CEEECI in 2008 was 1.42, while in 2015 it was almost 40% less. The lowest values (less than 0.5) for CEEECI cities in the study period were those of Pardubice, Brasov, and Tallinn. The largest increases wereobservedforsmallercitiessuchasŁódź(240%),NovoMesto(189%),Poznań(176%),Pardubice(158%), and Kraków (153%). However, only Poznań exceeded a value of one in 2015. Several cities were classified as CEEECI cities in one or two years of the study period: Riga, Wrocław, Sofia in 2008 and 2012; Donetsk, Ostrava, Pardubice, Dnepropetrovsk in 2012. It may be argued that these cities were not affected by the global economic crisis of 2008, as they had lost their rank after 2012 (Figure 1). In 2015 a total of 18 Central and Eastern European economic centres were identified and divided into four categories based on their ability to resist an economic crisis. Three cities were given the top CEEECI ranking. Six cities were ranked major. Seven cities were ranked midsize. Two cities were ranked minor. TheRomaniaandSloveniafeaturetwocitieseach.AnothersevencountriesinEasternEuropefeatureonly onetop-rankedeconomiccentreeach.Finally,norankedeconomiccentreswerenotedinAlbania,Bosnia Figure 2:CentralandEasternEuropeaneconomiccentres in2015based onDeloitte’sCentralEuropeTop500Reports(author:AnnaWiniarczyk-Raźniak). andHerzegovina,Montenegro,Moldova,NorthernMacedonia,andSerbia (Figure2).All of theabovesix countries are characterised by small geographic area, small population, low total GDP, and low GDP per capita (Internet 2). Most of these countries became independent only in the 1990s following the dissolu­tion of the much larger Socialist Federal Republic of Yugoslavia. Warsaw, Prague, and Budapest are now the three most significant Central and Eastern European Economic Centres, and are classified in the top category due to their economic dominance in the region and very strong resistance to economic crisis. At the same time, only Prague is home to all nine of the studied sectors (Warsaw and Budapest are home to eight). These three top CEEECI cities differ in terms of their dominant economic sector. In Warsaw, it is consumer business and transportation, while in Prague energy and resources, and lastly in Budapest itismanufacturing.CitiessuchasBratislava,Bucharest,Ljubljana,Zagreb,PoznańandtheKatowicewere classified in the major category due to their strong ability to resist economic crisis (Figure 2). The only non-capitals in this group (Poznań and Katowice) are located in Poland, but they are »balanced« regional centres (Páthy 2017). The region continues to develop in an economic sense as well as in the sense of its residents’well-being(EgriandTánczos2018)andhometoanumberofrobustforeigncorporations(Nazarczuk and Umiński 2018). However, each of thethree cities is dominated by a different sector of the economy except Zagreb and Poznań (consumer business and transportation). Hence, it may be argued that the rank of these six cities may change in different directions in the event of crisis in a particular sector. The strength of Polish cities (7cities)isapparentinthisstudy,with38.9%ofCEEECIcitieslocatedinPoland.Polishcitiesarenotstrong­ly differentiated by main sector and are classified mostly as major and medium type. Three Polish cities aredominatedbythelifesciencesandanotherthreearedominatedbyconsumerbusinessandtransportation. Cities described as minor-type usually feature three or four sectors. In the event of crisis in a dominant sector, midsize-type cities are likely to decline substantially in economic terms (Figure 2). ThemostfrequentlyencounteredsectorinCentralandEasternEuropeaneconomiccentresisthecon­sumerbusinessandtransportationsector–eightcities(44.4%).Thisisespeciallythecaseincountriesthat usedtobepartoftheSovietUnionuntiltheearly1990sincludingLithuania,Latvia,Estonia,andUkraine. Manufacturing is dominant (22.2%) in four key urban regions – usuallyold industrial regions such as the Tri-city (Gdańsk, Gdynia and Sopot) and Łódź in Poland. However, Eastern European companies in this sectorarenotimportantplayersontheglobalscene(Krätke2014).Insummary,CentralandEasternEuropean EconomicCentresaredominatedbyconsumerbusiness,transportation,andmanufacturing(66,7%).The next key sector is banking (3 cities; Figure 2). 4 Discussion The largest companies in Central and Eastern Europe became concentrated in a larger number of cities intheperiod2008–2015(Internet1).Thisistheoppositetrendtowhatisnowobservedintheworldecon­omy,withfewercitieshostingthelargestcorporationsaroundtheworld(Internet2).Thismaybetheresult of the opening of some offices of foreign corporations headquartered in Eastern Europe in order to limit operating costs and invest more profits in the home country. What is important is that this occurred sev­eral years after the global economic crisis of 2008. Economicstabilityofthecommandandcontrolfunctionofcitiesintimeofcrisiswasanalysedinterms oftheeffectsofstructuralchangeontheeconomy.Thesmallestdifferencesinthelevelofeconomicdevel­opment based on sector specialisation and corporate financial performance were noted in Prague, Warsaw,Ljubljana,Vilnius,Zagreb,andPoland’sTri-city.However,thelatterfourcitieswerecharacterised by low standardised values, which is why the difference in values triggered by the loss of a major sector of the economy was quite small. A high degree of specialisation and a close link with a single sector were observed in the case of Donetsk, Bucharest, Kiev, and Bratislava. Two thirds of the analysed cities are economically dominated by consumer business and transporta­tion and manufacturing. A crisis in any of these key sectors could trigger a decline in the city command and control function. Consequently, the cities of Warsaw, Prague, and Budapest may be called the lead-ingCentralandEasternEuropeanEconomicCentres.Thethreecitiesarehighlyresistanttocrisisandhave significant international connectivities. At the same time, these three cities possess an adequate amount of potential to play a meaningful role in the world economy. However, Prague outpaced Budapest in 2012 and 2015 compared with 2008. GiventhepoliticalhistoryofCentralandEasternEuropeanditslinkstoglobalisation,itremainsaunique region where many companies operating in this part of the world are in fact just regional offices of lead­ing corporations. Śleszyński (2015) argues that strategic decisions regarding their key functions are made outside of Central and Eastern Europe, and argues that the largest companies, even those whose head­quartersarelocatedinaforeigncountry,maygeneratethecommandandcontrolfunctionsofagivencountry’s economy. DespitebeinglistedintheDeloitteCentralEuropeTop500Report,manycompaniesinthestudyarea do not fully perform the command and control function in their home cities. However, regional offices located in the Central and Eastern Europe may have some decision power at the local and regional level. Regional managers thus may perform a limited regional or local command and control function, but not higher-level functions that remain reserved for main offices located outside of the study area. For exam­ple, the decision to close a plant or regional office made outside of the studied region may strongly affect thestabilityofacity’sregionalcommandandcontrolfunction.Hence,itisimportanttostudycitiesnotonly intermsofthefinancialresultsofkeycorporations,butalsointermsoftheabilityofthecommandandcon­trol function to remain stable in times of economic crisis in a principalsector generating thisfunction. It may be argued that, for a city, important is not only the magnitude of the C&C function, but also its ability to weather a crisis. This is important, as the economy of a city and its C&C function are not only affected by economic processes, but also by historial problems and ongoing warfare in some cases. 5 Conclusion In a global world, it appears that a city’s global rank in a periodof economic growth is not the only impor­tant issue. What is also important is its global rank in periods of economic recession, which may affect its primaryfunctions.Themethodologypresentedinthepaperanswersthefollowingquestion:»Whatwould happen if something goes wrong in the realm of a city’s C&C function?« Similar research needs to be per­formed as part of other urban conceptual modelssuch as the concept of the world city. Standard forecasts arenotsufficient(Neal,DerudderandTaylor2019);insteadwhatisneededisanalysisonwhatwouldhap-pen to intercity connectivities in the event that the number of linkages between a given city and its main connectivity partner city would strongly decrease. Would it continue to be connected as well as before? Would it continue to be a world city? Research has shown that cities can resist economic crisis in their command and control function bet­teriftheyarehometoalargernumberofeconomicsectorsandfollowamoresustainablepathofeconomic development.Inordertoincreasetheattractivenessofacitytoglobalinvestors,itwouldbehelpfultohelp streamlinecitymanagementpracticesbyreducingredundanciesbetweenlocalgovernmentsandthenation­al government, and increasing the effectiveness of the local job market. These steps would reduce differences between cities (Wolman et al. 1992) and may also help cities weather any potential crisis that may occur. Alderson and Beckfield (2004) speak of a certain type of prestige associated with a city’s com-mandandcontrolfunction.Citiesinthestudiedregioncouldattempttobuildtheirbrandascitiesperforming a regionalcommandandcontrolfunction resistant to economiccrisisinamainsectorof theirlocalecon-omy.Astronglocalorregionalbrandwouldthenhelpthemtotriggeravirtuouscycleofbrandrecognition and continuous foreign direct investment. 6 References Alderson,A.,Beckfield,J.2004:Powerandpositionintheworldcitysystem.AmericanJournalofSociology 109-4. DOI: https://doi.org/10.1086/378930 Anh,V.,Thuy,H.,Khanh,B.2018:WhatCausesFinancialCrisisinAsianCountries?JournalofEconomics and Development 20-3. DOI: https://doi.org/10.33301/JED-P-2018-20-03-01 Bacsosz, S. 2019: Analysis of the geographical diversification of financial instruments. Regional Statistics 9-1. DOI: https://doi.org/10.15196/RS090110 Barnett, J. 2001: Adapting to Climate Change in Pacific Island Communities. World Development 29-6. DOI: https://doi.org/10.1016/S0305-750X(01)00022-5 Beaverstock, J., Smith, R., Taylor, P. J. 1999: A roster of world cities. Cities 16-6. DOI: https://doi.org/ 10.1016/S0264-2751(99)00042-6 Benos,N.,Karagiannis,S.,Karkalakos,S.2015:ProximityandgrowthspilloversinEuropeanregions:The role of geographical, economic and technological linkages. Journal of Macroeconomics 43. Charap,S.,ColtonT.J.2017:EveryoneLoses:TheUkraineCrisisandtheRuinousContestforPost-Soviet Eurasia. Abingdon, Oxon. Cornett, M., McNutt, John, J., Hassan, T. 2009: The Financial Crisis, Internal Corporate Governance, and the Performance of Publicly-Traded U.S. Bank Holding Companies. DOI: https://doi.org/10.2139/ ssrn.1476969 Csomós, G. 2013: The Command and Control Centers of the United States (2006/2012): An Analysis of Industry Sectors Influencing the Position of Cities. Geoforum 50. DOI: https://doi.org/10.1016/ j.geoforum.2013.09.015 Csomós, G. 2017: Cities as command and control centres of the world economy: An empirical analysis, 2006–2015. Bulletin of Geography, Socio-economic Series 38-38. DOI: https://doi.org/10.1515/ bog-2017-0031 Csomós, G., Derudder, B. 2014: Ranking Asia-Pacific cities: Economic performance of multinational corporations and the regional urban hierarchy. Bulletin of Geography, Socio-economic series 25-25. DOI: https://doi.org/10.2478/bog-2014-0030 Delbianco,F.,Fioriti,A.,Tohmé,F. 2019:Quantifyingworldwideeconomicdistress. RegionalStatistics9-1. DOI: https://doi.org/10.15196/RS090108 Derudder, B., Cao, Z., Liu, X., Shen, W., Dai, L., Zhang, W., Taylor, P. J. 2018: Changing Connectivities of Chinese Cities in the World City Network, 2010–2016. Chinese Geographical Science 28-2. DOI: https://doi.org/10.1007/s11769-018-0938-6 Derudder,B.,Taylor,P.J.2018:Centralflowtheory:comparativeconnectivitiesintheworld-citynetwork. Regional Studies 52-8. DOI: https://doi.org/10.1080/00343404.2017.1330538 Dorocki, S., Raźniak, P., Winiarczyk-Raźniak, A. 2018: Zmiany funkcji kontrolno-zarządczych w mias­tach europejskich w dobie globalizacji. Studies of the Industrial Geography Commission of the Polish Geographical Society 32-3. Dorocki, S., Raźniak, P., Winiarczyk-Raźniak, A. 2019: Changes in the command and control functions of European cities in 2006-2016. Geographia Polonica 92-3. DOI: https://doi.org/10.7163/GPol.0149 Egri,Z.,Tánczos,T.2018:ThespatialpeculiaritiesofeconomicandsocialconvergenceinCentralandEastern Europe. Regional Statistics 8-1. DOI: https://doi.org/10.15196/RS080108 Feklyunina,V.,Romanova,V.2017:UkraineandTriangularDiplomacy:Kyiv’sLegitimacyDilemmasinthe MidstoftheCrisisValentinaandValentyna.TriangularDiplomacyamongtheUnitedStates,theEuropean Union, and the Russian Federation. Cham. DOI: https://doi.org/10.1007/978-3-319-63435-7_7 Foster,K. A.2007:ACaseStudyApproachtoUnderstandingRegionalResilience. WorkingPaper2007-08. Berkeley. Friedmann,J.,Wolff,G.1982:Worldcityformation:anagendaforresearchandaction(urbanizationprocess). International Journal of Urban and Regional Research 6-3. DOI: https://doi.org/10.1111/j.1468-2427. 1982.tb00384.x Friedmann,J.1986:Theworldcityhypothesis.DevelopmentandChange17-1.DOI:https://doi.org/10.1111/ j.1467-7660.1986.tb00231.x Hall, P. 1966: The World Cities. London. Heenan, D. 1977: Global cities of tomorrow. Harvard Business Review 55. Hooling,C.S.2001:Understandingthecomplexityofeconomic,ecologicalandsocialsystems.Ecosystems 4-5. DOI: https://doi.org/10.1007/s10021-001-0101-5 Huang, Y., Leung, Y., Shen, J. 2013: Cities and Globalization: An International Cities Perspective. Urban Geography 28-3. DOI: https://doi.org/10.2747/0272-3638.28.3.209 Internet 1: http://www.deloitte.com (10.9.2017) Internet 2: http://www.forbes.com (23.3.2019) Internet 3: https://www.lboro.ac.uk/gawc/gawcworlds.html (18.5.2019) Krätke, S. 2014: How manufacturing industries connect cities across the World: Extending research on ‘multiple globalizations’. Global Networks 14-2. DOI: https://doi.org/10.1111/glob.12036 Marino, D., Tebala, D. 2016: Measuring the creative province: A synthetic index for Italy. Procedia-Social and Behavioral Sciences 223. DOI: https://doi.org/10.1016/j.sbspro.2016.05.375 Morrison, D. 1967: Multivariate statistical methods. New York. Nazarczuk,J.M.,Umiński,S.2018:ThegeographyofopennesstoforeigntradeinPoland:Theroleofspe­cial economic zones and foreign-owned entities. Bulletin of Geography, Socio-economic Series 39. Neal, Z., Derudder, B., Taylor P. J. 2019: Should I stay or should I go: Predicting advanced producer ser­vicesfirmexpansionandcontraction.InternationalRegionalScienceReview42-2.DOI:https://doi.org/ 10.1177/0160017618784739 Páthy,Á.2017:TypesofdevelopmentpathsandthehierarchyoftheregionalcentresofCentralandEastern Europe. Regional Statistics 7-2. DOI: https://doi.org/10.15196/RS070202 Raźniak,P.,Dorocki,S.,Winiarczyk-Raźniak,A.2017:Permanenceoftheeconomicpotentialofcitiesbased onsectordevelopment.ChineseGeographicalScience1-27.DOI:https://doi.org/10.1007/s11769-017­0850-5 Raźniak, P., Dorocki,S., Winiarczyk-Raźniak, A. 2018a: Eastern European cities as command and control centres in time of economic crisis. Acta geographica Slovenica 58-2. DOI: https://doi.org/10.3986/ AGS.3124 Raźniak, P., Dorocki, S., Winiarczyk-Raźniak, A. 2018b: Zmiany funkcji kontrolno-zarządczych w mias­tach europejskich w dobie globalizacji. Studies of the Industrial Geography Commission of the Polish Geographical Society 32-3. Raźniak, P., Dorocki, S., Winiarczyk-Raźniak, A. 2019: Ewolucja funkcji kontrolno-zarządczych w mias­tach Europy Środkowej i Wschodniej, Prace Komisji Geografii Przemysłu Polskiego Towarzystwa Geograficznego 33-2. Raźniak,P.,Dorocki,S.,Winiarczyk-Raźniak,A.,PłaziakM.,Szymańska,I.A.2019:Changesinthefinan­cial results of corporations in cities in Eastern Europe. Acta Geobalcanica 5-1. DOI: https://doi.org/ 10.18509/AGB.2019.02 Raźniak,P.,Winiarczyk-Raźniak,A.2015:Didthe2008globaleconomiccrisisaffectlargefirmsinEurope? Acta geographica Slovenica 55-1. DOI: https://doi.org/10.3986/AGS.740 Sassen, S. 1991: The Global City: New York, London, Tokyo. Princeton. Shiller, R. J. 1991: Arithmetic repeat sales price estimators. Journal of Housing Economics 1-1. DOI: https://doi.org/10.1016/S1051-1377(05)80028-2 Śleszyński,P.2015:EconomiccontrolfunctionsinPolandin2013.GeographiaPolonica88-4.https://doi.org/ 10.7163/GPol.0041 Śleszyński, P. 2018: Research topics of geography of enterprise and decision-control functions in Poland againstglobaltrends. PraceKomisjiGeografiiPrzemysłuPolskiegoTowarzystwaGeograficznego32-4. Taylor, P. J. 2004: World City Network: A global urban analysis. London. Taylor P. J., Hoyler, M., Pain, K., Vinciguerra, S. 2014: Extensive and intensive globalizations: Explicating thelowconnectivitypuzzleofU.S.citiesusingaCity-Dyadanalysis.JournalofUrbanAffair36-5.DOI: https://doi.org/10.1111/juaf.12077 Taylor, P. J., Ni, P., Derudder, B., Hoyler, M., Huang, J., Lu, F., Pain, K., Witlox, F., Yang, X., Bassens, D., Shen, W. 2009: The way we were: Command and Control Centres in the global space-economy on the eve of the 2008. Environment and Planning A 41-1. DOI: https://doi.org/10.1068%2Fa41318 Wolman,H.,Hanson,R.,Hill,E.,Howland,M.,Ledebur,L.1992:NationalUrbanEconomicDevelopment Policy. Journal of Urban Affairs 14, 3-4. https://doi.org/10.1111/j.1467-9906.1992.tb00288.x MANAGEMENT OF SMALL RETENTION PONDS AND THEIR IMPACT ON FLOOD HAZARD PREVENTION IN THE SLOVENSkE GORICE HILLS Mateja Ferk, Rok Ciglič, Blaž Komac, Dénes Lóczy Retention pond at the former Benedictine monastery at Jareninski Dvor. DOI: https://doi.org/10.3986/AGS.7675 UDC: 911:556.18(497.41) 627.13:556.166(497.41) COBISS: 1.01 Mateja Ferk1, Rok Ciglič1, Blaž Komac1, Dénes Lóczy2 Management of small retention ponds and their impact on flood hazard prevention in the Slovenske Gorice Hills ABSTRACT: One of the methods of water resource management is to construct small retention ponds. Within the framework of the »Possible ecological control of flood hazard in the hilly regions of Hungary and Slovenia» project the management of small ponds and their impact on flood prevention were studied in selectedcatchments.Dataonpondmanagementweregatheredfrominterviewswithpondowners.Inapilot study, we conducted aninventarisation and classification of all retention ponds. Primarily they were con­structed for more specific use: fishing, irrigation, watering livestock. These functions have been gradually replaced by leisure-time activities, aesthetics, and tourism. Spring, stream and rainfall-fed ponds prevail in the pilot area and reduce the flood risk. Due to the increased variability of precipitation patternsponds are also becoming an important measure to limit drought consequences at a local level. KEYWORDS:hydrogeography,naturalhazards,floods,watermanagement,dams,detentionponds,Slovenia Upravljanje malih vodnih zadrževalnikov in njihov vpliv na poplavno varnost Slovenskih goric POVZETEK: Eden od načinov gospodarjenja z vodnimi viri je izgradnja majhnih zadrževalnih ribnikov. Vokviruprojekta»PrimerniekološkiukrepinapodročjupoplavnenevarnostivhribovitemobmočjuMadžarske inSlovenije«smopreučiliupravljanjemajhnihribnikovinnjihovvplivnapreprečevanjepoplavvizbranih porečjih.Podatkeoupravljanjuribnikasmozbralispomočjointervjujevzlastnikiribnikov.Vpilotništudi­jismoizvedliinventarizacijoinrazvrščanjezadrževalnihribnikov.Vglavnemsobilizgrajenizanamensko uporabo, kot je ribolov, namakanje, napajanje živine. Te funkcije so postopoma nadomestile prosti čas, estetika in turizem. Na pilotnem območju prevladujejo zadrževalniki, ki jih polnijo izviri, vodotoki in padavine,inzmanjšujejopoplavnoogroženost.Zaradivečjevariabilnostipadavinpostajajoribnikinalokalni ravni tudi pomemben ukrep za omejevanje posledic suše. KLJUČNE BESEDE: hidrogeografija, naravnenesreče, poplave, upravljanje voda, pregrade, vodna zajetja, Slovenija The paper was submitted for publication on November 12th, 2019. Uredništvo je prejelo prispevek 12. novembra 2019. 1 Research Centre of the Slovenian Academy of Sciences and Arts, Anton Melik Geographical Institute, Ljubljana, Slovenia mateja.ferk@zrc-sazu.si, rok.ciglic@zrc-sazu.si, blaz.komac@zrc-sazu.si 2 University of Pécs, Institute of Geography and Earth Sciences, Pécs, Hungary loczyd@gamma.ttk.pte.hu 1 Introduction Inthepast,watermanagementfocusedonbuildingandmanagingwatersupplyinfrastructure.Thisapproach brought into use large-scale centralized water storage infrastructure systems for irrigation, sewage, and energyproductionwhichwerealsousedforfloodcontrol.AccordingtothedataoftheWorldCommission on Dams (World register…2020) there were more than 50,000 dams globally in 2019. In Slovenia, there are47largedams(KomacandZorn2016).Thisso-calledhardpathwatermanagementhasimprovedhuman watersecurityworldwide.However,thediminishingcapacityofthehardpathtosolveemergingwaterprob­lemsledwatermanagerstoseeknewapproaches.Softpathsolutionsfocusoninstitutionalreforms,small-scale interventions,theintroductionofwater-efficienttechnologies,andthemanagementofagricultural,indus­trial, and residential water use. They can better address future water scarcity where hard path approaches havenotbeensuccessful(Wutichetal.2014).Retentionpondsonfluvialsystemsareausefulsoftpathapproach andcontributetowaterresourcemanagementbyinfluencingwaterdischargeandsedimenttransportdynam­ics (Verstraeten and Poesen 2001; Koskiaho 2003), and water chemistry (Fairchild and Velinsky 2009). Retention ponds can be divided by their size into two categories. Large ponds or reservoirs are flood retentiondams,floodcontroldams,waterretentionobjects,andsedimenttraps,whilesmallretentionponds are fishing ponds, watering holes, and pools. The dams of large retention ponds are mostly of concrete or combined construction, while small retention ponds involve simpler earthworks (Steinman and Banovec 2008; Table 1). In Slovenia, large ponds were mostly built for energy production, drinking and techno-logicalwaterstorage,floodanddroughtmanagement,andirrigationforfoodproduction(Širca2010).Small retention ponds, on the other hand, provide water for irrigation and support secondary uses, such as fish-ingand tourism. With large retention ponds, the detention time is from one to several years, while it lasts from one to several days in the small ones. InSlovenia,waterinfrastructuremanagementisbeyondindividualinterestsandisapublicutilityser­viceconcessedbytheSlovenianEnvironmentAgency.However,theListoftheexistingwaterinfrastructure excludes sediment retention objects, their inflow and outflow channels, and irrigation and drainage sys­tems, while only the barriers and dams are included (Seznam obstoječe…2006). The Waters Act (Zakon o vodah 2002) lists 40 dams as water infrastructure (Globevnik 2012) while the Rules to determine water infrastructure (Pravilnik o določitvi…2005) declare the formal status of water infrastructure, especially relatedtomaintenance.WaterinfrastructureispartofgeodeticdataandgovernedintheEUbytheINSPIRE Directive (Infrastructure for…2017). In Slovenia, the water infrastructure is governed by the National water management program, water management plans, remediation programs, and other water management programs. The National water management programdetermines watermanagementpolicy,aswell asthegoals,directives,and priorities Table 1: Types of water ponds according to the construction and type of distribution of water pressure (Steinman and Banovec 2008). for water use, protection, and management. At the catchment level, the period 2016–2021 is regulated by the Adriatic Sea Watershed Management Plan (Načrt upravljanja voda na vodnem območju Jadranskega…2016; Uredba o načrtih upravljanja 2016) and the Danube River Catchment Management Plan (Načrt upravljanja voda na vodnem območju Donave…2016; Uredba o načrtih upravljanja 2016). Water infrastructure objects are documented in the Water Register, the official record of the SlovenianWaterAgency(E-vode2019).Itencompasses55databasesonwater,includinghydrology,water typology, water areas, nature protection areas and flood hazard maps (Pravilnik o vodnem…2017). Although new data regarding hydrology and use of the water areas are available they lack coherence as they were created through desktop work, without any field research (Barborič et al. 2017). Significant differenceswereindicatedbetweentheofficialdocumentationonwaterinfrastructureandtheactualstate innature(Sodnik,KogovšekandMikoš2014).Thisespeciallyappliestosmallwaterinfrastructureobjects, such as ponds. Two regulations on retention ponds exist in Slovenia, but they originally support the management of large retention ponds and they are only applied to hydroelectricity dams. These are the Rules on the tech­nical monitoring of high water dams(Pravilnik o tehničnem…1966) and the Rules on the monitoring of seismicity in the area of large dams (Pravilnik o opazovanju…1999). The Instructions for preparing riskassessments for dam barrier failures (Lenart, Rajar and Širca 2017) were never passed in the Republic of Slovenia,sotoday’spracticeisbasedonanalmosthalf-a-century-oldYugoslavregulation(Uputstvo…1975). AttheEuropeanlevel,thecoredocumentregulatingretentionpondsistheManifestoonDamsandReservoirs (Manifesto…2015),whiletheKyotoProtocol(Worlddeclaration…2012)governstheissueatgloballevel. In general, the situation is rather bad as regards the comprehensive approach to planning, construc­tion, operation and safety rules for water dams in Slovenia. The rules are scattered across constructionandotherlegislation,welackorderlinessandevenacomprehensiveoverviewofthesituation(Širca,Ravnikar TurkandZadnik2010).Largeretentionponds(hydroelectricitydamsareexcluded)arepoorlymaintained andnotregularlyexamined.Pastconstantchangesandscatteredorganizationofretentionpondmanagement causedbadmanagementpractices.Archivaldatawereoftenlostandsometimeseventheconstructiondataare missing (Širca, Ravnikar Turk and Zadnik 2010). Many so-called sediment trap objects are not regu­larlycleanedandnolongerretainsediments(Papež2010),leadingtoerosion(Kračun2010)andincreasing flash flood hazard (Komac and Zorn 2011). Since their construction, the barriers have not been adjusted to the current hydrological, climate, and land use conditions (Zemeljske…2016), exposing the regions to combined and cascade disasters (Komac 2015). Therefore,thehazardintheSlovenskeGoriceHillsshouldnotbeignoredalthoughtheregionhaslower dams than other regions in Slovenia. Their less cohesive building material and structure need to be con-sidered.Asevenlargeretentionfacilitiesfaceseveralworryingissues,itisevenmorechallengingtoenforcelegislationforsmallretentionponds(Širca2010).Furthermore,incompleterecordsonsmallretentionponds and bodies of water are an important, even pressingissue, urgently calling for their comprehensive analy­sisandmanagement.Themanagementoflargedamsfacesnumerousissuesandhasbeenputtotheagendas ofdifferentnationalandinternationalorganizations.Smallretentionpondmanagement,ontheotherhand, encounters several challenges that have not been properly addressed yet. An important issue is the rapid fillingofthepondswithsedimentwhichincreasesmaintenancecosts(VerstraetenandPoesen1999;2000). Also, the improved accessibility of water in the last century for people and their livestock decreased the need for ponds in rural areas, consequently they are increasingly abandoned (Mioduszewski 2012). TheprojectentitledPossibleecologicalcontroloffloodhazardinthehillyregionsofHungaryandSlovenia is one of the attempts to address this question in the Pannonian Basin. It studies the suitability of ecolog­ical measures for decreasing floods hazard in the hilly regions of Eastern Slovenia. Namely, while sustainable reduction of flood risk can be achieved by large scale spatial planning and land use adapta­tioninthedownstreamrivervalleys,waterretentionareasinsmallbasinscaneffectivelylowerthefrequency offloods (Hooijer et al. 2004; Richert et al. 2011; Kijowska-Strugała and Bucała-Hrabia 2019). It has been establishedthatsmallretentionpondsareespeciallyeffectiveinpeakflowreductiononalocalscale(Chrétien etal.2016).Furthermore,studieshaveshowntheyhaveabeneficialimpactonlimitingerosion(Verstraeten and Poesen 1999; Koskiaho 2003), as well as improving runoff quality (Chrétien et al. 2016) and are an added ecological value of the environment (Mioduszewski 2012). In the research, we focused on ponds that are defined as small artificial structures to retain freshwa-ter.WeinvestigatedthemanagementofsuchpondsinthePesnicaRivercatchment(partofSlovenskeGorice Hills).Thedetailedanalysisoftheretentionpondsintworepresentativelower-rankedcatchments–Jarenina and Vukovje creeks consisted of spatial analysis and in-depth interviews. The aim of the paper is to pre­sent a comprehensive assessment of management practices of small ponds, and their impact on flood prevention in the Slovenske Gorice Hills. 2 Hydrological features of the Slovenske Gorice Hills The Slovenske Gorice Hills are a hilly region in north-east Slovenia (Perko 1998). The area is located in the west of the Pannonian Basin between the Drava River to the south and the Mura River to the north. The hills are composed of Neogene marine sediments: mostly clays, sandy marl, sandstone, and con­glomerates, with local outcrops of limestone (Belec 1998; Kert 1998). According to the calculations using a version of the Gavrilović equation according to Pintar, Mikoš and Verbovšek (1986), the annual sedi­ment production in Slovenske Gorice Hills is 1031.6m3/km2 or 16.5t/ha and the annual sediment yield is 639.7m3/km2 or 10.2t/ha (Hrvatin et al. 2019). Water flows quickly from the impermeable bedrock to the lowlands and the Slovenske Gorice Hills have a dense stream network (2.1km/km2). Relatively high precipitation and high temperatures contribute to high evapotranspiration during summer when many smallwatercoursesdryup.Specificrunoffanddischargecoefficientsarebelowaverageinthearea(Kolbezen 1998;Frantar2008a;2008b).Inrecentyears,asmallershareofsnowprecipitationhasincreasedwaterrunoffduring winter (Žiberna 2017). The share of forest below the Slovenian average covering about a third of thearea(Kert1998)decreaseswaterretentioninsourceareas.However,inthelasthundredyears,theland-scape has changed considerably: the share of forests has increased, replacing orchards and vineyards on steep slopes (Ciglič and Nagy 2019; Deriaz et al. 2019). In addition, modernization of agriculture led to terrace abandonment in viticulture (Pipan and Kokalj 2017). Small amount of precipitation, high evapo-transpiration,quickrunoff,andpoorretentioncapacityincreasethefrequencyofdroughts(Frantar2008a;2008b; Kozjek, Dolinar and Skok 2017; Žiberna 2017). The valley floors had often been flooded before the regulation of watercourses in the second half of the 20th century. These measures have decreased flood hazard in the valleys but as the streams have been changed by man, water runs off faster. Water discharge varies significantly; it rises during downpours and snowmeltsandlowersduringdroughtswhenthestreamsevendesiccate(Kert1998).Torrentialfloodsare common and occur during local downpours in summer and autumn (Trobec 2016). As noted elsewhere (Frantar and Hrvatin 2005; Kovačič 2016; Hrvatin and Zorn 2017), the precipitation trend in the period1961–2016ispositiveintheautumnandwintermonths(Žiberna2017).Theincreaseofprecipitationdur­ing the colder months with limited evapotranspiration poses a threat to flood security and increases the importance of maintaining small retention ponds for the future. Duetosignificantvariabilityofprecipitationandwaterdischarge,theareaoftheSlovenskeGoriceHills issubjecttohighuncertaintyofwatersupplymanagement:ononehand,itreceivesshort-lived,heavydown-pours and torrential floods with a quick water discharge, while on the other hand, long periods with very scarce and low amounts of precipitation occur leading to water shortages. This is why innovative man­agementpracticesareneededinordertoincreasewaterretentionduringdroughtsandpreventhighwater runoff. One of the measures for water retention in the periods of drought and for preventing runoff dur­ing the period of more abundant precipitation is ensuring proper land use (for example, by afforestation) andwaterinfrastructuremanagement.Thelatterinvolvesriverbanksmaintenanceandexcavatingorbuild­ing small and large water retention ponds. 2.1 The Pesnica River Valley ThePesnicaRiverValleyislocatedinthecentralpartoftheSlovenskeGoriceHills.Itrunsfromthenorth-west to the southeast collecting most of the waters in the area. The catchment is of asymmetrical shape withtheleft,north-easternbankmorehydrologicallydeveloped.Thevalleyfloorisafew100mwideabove ZgornjaKungota,anditswidthextendstoabout3kminthelowerreachesnearPtuj.Thevalleywasdevel­opedinMioceneclasticsediments,mainlysandstoneandmarl.Theflatvalleybottomisfilledwithfluvial deposits and was shaped by the river’s frequent floods in a wetland environment. In the past, a very low stream gradient –1.7‰ (Kobold 2012) caused the meandering of the river. This can be clearly observed on the Josephine Military Map (e.g. Zorn 2007) or on the map of the Franciscean Cadastre (Natek 1992; Gabrovec, Bičík and Komac 2019; Kladnik et al. 2019). The meandering river channel was channelized in the 1960s in order to support agriculture. Of its approximately 69km course (65km on the Slovenianter­ritory), about 50km were regulated. In order to prevent flash floods about 90km of the tributaries were alsoregulated(Juvanetal.1997).Intotal,13%ofthestreamnetworksurfacehasbeenmeliorated(Leitinger 2012) and only the headwaters of some small tributaries and small narrow valleys remained undisturbed. With channelization the wetland was converted to farmland, on the other hand, channelization increas­es flood hazard (Lóczy, Kis and Schweitzer 2009; Lóczy and Dezső 2013). The flood hazard prevention measures in the Pesnica River Valley included the building of several large retention ponds. The largest retentionpondalongthePesnicaRiveristhePernicaaccumulationlake(Figure1).Itconsistsoftwoparts; the Pernica 1 and Pernica 2 retention ponds, divided by a dam and afloodgate. The Jarenina and Vukovje creeksflowintothePernica1retentionpond.ThePesnicaRivercontributeswaterintothePernica2reten­tion pond. Also the Pristava retention pond lies on the Pesnica River, while all other retention ponds are located on its tributaries (Figure 2). Figure 1: Pernica accumulation lake in 2006 and 2016. 2.2 Spatial analysis of the Jarenina Creek and Vukovje Creek catchments The catchments of Jarenina and Vukovje Creeks are similar by their average elevation (306m and 318m, respectively)whichisabout50mhigherthantheaverageelevationoftheSlovenskeGoriceHills.Theval­ley floors lie at an altitude of 250m, while the highest peaks of the hills exceed 400m in the north-western part of the JareninaCreek catchment. Slope gradients, which were calculated with a 5m resolution, range between 13° and 14°, with a maximum of about 50°. Landslide susceptibility level on a scale from 0 to 5 (after the landslide index method; Zorn and Komac 2008) is around 3, while areas with the highest pos­sible level of landslide susceptibility (scale level 5) can be found in both catchments. AccordingtotheSlovenianEnvironmentAgency,theaveragetemperaturewas–1.1°CinJanuaryand 20.1°C in July in the period from 1981 to 2010. The area gets about 980mm of precipitation on average, withjustunder4200MJ/m2ofinsolation.Annualsnowcoverspansfrom42to56days.Theaverageevap­otranspiration is just over 630mm and, on average, about 350mm of water drains from the area. The net groundwater recharge is about 80mm on average but it can vary significantly. The variation coefficient of groundwater recharge is 82% in the Jarenina Creek catchment and 52% in the Vukovje Creek catch-ment.Partsofthepilotareaexperiencesoilwatershortage(over80%)upto20daysormore.Bothcatchments have watercourses of the first, second, and third order (Figure 6). Land use is similar in both catchments. About one-third of the area is covered by meadows, followed by forests and arable fields. The percentage of the forested area is lower in the Jarenina Creek catchment than in the Vukovje Creek catchment, while the percentage of arable land is somewhat higher. The fourth categoryisbuilt-upareas,whileotherareasincludevineyards,permanentcrops,andovergrownareas.There are only a few tenths of a percent of water surfaces and wetlands. The catchments considered have identical natural geographical features. However, it was confirmed that the Jarenina Creek catchment has been more reshaped by human activity. This is confirmed by the higher share of built-up areas and agricultural land use. However, today there is more forest than in the firsthalfofthe19th centuryandfewerfieldsandvineyards(Deriazetal.2019;GabrovecandKumer2019). 3 Methods WeselectedsevensmallretentionpondsinthePesnica Rivercatchmentto conductdetailedanalyses using geoinformationtools andstructuredinterviewswiththewaterpondowners(Figure2). Astructuredinter-view(ŠmidHribarandLedinekLozej2013;PipanandKokalj2017)isatechniqueforthesystematicgathering ofverbalinformation.Itwasusedtorecordopinionsanddeterminetheinterviewee’sposition(Nared2007) on the selected examples of anthropogenic bodies of water in order to analyze their common features. We studied theseexamplesto determine themain characteristicsof themanagementsystemofsmallretention ponds,theirversatility,andchallenges.Theinterviewswerestructuredwiththefollowingsequenceofquestions: • Who is the landowner? • Who manages the water body? • What year was the pond established? • What was it primarily used for? • What is its current purpose? • What is the depth of the water? • What is the speed of sediment accumulation and how often does it have to be removed? • Do the pond banks have to be maintained? • Has the pond ever been (over)flooded? The results enabled us to evaluate the sustainability of the ponds from the management perspective (i.e. maintenance efforts, the quantity of sediment input, primary purpose, frequency of flooding). Basedonfieldobservationsandtheinformationgainedfromtheinterviews,apilotstudywasdesigned tomapandclassifyalltheretentionpondsinthecatchmentsofthetwoPesnicaRivertributaries:theJarenina and Vukovje Creeks which both contribute to the Pernica 1 retention pond (Figures 1 and 2). Alltheretentionpondsinthepilotareaweremappedinthegeographicinformationsystemandapond inventory was created. The collected data included: • Catchment name; • Location (coordinates); • ID number and name of the pond; • Pond surface area (m2); • Description of the hinterland – inflow; • Description of the area of the water reservoir – outflow. Wecalculatedthepercentageofsmallretentionpondsthatareconstructedonthemainwatercourseand influence water discharge and sediment yield anda percentage of ponds that work with a bypass channel. 4 Results and discussion 4.1 Interviews with the pond owners The structured interviews with the owners of the small retention ponds served as an insight into the com­mon challenges regarding their management observed in previous studies: from legislation issues (Širca 2010) to costs related to maintenance (Verstraeten and Poesen 1999; 2000; 2001) and the motivation to maintain or abandon the ponds (Mioduszewski 2012). The results show that in the Pesnica River catch­ment the ponds are usually privately owned and the owners manage them by themselves. Although they are responsible for the operative as well as financial aspects of management they mostly lack experience with any possible authorities responsible for water-related topics. Understandably, since also the review oflegislationaboutsmallretentionpondsinSloveniashowedtherearenospecificandclearlawsandrules (they apply only for large retention ponds). In most cases, the uncertainties and confusion of the legisla­tion do not create problems for individual landowners to establish private ponds. In only one case, the interviewee reported an issue with acquiring a permit to remove sediments from his retention pond in the Vukovje Creek catchment. Consequently, the pond was abandoned and is currently empty. Figure 2: Location of small retention ponds included in the interviews with the pond owners and the pilot study area. Based on their primary function when they were created and also their current function the ponds canbedividedintotwomaingroups:fishingponds(commercialandnon-commercial)andirrigationponds. Some retention ponds have existed for hundreds of years. They were made for aesthetic reasons and for fish farming near castles (e.g. Hrastovec Castle; Figure 3), countryside mansions (e.g. former Rittersberg in Spodnji Jakobski Dol), or monasteries (e.g. the Benedictine Monastery at Jareninski Dvor; Figure 4). They can be distinguished from most other ponds by their larger size, which is a consequence also of the favorable environmental conditions: they were created in natural stream valleys, at large natural springs, and in lower basins with permanently high groundwater levels. Through time and the changes in own­ership, their function was adapted to the owners’ needs. However, the ornamental and fishing ponds have mostly preserved their original function to this day. Decades ago, fish cultivation was an additional source of income for farmers in the Slovenske Gorice Hills.Nowadays,thefinancialimpactisminimal,leadingmanyoftheownerstoabandonthefishingponds. According to the interviewees, the lower income from fish sales in the past decade is most likely a conse­quenceofthedominatinglow-pricedproductsbymajormanufacturersonthemarket.Theirobservations are in line with the findings at the European Union level where the economic performance of fish farm­ing was linked to the heavy global competition but also to market requirements for the constant supply and quality with guaranteed environmentally-friendly production chains (Review of the EU…2009). Competingwithsuchrequirementsisimpossibleforindividual,non-alignedfarmers.Furtherfactorsneg­ativelyinfluencingaquaculturedevelopmentarewateruserconflictsandincreasinglycomplexregulations (ReviewoftheEU…2009).Consequently,commercialfishfarmingwillbecomeeconomicallybeneficialonly if it is integrated into national strategic development plans, providing the farmers with economic stimula­tionsandlegaladvice,andconnectingthemintolocalproductionchains(Adámek,MosserandHauber2019). Theothergroupofretentionpondsisintendedforirrigationtomaximizeorstabilizecropyields.They were constructed in the 1990s when state subventions were made available. The irrigation water is used in orchards, vineyards and private gardens. The interviews revealed that the pond maintenance costs for the owners are comparable with the gains of irrigation. Therefore, the owners are looking for additional possibilities to use the existing ponds, otherwise, the ponds will be abandoned. Consequently, the trend of retention pond use is shifting towards an increasingly multi-functional role: irrigation, watering live­stock, tourism with non-commercial sports fishing, and the aesthetic function is gaining in importance. Also, other research has shown that public awareness of the benefits of multi-functional ponds encour­ages local people to properly maintain them (Oda et al. 2019). Due to increased climate variability and Figure 3: The tradition of fishing in the castle ponds at Hrastovec Castle has been maintained for several centuries. change in precipitation patterns (see chapter 2) the need for irrigation ponds will most likely increase in thefuture.Furthermore,smallon-farmpondsareamoresustainablewatersource,comparedtolargescale groundwater extraction for irrigation (Sanfo et al. 2017; Vico, Tamburino and Rigby 2020). Usually,retentionpondsareconstructedinareasoflocalspringsandstreams.Consequently,constructing them requires simply digging out or deepening a small basin in the valley floor, with minimal construc­tion of barriers or dams. Simple dams are constructed by piling up clay sediment and fortified in places withwoodenstakes.Someinterviewedpondownersstatedthepondswereconstructedmanydecadesago by previous landowners. In such cases, the knowledge about motivation and reasons for constructing the ponds at a specific location in a specific way is lost. ThebedrockoftheSlovenskeGoriceHillsispronetoquickweatheringanderosion.Furthermore,despite the fact that these shallow retention ponds are no more than 5m deep, they do not need to be frequently cleaned as the accumulation of sediment is still slow. The owners remove the sediment from the ponds onlyonceinevery10to30years.Thereareseveralreasonsforthat,demonstratingthedeepunderstanding and knowledge of local people about their environment: • The retention ponds are usually located in areas with low inclination slopes, where the torrential char­acter of the watercourses is decreased. • Low sediment accumulation occurs also because the retention ponds were generally created above the main watercourses and are not subject to flooding. • Many ponds are built on springs or small tributaries transporting low amounts of bedload. • When the ponds are built next to the streams, to minimize pond sedimentation, the main watercourse iscommonlydivertedtobypassthepond(Figure5).Theinundationofpondsiscontrolledbywaterchan­neled from the main watercourse which transports significantly less bedload material. In this way, the owners avoided possible damage during floods and high sediment flow along the main streams. Figure 4: Fishing pond (marked with red rectangle) at the Benedictine Monastery at Jareninski Dvor as shown on the Franciscean Cadastre from the early 19th century (Franciscejski kataster 1824). These adaptation methods helping the owners to more efficiently manage and maintain the ponds, influence also the effect of the ponds as possible natural ecological measures for preventing floods. For this reason, we analyzed the percentage of retention ponds in the pilot areas that are located directly on thestreamnetworkandthenumberofpondsthatarephysicallyseparatedfromthewatercoursesandonly filled with groundwater or precipitation (chapter 4.2). Sincetheissueofpondsedimentationwasfrequentlyalreadylimitedbycreatingspecificresilientpond types, some interviewed pond owners reported the removal of aquatic weeds from the retention ponds as themosttimeandcostconsumingmanagementissue.Overgrowthbyvegetationcausesproblemsbecause italterstheecologicalconditionsinthepondsveryquicklybyinducingeutrophication,overgrowthofthe surface with algae, leading to low visibility and low oxygen level (Vanacker et al. 2016). Such changes in water properties are harmful to fish populations and, consequently, need to be avoided especially in fish-ingponds(Vanackeretal.2016;Adámek,MosserandHauber2019).Thisiswhythevegetationintheretention ponds, and especially on their banks, has to be regularly removed every few years (Figure 5). 4.2 Pond inventarisation in the catchments of Jarenina and Vukovje Creeks Atotalof41artificialstructurestoretainfreshwater(i.e.retentionponds)weredetectedduringfieldmap-ping in the Jarenina Creek and Vukovje Creek catchments (Figure 6, Table 2). The identified ponds cover a total area of 15,014.73m2 and the average pond size is 366.21m2. Despite the comparable size of both catchments, there are significantly fewer ponds in the Jarenina Creek catchment (15 ponds) than in the Vukovje Creek catchment (26 ponds). However, the largest pond of all (JC1 – 5,125m2) and the average pondsizeintheJareninaCreekcatchment(642.81m2)considerablyexceedthepondsintheVukovjeCreek catchment (average size is 206.64m2). Also, the total area of ponds in the Jarenina Creek catchment (9,642.10m2) is larger than in the Vukovje Creek catchment (5,372.63m2). Figure 5: Vegetation is removed from the banks of the retention ponds every few years. White arrow is indicating the location of the diversion channel which diverts the excess water and bedload around the pond. Ponds differ according to the material and method of construction, their source of water, and the way in which water can be drained from the pond (Fish pond…2005). Additionally, the possible influence of ponds to decrease flood hazard was evaluated. Figure 6: The catchment of Jarenina and Vukovje creeks with the location of small retention ponds classified according to their influence on flood hazard reduction. The figure is showing also the drainage basin network with the order classification (Strahler 1957). The data on the stream network was acquired from the Surveying and Mapping Authority of the Republic of Slovenia. All identified ponds in both catchments were built using local natural material (i.e. gravel, sand, clay, soil,andwood)andcanbedefinedassimpleearthenponds.Noconcreteormetalconstructionswereused. Accordingtotheconstructionmethods,thepondsareeitherdug-outpondsorcut-and-fillponds(Table 3). The dug-out ponds (also sunken ponds) are constructed in flat areas by excavating bedrock material to formaholeintheground(Fishpond…2005).Theexcavatedmaterialcanbeusedtoadditionallystrength-enorrisepondwallsabovethesurroundingsurface.Inbothcatchments15ponds(37%)aredug-outponds: 5 ponds (33,3%) in the Jarenina Creek catchment, and 10 ponds (38,5%) in the Vukovje Creek catchment. The cut-and-fill ponds (also barrage ponds) are constructed on slopes by the excavation of the bedrock material and using it to embank the pond on the downslope side (Fish pond…2005). In this way, a bar-rierordamisbuilttoretainthewater.Inbothcatchments,26ponds(63%)arecut-and-fillponds:10ponds (66,6%) in the Jarenina Creek catchment, and 16 ponds (61,5%) in the Vukovje Creek catchment. Accordingtothesourceofwaterrecharge,thepondscanbefedbygroundwaterorsurfacewater(Fish pond…2005). The groundwater can flow to the pond as seepage from the matrix porosity of the bedrock (the level of water will vary with the groundwater-table) or from a spring in or close to the pond (the level of water will vary according to the wet/dry seasons). The surface water can be supplied by rainfall or from surface run-off (e.g. surface stream) both dependent on the wet/dry seasonality. All identified ponds are fed by a combination of water sources (e.g. rainfall and seepage affect all ponds). However, for analytical reasons, the main source of water for each pond was considered in the study (Table 3). The most common source of water in all identified ponds is groundwater (73%): 67% of ponds in the Jarenina Creek catchment and 77% of ponds in the Vukovje Creek catchment. Approximately two-thirds of the groundwater and 46% of all the water in both catchments comes from springs. Only three ponds in each catchment are supplied by surface water coming from a stream (15% of all water supply). 12% of the ponds are mainly supplied by rainfall. Like it was discussed through the interviews with pond own­ers (see chapter 4.1), also the pond inventarisation confirmed that the pond location (from the aspect of water supply) is well adjusted to the environmental conditions of the Slovenske Gorice Hills. Ponds fed byseepageandspringwatertransportaninsignificantamountofbedloadandneedlessmaintenance.From theaspectofsustainability,pondsfedbygroundwater(seepageorsprings)havealoweramplitudeofwater-table variability and are less affected byprecipitation seasonality. In rural areas, the reliability of the water sources was especially important before the construction of the water distribution system and remained important for livestock and plantation farmers until the present. Moreover, the importance of sustainable small ponds is increasing due to the current climate variability (Sanfo et al. 2017; Vico, Tamburino and Rigby 2020). Another method (used also by some interviewed pond owners) to prevent flooding and/or filling of ponds with bedload transported by streams is to construct diversion channels to divert excess water and bedload around the pond (Fish pond…2005). Out of 6 ponds fed by a surface stream, 4 ponds (2 in each Table 2: Pond inventory of Jarenina (JC) and Vukovje Creek (VC) catchments showing the area of the identified ponds. ID area (m2) ID area (m2) ID area (m2) JC1 5125.00 VC1 929.00 VC16 312.00 JC2 801.00 VC2 117.00 VC17 346.00 JC3 1103.00 VC3 91.20 VC18 113.00 JC4 125.00 VC4 301.00 VC19 204.00 JC5 39.00 VC5 64.90 VC20 440.00 JC6 31.40 VC6 120.00 VC21 263.00 JC7 250.00 VC7 33.00 VC22 138.00 JC8 837.00 VC8 6.23 VC23 118.00 JC9 271.00 VC9 150.00 VC24 83.80 JC10 557.00 VC10 129.00 VC25 141.00 JC11 109.00 VC11 267.00 VC26 532.00 JC12 98.40 VC12 54.50 JC13 88.60 VC13 166.00 JC14 124.00 VC14 106.00 JC15 82.70 VC15 147.00 Table 3: Pond classification by construction method, water supply, and water outlet. The table shows the ID values of identified ponds in the Jarenina Creek catchment (JC1, JC2,…JC15) and Vukovje Creek catchment(VC1, VC2,…VC26). GROUNDWATER SURFACE WATER Seepage Spring Rainfall Surface stream Surface channel Underground Pumped Undrainable channel DUG-OUT POND 8, 9, 13 / 5, 12 / / 8, 9 5, 12 13 CUT-AND-FILL POND 2 6, 7, 10, 11, 14, 15 / 1, 3, 4 3, 4, 14, 15 1, 2, 6, 7, 10, 11 / / DUG-OUT POND 4, 5, 6, 9, 24, / 2, 8, 13 / 6 4, 9, 26 2, 8, 13 5, 24, 25 25, 26 CUT-AND-FILL POND / 1, 7, 11, 12, 15, / 3, 10, 14 3, 14, 19 1, 7, 10, 11, 12, 18 / 16, 17, 18, 19, 15, 16, 17, 20, 20, 21, 22, 23 21, 22, 23 catchment;JC3andJC4,andVC3andVC14)haveadiversionchannel.BothpondsintheVukovjeCreek catchmentarecurrentlyoutofoperation(surroundedbymeadowsandpartlyovergrownbybush).Weassume they are several decades old and that the diversion channels played a greater role in the past when they wereusedaswateringholesforlivestock.BothpondsintheJareninaCreekcatchmentwerebuiltinthelast twodecadesandhavecurrentlyanaestheticandtourismfunction.Thediversionchannelsforbothponds are regularly cleaned and maintained. Land use changes (afforestation reflecting the changes in farming economy)alteredthefunctionofthemajorityofponds.Theirmaintenancestronglydependsontheaware­nessofpondownersofthepossiblebenefitsofmulti-functionalponds:irrigation,tourism,andaesthetics. Considering the supply of water, the surface stream-fed ponds have the most direct impact on water and sediment discharge (Koskiaho 2003; Muendo et al. 2014; Chrétien et al. 2016) by slowing down the run-off and consequently on flood reduction. This effect is lessened by the diversion channels, especial­ly for the bedload which can be transported around the ponds. For water retention in the source area of the catchments also rainfall and spring-fed ponds are important as both types are correlated to precipi­tation patterns; some excess water is stored in the source area during peak flows decreasing fast drainage towardslowervalleys(Koskiaho2003;Chrétienetal.2016).Stream,rainfall,andspring-fedpondsarebecom­ing more and more important also for limiting on the impacts of drought at individual and local levels due to increased variability of precipitation patterns (Oda et al. 2019; Vico, Tamburino and Rigby 2020). Ponds fed through seepage do not influence water discharge (i.e. movement of groundwater). Moreover, during floods, these ponds can be completely submerged and are not impacting the extent of floods or sediment transport. On the other hand, because they are excavated below the water-table they represent vulnerablelocationsforchemicalorbiologicalcontaminationofthegroundwater(LóczyandDezső2013). For sustainable management of ponds also the outlet of water is important which can be drainable or undrainable (Table 3). Drainable outlets can be driven by gravitation (surface streams and underground channels) or the water can be mechanically pumped from the pond. Gravitational outlets are typical for cut-and-fill ponds supplied by springs and surface streams. Mostly they have controlled outlets because they were built by embankment (e.g. barrier or dam) through which a surface or underground channel enablesthedrainageofexcesswater.IntheJareninaCreekcatchment12outof15pondshaveacontrolled surface (4 ponds) or an underground (8 ponds) outlet channel. In the Vukovje Creek catchment, 19 out of26pondshaveacontrolledsurface(4ponds)oranunderground(15ponds)outletchannel.Thesetypes ofoutletsdirectlyinfluencedrainagedownstream(e.g.discharge,bedloadandsuspendedsediment,aswell as water quality). Therefore, their proper management reduces flood and drought risks. All ponds fed by rainfall (5 ponds in both catchments) were excavated on higher ground above the valley floor but have nosurfaceorundergroundoutlet.However,theycanbeemptiedbypumpingoutwater,ifneeded.Ofaspe­cialtypearepondsfedbyseepage,whicharedeepenedbelowthesurroundingsurfaceandcannotbedrained (watertabledictatesthewaterlevelofthepond).Asmentionedbefore,theyhavethelowestimpactonflood reduction but are the most vulnerable point for water pollution, e.g with increased nutrient presence (Ilić and Panjan 2018). 5 Conclusion The paper discusses small water ponds in north-eastern Slovenia on the example of the Slovenske Gorice Hills. The main conclusion is that small retention ponds, unlike large ponds, are not included in the regional, national and international legislation and the management is left to the landowners. Poor and uncoordinated management strategies and low investments contribute to the state of the water ponds and furtherminimizetheirpossibleuseinfloodprotection.Wealsonotedthatthelong-termlandscapechanges, especially land use changes, considerably alter the functioning of the small ponds as an important part of the hydrological system. Inthe last decades, afforestation, related to the changes in the farming economy, lowered the sediment input and shifted the prevailing use of water ponds from commercial fishing, irri­gationandwateringlivestocktootherfunctions,suchastourism,non-commercialfishing,andaesthetics. More than half of the retention ponds in the catchments of Jarenina and Vukovje creeks are supplied bygroundwater(throughseepageorsprings)whichincreasesthesustainabilityofpondsbecauselesseffort is needed to maintain them (low sedimentation rates, reliable source of water). Only 15% of the ponds are suppliedbysurfacestreamsand12%byrainfall.Thepondsconstructedonsurfacestreamslargelyimpact the water discharge and sediment transport and reduce the flood hazard. However, their maintenance is moredemandingthanthatofotherpondtypes.Tocopewithsedimentinfillingandavoidpeakflowdam­agestoponds,diversionchannelswerebuiltontwo-thirdsofsuchponds,divertingthemajorityofbedload and excess water around the ponds. This technique lowers the maintenance efforts and increases pond sustainability, however, it lowers the pond impact on flood hazard reduction. Forwaterretentioninthesourceareaofthecatchmentsbothrainfallandspring-fedpondsareimpor­tant as both types are correlated to precipitation patterns; some excess water is stored in the source area during peak flows decreasing fast drainage towards lower valley sections. Stream, rainfall, and spring-fed ponds are due to increased variability of precipitation patterns becoming more and more important also for limiting drought consequences at an individual and local level. Ponds fed by seepage have an egligi­ble impact on flood risk reduction . Nevertheless, they are vulnerable locations for chemical or biological contamination of the groundwater because they are deepened below the local water-table level. Since the owners’ investments to maintain the retention ponds are relatively high compared to their economic benefits, the owners often choose to abandon retention ponds. Although retention ponds do not prevent floods, their future abandonment would destabilize the hydrological and agriculture system and increase flood risk and drought impact in the area. ACKNOWLEDGEMENTS: The authors acknowledge the study was performed in the frame of the pro­ject Possible ecological control of flood hazard in the hilly regions of Hungary and Slovenia. The project was financially supported by the Slovenian Research Agency (ARRS, N6-0070) and the Hungarian National Research, Development and Innovation Office (NKFIH, SNN 125727 and the programme Excellence in Higher Education, Theme II. 3. »Innovation for sustainable life and environment«). The study was per­formed also in the frame of a research programme Geography of Slovenia (ARRS, P6-0101). 6 References Adámek, Z., Mösser, M., Hauber, M. 2019: Current principles and issues affecting organic carp (Cyprinus carpio) pond farming. Aquaculture 512. DOI: https://doi.org/10.1016/j.aquaculture.2019.734261 Barborič,B.,TriglavČekada,M.,Bric,V.,Kete,P.,DežmanKete,V.2017:Novepodatkovnepodlagezaboljše upravljanje z vodami. Naravne nesreče 4. Ljubljana. Belec, B. 1998: Panonski svet. Slovenija, pokrajine in ljudje. Ljubljana. Chrétien, F., Gagnon, P., Thériault, G., Guillou, M. 2016: Performance analysis of a wet-retention pond in a small agricultural catchment. Journal of Environmental Engineering 142-4. DOI: https://doi.org/ 10.1061/(ASCE)EE.1943-7870.0001081 Ciglič,R.,Nagy,G.2019:Naturalnessleveloflanduseinahillyregioninnorth-easternSlovenia.Geografski vestnik 91-1. DOI: https://doi.org/10.3986/GV91101 Deriaz, J., Ciglič, R., Ferk, M., Loczy, D. 2019: The influence of different levels of data detail on land use changeanalyses:acasestudyofFranciscanCadastreforapartofthePannonianHills,Slovenia.European countryside 11-3. DOI: https://doi.org/10.2478/euco-2019-0019 E-vode, 2019. Medmrežje: http://www.evode.gov.si/index.php?id=1 (4.11.2019). Fairchild,G.W.,Velinsky,D.J.2009:Effectsofsmallpondsonstreamwaterchemistry.LakeandReservoir Management 22-4. DOI: https://doi.org/10.1080/07438140609354366 Fish pond construction and management, 2005. A field guide and extension manual. FranciscejskikatasterzaŠtajersko,k.o.Vajgen[Weigen],listB03.ArhivRepublikeSlovenije.Ljubljana,1824. Frantar, P. 2008a: Odtočni količniki. Vodna bilanca Slovenije 1971–2000. Ljubljana. Frantar, P. 2008b: Specifični odtoki 1971–2000. Vodna bilanca Slovenije 1971–2000. Ljubljana. Frantar,P.,Hrvatin,M. 2005:PretočnirežimivSlovenijimedletoma1971in2000. Geografskivestnik 77-2. Gabrovec,M.,Bičík,I.,Komac,B.2019:Landregistersasasourceofstudyinglong-termland-usechanges. Acta geographica Slovenica 59-2. DOI: https://doi.org/10.3986/AGS.7349 Gabrovec, M., Kumer, P. 2019: Land-use changes in Slovenia from the Franciscean Cadaster until today. Acta geographica Slovenica 59-1. DOI: https://doi.org/10.3986/AGS.4892 Globevnik, L. 2012. Vzdrževanje vodne infrastrukture in vodotokov – pomen, realnost in perspektive. 1. kongres o vodah Slovenije, 22. marec 2012. Ljubljana. Hooijer, A., Klijn, F., Pedroli, G. B. M., Van Os, A. G. 2004: Towards sustainable flood risk management in the Rhine and Meuse river basins: synopsis of the findings of IRMA-SPONGE. River Research and Applications 20. DOI: https://doi.org/10.1002/rra.781 Hrvatin, M., Ciglič, R., Lóczy, D., Zorn, M. 2019: Določanje erozije v gričevjih severovzhodne Slovenije z Gavrilovićevo enačbo. Geografski vestnik 91-2. DOI: https://doi.org/10.3986/GV91206 Hrvatin,M.,Zorn,M.2017:TrendipretokovrekvslovenskihAlpahmedletoma1961in2010.Geografski vestnik 89-2. DOI: https://doi.org/10.3986/GV89201 Ilić, D., Panjan, J. 2018: Nitrogen and phosphorus pollution in Goričko Nature Park. Acta geographica Slovenica 58-1. DOI: https://doi.org/10.3986/AGS.727 InfrastructureforSPatialInfoRmationinEurope–INSPIRE.Brussels,2017.Internet:https://inspire.ec.europa.eu/ about-inspire/563 (21.10.1019). Juvan,S.,Hojnik,T.,Kaligarič,S.,Trop,M. 1997:PresojavplivovnačrtovanegapotekaACpopesniškidolini na vodni režim in vodno okolje. Mišičev vodarski dan, zbornik referatov. Maribor. Kert, B. 1998: Slovenske gorice. Slovenija, pokrajine in ljudje. Ljubljana. Kijowska-Strugała,M.,Bucała-Hrabia,A.2019:Floodtypesinamountaincatchment:TheOchotnicaRiver, Poland. Acta geographica Slovenica 59-1. DOI: https://doi.org/10.3986/AGS.2250 Kladnik,D.,Geršič,M.,Pipan,P.,VolkBahun,M.2019:Land-usechangesinSlovenianterracedlandscapes. Acta geographica Slovenica 59-2. DOI: https://doi.org/10.3986/AGS.6988 Kobold, M. (ur.) 2012: Hidrološki letopis Slovenije 2009. Ljubljana. Kolbezen, M. 1998: Kopenske vode. Geografski atlas Slovenije. Ljubljana. Komac, B. 2015: Modeliranje obpotresnih pobočnih procesov v Sloveniji. Geografski vestnik 87-1. DOI: https://doi.org/10.3986/GV87107 Komac, B., Zorn, M. 2011: Vloga zavarovanih območij pri blažitvi naravnih nesreč. Razvoj zavarovanih območij v Sloveniji. Regionalni razvoj 3. Ljubljana. Komac, B., Zorn, M. 2016: Naravne in umetne pregrade ter z njimi povezani hidro-geomorfni procesi. Geografski vestnik 88-2. DOI: https://doi.org/10.3986/GV88204 Koskiaho, J. 2003: Flow velocity retardation and sediment retention in two constructed wetland–ponds. Ecological Engineering 19-5. DOI: https://doi.org/10.1016/S0925-8574(02)00119-2 Kovačič, G. 2016: Discharge trends of the Adriatic Sea basin rivers in Slovenia, excluding the Soča river basin. Geografski vestnik 88-2. DOI: https://doi.org/10.3986/GV88201 Kozjek, K., Dolinar, M., Skok, G. 2017: Objective climate classification of Slovenia. International journal of climatology 37-S1. DOI: https://doi.org/10.1002/joc.5042 Kračun, M. 2010: Analiza zamuljevanja akumulacije Pernica z vidika njenega namena. Varnost pregrad vSloveniji.12.posvetovanjeSlovenskeganacionalnegakomitejazavelikepregrade(SLOCOLD).Krško. Leitinger,V.2012:HidrografskeznačilnostiporečjarekePesnicespoudarkomnavodnogospodarskihure­ditvah. Diplomsko delo, Fakulteta za humanistične študije Univerze na Primorskem. Koper. Lenart, S., Rajar, R., Širca, A. 2017: Prispevek projekta VODPREG 2 k oceni porušitev vodnih pregrad. Trajnostni razvoj mest in naravne nesreče. Naravne nesreče 4. Ljubljana. Lóczy,D.,Dezső,J.2013:Groundwaterfloodinghazardinrivervalleysofhillregions:exampleoftheKapos River, Southwest-Hungary. Hungarian Geographical Bulletin 62-2. Lóczy, D., Kis, E., Schweitzer, F. 2009: Local flood hazards assessed from channel morphometry along the Tisza River in Hungary. Geomorphology 113. DOI: https://doi.org/10.1016/j.geomorph.2009.03.013 Manifesto: Dams & Reservoirs 2015. Issue 1. European ICOLD Club. October 2015. Mioduszewski, W. 2012: Small water reservoirs – their function and construction. Journal of Water and Land Development 17-VII-XII. Muendo, P. N., Verdegem, M. C. J., Stoorvogel, J. J., Milstein, A., Gamal, E., Minh Duc, P., Verreth, J. A. J. 2014: Sediment accumulation in fish ponds; Its potential for agricultural use. International Journal of Fisheries and Aquatic Studies 1-5. NačrtupravljanjavodanavodnemobmočjuDonavezaobdobje2016–2021,2016.Internet:https://www.gov.si/ assets/ministrstva/MOP/Dokumenti/Voda/NUV/63dbe4066b/NUV_VOD.pdf (5.10.2019). Načrt upravljanja voda na vodnem območju Jadranskega morja za obdobje 2016–2021, 2016. Internet: https://www.gov.si/assets/ministrstva/MOP/Dokumenti/Voda/NUV/4195091b63/NUV_VOJM.pdf (28.2.2020). Nared, J. 2007: Prostorski vplivi slovenske regionalne politike. Geografija Slovenije 16. Ljubljana. Natek, K. 1992: Franciscejski kataster in geografski informacijski sistem. Traditiones 21. Oda,T.,Moriwaki,K.,Tanigaki,K.,Nomura,Y.,Sumi,T.2019:Irrigationpondsinthepast,present,andfuture: A case study of the Higashi Harima Region, Hyogo Prefecture, Japan. Journal of Hydro-environment Research 26. DOI: https://doi.org/10.1016/j.jher.2018.11.004 Papež, J. 2010: Obseg in pomen rednega urejanja voda. Varnost pregrad v Sloveniji. 12. posvetovanje Slovenskega nacionalnega komiteja za velike pregrade (SLOCOLD). Krško. Perko, D. 1998: The regionalization of Slovenia. Acta geographica 38. Pintar, J., Mikoš, M., Verbovšek, V. 1986: Elementi okolju prilagojenega urejanja vodotokov: alternativa utesnjevanju živih naravnih procesov v toge objekte. Drugi kongres o vodama Jugoslavije. Beograd. Pipan, P. Kokalj, Ž. 2017: Transformation of the Jeruzalem Hills cultural landscape with modern vineyard terraces. Acta geographica Slovenica 57-2. DOI: https://doi.org/10.3986/AGS.4629 Pravilnik o določitvi vodne infrastrukture. Uradni list Republike Slovenije 46/2005. Ljubljana. Pravilnikoopazovanjuseizmičnostinaobmočjuvelikepregrade.UradnilistRepublikeSlovenije92/1999, 44/2003 in 58/2016. Ljubljana. Pravilnikotehničnemopazovanjuvisokihjezov.UradnilistSocialističnefederativnerepublikeJugoslavije 7/1966. Beograd. Pravilnik o vodnem katastru. Uradni list Republike Slovenije 30/2017. Ljubljana. Review of the EU aquaculture sector and results of costs and earnings survey 2009. Definition of data col­lection needs for aquaculture, Final report. Richert, E., Bianchin, S., Heilmeier, H., Merta, M., Seidler, C. 2011: A method for linking results from an evaluation of land use scenarios from the viewpoint of flood prevention and nature conservation. Landscape and Urban Planning 103. DOI: https://doi.org/10.1016/j.landurbplan.2011.07.001 Sanfo, S., Barbier, B., Dabiré, I. W. P., Vlek, P. L. G., Fonta, W. M., Ibrahim, B., Barry, B. 2017: Rainfall vari­ability adaptation strategies: An ex-ante assessment of supplemental irrigation from farm ponds in southern Burkina Faso. Agricultural Systems 152. DOI: https://doi.org/10.1016/j.agsy.2016.12.011 Seznam obstoječe vodne infrastrukture. Uradni list Republike Slovenije 63/2006, 96/2006. Medmrežje: http://www.pisrs.si/Pis.web/pregledPredpisa?id=DRUG2548 Sodnik, J., Kogovšek, B., Mikoš, M. 2014: Vodna infrastruktura v Sloveniji: kako do ocene realnega stan­ja? Naravne nesreče 3. Ljubljana. Steinman, F., Banovec, P. 2008: Hidrotehnika, Vodne zgradbe I. Univerza v Ljubljana, Fakulteta za grad-beništvo in geodezijo. Ljubljana. Medmrežje: http://fgg-web.fgg.uni-lj.si/kmte/documents/academic/ skripta/Hidrotehnika_sept.2008.pdf Strahler, A. N. 1957: Quantitative analysis of watershed geomorphology. Transactions of the American Geophysical Union 38-6. DOI: https://doi.org/10.1029/tr038i006p00913Širca,A.2010:Slovenskevelikepregrade:stanje,perspektiveinovire.VarnostpregradvSloveniji.12.posve­tovanje Slovenskega nacionalnega komiteja za velike pregrade (SLOCOLD). Krško. Širca, A., Ravnikar Turk, M., Zadnik, B. 2010: Vloga strokovnih združenj pri izboljševanju varnosti pre­grad.VarnostpregradvSloveniji.12.posvetovanjeSlovenskeganacionalnegakomitejazavelikepregrade (SLOCOLD). Krško. ŠmidHribar,M.,LedinekLozej,Š.2013:Theroleofidentifyingandmanagingculturalvaluesinruraldevel­opment. Acta geographica Slovenica 53-4. DOI: https://doi.org/10.3986/AGS53402 Trobec,T.2016:Prostorsko-časovnarazporeditevhudourniškihpoplavvSloveniji.Dela46.DOI:https://doi.org/ 10.4312/dela.46.5-39 Uputstvo o izradi dokumentacije za određivanje posledica usled iznenadnog rušenja ili prelivanja visokih brana. Savezni komitet za poljoprivredu, 1975. Beograd. UredbaonačrtihupravljanjavodanavodnihobmočjihDonaveinJadranskegamorja.UradnilistRepublike Slovenije 67/2016. Ljubljana. Vanacker, M., Wezel, A., Arthaud, F., Guérin, M., Robin, J. 2016: Determination of tipping points for aquatic plantsandwaterqualityparametersinfishpondsystems:Amulti-yearapproach.EcologicalIndicators64. DOI: https://doi.org/10.1016/j.ecolind.2015.12.033 Verstraeten,G.,Poesen,J.1999:Thenatureofsmall-scaleflooding,muddyfloodsandretentionpondsed­imentation in central Belgium. Geomorphology 29. Verstraeten, G., Poesen, J. 2000: Estimating trap efficiency of small reservoirs and ponds: methods and implications for the assessment of sediment yield. Progress in Physical Geography 24-2. Verstraeten,G.,Poesen,J.2001:Variabilityofdrysedimentbulkdensitybetweenandwithinretentionponds anditsimpactonthecalculationofsedimentyields.EarthSurfaceProcessesandLandforms26-4.DOI: https://doi.org/10.1002/esp.186 Vico, G., Tamburino, L., Rigby, J. R. 2020: Designing on-farm irrigation ponds for high and stable yield fordifferentclimatesandrisk-copingattitudes.JournalofHydrology584.DOI:https://doi.org/10.1016/ j.jhydrol.2020.124634 World declaration on water storage for sustainable development. 24th Congress of the International Commission on Large Dams, 5.6.2012. Internet: https://www.icold-cigb.org/userfiles/files/World% 20declaration/World%20Declaration-ENG.pdf (15.2.2020). Worldregisterofdams.InternationalCommissiononLargeDams,September2019.Internet:https://www.icold­cigb.org/ (24.2.2020). Wutich, A., White, A. C., White, D. D., Larson, K. L., Brewis, A., Roberts, C. 2014: Hard paths, soft paths or no paths? Cross-cultural perceptions of water solutions. Hydrology and Earth System Sciences 18. DOI: https://doi.org/10.5194/hess-18-109-2014 Zakon o vodah. Uradni list Republike Slovenije 67/2002, 2/2004, 41/2004, 57/2008, 57/2012, 100/2013, 40/2014 in 56/2015. Ljubljana. ZemeljskeinbetonskevodnepregradestrateškegapomenavRepublikiSloveniji,VODPREG2,2016.Razvojno raziskovalni projekt, Končno poročilo. Zavod za gradbeništvo, IBE d.d., Fakulteta za gradbeništvo in geodezijo, Ljubljana, Hidrotehnik d.d. Ljubljana. Zorn, M. 2007: Jožefinski vojaški zemljevid kot geografski vir. Geografski vestnik 79-2. Zorn, M., Komac, B. 2008: Zemeljski plazovi v Sloveniji. Georitem 8. Ljubljana. Žiberna,I.2017:TrendivodnebilancevseverovzhodniSlovenijivobdobju1961-2016.GeografijaPodravja. Maribor. SEDIMENT PRODUCTION IN FLYSCH BADLANDS: A CASE STUDY FROM SLOVENIAN ISTRIA Gregor Kovačič Measuring the catchment areas of erosion plots in the Strane Badlands in the Rokava River headwaters, April 28th, 2008. DOI: https://doi.org/10.3986/AGS.6721 UDC: 911.2:551.3.053(497.4Istra) COBISS: 1.01 Gregor Kovačič1 Sediment production in flysch badlands: A case study from Slovenian Istria ABSTRACT: This article deals with the results of seven years of measurements of sediment release from the flysch badlands in the Rokava Riverheadwaters. Measurements of sediment production were carried out in erosion plots, and measurements of cliff (or rockwall) retreat using erosion pins. Selected meteo­rological time series from the Portorož Airport meteorological station were included in the analysis. The calculation showed that from 2008 to 2015 (149 measurements) sediment production was 36kg/m2 per yearandtheflyschcliffretreatedby146mmor21mmperyear.Theamountofsedimentproducedismod­erately positively correlated with the number of days between successive measurements (r=0.51), with a recorded daily transition of air temperature over/below 0°C (r=0.56) and slightly more weakly corre­lated with the precipitation amount (r=0.45). On the other hand, the amount of sediment produced has a low negative correlation withaverage air temperature (r=-0.29) and average minimum air temperature (r=-0.30). However, no statistically significant correlation was calculated between the amount of sedi­ment produced and average wind speed. KEYWORDS:geomorphology,geomorphicprocesses,erosionprocesses,sedimentproduction,cliffretreat, geography, Slovenian Istria, Slovenia Sproščanje gradiva na erozijskih žariščih v flišu v slovenski Istri POVZETEK: Prispevek obravnava sedemletne meritve sproščanja fliša z erozijskega žarišča Strane vpovirjuRokave.Meritvesproščanjakamninskegagradivasmoopravljalispomočjoerozijskihpolj,meritve umikanja flišne stene pa tudi s pomočjo erozijskih žebljičev. V analizo smo vključili podatke izbranih meteorološkihčasovnihvrstspostajePortorož–letališče.Izračunikažejo,dajebilavobdobju2008–2015 (149 meritev) intenzivnost sproščanja fliša 36kg/m2/leto, stena pa se je umaknila za 146mm oziroma 21mm/leto. Količina sproščenega gradiva je zmerno pozitivno povezana s številom dni med meritvami (r=0,51)terštevilomdnisprehodomtemperaturezrakapreko/pod0°Cmeddvemameritvama(r=0,56), manj z višino padavin (r=0,45), medtem ko s povprečno dnevno temperaturo (r=–0,29) in povprečno minimalno temperaturo zraka (r=–0,30) kaže nizko negativno povezanost. S povprečno hitrostjo vetra količina ne kaže statistično značilne povezanosti. KLJUČNE BESEDE: geomorfologija, geomorfni procesi, erozijski procesi, sproščanje gradiva, umikanje pobočij, geografija, slovenska Istra, Slovenja This article was submitted for publication on December 27th, 2018. Uredništvo je prejelo prispevek 27. decembra 2018. 1 University of Primorska, Faculty of Humanities, Koper, Slovenia gregor.kovacic@fhs.upr.si 128 1 Introduction Studying erosion in its broadest sense – that is, all exogenous processes of rock and eluvium removal and transportation (Kladnik et al. 2005) – is difficult because these processes are usually slow and most land-formstakealongtimetodevelop(HowardandKerby1983).Directobservationandmeasurementorerosion prediction models (erosion models; Stroosnijder 2005, cf. Zorn 2008b) can be used at different time and spatialscales(TurowskiandCook2017).However,theproductionandcalibrationoferosionmodels,which allowfasteracquisitionandinterpretationoferosiondataforlargerareas,alsorequiredataobtainedthrough direct measurement in the natural environment or laboratories (Stroosnijder 2005). Direct measurement methods are technically complex and time consuming, and the results obtained are difficult to extrapo­late to longer time periods and larger spatial units (Zorn 2008b). Due to their complexity, measurements areusuallypossibleonlyinerosionplotsofvarioussizes,andtheresultsaredifficulttoextrapolatetolarg­er areas (for more, see Zorn 2008b). Due to greater storage capacity, the amount of sediment produced usually decreases with increasingsize of the study area (Nadal-Romero et al. 2011). Erosion process mea­surements are rare in Slovenia (Zorn 2015). Measurements in flysch badlands were performed by Zorn(2008a; 2009) and in dolomite badlands by Komac (2003) and Švigelj (2015). There have been a few stud-iesexaminingtheintensityofsoilerosiononvarioustypesoffarmland;thefindingsarepublishedinKomac and Zorn (2005; 2007), Zorn and Komac (2005), and Zorn (2008a). Extremely well suited for studying geomorphic systems at small spatial and time scales (Wainwright and Brazier 2011) are badlands, where geomorphic processes are intense and landform changes are rapid, allowingeffectiveuseofdirectobservationandmeasurementmethods(Campbell1997;Gulametal.2014). Generaloverviewsofbadlandsandbadland-formingprocesseshavebeenprovidedbyvariousauthors(e.g., Scheideggeretal.1968;BryanandYair1982;Howard1994;1997;2009;Campbell1997;SaundersandYoung 1983; Torri et al. 2000; Gallart et al. 2002; Harvey 2004). Badlands can be defined as landforms with or without sparse vegetation, with steep slopes, a dense network of erosion rills and gullies, and few or no eluvial deposits produced in loose or poorly consoli­dated,impermeable,orpoorlypermeablerocks,usuallywithasignificantshareofclay(Gulametal.2014). Inbadlands,rocksaredirectlyexposedtoweatheringprocesses,theeffectsofraindrops,runoff,andwind erosion (Zorn 2008a; 2009; 2012). This results in the formation of taluses under steep rockfaces (Jurak andFabić2000;Zorn2009;2012).Badlandformationisinfluencedbyvariousfactors,withgeologicaland climate factors being the predominant ones (Bryan and Yair 1982; Gulam et al. 2014). An important fac­torfortheirpreservationistheabsenceofvegetationorsparsevegetationcover,whichisalsodisappearing in some places due to human impact (Harvey 2004; Nadal-Romero and Regüés 2010; Gulam et al. 2014). In some areas, the abandonment of farming on steep slopes and afforestation have resulted in a signifi­cant (i.e., 70–85%) decrease in the area of badlands (Ciccacci et al. 2008; Staut and Mikoš 2008). Badlandsarebeingstudiedacrosstheglobe(e.g.,Schumm1956;1962;Liuetal.1985;Feolietal.2002; Boardman et al. 2003; Eriksson et al. 2003; Curry and Morris 2004; Poesen et al. 2006; Achten et al. 2008; Maerker et al. 2008; Joshi et al. 2009), including the Mediterranean (e.g., De Ploey 1974; Yair et al. 1980; Alexander 1982; Imeson et al. 1982; Rendell 1982; Wise et al. 1982; Clotet et al. 1988; Benito et al. 1992; 1993; Torri et al. 1994; Wainwright 1994; Poesen and Hooke 1997; Sirvent et al. 1997; Clarke and Rendell 2000; 2006; Nogueras et al. 2000; Cantón et al. 2001; Gallart et al. 2002; Wainwright and Thornes 2003; RegüésandGallart2004;Díaz-HernándezandJuliá2006;Piccarettaetal.2006;Nadal-Romeroetal.2007; 2008; 2011; Ciccacci et al. 2008; Miščević et al. 2009; Nadal-Romero and Regüés 2010; Martínez-Murillo et al. 2013). They are also typical of the flysch areas in Istria (southwest Slovenia and northwest Croatia), whichhavebeenstudiedbyvariousresearchers(e.g.,JurakandFabić2000;Juraketal.2002;2003;Petkovšek 2002; Petkovšek and Mikoš 2003; Staut and Mikoš 2008; Zorn 2007; 2008a; 2008c; 2009; 2010; 2012; Zorn and Mikoš 2008; Zorn and Komac 2011; Gulam et al. 2014; 2018). They usually cover up to ten hectares andaresmallcomparedtootherselsewherearoundtheglobe,whichcancoveruptoseveraldozensquare kilometers (Bryan et al. 1987; Howard 1994). However, in Istria they are very common. Gulam, Pollak, andPodolszki(2014)inventoried5,568badlandswithatotalareaof10.7km2intheflyschareasofCroatian Istria. In Istria, sediment release in badlands can be up to eight hundred times greater than in areas cov­ered with vegetation (Jurak and Fabić 2000). The percentage of badlands in individual Istrian watersheds can reach up to 12% (Gulam et al. 2014). A simplified model of theformation and development of a bad­land on flysch rocks is presented in Gulam (2012) and Gulam, Pollak and Podolszki (2014). This article presents the analysis results of measurements of sediment release from the cliff in the fly-schintheStraneBadlandsintheRokavaheadwatersinSlovenianIstria,conductedbetween2008and2015. Someresultsfortheperiodfrom2008to2012havealreadybeenpublishedinZornetal.(2017).Theanaly­sisincludedselectedmeteorologicalvariables,andtheircorrelationwithorimpactonsedimentproduction was established using data retrieved from the Portorož Airport meteorological station. 2 Geographical location of the test site, methods, and materials The Strane Badlands are located on the left, shady side of the Rokava Valley, above one of the left mean­ders of the Rokava (or Pinjevec) River, 1.5km south of Marezige and 300m west of Rokavci (Figure 1). Like the entire Dragonja Basin, which the Rokava River is part of, the badlands extend in an east–west direction. The area is made of Eocene flysch, with alternating layers of siliciclastic sediment and carbon-ate-siliciclastic turbidites of sandstone and marl (Placer et al. 2004). The flysch areas in Istria are also characterized by calciturbidites 1.5 meters or more thick (megalayers), but there are none in these bad­lands. The barren flysch marlstone is highly non-resistant to exogenous processes (Vivoda Prodan and Arbanas 2016; Vivoda Prodan et al. 2017), which is also the case in the badlands studied, where the marl-stone is fractured to a depth of 5 to 10cm (Zorn 2008a). Figure 1: Location of the Strane Badlands in the Rokava headwaters. Over the past twenty years, the Dragonja Basin has frequently been the subject of studies examining geomorphicprocesses,especiallyusingvariouserosionmodels,whereasdirectmeasurementsofgeomorphic processeshavebeenlesscommon(Globevniketal.1998;Globevnik2001;Petkovšek2002;Petkovšekand Mikoš 2003; 2004; Staut 2004; Keesstra 2006; 2007; Keesstra et al. 2005; 2009; Staut and Mikoš 2008; Zorn 2008a; 2009; 2012). The study area has a transitional climate between the coastal and mainland temperate Mediterranean climates, with annual precipitation between 1,100 and 1,200mm (Ogrin and Plut 2009). The total area of the central part of the badlands composed of a flysch cliff and a largely overgrown talus beneath it covers approximately 0.8 hectares, and the wider area of more intense erosion and denudation processes covers approximately1.8hectares.Thehighestpartofthecliffedgeisatanelevationofapproximately203to210 m,andtheRokavariverbedisatanelevationof150m.Thebadlandslieabovetheouteredgeofthemean­der and are thickly overgrown, which is why the Rokava River washes away material from its lower parts only when precipitation is very intense (Zorn 2008a), keeping the badlands active. The contact between the running water and the talus is key to preserving the badlands (Gulam et al. 2014). On March 19th, 2008, four semi-open erosion plots for measuring the amount of sediment released fromtheflyschcliffweresetupatapproximatelythesamemicro-locationsasusedbyZorn,whoperformed the measurements from 2005 to 2006. The plots lay at an elevation of 198 to 203m and were delimited on the upper side by the edge of the badlands and were open on the sides. They had a northeast orientation. The erosion plot barriers were made of wood and plastic (Figure 2) and were placed approximately 2.0 to 2.5m from the wall. The plastic edges of individual erosion plots, which were placed on the slope to pre­ventthematerialproducedfrommixingwiththeoldermaterialonthetalus,weresetapproximately0.5m from the wall. To prevent material from slipping past the barriers, wooden slats were placed on the sides of the plots. The catchment areas of individual erosion plots that followed one another in a north–south Figure 2: Location of erosion plots in the Strane Badlands (March 19th, 2008). direction measured 9.63m2 for the first plot, 8.38m2 for the second, 4.13m2 for the third, and 2.61m2 for thefourth,oratotalof24.74m2altogether,whichwastwiceasmuchaswithZorn’smeasurements(2008a; 2009). The inclinations of the plot barriers ranged from 32° to 55°, which is more than the natural angle of repose of loose material, and the inclination of the erosion plots ranged between 80° and90°. The slope above the edge of the cliff is less steep and it is overgrown with mixed forest consisting of Austrian pine and downy oak. In the badlands studied, the sediment release from the flysch cliff was measured by weighing the sed­iment accumulated behind the erosion plot barriers (Figure 3). A digital hanging scale (with a precision of 10 g) was used to measure both the total mass of the sediment and the masses of three different grain sizes of sediments (up to 13, 13 to 25, and over 25mm; Figure 4). This article presents the results for the total mass of sediment produced in individual erosion plots. The analysis took account of the total mass of dry material. The ratio between the wet and dry material was defined in the laboratory; with the same volume of material, the mass of wet material is 7% greater. In parallel, direct cliff retreat was measured using erosion pins placed in the cliff’s marlstone and sandstone (the measurements are conducted every sixmonths,Figure5).Whenapinfallsfromthecliff,ithastobereplaced.Occasionallythepinsinthelower part of the cliff or the upper part of the talus are covered by debris. During the measurements, the mea­surement plots were damaged several times, which to some extent affected the results. The rate of cliff retreat and the amount of sediment produced per area were calculated using the data onthesizeofthecatchmentareasintherockfacesabovetheerosionplotsandtheflyschmass(1,712.04kg/m3) that Zorn (2008a) had used in his calculations. This article presents the results of 149 measurements conductedat various intervals between March 19th, 2008 and February 17th, 2015 (a total of 2,527 days). The measurement data were correlated with the data of selected meteorological variables measured and processed at the Portorož Airport meteorological station just over 14km west of the badlands studied at an approximately 180m lower elevation (Table 2, Figure1).Dataonthemassofthesedimentaccumulatedintheerosionplotsbetweentwosuccessivemea­surements were correlated with data on the average air temperature, average minimum air temperature, Figure 3: Material accumulated behind the barrier of an erosion plot (June 6th, 2012). Figure 4: Weighing the accumulated sediment with a hanging scale, October 23rd, 2017. Figure 5: Erosion pins for direct measurements of flysch cliff retreat, April 28th, 2008. average wind speed, total precipitation and average daily precipitation, and the total and average number of days with a recorded transition of air temperature over/below 0°C calculated for the periods betweenthe two successive measurements. The information on changes in temperature is important for studying the effect of frost weathering on the production of flysch sediment during the cold half of the year. This effect on the amount of the sediment produced in badlands has been demonstrated by various authors(Zorn 2008a; 2012; Regüés et al. 1995; 2000). The direction and strength of correlation between various pairs of variables were analyzed using the Pearson correlation coefficient (p<0.05). From October 20th, 2010 to November 13th, 2012 (742 days), air temperature was measured at one-hourintervalsabovethebadlandtalus(facingeast)andontheslopeabovetheedgeofthebadlands(facing north). A regression analysis between the average and minimum daily air temperatures measured at the Portorož Airport station and the data obtained in this study showed exceptionally high correlation(r2=0.98–0.99). Regression functions were used to calculate the average and minimum daily air temper-aturesinthebadlands,whichwereincludedintheaforementionedanalysisofcorrelationbetweenthemassof sediment produced and the values of meteorological variables. 3 Results and discussion 3.1 Weather conditions: comparison with long-term average values The typical air temperatures in the badlands were as follows: the average air temperature was 12.88°C, the absolute minimum air temperature was -9.5°C (February 14th, 2012), and the absolute maximum air temperature was 38.3°C (July 2nd, 2012). Over the seven-year period of measurements, there were 335dayswitharecordedtransitionofairtemperatureover/below0°C,whichis36%morethanwhatwasmea-sured at the Portorož Airport station. These changes in temperature are only typical of the cold half of the year (from October to March). The average number of days with a recorded transition of air temperatureover/below 0°C per day for 149 observation periods was 0.11 and in the cold half of the year it was 0.36. During the measurement period, the annual precipitation at the Portorož Airport station was 1,027mm, which is 6% more than between 1981 and2010 (Agencija…2017). Due to higher annual precipitation inIstria’s interior, another 100 to 200mm can be added to the total annual precipitation in the measurement area. Compared to the period from 1981 to 2010, the following months were wetter in the period stud­ied: July (59% more precipitation), February (41% more), and November and May (both with 15% moreprecipitation). The average August precipitation amounted to only 66% of the long-term average, and the average precipitation in October only 74%. Average daily precipitation was 2.7mm. The average annual and daily precipitation are not direct indicators of the erosive force of precipitation (which was not mea­sured), but in general erosivity is greater during the wet part of the year (Petkovšek 2002; Zorn 2008a). Precipitationintensityaffectstherateoftransferofthealreadyweatheredmaterialdowntheslope.Inturn, the wetness or saturation of the rock affects mechanical weathering by wetting and drying, which is typ­ical of fine-grained sedimentary rock, including flysch marl. The average wind speed during the period studiedwas2%higherthanfrom1981to2010.ThewindiermonthsincludedFebruary,March,December,andOctober, with 12%, 7%, 5%, and 4% higher average wind speeds, respectively. The greatest downward deviations were recorded in April (6%) and July (2%). Through wetting and drying, wind is an important factor in flysch weathering, and it also triggers the release of weathered material down the slope. 3.2 Sediment production Over the course of seven years, just over 5,733kg of sediment was weighed in the four erosion plots, of which 66.7% had a grain size up to 13mm, 24% had a grain size between 13 and 25mm, and 9.2% had a grain size larger than 25mm (Table 1). Flysch mostly weathers into smaller-grained material, which isa result of the fine-grained composition of sandstone and marl. Occasionally larger pieces of sandstone with a typical oblique rectangular prism shape accumulated in the erosion plots. The largest share of the finest-grained material was recorded in Plot 4 (74%) and the smallest share was recorded in Plot 3 (60%),which can be explained by the different flysch composition in the catchment areas, where there were no thicksandstonelayersabovePlot4.Incontrast,thelargestshareofthecoarsest-grainedmaterialwasweighed in Plot 3 (11%) and the smallest in Plot 4 (9%). The largest share in the total mass of the sediment in the badlands came from Plot 1 (2,072kg or 36%) andthesmallestfromPlot4(824kgor14%;Table1).Sedimentproductionrangedfrom213kg/m2(Plot 2) to 316 (Plot 4)kg/m2, with an average of 250kg/m2. Over the seven years, the annual average sediment productionwas36kg/m2/year(rangingfrom31to46kg/m2/year).Overthecourseoftwoyears,Zorn(2008a, 2009, 2010, 2012) calculated sediment production twice as high, ranging from 58 to 122kg/m2/year for anindividualerosionplot(anaverageof80kg/m2/year).Thisdifferencemayreflectdifferentweathercon­ditionsbetweenthetwoperiodsobservedand/oradifferentfrequencyoffieldmeasurements,whichwere performed weekly from 2005 to 2006 and every two to seventy-four days (or on average every seventeen Table 1: Selected indicators of values measured and calculated in the Strane Badlands in the Rokava headwaters between March 19th, 2008 and February 17th, 2015. Indicator Plot 1 Plot 2 Plot 3 Plot 4 -Total (. sum, x mean) Catchment area (m2) 9.63 8.38 4.13 2.61 .=24.75 Accumulated sediment mass (kg) 2,072.71 1,781.62 1,054.50 824.33 .=5,733.16 Accumulated sediment mass (%) Calculated flysch cliff retreat, period (mm) Calculated flysch cliff retreat, annual (mm) Sediment production (kg/m2) Sediment production (kg/year) Sediment production (kg/m2/year) Sediment production (t/ha) Sediment %, grain size < 13mm Sediment %, grain size 13–25mm Sediment %, grain size>25mm 36.15 125.67 18.16 215.14 299.52 31.09 2,151.4 69.11 21.64 9.26 31.08 124.24 17.95 212.70 257.46 30.74 2,127 64.04 27.27 8.69 18.39 149.18 21.56 255.41 152.38 36.91 2,554.1 60.35 28.38 11.27 14.38 184.80 26.71 316.39 119.12 45.72 3,164 74.25 17.90 7.85 .=100 -x=145.88 -x = 21.30 -x = 249.91 -x= 207.12 -x = 36.11 -x = 2,499 -x = 66.71 -x = 24.05 -x = 9.24 Figure 6: Damage to a barrier of erosion Plot 2 caused by sandstone breaking off, October 12th, 2017. days) from 2008 to 2015. Hence, it is possible that the measurements performed after longer breaks do not cover all the material produced because it may have managed to slip past the barriers due to the ero­sion plots being overfilled. The same applies to measurements performed after the erosion plots had been damaged (Figure 6). Sediment production is inversely proportional to the size of the erosion plots’ catch­ment areas (Table 1), which may again be the result of sediment slipping past the barriers in erosion Plots 1and2.Theannualamountofsedimentaccumulatedinerosionplotsrangedfrom119(Plot4)to300kg/year (Plot 1), with an average of 207kg/year. DataonsloperetreatacrosstheglobewerecollectedbyYoung(1969,1974),SaundersandYoung(1983), andYoungandSaunders(1986),whoreportapredominantretreatofupto1mm/year.PoesenandHooke (1997)reportsignificantlyhighervaluesfortheMediterranean,rangingfrom0.05to30mm/year,andmore recent research even reports retreats of up to 65mm/year (Gulam et al. 2018) and surface lowering of up to 75mm/year (Ciccacci et al. 2008). It makes sense to compare the results of this study with the results obtained in flysch badlands because the rate of cliff retreat depends strongly on the lithological charac­teristicsofbadlands(BryanandYair1982;Gulametal.2014).Theflyschcliffinthebadlandsstudiedretreated 18 to 27mm/year or 21mm/year on average. This is half as much as Zorn calculated for individual ero­sion plots (2008a, 2009, 2010, 2012) – that is, 35–50mm. Gulam et al. (2018) calculated an annual flysch cliff retreat of 27–33mm using profilometers and a retreat of up to 65mm/year using photogrammetry. Jurak et al. (2003) used the same method to calculate a retreat of 21mm/year, Ogrin (1992) determined a retreat of 20mm/year using dendrochronology, and Petkovšek (2002) and Petkovšek and Mikoš (2003) reportedaretreatof40to50mm/year.Erosionpinswereusedtomeasureacliffretreatof19.4to41.6mm/year or31mm/yearonaverage.Marl retreats 3mm/yearfasterthansandstone.Theretreat measuredusingthe erosion pins was larger than the calculated retreats. The differences arise from the measurement method, butpartlytheycanalsobeexplainedbymaterialslippingpasttheerosionplotbarriers.Thedifferentrates of flysch cliff retreats in the badlands in Istria (Petkovšek 2002; Jurak et al. 2003; Zorn 2008a, 2009, 2010, 2012; Gulam et al. 2018) and a comparison with the results of this study show that sediment production strongly depends not only on lithological characteristics, but also on terrain and climate characteristics. 3.3 Correlation between sediment production and meteorological variables Thesedimentmassesmeasuredintheerosionplotsshowhighsignificantpositivecorrelationswithmete­orologicalvariables(0.70–0.83).Inaddition,theyshowmoderatesignificantcorrelationswiththenumber of days between measurements (r=0.43–0.70), which indicates that sediment production on the cliff is not the same throughout the year. The seasonal aspect of erosion processes and the predominant effect of frost weathering during the cold half of the year and rain erosion during the warm half of the year were reported by Regüés et al. (1995; 2000), Regüés and Gallart (2004), Nadal-Romero and Regüés (2010), and Zorn (2008; 2012). The Pearson correlation coefficient between the total amount of sediment measured in all four plots and the duration of the intervals between measurements was 0.51. The strongest correla­tion with this indicator was established for Plot 4 (Table 2). Theaverageairtemperatureatthemeasurementsiteshowsalownegativecorrelationwiththeamount of sediment produced (r=-0.29), and approximately the same values were established for the correlation between the amount of sediment produced and the average minimum air temperature at individual plots (r=-0.12 to -0.40), which is comparable to Zorn’s (2008a; 2012) findings (r=-0.31). The fact that frost weathering is an important process affecting flysch release in badlands is proven by the moderate posi­tive significant correlation of the amount of sediment produced with the number of days with a recorded transition of air temperature below/over 0°C (r=0.59) and the average number of days with a recorded transitionofairtemperaturebelow/over0°C(r=0.54,Table2).ThesamewasestablishedbyZorn(2008a; 2012), who established a weak significant positive correlation between the number of days with temper­atures below freezing and the amount of sediment produced (r=0.25). The measurements for this study were performed on the shady slope, for which Nadal-Romero et al. (2007) determined that alternation between freezing and thawing is a key factor in weathering and sediment production. Thetotalprecipitationbetweenmeasurementsshowslowtomoderatesignificantpositivecorrelations withthesedimentmassmeasuredintheerosionplots(r=0.35–0.67).Again,thestrongestcorrelationwith this indicator was established for Plot 4; the correlation with the total mass of the sediment accumulated was 0.45, which is completely comparable with the 0.43 reported by Zorn (2008a, 2012). Zorn established insignificant positive correlations between the total mass of sediment produced and precipitation indi­cators (r=0.21–0.26) and a weak positive correlation with the precipitation erosion indicator (r=0.35). Theresultsofthisstudyshownostatisticallysignificantcorrelationbetweenthemassofthesedimentaccu-mulatedintheerosionplotsandtheaveragedailyprecipitation(r=0.04–0.16).Basedonanextensivedatabase on erosion plots in the Mediterranean, Nadal-Romero et al. (2011) report no clear correlations between theamountofsedimentproducedandtemperaturesorprecipitation,whereastheresultsofthisandZorn’s studies (2008a, 2008b, 2012) show the opposite. Zorn(2008a;2012)calculatedalowsignificantpositivecorrelation(r=0.34)betweentheaveragewind speed at the Koper meteorological station and the mass of sediment produced, and a moderate signifi­cantpositivecorrelationwithmaximumgusts(r=0.42).Theresultsofthisstudydonotshowastatistically significantcorrelationbetweentheaveragewindspeedatthePortorožAirportmeteorologicalstationand themassofthesedimentaccumulated(r=0.06–0.14).ThePearsoncorrelationcoefficientbetweenthetotal amount of sediment in all four plots and the average wind speed is 0.11 (Table 2). Inthiststudy,theamountofsedimentproducedshowsamoderatepositivecorrelationwiththenum­ber of days between two consecutive measurements with a recorded transition of air temperature over/below 0°C (r=0.59) and a weaker correlation with the precipitation (r=0.45). In addition, it shows alownegativecorrelationwiththeaveragedailytemperature(r=-0.29)andtheaverageminimumairtem-perature(r=-0.30).Thefindingsofthestudyshowtheimportantimpactoffrostweatheringonflyschsediment production.Acorrelationbetweenthemassofsedimentproducedandtheaveragewindspeedwasnotcon­firmed. Table 2: Pearson correlation coefficients between the mass of sediment accumulated in the erosion plots and selected meteorological variables (n=149 measurements; statistical significance is tested for p<0.05; *statistically insignificant). Meteorological variable Plot 1 Plot 2 Plot 3 Plot 4 Total Strane: average air temperature -0.24 -0.39 -0.23 -0.12* -0.29 Strane: average minimum air temperature -0.25 -0.40 -0.23 -0.12* -0.30 Strane: days with air temp. transition over/below 0°C 0.56 0.60 0.53 0.41 0.59 Strane: avg. days with air temp. transition over/below 0°C 0.51 0.60 0.41 0.36 0.54 Portorož Airport: average wind speed 0.10* 0.14* 0.06* 0.09* 0.11* Portorož Airport: precipitation 0.39 0.35 0.43 0.67 0.45 Portorož Airport: average daily precipitation 0.13* 0.04* 0.14* 0.16* 0.11* Days between measurements 0.45 0.43 0.47 0.70 0.51 4 Conclusion The results of the seven-year measurements of weathered flysch sediment released from the rockface in the Strane Badlands of the Rokava headwaters in Slovenian Istria showed that on average 36kg of sedi­ment per m2 is produced a year, which is comparable with the results obtained in other Istrian badlands (e.g., Jurak and Fabić 2000; Jurak et al.2002; Petkovšek 2002; Jurak et al. 2003; Petkovšek and Mikoš 2003; StautandMikoš2008;Zorn2007;2008a;2008b;2008c;2009;2012;ZornandMikoš2008;ZornandKomac 2011; Gulam et al. 2014, 2018). The cliff retreats at a rate of 21mm per year. In the future, research measuring flysch sediment production should be expanded with continuous measurements of selected meteorological parameters at this site, which would allow easier study of the impact of individual climate elements on sediment production during the year and more reliable conclu­sions about their impact on the rate of erosion-denudation processes. It would make sense to supplement directmeasurementswithothermethods,suchasterrestrialphotogrammetry.Inmeasuringthedirectretreat of the cliff, the pins could be replaced by more accurate profilometers, such as reported by Gulam et al. (2018). Attention should also be directed to dissolved weathering products and fine-grained suspended particlesthatdisappearwithwaterthroughtheweatheredmaterialanddonotaccumulatebehindtheero­sion plot barriers. Studyinggeomorphologicalprocessesinbadlandsisimportantforgainingbasicinsightsintothedevel­opmentofterrain.Becauseerosion-denudationprocessestakeplacerelativelyquicklyinflyschbadlands,they can also serve as a good case study for exploring changes in the intensity of geomorphological processes during this period of increasingly evident planetary climate and environmental change, in view of which theimportanceofstudyingthemisboundtoincrease.Thefindingsofthisstudyhaveundoubtedlyshown that slope processes in badlands are very effective in shaping flysch slopes. ACKNOWLEDGEMENTS:IwouldliketothankallthegeographystudentsattheUniversityofPrimorska FacultyofHumanities,whohavebeenperformingfieldmeasurementssince2008,whentheerosionplots were set up, and thus contributed their share of data to the extensive database. I would also like to thank fellowgeographersMihaStaut,SimonKerma,andJanezBerdavs,andespeciallySašoTrnovec,whohelped set up the erosion plots in the Rokava headwaters in 2008, and Matija Zorn, whose dissertation research inspired interest in exploring erosion processes in flysch areas and further research. 5 References Achten, W. M. J., Dondeyne, S., Samweli, M., Elly, K., Jean, P., Jozef, D., Bart, M. 2008: Gully erosion in SouthEasternTanzania:Spatialdistributionandtopographicthresholds.ZeitschriftfürGeomorphologie 52-2. DOI: https://doi.org/10.1127/0372-8854/2008/0052-0225 AgencijaRepublikeSlovenijezaokolje:DnevnipodatkiometeorološkihspremenljivkahzapostajoPortorož– letališče v obdobju 2008–2015 in mesečna povprečja za obdobje 1981–2010. Ljubljana, 2017. Alexander,D.E.1982:DifferencebetweencalanchiandbiancanebadlandsinItaly.BadlandGeomorphology and Piping. Norwich. Benito,G.,Gutiérrez,M.,SanchoC.1992:ErosionratesinbadlandareasofthecentralEbroBasin(NE-Spain). Catena 19, 3-4. DOI: https://doi.org/10.1016/0341-8162(92)90002-S Benito,G.,Gutiérrez,M.,SanchoC.1993:Theinfluenceofphysico-chemicalpropertiesonerosionprocesses in badland areas, Ebro Basin, NE-Spain. Zeitschrift für Geomorphologie 37-2. Boardman, J., Parsons, A. J., Holland, R., Holmes, P. J., Washington, R. 2003: Development of badlands andgulliesintheSneeuberg,GreatKaroo,SouthAfrica.Catena,50,1-2.DOI:https://doi.org/10.1016/ S0341-8162(02)00144-3 Bryan, R. B., Campbell, I. A., Yair, A. 1987: Postglacial geomorphic development of the Dinosaur ProvincialParkbadland,Alberta.CanadianJournalofEarthScience24-1.DOI:https://doi.org/10.1139/ e87-011 Bryan, R. B., Yair, A. 1982: Perspectives on studies of badland geomorphology. Badland geomorphology and piping. Norwich. Campbell,I.A.1997:Badlandsandbadlandgullies.Aridzonegeomorphology:Process,formandchange in drylands. Chichester. Cantón, Y., Domingo, F., Solé-Benet, A., Puigdefábregas, J. 2001: Hydrological and erosion response of a badlands system in semiarid SE Spain. Journal of Hydrology 252. DOI: https://doi.org/10.1016/ S0022-1694(01)00450-4 Ciccacci, S., Galiano M., Roma, M. A., Salvatore, M. C. 2008: Morphological analysis and erosion rate evaluationinbadlandsofRadicofaniarea(SouthernTuscany–Italy).Catena74-2.DOI:https://doi.org/ 10.1016/j.catena.2008.03.012 Clarke, M. L., Rendell, H. 2000: The impact of the farming practice of remodelling hillslope topography onbadlandmorphologyandsoilerosionprocesses.Catena40-2.DOI:https://doi.org/10.1016/S0341­8162(99)00047-8 Clarke, M. L., Rendell, H. M. 2006: Process-form relationships in Southern Italian badlands: erosion rates andimplicationsforlandformevolution.EarthSurfaceProcessesandLandforms31.DOI:https://doi.org/ 10.1002/esp.1226 Clotet,N.,Gallart,F.,Balasch,C.1988:Mediumtermerosionratesinasmallscarcelyvegetatedcatchment in the Pyrenees. Catena Suppl. 13. Curry, A. M., Morris, C. J. 2004: Lateglacial and Holocene talus slope development and rockwall retreat on Mynydd Du, UK. Geomorphology 58. DOI: https://doi.org/10.1016/S0169-555X(03)00226-5 De Ploey, J. 1974: Mechanical properties of hillslopes and their relation to gullying in central semi-arid Tunisia. Zeitschrift für Geomorphologie Supplementband 21. Díaz-Hernández, J. L., Juliá, R. 2006: Geochronological position of badlands and geomorphological patternsintheGuadix–Bazabasin(SESpain).QuaternaryResearch65.DOI:https://doi.org/10.1016/ j.yqres.2006.01.009 Eriksson,M.,Reuterswärd,K.,Christiansson,C.2003:ChangesinthefluvialsystemoftheKondoaIrangi Hills,central Tanzania,since1960.Hydrological Processes 17.DOI:https://doi.org/10.1002/hyp.1386 Feoli, E., Gallizia Vuerich, L., Woldu, Z. 2002: Processes of environmental degradation and opportunities for rehabilitation in Adwa, Northern Ethiopia. Landscape Ecology 17. DOI: https://doi.org/10.1023/ A:1020566801067 Gallart, F., Solé, A., Puigdefabregas, J., Lázaro, R. 2002: Badland systems in the Mediterranean. Dryland Rivers: Hydrology and Geomorphology of Semi-arid Channels. Chichester. Globevnik, L. 2001: Celosten pristop k urejanju voda v povodjih. Ph.D. thesis, Fakulteta za gradbeništvo in geodezijo Univerze v Ljubljani. Ljubljana. Globevnik,L.,Sovinc,A.,Fazarinc,R.1998:LanddegradationandenvironmentalchangesintheSlovenian Submediterranean (the Dragonja River catchment). Geoökodynamik 19, 3-4. Gulam, V. 2012: Erozija ogolina u flišu središnje Istre. Ph.D. thesis, Rudarsko geološko naftni fakultet Sveučilište u Zagrebu, Zagreb. Gulam, V., Gajski, D., Podolszki, L. 2018: Photogrammetric measurement methods of the gully rock wall retreat in Istrian badlands. Catena 160. DOI: https://doi.org/10.1016/j.catena.2017.09.024 Gulam, V., Pollak, D., Podolszki, L. 2014: The analysis of the flysch badlands inventory in central Istria, Croatia. Geologica Croatica 67-1. DOI: https://doi.org/10.4154/GC.2014.01 Harvey, A. 2004: Badland. Encyclopedia of Geomorphology 1. London. Howard, A. D. 1994: Badlands. Geomorphology of Desert Environments. London Howard, A. D. 1997: Badland morphology and evolution: Interpretation using a simulation model. Earth SurfaceProcessesandLandforms22.DOI:https://doi.org/10.1002/(SICI)1096-9837(199703)22:3<211:: AID-ESP749>3.0.CO;2-E Howard, A. D. 2009: Badlands and gullying. Geomorphology of Desert Environments. Dordrecht. Howard,A.D.,Kerby,G.1983:Chanellchangesinbadlands.BulletinoftheGeologicalSocietyofAmerica 94. DOI: https://doi.org/10.1130/0016-7606(1983)94<739:CCIB>2.0.CO;2 Imeson,A.C.,Kwaad,F.J.P.M.,Verstraten,J.M.1982:Therelationshipofsoilphysicalandchemicalprop­erties to the development of badlands in Morocco Badland geomorphology and piping. Norwich. Joshi, V., Tambe, D., Dhawade, G. 2009: Geomorphometry and fractal dimension of a riverine badland in Maharashtra.JournaloftheGeologicalSocietyofIndia73.DOI:https://doi.org/10.1007/s12594-009-0015-y Jurak, V., Aljinović, D., Petraš, J., Gajski, D. 2003: Field intensity measurements of erosion instability in flyschdepositsinIstria,Croatia.GeotechnicalMeasurementsandModelling.Lisse/Abingdon/Exton/Tokyo. Jurak, V., Fabić, Z. 2000: Erozija kišom u slivu bujičnog vodotoka u središnjoj Istri. Zbornik radova 2. Hrvatskog geološkog kongresa. Cavtat – Dubrovnik. Jurak, V., Petraš, J., Gajski, D. 2002: Istraživanje ekscesivne erozije na ogoljelim flišnim padinama u Istri primjenom terestičke fotogrametrije. Hrvatske vode 10-38. Keesstra,S.D.2006:Theeffectsofnaturalreforestationonthehydrology,rivermorphologyandsediment budget of the Dragonja river SW Slovenia. Ph.D. thesis, Vrije Universiteit. Amsterdam. Keesstra, S. D. 2007: Impact of natural reforestation on floodplain sedimentation in the Dragonja basin, SW Slovenia. Earth Surface Processes and Landforms 32-1. DOI: https://doi.org/10.1002/esp.1360 Keesstra, S. D., van Dam, O., Verstraeten, G., van Huissteden, J. 2009: Changing sediment dynamics due to natural reforestation in the Dragonja catchment, SW Slovenia. Catena 78-1. DOI: https://doi.org/ 10.1016/j.catena.2009.02.021 Keesstra, S. D., van Huissteden, J. Vandenberghe, J., van Dam, O., de Gier, J., Pleizier, I. D. 2005: Evolution of the morphology of the river Dragonja (SW Slovenia) due to land-use changes. Geomorphology 69, 1-4. DOI: https://doi.org/10.1016/j.geomorph.2005.01.004 Kladnik, D., Lovrenčak, F., Orožen Adamič, M. 2005: Geografski terminološki slovar. Ljubljana. Komac,B.2003:DolomitereliefintheŽibršeHills.ActageographicaSlovenica43-2.DOI:https://doi.org/ 10.3986/AGS43201 Komac, B., Zorn, M. 2005: Soil erosion on agricultural land in Slovenia – measurements of rill erosion in the Besnica valley. Acta geographica Slovenica 45-1. DOI: https://doi.org/10.3986/AGS45103 Komac, B., Zorn, M. 2007: Meritve in modeliranje erozije v Sloveniji. Strategija varovanja tal v Sloveniji: zbornik referatov. Ljubljana. Liu,T.,An,Z.,Yuan,B.,Han,J.1985:Theloess-palaeosolsequenceinChinaandclimatichistoriy.Episodes 8-1. Maerker, M., Castro, C. P., Pelacani, S., Baeurel, M. V. S. 2008: Assessment of soil degradation suscepti­bilityintheChacabucoProvinceofCentralChileusingamorphometrybasedresponseunitsapproach. Geografia Fisica e Dinamica Quaternaria 31-1. Martínez-Murillo,J.F.,Nadal-Romero,E.,Regüés,D.,Cerdad,A.,Poesen,J.2013:Soilerosionandhydrology ofthewesternMediterraneanbadlandsthroughoutrainfallsimulationexperiments:Areview.Catena 106. DOI: https://doi.org/10.1016/j.catena.2012.06.001 Miščević, P., Števanić, D., Štambuk-Cvitanović, N. 2009: Slope instability mechanisms in dipping conglomerates over weathered marls: Bol landslide, Croatia. Environmental Geology 56-7. DOI: https://doi.org/10.1007/s00254-008-1236-x Nadal-Romero,E.,Larton,J.,Martí-Bono,C.,Regüés,D.2008:Temporaldistributionofsuspendedsediment transport in a humid Mediterranean badland area: The Araguás catchment, Central Pyrenees. Geomorphology 97, 3-4. DOI: https://doi.org/10.1016/j.geomorph.2007.09.009 Nadal-Romero,E.,Martínez-Murillo,J.F.,Vanmaercke,M.,Poesen,J.2011:Scale-dependencyofsediment yieldfrombadlandareasinMediterraneanenvironments.ProgressinPhysicalGeography35-3.DOI: https://doi.org/10.1177/0309133311400330 Nadal-Romero,E.,Regüés,D.2010:Geomorphologicaldynamicsofsubhumidmountainbadlandareas– weathering, hydrological and suspended sediment transport processes: A case study in the Araguás catchment(CentralPyrenees)andimplicationsforalteredhydroclimaticregimes.ProgressinPhysical Geography 34-2. DOI: https://doi.org/10.1177/0309133309356624 Nadal-Romero, E., Regüés, D., Martí-Bono, C., Serrano-Muela, P. 2007: Badland dynamics in the Central Pyrenees:temporalandspatialpatternsofweatheringprocesses.EarthSurfaceProcessesandLandforms 32-6. DOI: https://doi.org/10.1002/esp.1458 Nogueras, P, Burjachs, F., Gallart, F., Puigdefabregas, J. 2000: Recent gully erosion in the El Cautivo bad­lands (Tabernas, SE Spain). Catena 40-2. DOI: https://doi.org/10.1016/S0341-8162(99)00048-X Ogrin,D.1992:Dendromorphologicalanalysisoferosionprocesses–twocasestudiesfromKoprskoprimorje (Slovenia). Geomorphology and sea. Zagreb. Ogrin, D., Plut. D. 2009: Aplikativna fizična geografija Slovenije. Ljubljana. Petkovšek, G. 2002: Kvantifikacija in modeliranje erozije tal z aplikacijo v porečju Dragonje. Ph.D. thesis, Fakulteta za gradbeništvo in geodezijo Univerze v Ljubljani. Ljubljana. Petkovšek, G., Mikoš, M. 2003: Measurements of erosion processes in the experimental catchment of the Dragonja river, SW Slovenia. Acta Hydrotechnica 21-34. Petkovšek, G., Mikoš, M. 2004: Estimating the R factor from daily rainfall data in the Sub-Mediterranean climateofsouthwestSlovenia.EstimatingtheRfactorfromdailyrainfalldataintheSub-Mediterranean climate of southwest Slovenia 49-5. DOI: https://doi.org/10.1623/hysj.49.5.869.55134 Piccaretta,M.,Faulkner,H.,Bentivenga,M.,Capolongo,D.2006:Theinfluenceofphysico-chemicalmate-rial properties on erosion processes in the badlands of Basilicata, Southern Italy. Geomorphology 81, 3-4. DOI: https://doi.org/10.1016/j.geomorph.2006.04.010 Placer, L., Košir, A., Popit, T., Šmuc, A., Juvan, G. 2004: The Buzet Thrust Fault in Istria and overturned carbonate megabeds in the Eocene flysch of the Dragonja Valley (Slovenia). Geologija 47-2. Poesen, J. W. A., Hooke, J. M. 1997: Erosion, flooding and channel management in Mediterranean envi­ronments of southern Europe. Progress in Physical Geography 21-2. DOI: https://doi.org/10.1177/ 030913339702100201 Poesen, J., Vanwalleghem, T., de Vente, J., Knapen, A., Verstraeten, G., Martínez- Casasnovas, J. A. 2006: GullyerosioninEurope.SoilErosioninEurope.Chichester.DOI:https://doi.org/10.1002/0470859202.ch39 Regüés,D.,Gallart,F.2004:Seasonalpatternsofrunoffanderosionresponsestosimulatedrainfallinabad­land area in Mediterranean mountain conditions (Vallcebre, southeastern Pyrenees). Earth Surface Processes and Landforms 29. DOI: https://doi.org/10.1002/esp.1067 Regüés,D.,Guárdia,R.,Gallart,F.2000:Geomorphicagentsversusvegetationspreadingascausesofbad­land occurrence in a Mediterranean subhumid mountainous area. Catena 40-2. DOI: https://doi.org/ 10.1016/S0341-8162(99)00045-4 Regüés, D., Pardini, G., Gallart, F. 1995: Regolith behaviour and physical weathering of clayey mudrock asdependentonseasonalweatherconditionsinabadlandareaatVallcebre,EasternPyrenees.Catena 25, 1-4. DOI: https://doi.org/10.1016/0341-8162(95)00010-P Rendell, H. M. 1982: Clay hillslope erosion rates in Basento Valley, S. Italy. Geography Annales 64, 3-4. DOI: https://doi.org/10.1080/04353676.1982.11880061 Saunders,I.,Young,A.1983:Rateofsurfaceprocessesonslope,sloperetreatanddenudation.EarthSurface Processes and Landforms 8-5. DOI: https://doi.org/10.1002/esp.3290080508 Scheidegger,A.E.,Schumm,S.A.,Fairbridge,R.W.1968:Badlands.TheEncyclopediaofGeomorphology. New York. Schumm, S. A. 1956: Evolution of drainage systems and slopes in badlands at Perth Amboy, New Jersey. GeologicalSocietyofAmericaBulletin67-5.DOI:https://doi.org/10.1130/0016-7606(1956)67[597:EOD­SAS]2.0.CO;2 Schumm, S. A. 1962: Erosion on miniature pediments in Badlands National Momunent, South Dakota. GeologicalSocietyofAmericaBulletin73-6.DOI:https://doi.org/10.1130/0016-7606(1962)73[719:EOM­PIB]2.0.CO;2 Sirvent, J., Desir, G, Gutierrez, M, Sancho, C., Benito, G. 1997: Erosionrates in badland areas recorded by collectors,erosionpinsandprofilometertechniques(EbroBasin,NE-Spain).Geomorphology18.DOI: https://doi.org/10.1016/S0169-555X(96)00023-2 Staut, M., Mikoš, M. 2008: Spremembe intenzivnosti erozije v porečju Dragonje v drugi polovici 20. sto­letja. Annales Series Historia Naturalis 18-1. Stroosnijder, L. 2005: Measurenments of erosion: It is possible. Catena 64, 2-3. DOI: https://doi.org/ 10.1016/j.catena.2005.08.004 Švigelj, J. 2015: Intenzivnost preperevanja na dolomitu in flišnih kamninah. B.Sc. thesis, Fakulteta za humanistične študije Univerze na Primorskem. Koper. Torri,D.,Calzolari,C.,Rodolfi,G. 2000:Badlandsinchangingenvironments:anintroduction. Catena40-2. DOI: https://doi.org/10.1016/S0341-8162(00)00119-3 Torri,D.,Colica,A.,Rockwell,D.1994:Preliminarystudyoftheerosionmechanismsinabiancanabadland (Tuscany, Italy). Catena 23, 3-4. DOI: https://doi.org/10.1016/0341-8162(94)90073-6 Turowski, J. M., Cook, K. L. 2017: Field techniques for measuring bedrock erosion and denudation. Earth surface processes and landforms 42-1. DOI: https://doi.org/10.1002/esp.4007 VivodaProdan,M.,Arbanas,Ž.2016:WeatheringInfluenceonpropertiesofsiltstonesfromIstria,Croatia. Advances in Materials Science and Engineering 2016. DOI: https://doi.org/10.1155/2016/3073202 VivodaProdan,M.,Mileusnić,M.,MihalićArbanas,S.,Arbanas,Ž.2017:In.uenceofweatheringprocesses ontheshearstrengthofsiltstonesfroma.yschrockmassalongthenorthernAdriaticcoastofCroatia. Bulletin of Engineering Geology and the Environment 76. DOI: https://doi.org/10.1007/s10064-016­0881-7 Wainwright,J.1994:Anthropogenicfactorsinthedegradationofsemiaridregions:Aprehistoriccasestudy in Southern France. Effects of environmental changes on drylands. Chichester. Wainwright, J., Brazier, R. 2011: Slope systems. Arid zone geomorphology. 3rd Edition. Chichester. Wainwright, J., Thornes, J. B. 2003: Environmental issues in the Mediterranean: Processes and perspec­ tives from the past and present. London. Wise,S.M.,Thornes,J.B.,Gilman,A.1982:Howoldarethebadlands?AcasestudyfromsoutheastSpain. Badland geomorphology and piping. Norwich. Yair, A., Lavee, H., Bryan, R. B., Adar, E. 1980: Runoff and erosion processes and rates in the Zin valley badlands,northernNegev,Israel.EarthSurfaceProcesses5-3.DOI:https://doi.org/10.1002/esp.3760050301 Young, A. 1969: Present rate of land erosion. Nature 224. DOI: https://doi.org/doi:10.1038/224851a0 Young, A. 1974: The rate of slope retreat. Institue of British Geographers Special Publication 7. Oxford. Young, A., Saunders, I. 1986: Surface processes and denudation. Hillslope Processes. Boston. Zorn, M. 2007: Ali se zavedamo hitrosti erozijskih procesov – primer iz slovenske Istre. Dela 28. DOI: https://doi.org/10.4312/dela.28.13.183-195 Zorn, M. 2008a: Erozijski procesi v slovenski Istri. Ljubljana. Zorn, M. 2008b: Nekaj načinov preučevanja erozijskih procesov. Geografski vestnik 80-1. Zorn, M. 2008c: Erozijski procesi v hrvaškem delu »Sive Istre«. Geografski vestnik 80-2. Zorn,M.2009:ErosionprocessesinSloveneIstria,Part2:Badlands.ActageographicaSlovenica49-1.DOI: https://doi.org/10.3986/AGS49203 Zorn,M.2010:Erozijskiprocesi–alisezavedamonjihovehitrosti?Odrazumevanjadoupravljanja.Ljubljana. Zorn, M. 2012: Geomorphic activity in flysch badlands of the Istria peninsula (SW Slovenia). Zeitschrift für Geomorphologie 56-3. DOI: https://doi.org/10.1127/0372-8854/2012/S-00104 Zorn, M. 2015: Erozija prsti - prezrt okoljski problem. Geografski obzornik 63, 2-3. Zorn, M., Komac, B. 2005: Erozija prsti na kmetijskih zemljiščih v Sloveniji. Ujma 19. Zorn, M., Komac, B. 2011: The Importance of Measuring Erosion Processes on the Example of Slovenia. Hrvatski geografski glasnik 73-2. DOI: https://doi.org/10.21861/hgg.2011.73.02.02 Zorn,M.,Mikoš,M.2008:RockwallretreatonbadlandsinSloveneIstria.Geologija51.DOI:https://doi.org/ 10.5474/geologija.2008.012 Zorn,M.,Staut,M.,Kovačič,G.2017:NekajgeomorfološkihposebnostislovenskeSiveIstre.SlovenijaVII. Ljubljana. IMMIGRANT INTEGRATION REGIMES IN EUROPE: INCORPORATING THE WESTERN BALkAN COUNTRIES Vesna Lukić, Aleksandar Tomašević Ensuring timely policy responses to integration is important. DOI: https://doi.org/10.3986/AGS.7286 UDC: 316.4.063.3-054.72(497) COBISS: 1.01 Vesna Lukić1, Aleksandar Tomašević2 ImmigrantintegrationregimesinEurope:IncorporatingtheWesternBalkancountries ABSTRACT: This article discusses immigrant integration policies in Europe. We analyzed data from the 2015MigrationIntegrationPolicyIndextoidentifysimilarimmigrantintegrationregimesinEuropeaccord­ing to policy priorities related to immigrants’ socioeconomic rights. The results of a latent class analysis demonstrated that there are two immigrant integration policy regimes among the EU 28, Albania, Bosnia and Herzegovina, Montenegro, Macedonia, and Serbia, with variation between the old EU member states since 1995 (without Greece) and Estonia versus the new EU member states since 2004 (without Estonia), Greece, and the Western Balkan countries. Based on our classification, we conclude that there is a trend of convergence in integration policy regimes in Europe, in which the effects of spatial/geographical and temporal dimensions are manifested. KEYWORDS:immigration,integrationpolicy,MigrantIntegrationPolicyIndex,latentclassanalysis,Europe Režimi integracije priseljencev v Evropi: vključevanje držav Zahodnega Balkana POVZETEK:Včlankuavtorjaobravnavataevropskopolitikovključevanjapriseljencevvvečinskodružbo. NapodlagianalizepodatkovIndeksapolitikevključevanjamigrantov(MIPEX)zaleto2015inobhkratnem upoštevanju prednostnih nalog, ki se nanašajo na družbenogospodarske pravice priseljencev, ugotavljata, katerirežimivključevanjapriseljencevvEvropisosimedsebojpodobni.Rezultatianalizelatentnihrazre­dovsopokazali,dasev28državahčlanicahEU,Albaniji,BosniinHercegovini,Črnigori,SeverniMakedoniji inSrbijiuporabljatadvarežimapolitikevključevanjapriseljencev,pričemersoopaznerazlikemedstarejši­midržavami EU (članicami od leta 1995, a brez Grčije) in Estonijo ter novejšimi članicami (od leta 2004, brez Estonije),GrčijoindržavamiZahodnegaBalkana. Na podlagi izdelane klasifikacijeavtorja ugotavljata, dapostajajoevropskirežimiintegracijskepolitikemedsebojčedaljeboljpodobni,navedenopaimaopazne prostorske oziroma geografske in časovne posledice. KLJUČNE BESEDE: priseljevanje, integracijska politika, indeks politike vključevanja migrantov, analiza latentnih razredov, Evropa The paper was submitted for publication on March 4th, 2019. Uredništvo je prejelo prispevek 4. marca 2019. 1 Institute of Social Sciences, Belgrade, Serbia lukicbodirogav@gmail.com 2 University of Novi Sad, Faculty of Philosophy, Department of Sociology, Novi Sad, Serbia atomashevic@ff.uns.ac.rs 1 Introduction In Europe today, there has been an increasing focus on immigration-related issues by both scholars and policymakers. A literature review suggests that in the academic world much attention has been given to thediscussionofimmigrationandintegrationpoliciesacrosscountries(Bjerreetal.2014;deHaas,Natter and Vezzoli 2015). However, most studies have focused on northwestern Europe, with its long immigra­tion and integration experience, the EU, or OECD countries (Joppke 2007; Garibay and Cuyper 2013;Gregurović and Župarić-Iljić 2018), whereas this topic remains under-researched in the non-EU coun­tries and particularly in the Western Balkans. Therefore, tackling migration and developing immigration and integration policies in European countries that are not immigrant destinations is challenging. Migrationcutsacrossdisciplines,whereasmigrantscrossgeographicallyandsociallyconstructedbor­dersandboundaries.AsHardwick(2008)andJosipovičandRepolusk(2003)emphasize,itisimportantto consider the spatial perspective of migration when theorizing about assimilation. The social and political challenges of immigration make immigrant integration a very important policy domain for governments. European countries formulated national immigrant integration policies in the past primarily in countries in northwestern Europe, whereas the institutional framework for migration is relatively new in eastern Europe.SeveralnationalmodelsofintegrationforthecountriesofnorthwesternEuropeweredistinguished inmigrationliteratureaccordingtogovernmentpolicyprinciplesandresponsestoimmigrationandinte­gration: the assimilationist model, the multicultural model, and the differential exclusionist model (Brubaker 1992; Castles 1995). Thesemodelshavebeenquestionedandcriticizedbyresearchersforvariousreasons.Criticaldiscussions suggestthatthetypologyofnationalimmigrantintegrationmodelsdoesnotrecognizethedynamicchar­acterofintegrationpolicies.It,therefore,under-recognizesthetemporalityofcategoriesofintegrationregimes (Meuleman and Reeskens 2008; Finotelli and Michalowski 2012). Entzinger and Biezefeld (2003) empha­sizethelimitationsoffocusingonalimitednumberofdimensionsinthemostcommonlyusedtypologies. Other shortcomings of the typologies refer to the validity of indicators as well as to ignoring the poten­tial immigration-integration policy nexus (Boucher and Gest 2014). Contemporary Europeanization, as Knill and Lehmkuhl (2002) define it, and globalization process-esinfluencemigrationmanagement(AppaveandLaczko2011).Therefore,effortstowardtheEuropeanization ofintegrationpolicieshavebeenevidentinthelastdecade(Parkes2008).Since2004,theEuropeanUnion has been developing the legal framework and principles of a coherent policy for integrating immigrants (Council…2004,2008). Thedocumentscreateacommonunderstandingofintegrationasaprecondition forharmonization ofimmigrant integration legislation(Martiniello 2006; Lozano etal. 2014). Integration is defined as a »dynamic, two-way process of mutual accommodation by all immigrants and residents of Member States« (Council…2004). Although the EU supports national policies with policy funds, coor­dination and exchange of knowledge among the integration policies is a responsibility of EU countries. Previousresearchhasidentifiedthetransformationofnationalintegrationmodelsandconvergenceininte­gration policies and practices across the EU (Joppke 2007; Penninx and Garcés-Mascarenas 2015), as well as convergence in integration models (Doomernik and Bruquetas-Callejo 2015). Researchers, policymakers, and decision-makers are increasingly preoccupied with immigration and integrationpoliciesacrosscountries.Variousdatabasesofmigrationpoliciesandadoptedlegislationhave been created that cover a number of policy fields, countries, and timespans. These databases offer many opportunities to researchers and policymakers for comparative analysis or evaluation of integration poli­cies across countries. The immigrant integration indicators are part of the International Migration Policy andLawAnalysis(IMPALA)database(Beineetal.2016)trackingimmigrationpolicies.Theyarealsopart oftheDeterminantsofInternationalMigration(DEMIG)databasetrackingmigrationpolicychanges(de Haas,NatterandVezzoli2018).TheBarcelonaCentreforInternationalAffairs(CIDOB)andtheMigration Policy Group (MPG) have produced the Migrant Integration Policy Index (MIPEX). This index covers thirty-eight countries and eight policy areas (labor market, family reunion, education, political participa­tion, long term residence, access to nationality, anti-discrimination, and health) relevant for immigrant integrationfor2004–2014(Huddlestonetal.2015).ThedataarelimitedtoEUcountries,Australia,Canada, Iceland, Japan, South Korea, New Zealand, Norway, Switzerland, Turkey, and the United States. Data for the Western Balkans are not included. The MIPEX data are available for Bosnia and Herzegovina, North Macedonia, and Serbia (2013 and 2015), whereas Albania and Montenegro were assessed only in 2015. In the Western Balkan countries, immigration and integration issues are relatively new. Scholars interested in comparing policies for integrating immigrants across countries and classifica­tion of integration regimes (Meuleman and Reeskens 2008; Zamfir et al. 2014) frequently refer to MIPEXdata.Someresearcherscombinethesedatawithotherindicators;GregurovićandŽuparić-Iljić(2018)used the MIPEX overall index together with the OECD indicators of integration. Recently scholars have linked integration policy models with other social phenomena such as the political participation of immigrants (Helblingetal.2016)orperceptionsofeconomicandculturalthreats(CallensandMeuleman2016).However, only the EU countries were studied. Considering the diversity of European societies, there is a need to study and analyze integration mod-elsofcountriesthathavenotbeenstudiedsofar.Thisisparticularlyimportantforcountriesthataccording toMelegh et al. (2014) are stillcharacterized by major emigration flows and could transform into »coun­triesofimmigration.«Thereisalsoaneedtoenhancetheunderstandingofdifferenttypologiesofintegration regimes that represent the integration of immigrants. Understanding the concept of integration of immigrants is very important given its complex nature involving multiple processes. There is an increasing number of thoughts about alternative approaches to the concept of integration (Grzymala-Kazlowska and Phillmore 2018) due to the era of super-diversity as a new demographic reality (Vertovec 2007). However, regarding the data referred to in this article, immi­grant integration is understood as an opportunity for gaining equal socio-economic rights. This article identifies similar clusters of immigrant integration regimes in Europe according to poli­cyprioritiesrelatedtoimmigrants’socioeconomicrights.Basedontheresearchgapidentifiedinprevious literature, this article answers two main research questions. First, it examines how many different types of integration regimes can be distinguished in Europe when the Western Balkan countries are taken into account. Second, it looks at what the differences are between clusters of immigrant integration regimes and which ones do better in integrating migrants based on the policy areas discussed. 2 Data and methods The data used in this paper are from the 2015 Migration Integration Policy Index (MIPEX) database for thirty-threecountries,collectedin2014.Theindexrepresentstheintegrationpolicybytheindicatorsbased on qualitative expert evaluation of existing national laws and policies. The MIPEX database covers eight policy areas relevant to the integration of immigrants. Every policy area presents diverse components and policy dimensions that are related to policy indicators. Thus, the policy area scores are based on the aver­age scores of policy dimensions that favor the integration of immigrants (Huddleston 2016). The aim of our analysis was the construction of an exclusive and exhaustive typology of integration policy regimes among the EU 28, Albania, Bosnia and Herzegovina, Montenegro, North Macedonia, and Serbia. Because the integration regime is represented by eight categorical policy indicators, our analysis requires a model that will reveal the latent classification of these regimes into clusters based on similar patternsofvaluesforeachindicator.Inotherwords,thismodelwouldrevealadiscretelatentvariablethat will indicate the membership of each country in one of the regime types in such a manner that countries belonging to different types will exhibit substantial differences in the majority of MIPEX indicators. This model was constructed using polytomous latent class analysis, which estimates the conditionalprobabil­ityofeachcountrybelongingtooneofthedifferentregimes(latentclasses)giventhevaluesofpolytomous categoricalvariablesbasedontheMIPEXindicators.Thisprocedurerequiresthespecificationofthenum­berofclasses,andsotheanalysisproceedsbyestimatingseveralmodels,wherebyeachmodelhasoneclass morethanthepreviousone.Themaximumnumberofestimatedclasses dependsonthenumberofcoun­triesanalyzed(samplesize),thetotalnumberofvariablesobserved,andthenumberoflevelsofcategorical variablesobserved.Themodelwiththebestfittotheempiricaldatawaschosenasthemostsuitabletypol­ogy. This also means that the model output results in a conditional probability distribution in which for eachcountrythereisahighprobabilityofmembershipinoneoftheclassesandalowprobabilityofmem­bershipinanyotherclass,whichsatisfiesourgoalofcreatingatypologythatisstraightforwardandexclusive. Our analysis included data that consist of scores for eight policy areas on migrant integration, with thirty-three observations for each policy area. For each policy area score, the minimum value is 0 and the maximum is 100, with each value being one categorical level of policy area variable. Therefore, given the sample size, the reduction of categories (data recoding) was necessary in order to perform polytomous latent class analysis. We reduced the number of categories of each policy area to three so that the new val­ues are as follows: 1 if the original value of the score was less than the first tercile, 2 if the original value of the score was between the first and second tercile, and 3 if the original value of the score was higher than the second tercile, as shown in Table 1. TheanalysiswasperformedusingRstatisticalsoftwareandthepoLCApackage(DrewandLewis2011). 3 Results As noted above, the total number of estimated parameters in the latent class model is a function of the number of variables, the number of categories for each variable, and the number of latent classes speci­ fied by the model. A model that consists of seven policy areas, three categories for each policy area, and two latent classes has twenty-nine estimated parameters. Given our sample size of thirty-three, the model with eight poli­ cy areas could handle only one class, which is not suitable for our research goals (for a further reference Table 1: Policy area indicator values for thirty-three European countries. Country I1 I2 I3 I4 I5 I6 I7 I8 (labor (family (education) (political (permanent (access (anti-(health) market) reunion) participation) residence) to nationality) discrimination) Albania 23122 3 23 Austria 31322 1 13 Belgium 33333 3 32 Bosnia and Herzegovina 2 1 1 1 3 2 2 2 Bulgaria 22113 1 32 Croatia 23113 1 23 Cyprus 11221 2 12 Czech Republic 2 22 2 1 2 13 Denmark 31333 2 12 Estonia 33323 1 11 Finland 33333 3 32 France 21231 3 32 Germany 32332 3 22 Greece 21222 1 23 Hungary 12123 1 33 Ireland 11231 3 22 Italy 33233 2 21 Latvia 12112 1 12 Lithuania 2 11 1 1 1 12 Luxembourg 1 23 3 2 3 13 Malta 11121 1 12 Montenegro 2 12 1 2 2 32 Netherlands 3 23 2 2 3 22 North Macedonia 1 2 1 1 1 2 3 2 Poland 12113 2 11 Portugal 33333 3 33 Romania 23112 1 33 Serbia 22211 2 23 Slovakia 12212 1 21 Slovenia 13222 2 21 Spain 33233 2 13 Sweden 33333 3 31 United Kingdom 2 13 2 1 3 33 regardingtherelationshipbetweenmodelparameters,numberofclasses,samplesize,anddegreesoffree­dom in a latent class model, see Hagenaars and McCutcheon 2002). Therefore, we decided to exclude the data for the policy area health, which was most recently included in the MIPEX database (in 2015). With seven remaining policy areas (I1–I7), the model can be estimated for one or two classes, but not for three or more. We constructed two models: Model A with seven policy areas, three categories per policy area and one class; and Model B with seven policy areas, three categories per policy area, and two classes in order for classes to encompass different integration regimes of countries. Given the data restrictions imposed on the estimated models, our analysis focused on the distinction between Model A, which classifies all countries into the same type, therefore disregarding the concept of different integration regime types, and Model B, which classifies countries into two different integration regime types. As shown in Table 2, Model B (two classes) has lower values for both AIC (the Akaike Information Criterion) and BIC (the Bayesian Information Criterion), and also higher relative entropy, and so it has Table 2: Model comparison. Model Number of classes Degrees of freedom Number of estimated parameters Log-likelihood Akaike Information Criterion Bayesian Information Criterion Relative entropy A 1 19 14 -253.33 534.64 555.60 0.8804 B 2 4 33 -221.8322 501.66 545.06 0.8953 Figure 1: Conditional probabilities of outcomes given class membership (Model B). Note: I1 = labor market, I2 = family reunion, I3 = education, I4 =political participation, I5=permanent residence, I6=access to nationality, I7=anti-discrimination a better fit to empirical data and the distribution of policy area score values compared to the model with oneclass(Nylund,AsparouhovandMuthén2007).Inotherwords,theclassificationofcountriesintotwo classes has a higher fit than the classification of all countries into one class. Theresultsofthelatentclassanalysis(Figure1)showthatforModelB45.8%ofthecountriesanalyzed belongtoClass1,and54.2%ofcountriesbelongtoClass2.InModelB,Class2includescountriesforwhich the distribution of the outcomes of the variables (policy areas) has a high probability of each of the seven variables (policy areas) having the lowest or median score (1 or 2), and a very small probability of them having the highest score (3). There are higher probabilities for the countries grouped in Class 2 to have the highest value for variables I2, I5, and I7 (family reunion, permanent residence, and anti-discrimina-tion),whereasfortheothersthecountriesfromthisclassdonothavethehighestvalue.Ontheotherhand, countries belonging to Class 1 have high probabilities of each variable’s value being the highest one. In summary, Class 2 consists of the countries where one can expectlow or medium levels of integration pol-icyinthemajorityofareas,whereasClass1containscountriesforwhichonecanexpectthehighestscores for integration policy. The most striking differences between the two classes are in regard to areas I1, I3, I4,andI6(labormarket,education,politicalparticipation,andaccesstonationality),andthesepolicyareas may serve as clear demarcation lines between countries belonging to the highest tier of integration stan­dards and others. Figure 2: Integration regimes according to latent class prevalence membership for the EU 28 and Western Balkan countries. Table 3: The latent class membership for EU 28 and Western Balkan countries. Class 1 Class 2 Austria, Belgium, Denmark, Estonia, Finland, France, Germany, Ireland, Italy, Luxembourg, Netherland, Portugal, Spain, Sweden, United Kingdom Albania, Bosnia and Hercegovina, Bulgaria, Croatia, Cyprus, Czech Republic, Greece, Hungary, Latvia, Lithuania, North Macedonia, Malta, Montenegro, Poland, Romania, Serbia, Slovakia, Slovenia Intermsoflatentclassprevalencemembership(Table3),weidentifiedtwodifferenttypesofintegration regimes in Europe when the Western Balkan countries are included. Class 1 includes EU countries since 1995 (the EU 15) without Greece and with Estonia, and Class 2 includes EU countries since 2004 (EU 13) withGreeceandwithoutEstoniabutwiththeWesternBalkancountries.AsshowninFigure2,aclearpat-tern of class division across Europe can be seen. We find high probabilities for the countries in Class 2 of having achieved high standards regarding immigrants’equalrightsandopportunitiesintheareasoffamilyreunion,permanentresidence,andanti­discrimination(I2,I5,andI7).However,theEU13(withtheexceptionofEstonia)andtheWesternBalkan countries will need to make further efforts in developing integration policies in other areas relevant for integration of immigrants in the host society such as the labor market (I1), education (I3), political par­ticipation (I4), and access to nationality (I6). According to Huddleston (2016), political participation is seen as an area of weakness for integration policy in all Western Balkan countries and most new coun­triesofimmigration,suchastheCzechRepublicandGreece.ThehighestprobabilityfortheEU13(except Estonia) and the Western Balkan countries for achieving standards of integration equivalent to those in the EU 15 countries is in anti-discrimination (I7). The differences between the two classes of immigrant integrationregimespointtoconvergenceinintegrationpoliciesandpracticesinEurope.ThenewEUmem­berstatesandWesternBalkancountriesneedmoretimetocreateandimplementlawsandpolicesinorder toachievecomprehensivehighstandardsdefinedbytheEUregardingimmigrants’equalrightsandoppor­tunities. 4 Discussion The validity of traditional national models of integration is currently being questioned and reviewed due tothetransformationofnationalintegrationpoliciesaffectedbytheprocessofEuropeanizationanddiverse recent migration challenges (Joppke 2007; Choquet 2017). Calling for more international and compara­tive migration research (Bommes and Thränhardt 2010) considers paradigms of migration research to be »national paradigms« shaped with states’ reactions toward international migration and its social effects identifiedinnationalmodels.Inthissense,Pajnik(2007)advocatesthinkingofmigrationfromatransna­tional citizenship perspective. Variousimmigrationexperiencesofcountries,theshareofimmigrantsinthenationalpopulation,and diverseinternationalmigrationpatternsarereflectedinpoliciesandmeasuresofimportanceforintegrating immigrants. The comparison and evaluation of different countries’ migration policies make it possible to monitorthebestintegrationpractices.Basedontheoverallscoreonthe2015MIPEXIndexwithouthealth (index value 45) for the Western Balkan countries, immigrant integration policies in those countries are barely halfway favorable for social integration (Huddleston 2016). Although this value is in line with the MIPEXindexvalue(42)forthenewEUmemberstates(2004–2013),thereisroomforimprovementcom­paredtotheEU15countries(61),whichdobetterinprovidingrightsthatmayproducebetterintegration outcomes.OurresultspointtotwoclassesofintegrationregimesfortheEU28andWesternBalkancoun­tries,withacleardistinctionbetweentheEU15countriesononehandandtheEU13andWesternBalkan countries on the other. The exceptions are Greece and Estonia, which, regarding their EU membership group,doworseandbetterinthegovernanceofintegratingimmigrants,respectively.Lowerscoresinsome policy areas in Greece in comparison to Estonia show that targeted support in Greece is missing. Limited fundsforintegrationarenotinlinewithalargenumberofasylumseekersandillegalarrivalsofimmigrants. Lookingattargetedsupportineducationandthelabormarket,thereismoreresponsivenesstoimmigrants’ specificneedsinEstonia,whichhasasmallnumberofimmigrantsperyear(Huddlestonetal.2015).High standards in anti-discrimination in Greece (Huddleston et al. 2015) are in line with the probability val­ues of the class it belongs to. The Western Balkan countries are continuously harmonizing their legislation relevant to migration with EU standards. However, according to the MIPEX report for the countries of the Western Balkan (Huddleston 2016), indicators of targeted support for immigrants in the labor market and in education express that those measures are completely nonexistent. There are no proper immigrant integration poli­cies for immigrants’ participation in political life, consultative bodies, and implementation policies. Nevertheless,customizingdifferentservicesforasmallnumberofimmigrantsisnoteasilyfeasible.Likewise, the creation of targeted measures to support the integration of immigrants in certain sectors, such as the labormarket,education,orpoliticalparticipation,dependsonthefinancialcapacityofthestateandaddi­tional funds for these purposes. For EU member states, financial resources for integration measures are available through EU funds, whereas other countries face more challenges in this regard. Doomernik and Bruquetas-Callejo(2015)pointoutthatintegrationmeasuresincentralEuropelargelydependonEUfunds. Comparing our classification with the results obtained by Meuleman and Reeskens (2008) as well as with the results obtained by Zamfir et al. (2014) for twenty-seven EU countries, based on the 2007 and 2010 MIPEX data, respectively, we find that the outcomes are not consistent when the Western Balkan countries are incorporated into the classification of integration regimes. An important difference is that, based on the same policy areas – except for Meuleman and Reeskens (2008), who did not cover educa-tion–MeulemanandReeskens(2008)andZamfiretal.(2014)classifiedintegrationregimesintooneclass more (i.e., three). Latent Class 2 from our analysis is similar to latent Class 2 of Meuleman and Reeskens (2008),whichencompassedalleasternEuropeancountriesplusAustria,Denmark,Greece,andMalta,which were considered to have exclusionist integration policies, characterized by difficult access to nationality andpoliticalparticipation.Inthisregard,Denmarkhasmadesignificantprogresssince2007,whereasprogress in integration policy in Austria is related to labor market mobility as the major political priority for new integrationpolicies(Huddlestonetal.2015).Inourclassification,basedonthe2014MIPEXdata,Austria andDenmarkareinClass1,pointingtotheimprovedlegalenvironmentofimmigrantintegration.According toourresearch findings,differences andinequalities inthegovernanceofimmigrant integrationbetween the old and the new EU member states are more visible when a broader territory is covered. InthecourseoftheanalysisoftheMIPEXdata,asLukić(2018)pointsout,itshouldbeborneinmind that these data point only to the legal and political framework of immigrant integration, and not to more complex integration policy implementation and effectiveness. According to Czaika and de Haas (2013), thecomplexityofperceptionoftheeffectivenessofimmigrationpoliciesmightbetheoutcomeoftheunclear boundarybetweenmigration/integrationpolicyaswellasbetweenmigrationpolicyandotherpublicpoli­cies.Indeed,theintertwiningofdifferentareaswithintheinstitutionalframeworksignificantlyaffectsthe integration of immigrants. Given the diverse integration experience and immigration patterns between countries, the appropriateness of comparing integration policies through MIPEX has recently been ques­tioned(GregurovićandŽuparić-Iljić2018).Nonetheless,MIPEXisthemostcomprehensivedatabaseand the most widely used tool for measuring immigrant integration policies. 5 Conclusion This article examined immigrant integration regimes in Europe according to policy priorities related to immigrants’ socioeconomic rights. A latent class analysis showedthat a model with two classes has a bet­ter fit to the MIPEX data than a baseline model in which no distinction between classes is made. Due to the data and method limitations, no further models could be tested. The results of the latent class analysis demonstrate that two distinct types of integration regimes can be distinguished across the EU 28 and the Western Balkan countries. There are clear differences between the EU 15 group of countries versus the EU 13 and Western Balkan countries group regarding integra-tionregimes.Thecountriesinthefirstgrouphaveachievedhighstandardsinpolicyareasrelevanttointegrating immigrants. Furthermore,intheWesternEuropeancountriesvariationsinimmigrantintegration-related legislationandpolicyinstrumentsproducedbyhistoricalandnationalcontextsaredecreasing.Thecountries from the more numerous second group are lagging in developing integration policies and targeted mea­suresforsupportingtheintegrationofimmigrants.Thisisinlinewiththelong-termprocessofintegrating immigrants into the countries’ societies, whereas creating targeted measures for supporting the integra­tionofimmigrantsisnoteasytoachieveincountrieswithasmallnumberofimmigrantsandlimitedfinancial means. Continued comparative analysis of national policies toward the integration of immigrants can shed light on how particular countries manage to incorporate immigrants in various social spheres while mak­ing changes in those policies at the same time. The outcomes provide inputs for policies aiming at best practices in immigrant integration. Monitoring differences within a broader geographical and social set­tingmakesitpossibletobetterunderstandthestrengthsandweaknessesofintegrationregimes.Themain conclusion is that there is convergence in integration regimes where the temporal dimension of migra­tion policy development and the geographical dimension are visible. ACKNOWLEDGEMENT: This paper was written as part of the 2020 Research Program of the Institute ofSocialScienceswiththesupportoftheMinistryofEducation,ScienceandTechnologicalDevelopment of the Republic of Serbia. 6 References Beine,M.,Boucher,A.,Burgoon,B.,Crock,M.,Gest,J.,Hiscox,M.,McGovern,P.,Rapoport,H.,Schaper,J., Thielemann, E. 2016: Comparing immigration policies: An overview from the IMPALA database. International Migration Review 50-4. DOI: https://doi.org/10.1111/imre.12169 Appave, G., Laczko, F. 2011: World migration report 2011. Geneva. Internet: http://publications.iom.int/ system/files/pdf/wmr2011_english.pdf (5.12.2019). Bjerre, L., Helbling, M., Römer, F., Zobel, M. 2014: Conceptualizing and Measuring immigration policies: A comparative perspective. International Migration Review 49-3. DOI: https://doi.org/10.1111/ imre.12100 Bommes, M., Thränhardt, D. 2010: Introduction: national paradigms of migration research. National paradigms of migration research. Osnabrück. Boucher, A., Gest, J. 2014: Migration studies at crossroads: A critique of immigration regime typologies. Migration Studies 3-2. DOI: https://doi.org/10.1093/migration/mnu035 Brubaker, R. 1992: Citizenship and nationhood in France and Germany. Cambridge. Callens,M.S.,Meuleman,B.2016:Dointegrationpoliciesrelatetoeconomicandculturalthreatperceptions? AcomparativestudyinEurope.InternationalJournalofComparativeSociology58-5.DOI:https://doi.org/ 10.1177/0020715216665437 Castles, S. 1995: How nation-states respond to immigration and ethnic diversity. New Community 21-3. DOI: https://doi.org/10.1080/1369183X.1995.9976493 Choquet, S. 2017: Models of integration in Europe. European issues 449. Council of the European Union, 2004: Common basic principlesfor immigrant integration policy in the European Union, 14615/04. Internet: http://www.consilium.europa.eu/ueDocs/cms_Data/docs/ pressData/en/jha/82745.pdf (30.8.2018). Council of the European Union, 2008: European pact on immigration and asylum. Internet: https://register.consilium.europa.eu/doc/srv?l=EN&f=ST%2013440%202008%20INIT (3. 12. 2018). Czaika,M.,deHaasH.2013:Theeffectivenessofimmigrationpolicies.PopulationandDevelopmentReview 39-3. DOI: https://doi.org/10.1111/j.1728-4457.2013.00613.x de Haas, H., Natter, K., Vezzoli, S. 2015: Conceptualizing and measuring migration policy change. Comparative Migration Studies. DOI: https://doi.org/10.1186/s40878-015-0016-5 de Haas, H., Natter, K., Vezzoli, S. 2018: Growing restrictiveness or changing selection? The nature and evolution of migration policies. International Migration Review 52-2. DOI: https://doi.org/10.1111/ imre.12288 Doomernik,J.,Bruquetas-Callejo,M.2015:NationalimmigrationandintegrationpoliciesinEuropesince 1973. Integration processes and policies in Europe: Contexts, levels, actors. Dordrecht. Drew, A. L., Lewis, J. B. 2011: poLCA: An R package for polytomous variable latent class analysis. Journal of Statistical Software 42-10. DOI: https://doi.org/10.18637/jss.v042.i10 Entzinger, H., Biezefeld, R. 2003: Benchmarking in immigrant integration. Rotterdam. Finotelli, C., Michalowski, I. 2012: The heuristic potential of models of citizenship and immigrant integration reviewed. Journal of Immigrant & Refugee Studies 10-3. DOI: https://doi.org/10.1080/ 15562948.2012.693033 Garibay,M.G.,Cuyper,P.,2013:TheevaluationofintegrationpoliciesacrosstheOECD:areview.Antwerpen.Gregurović,S.,Župarić-Iljić,D.2018:Comparingtheincomparable?Migrantintegrationpoliciesandper­plexities of comparison. International Migration 56-3. DOI: https://doi.org/10.1111/imig.12435 Grzymala-Kazlowska, A., Phillimore J. 2018: Introduction: rethinking integration. New perspectives on adaptation and settlement in the era of super-diversity. Journal of Ethnic and Migration Studies 44-2. DOI: https://doi.org/10.1080/1369183X.2017.1341706 Hagenaars, J. A., McCutcheon, A. L. 2002. Applied Latent Class Analysis. New York. Hardwick, S. 2008: Place, space and pattern: Geographical theories in international migration. Migration theory: Talking across disciplines. New York. Helbling, M., Reeskens, T., Stark, C., Stolle, D., Wright, M. 2016: Enabling immigrant participation: Do integration regimes make a difference? Just ordinary citizens? Towards a comparative portrait of the political immigrant. Toronto. Huddleston,T.2016:AregionalMIPEXassessmentoftheWesternBalkans.Internet:http://www.mipex.eu/ sites/default/files/downloads/files/regional_overview_western_balkans.pdf (27.10.2018).Huddleston,T.,Bilgili,Ö.,Joki,A-L.,Vankova,Z.2015:MigrantIntegrationPolicyIndex2015.Barcelona, Brussels.Internet:http://mipex.eu/sites/default/files/downloads/files/mipex-2015-book-a5.pdf(13.8.2018). Joppke, C. 2007: Immigrants and civic integration in Western Europe. Belonging? Diversity, recognition andsharedcitizenshipinCanada.Internet:http://irpp.org/research-studies/immigrants-and-civic-inte­gration-in-western-europe/ (5.12.2019). Josipovič, D., Repolusk, P. 2003: Demographic characteristics of the Romany in Prekmurje. Acta Geographica Slovenica 43-1. DOI: https://doi.org/10.3986/AGS43105 Knill,C.,Lehmkuhl,D.2002:ThenationalimpactofEuropeanUnionregulatorypolicy:ThreeEuropeanization mechanisms.EuropeanJournalofPoliticalResearch41-2.DOI:https://doi.org/10.1111/1475-6765.00012 Lozano,A.A.,García-González,N.,Sebastiani,L.,Veinguer,A.A.,Araujo,S.G.2014:RevisitingtheEuropean Unionframeworkonimmigrantintegration:TheEuropeanintegrationforumasatechnologyofagency. Ethnicities 14-4. DOI: https://doi.org/10.1177/1468796814528692 Lukić,V.2018:Odimigracijedointegracije:Možemolidaučimoodzemaljasarazvijenimpolitikamainte­gracije migranata? [From immigration to integration – can we learn from countries with elaborated immigrantintegrationpolicies?]ZbornikMaticesrpskezadruštvene nauke167.DOI:https://doi.org/ 10.2298/ZMSDN1867639L Martiniello, M. 2006: Towards a coherent approach to immigrant integration policy(ies) in the European Union. Internet: https://www.oecd.org/dev/38295165.pdf (13.8.2018). Melegh, A., Gruber, K., Fassmann, H., Musil, E., Bauer, R., 2014: Longer-term demographic dynamics in South–EastEurope:Convergent,divergentanddelayeddevelopmentpaths.CentralandEasternEuropean Migration Review 3-2. Meuleman, B., Reeskens, T. 2008: The relation between integration policy and majority attitudes toward immigration.AnempiricaltestacrossEuropeancountries.SymposiumofInternationalSocietyofPolitical Psychology. 29th May 2008, Leuven. Nylund,K.L.,Asparouhov,T.,Muthén,B.O.2007:Decidingonthenumberofclassesinlatentclassanalysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling: A Multidisciplinary Journal 14-4. DOI: https://doi.org/10.1080/10705510701575396 Pajnik,M.2007:Integrationpoliciesinmigrationbetweennationalisingstatesandtransnationalcitizenship,withreferencetotheSloveniancase.JournalofEthnicandMigrationStudies33-5.DOI:https://doi.org/ 10.1080/13691830701359298 Parkes, R. 2008: Immigrant integration meets European integration. Berlin. Internet: https://www.swp-berlin.org/fileadmin/contents/products/arbeitspapiere/Immigrant_Update2_KS.pdf (13.8.2018). Penninx, R., Garcés-Mascarenas, B. 2015: The concept of integration as an analytical tool and as a policy concept. Integration processes and policies in Europe: Contexts, levels, actors. Dordrecht. Vertovec,S.2007:Super-diversityanditsimplications.EthnicandRacialStudies30-6.DOI:https://doi.org/ 10.1080/01419870701599465 Zamfir,A.M.,Mocanu,C.,Maer-Matei,M.,Lungu,E.O.2014:ImmigrationandintegrationregimesinEU countries. Journal of Community Positive Practices 14-1. COMMUNITY DEVELOPMENT: LOCAL IMMIGRATION PARTNERSHIPS IN CANADA AND IMPLICATIONS FOR SLOVENIA Mitja Durnik Maple leaf, the Canadian national symbol. DOI: https://doi.org/10.3986/AGS.5136 UDC: 316.4.063.3-054.72(71+497.4) COBISS: 1.01 Mitja Durnik1 Communitydevelopment:LocalImmigrationPartnershipsinCanadaandimplications for Slovenia ABSTRACT:Canadaisperceivedasastronglydesiredfinaldestinationformanyrefugeesandimmigrants due to its socio-economic advantages. The author assesses the Canadian praxis of the immigrant settle­ment from the community development standpoint, with a specific interest to present how successful Canadian immigration policy has been on the local level by using the established Local Immigration Partnershipsmodel.Ontheotherhand,byadoptingtheso-calledrestrictedmodelofimmigrantintegration Sloveniahasnotdevelopedaconsistentmodelofintegration,specificallyleavingasidethepotentialoflocal areas in resolving these complex issues. The paper is confirming that due to institutionalized multilevel partnership Canada has been more successful in immigrant integration than Slovenia. In both countries, however, integration into the health system has been evidently the most acute problem. In order to obtain more relevant results, a mixed-methods research was used combining interviews and integration index­es. In the majority of integration parameters, Canada shows significantly better results than Slovenia. KEY WORDS: geography, community, development, immigrants, integration, Canada, Slovenia Skupnostnirazvoj:lokalnapriseljenskapartnerstvavKanadiinimplikacijezaSlovenijo POVZETEK: Kanada je zaradi svojih socialno-ekonomskih prednosti ciljna država za številne begunce in priseljence. Avtor v članku ocenjuje kanadsko prakso naseljevanja priseljencev s stališča skupnostnega razvoja. V ospredju je predstavitev uspešne kanadske politike priseljevanja in njenega učinka na lokalni ravnizuporabomodelalokalnihintegracijskihpartnerstev. NasprotnopajedosedajSlovenijauveljavljala t.i. restriktivni model integracije priseljencev, pri čemer ni upoštevala potenciala lokalnih skupnosti pri reševanjutekompleksneproblematike.Vprispevkusmopotrdili,dajeKanadazaradiinstitucionaliziranega večstopenjskegapartnerstvauspešnejšavintegracijipriseljencevkotSlovenija.Vobehdržavahjeintegracija priseljencev v zdravstveni sistem najbolj akuten problem. Za namen raziskave smo uporabili kombiniran pristop, kjer smo izvedli intervjuje in primerjali nekatere kazalce integracije priseljencev. Kanada glede na večino integracijskih parametrov kaže bistveno boljše rezultate kot Slovenija. KLJUČNE BESEDE: geografija, skupnost, razvoj, priseljenci, integracija, Kanada, Slovenija The paper was submitted for publication on December 19th, 2018. Uredništvo je prejelo prispevek 19. decembra 2018. 1 University of Ljubljana, Faculty of Administration, Ljubljana, Slovenia mitja.durnik@fu.uni-lj.si 156 1 Introduction Slovenia joined the European Union in 2004 and despite being a member state for over a decade it has still not been recognized as the final country for immigrants, while Canada appears as the decisive goal for newcomers mainly for its rapid economic development. Moreover, comparing both countries, we can also notice different national approaches to integration. Slovenia has been largely following a restrictive approach based on the role of the nation-state (Pajnik 2007), while Canada’s borders have remained open to certain immigrant groups specifically needed for its development purposes (having in mind also the historicalimportanceofimmigration).InCanada,manymultilevelinitiativeshaveemergedofferingpoten­tial solutions for newcomers’ integration. The »welcoming community« approach tried to resolve integrationinsmallcitiesandruralareas(Gibson,BucklaschukandAnnis2017).Thefederalgovernment has launched a program called Community Partnership Settlement Plan partially responding to the con­temporary Syrian refugee crisis. Immigrantsoftenmigratefromhighlyconcentratedethniccommunitiestoraciallymixedplaces(Murdie andGosh2010)inhabitinglessexpensiveareas,whilelateron,theymovetoreputableneighborhoods(Waters andGersteinPineau2016).SegoviaGomez(2011)emphasizedhowcrucialtheethnicgroup’sconcentration for social inclusion is. Canadianauthorsmainlypointedouttherelevanceoflocalpartnershipsinsmallurbanareas(Ma2017; Tibe Bonifacio and Drolet 2017), the formation of minority neighborhoods in big cities (Hou 2004), or specific integration into the urban history (Anisef and Lanphier 2003). According to the Canadian case study concerning Local Immigration Partnerships (onwards LIPs), there has been an important debate onhowruralareashaveinitiateddifferentstrategiesofimmigrantintegration(Ma2017)thanhighlypop­ulated cities. The urban-rural debate and its relation to development strategies in an important manner determinedtheconceptofregionaldevelopmentinSlovenia(KozinaandClifton2019;VintarMally2018; Kozina, Poljak Istenič and Komac 2019). In the Slovenian context, Botrić (2016) emphasized the insufficient targeted support to immigrants by responsible Slovenian institutions. Medvešek and Bešter (2010) addressed crucial issues of integration suchasdemography,housing,andhealth,andVižintin(2017)explainedmulticulturalintegrationinrela­tion to educational challenges. There are three policy fields relevant in our comparative research: (1) labor: how immigrants are inte­gratedinthegenerallabormarket;(2)health:accesstohealthcarespecificallypointingoutculturalaspects of the medical treatment; (3) education: focus on integration of children in schools (see Vižintin 2017). Statesusuallyplaceimmigrantsintotwobroadcategories.Firstly,laborimmigrantsleavinghomeland on an economic basis, and, secondly, asylum seekers and refugees escaping from war zone areas, where their lives are possibly at risk (Kivisto and Faist 2009). 1.1 Community Development Models Communities are somehow a result of economic, cultural and social functions. In fact, the concept has beencaughtupbetweenconventionalperceptionofacommunityasaplace-basedstructureandtheso-called post-place communities having in mind also peoples’ social interactions. Geographical view within the firstoneisoftenrepresentedasarural-urbandebateregardingmodernizationinsociety(Bradshaw2008). Rothman(1996and2007)hasinitiatedthreecommunitydevelopmentapproachesthatmayinterconnect. Planning/policy (social planning) allows planners to use empirically proven data and act as »rationalist« thinkers.Itpresupposestechnocratictop-downactivitieswheregovernmentsusuallydirectpolicyactions (Chen and Ku 2016). Community capacity development (locality development) builds its efforts on peo­ples’ empowerment within communities and is participatory oriented. Planners try to shape the wide spectrum of different governmental and civic organizations in planning tasks. (Hardcastle, Powers and Wenocur2011).Socialadvocacy(socialaction)isoftenusedbymarginalgroupstryingtoradicallychange power relations regarding decision-makers. According to the above discussion, the following general hypotheses were examined: • Due to institutionalized multilevel partnerships, Canada is more successful in integrating immigrants in communities than Slovenia. Additional sub-hypotheses were formed in order to highlight the comparison between the countries: • Thecommunitycapacitydevelopmentmodelisimportantlyconnectedwithahighernationalintegration policy index. • Healthcare integration is the most acute problem in Slovenia and Canada. 2 Methods In the paper, macro (state) policies have been compared using the so-called Migrant Integration Policy Index(onwardsMipex)consistingof167differentpolicyindicatorsjoiningindifferentpolicyasindicated below (index developed by Huddleston et al. 2015). Using Mipex data, we calculated national integration policy in Canada and Slovenia as reflected through governmental decisions on the afore-mentioned pol­icy fields. In general, data are available from 2007 to 2014 for 35 different countries. Communities’ integration capacity research was based on various secondary sources such as reports and experts’ analyses concerning community development. Due to major shortcomings of available rel­evant studies in Slovenia ten semi-structured interviews were conducted with approximately 45-minutes length each, targeting different relevant policy actors like NGOs, policy experts, municipality represen­tatives,andpublicofficials.FormeasuringLjubljanamunicipality’scapacityinclusion,theCitiesofMigration Diagnostic tool was applied (the MyCOM Diagnostic tool is part of a larger project at Cities of Migration (How inclusive…2019) supported by the Open Society Foundations and led by the Cities of Migration initiativeatRyersonUniversity’sGlobalDiversityExchangewithinternationalpartners).In2018,byusing the MyCOM tool, we asked 55 individuals from the Ljubljana municipality area to fulfill parts of the diag­nostic tool (questions related to health, education and economic issues). 3 Building community capacity within Torontonian neighbourhood In 1998 the provincial Ontario government declared to merge six previously independent municipalities into a single one (Sturdy 2014). Local Immigration Partnerships have been initiated by the federal gov­ernment to improve the potential of local communities to resolve issues of immigrant integration (Local Immigration Partnerships… 2014). Torontohasthesecondlargestratioofforeign-borninhabitantsamongthebiggestworldcities(Murdie andGosh2010;AnisefandLanphier2003).Until2012,theTorontonianLIPmodelwasbasedontheneigh­borhood/communityconcept.Lateron,regionalpartnershipshavebeenestablisheddividingmunicipality intofourquadrants.Torontomunicipalitywasreplacedwithfundingagenciesmanagingactivities»onthe ground« (Sturdy 2014). The North Quadrant consisted of Bathurst-Finch, Don Valley,Lawrence Heights, and North York East. In the North York East lived 80,000 residents, 70% of whom were declared as immi­grants. Recent newcomers are often highly educated but they would still represent a significant ratio in the unemployment structure. Only 30% were born in Canada. The countries of origin were China and Hong Kong, Iran, India and the Philippines. 44% of all were born in China (The North York East Local immigration Partnership 2013). AfundamentalstepincreatinglocalpartnershipswastheestablishmentofLIPsCouncils,mainlycon­sistingofsettlementagencies,municipalities,schools,employmentagencies,groupsofnewcomers,provincial administrative bodies, regional governments, etc. ((Local Immigration Partnerships…2014). Important tasks of a LIP council are (Pero 2017):(1) creation of a three-year settlement and integration strategy; (2) determination of action plans and implementation strategies; (3) managing regular meetings with part­ners; (4) education and research. Some of its tasks are delegated to working groups (Pero 2017). Theprojectteamconsistedofdifferentexperts.Twoworkerswerehiredformanagingtheproject,while five animators were carrying outdifferent research activities. Eighteen community organizations became part of the team. Several consultants were identifying the needs and priorities (The North York…2013). Community psychical asset mapping demanded to locate geographical distribution of facilities, focus groups(involvedmorethan200immigrants)tookplacewithserviceprovidersaccordingtodifferentissues. Further activities shaped informant interviews asking management staff from community organizations about their needs. Local service delivery analysis detected the size and financial capabilities of community organizations.Communityconsultationsextendedpreviousdebatespecificallybuildingonafuturevision of community development. Imagining the community’s social and geographic space, newcomers were usingmapsofcommunityinroundtablesdiscussingdifferentobstaclesthattheyfoundproblematic.Then, a service provider meeting applied available information from newcomers’ consultations. A special task ofthecommunityadvisorypanelwastoredefineandreconstructprioritiesandareasofcommunitydevel­opment orienting to build up a new strategic plan (The North York…2013). Figure 1: LIP Multilevel Governance Structure focused on sub-regional LIPs in Toronto. Figure 2: North York East LIP strategic planning process (adapted from The North York…2013). Figure 3: Locality development model as an example of the semi-independent system. Table 1: North York East Service Analysis (selected categories) (adapted from The North York…2013 and Kobayashi et al. 2012). Categories Newcomer & service Area assets & strengths Service gaps & areas for providers’ experience improvement Labour and business • lack of working experience • employment agencies located • structural barriers for newcomers’ • fewer quality jobs within within the community employment the community • inclusion in co-decision-making • lack of assessment of immigrant skills Education and youth • no information about Canadian • full-day kindergarten schools • limited information in schools school culture • preschool training for parents • strict eligibility criteria for subsidies • financial problems for extra-curriculum activities Health • limitation in access to services • nearby health services • communication barriers due • additional fees for some services • information about services to different languages • stigmatization due to mental health problems 4 Social planning in the Slovenian context Somecivilgroupsorganizedpetitionsinfavorandagainstapotentialimmigrantsettlement–particularly, concerning the establishment of accommodation centers (Rijavec and Pevcin 2018). Accordingly, there have been many conflictual relations between municipalities and government concerning major respon­sibilitiesalthoughthestateformallyprovidedamajorshareofpublicservicesrecourses.Inreality,therole ofmunicipalitiesintheintegrationprocesshasbeenminimizedasmuchaspossible.Combiningdatagiven fromtheCitiesofMigrationdiagnostictool(measuringinclusioninLjubljana)andMipex(detectinginte­gration in Slovenia) the following matrix shows that health issue is the most critical policy in multilevel joint action. The Ljubljana municipality has supported major national policy initiatives under its jurisdiction. Integration policy has been a continuation of the state policy with some limited cooperation with non­governmental actors and the state (Interviews 2018) It needs to be highlighted that in terms of the national legislation newcomer workers still face many breaches. Specifically, in the city of Ljubljana, there have been some innovative ideas on how to include politicalrefugees,forinstance,toformakindofamulti-ethnicbusiness(e.g.»Skuhna«restaurant).Despite several successful attempts to start ethnic businesses the question always remains, as to how to continue when public funding is over (Interviews 2018). Table 2: State capacity of immigrant integration (measuring with Mipex) vs. municipal (city) inclusion policies (measuring with Cities of Migration online diagnostic tool). Cities of Migration 2018 (Ljubljana) Mipex 2014 (Slovenia) Labour Education Health Halfway favorable Slightly unfavorable Unfavorable Labour Intentional XXX Education Strategic XXXX Health Awareness X Overall index Intentional Overall index Halfway favorable XXX LEGEND: Mipex scale: 80–100 – favorable, 60–79 – slightly favorable, 41–59 – halfway favorable, 21–40 – slightly unfavourable, 1–20 – unfavorable, 0 – critically unfavorable. Cities of Migration scale (arithmetic mean): 80–100% – inclusive, 60–79% – strategic, 40–59% – intentional, 20–39% – awarenes, 0–19% – invisible. Matrix (level of cooperation): xxxxx highly cooperative, xxxx cooperative, xxx middle cooperative, xx little cooperative, x non-cooperative. NOTE: Percentages given by respondents were translated into scales (different ones for Mipex and MyCom). Table 3: Excerpts from interviews (2018) indicating the most acute issues (and positive aspects) of immigrant integration in Ljubljana municipality. Categories Newcomer and service Area assets and strengths Service gaps and areas providers’ experience for improvement Workplace integration • Hiring workers through working • Employment workshops • Improved target support for the employment agencies • Support of employment agency recognition of working skills • Limited legal protection • Mentoring programs for new entrepreneurs Education, children and youth • Language barriers between • Language training for immigrant • Higher financial stimulation parents and teachers children parents for teachers • Shortage of extra-curriculum • Two-stage model of integration • Developing approaches to activities understand diversity Health • Only emergency treatment • Vicinity of some health services • The need for translators for asylum seekers • Clinic for individuals without • Specific intercultural education • Out-of-pocket payments health insurance for workers for some chronic disease AsfarastheSlovenianeducationalsystemgoes,itstilldoesnotfullymeetmulticulturalcriteria.Some basicpolicydirectiveshavebeenacceptedatthenationallevelarrangingyouthintegrationintotheschool system.Forexample,Ljubljana’selementaryschoolLivadahashadaprominenthistoryinimmigrantchil­dren’sintegration.Themunicipalityalsosupportsaprojectarranginginformationandeducationalactivities for immigrant youth (Interviews 2018). Althoughthestateassuredbasichealthcareservicesasystemicapproachtotheissuewasmissing.Some veryfundamentalproceduralshortcomingswereidentified:limitedaccesstodoctorsandinterculturalbar­riers. Above all, medical personnel often did not recognize the equal health rights for persons holding international protection (Interviews 2018). 5 Comparative multilevel analysis Canada is still positioned among the countries which have carried out the most notable practices of inte­gration (see figure 5). The Canadian integration policy has not radically changed between 2008 and 2014. Comparatively, according to the Mipex index Slovenia did not achieve the results of most western demo­cratic states. FollowinganalysisofvariousLIPsacrossOntario,mentoringprogramsandinternshipsseemedimpor­tanttoimmigrantsusingmoreeffectivelytheireducationalcredentials(Kobayashietal.2012).Atthefederal level, permanent residents, reunited families and some temporary workers enjoy some of the best labor market opportunities in the developed world (Huddleston et al. 2015). In the case of Slovenia, in some parts of the private sector, non-EU immigrants have often had difficulties to access jobs in this sector (Huddleston et al. 2015). These issues are mostly related to the bureaucratic barriers in the employment process (Interviews 2018). The overall score of immigrant access to education in Canada is lower than in comparable countries, mostly since only a minority of programs offer immigrants equal access to higher and university educa­tionandtovocationaltraining(Huddlestonetal.2015).TheLIPsanalysissuggestedthatimmigrantswould need better coordination of education services (Kobayashi et al. 2012). The Slovenian school system has reactedslowlytothespecialneedsofthenewcomers–e.g.,therewasnospecializedcenterofferingexpert services (Huddleston et al. 2015). While most interviewees put educational integration very high among other policy groups, Slovenian data show low indexes of integration (Interviews 2018). Important shortcomings in health integration as an unresponsive healthcare system to immigrants’ needs are evident in Canada (Huddleston et al. 2015). Cross-sectional LIPs analysis mainly suggested the removalofsystemicbarriersapproachingmentalhealthservices,(Kobayashietal.2012).Canadalagsbehind somecountriesasAustralia,NewZealand,USAortheUnitedKingdom(Huddlestonetal.2015).Slovenia obviously does almost nothing to integrate and orient newcomer patients into the health system and to address any of their specific health needs (Huddleston et al. 2015). As specifically expressed in Slovenian Table 4: Parameters measuring labor market integration (calculated Mipex index: source of database: Huddleston et al. 2015). Indicators Access to general labor market Access to general support Targeted support Workers’ rights Labor market mobility (overall index) Canada (2007) Canada (2014) Slovenia (2007) Slovenia (2014) EU 25 (2007) EU 28 (2014) 90 90 30 30 56 61 58 83 42 50 55 62 50 50 10 20 29 36 100 100 50 50 69 71 75 81 33 38 52 57 Table 5: Parameters measuring education opportunities (calculated Mipex index: source of database: Huddleston et al. 2015) Indicators Access to education Targeting needs New opportunities Intercultural education Education (overall index) Canada (2010) Canada (2014) Slovenia (2007) Slovenia (2014) EU 25 (2007) EU 28 (2014) 50 50 N/A 33 42 34 80 80 N/A 17 90 47 60 60 N/A 15 80 23 70 70 N/A 40 80 43 65 65 N/A 26 73 37 Figure 7: Square area gives comparison of Canada and Slovenia (different policy groups) (calculated Mipex index: source of database: Huddleston et al. 2015). The figure presents the overall calculated Mipex index for Slovenia and Canada by various policy groups. The overall score inside every policy is100pointsasthemaximumandshowstheratiobetweenSloveniaandCanada. Forexample,withinsectionhealthCanadareaches80points,Slovenia only 20 points. Table 6: Parameters measuring overall scores (calculated Mipex index: source of database: Huddleston et al. 2015). Country Year Overall score Canada 2014 70 2013 71 2012 71 2011 71 2010 71 Slovenia 2014 48 2013 49 2012 49 2011 49 2010 48 EU 28 2014 52 2013 52 interviews, there has been no organized training of medical personnel and doctors for improving com­municationwithnewcomersandunderstandingtheirspecificculturalneedsinmedicaltreatment.Inmany cases, merely NGOs are those who offer translators and mediators in these activities (Interviews 2018). The Canadian federal integration policy has largely reflected local characteristics. In Slovenia, inco­herent national policy without a long-term vision resulted in individual local social action experiments lackingsignificantjointeffectsonalllevelsofpolicy-making.Putsimply,theshortcomingofseriousimmi-grant integration national program/strategy mainly corresponds with a lower level of the Mipex index. ThefinalcomparisonoftheoverallMipexindexshowsthatCanadianintegrationpolicyonthenation­al level has been far more successful than Slovenian. The general score for years 2014/2015 indicates that theoverallEUmembers’integrationpolicyismorerestrictivethanintheCanadiancase.Thismaybepar­tially explained also by geopolitical reasons. 6 Conclusion Therearecertainlysignificantcorrelationsbetweenpathsofcommunitydevelopmentandlevelsofimmi­grant integration. Canada demonstrates higher integration potential using the so-called capacity development approach allowing local communities to carry out the implementation of integration poli­cies. The Slovenian restrictive national policy without a coherent integration approach is reflected in the conventional social policy planning development model. EducationalpolicyinSloveniahasbeencaughtbetweenlowerMipexindexesandmorepositiveinter­viewee opinion. We can agree that the Slovenian educational system quickly corresponds with changes in immigrant integration patterns. In both countries, healthcare integration is the most acute issue among policy groups compared. We could argue that intercultural barriers represent the most important future challenges for policy-makers. Inside the Canadian debate, integration provides different social/spatial relations than assimilation. Using integration logic, communities may be integrated based on joint values and are less caught up in ghettoization. Ghettos are usually a result of assimilation processes where community members strong­ly build their identity on traditional cultural patterns from homeland (Vipond 2017). 7 References Anisef, P., Lanphier, M. 2003: The world in a city. Toronto. Botrić, V. 2016: Attitudes towards immigrants, immigration policies and labour market outcomes: ComparingCroatiawithHungaryandSlovenia.CroatianInternationalRelationsReview22-76.DOI: https://doi.org/10.1515/cirr-2016-0004 Bradshaw,T.K.2008:Thepost-placecommunity:Contributionstothedebateaboutthedefinitionofcom­munity. Community Development 39-1. DOI: https://doi.org/10.1080/15575330809489738 Chen,Y.-Y.,Ku,Y.-W.2017:Communitypracticeatacrossroad:theapproachesandchallengesinTaiwan. Community Development Journal 52-1. DOI: https://doi.org/10.1093/cdj/bsw046 Gibson, R., Bucklaschuk, J., Annis, R. C. 2017: Working together: Collaborative response to welcoming newcomersinBrandon,Manitoba.Canadianperspectivesonimmigrationinsmallcities.Cham.DOI: https://doi.org/10.1007/978-3-319-40424-0_3 Hardcastle,D.A.,Powers,P.R.,Wenocur,S.2011:Communitypractice:Theoriesandskillsforsocialworkers. New York. Hou, F. 2005: Recent immigrants and the formation of visible minority neighbourhoods in Canada’s large cities. Analyitical Studies Branch Research Paper Series 221. How inclusive is my city? 2019. Internet: https://my.citiesofmigration.org/ (3.8.2018). Huddleston,T.,Bilgili,Ö.,Joki,A.-L.,Vankova,Z.2015:Migrantintegrationpolicyindex2015.Barcelona. Internet: http://mipex.eu/sites/default/files/downloads/files/mipex-2015-book-a5.pdf (25.4.2018). Kivisto,P.,Faist,T.2009:BeyondaBorder:TheCausesandConsequencesofContemporaryImmigration. Thousand Oaks. Kobayashi,A.,Ratnayake,A.,Newbold,B.,Nicholson,C.,Andrew,C.,Zinga,D.,Hagar,D.,Acara,E.,Taraky, H.,Dam,H.,Nadeau,J.,Rocheleau,J.,Pottie,K.,Manning,L.M.,Tossutti,L.,Casagrande,M.,Gordon, M., Paquet, M., Rothstein, M., Brisson, N., Cliff-Jungling, S., Satzewich, V., Esses, V. 2012: Analysis of LIP strategic plans: Promising practices. Welcoming community’s initiative. Internet: http://p2pcana­da.ca/wp-content/blogs.dir/1/files/2013/05/The-Analysis-of-LIP-Strategic-Plans-Promising-Practices.pdf (8.5.2018). Kozina, J., Clifton, N. 2019: City-region or urban-rural framework: what matters more in understanding the residential location of the creative class? Acta geographica Slovenica 59-1. DOI: https://doi.org/ 10.3986/AGS.5137 Kozina, J., Poljak Istenič, S., Komac, B: 2019: Green creative environments: Contribution to sustainable urban and regional development. Acta geographica Slovenica 59-1. DOI: https://doi.org/10.3986/ AGS.7030 Local Immigration Partnerships: Outcomes 2008-2013, 2014: Internet: http://p2pcanada.ca/files/2014/ 07/Local-Immigration-Partnerships-Outcomes-2008-2013.pdf (3.4.2018). Ma, M. C. K. 2017: Local immigration partnerships. How is Peterborough engaged with immigrant inte­gration? Canadian perspectives on immigration in small Cities. Cham. DOI: https://doi.org/10.1007/ 978-3-319-40424-0_4 Medvešek,M.,Bešter,R.2010:Državljanitretjihdržavalitretjerazrednidržavljani?Integracijadržavljanov tretjih držav v Sloveniji. Ljubljana. Murdie, R., Gosh, S. 2010: Does spatial concentration always mean a lack of integration? Exploring ethnic concentration and integration in Toronto. Journal of Ethnic and Migration Studies 36-2. DOI: https://doi.org/10.1080/13691830903387410 Pajnik,M.2007:Integrationpoliciesinmigration:Betweennationalisingstatesandtransnationalcitizenship, withreferencetotheSloveniancase.JournalofEthnicandMigrationStudies33-5.DOI:https://doi.org/ 10.1080/13691830701359298 Pero, P. R. 2017: The new local governance of immigration in Canada: Local immigration partnerships and their role in immigrant settlement and integration in small- and medium-sized Ontarian cities. Ph.D. thesis, Queen’s University. Kingston. Rijavec, D., Pevcin, P. 2018: An examination and evaluation of multilevel governance during migration crisis.ThecaseofSlovenia.CentralEuropeanPublicAdministrationReview16-1.DOI:https://doi.org/ 10.17573/cepar.v16i1.359 Rothman,J.1996:TheInterweavingofcommunityinterventionapproaches.JournalofCommunityprac­tise 3, 3-4. DOI: https://doi.org/10.1300/J125v03n03_03 Rothman, J. 2007: Multi modes of intervention at the macro level. Journal of Community Practice 15-4. DOI: https://doi.org/10.1300/J125v15n04_02 Segovia, Gomez, A. 2011: Spatial research study of Peruvian immigrants in Santiago de Chile. Acta geo­graphica Slovenica 51-1. DOI: https://doi.org/10.3986/AGS51109 Sturdy, S. C. H. 2014: From government to governance: The city of Toronto’s role in immigrant settlement service coordination since the Canada-Ontario Immigration Agreement. MA Thesis, Simon Fraser University. Burnaby. The North York East Local Immigration Partnership: Strategic plan and report, 2013: Internet: http://torontonorthlip.ca/sites/torontonorthlip.ca/files/northyork_report.pdf (6.7.2019). Tibe Bonifacio, G., Drolet, J. L. 2016: Canadian perspectives on immigration in small cities. Cham. VintarMally,K.2018:RegionaldifferencesinSloveniafromtheviewpointofachievingEurope’ssustainable development. Acta geographica Slovenica 58-2. DOI: https://doi.org/10.3986/AGS.3309 Vipond, R. 2017: Making a global city: How one Toronto school embraced diversity. Toronto. Vižintin, M. A. 2017: Medkulturna vzgoja in izobraževanje: vključevanje otrok priseljencev. Ljubljana. Waters, M. C., Gerstein Pineau, G. 2016: The integration of immigrants into American society. Panel on the Integration into American society. Washington. DOI: https://doi.org/10.17226/21746 Guidelines for contributing authors in Acta geographica Slovenica EDITORIAL POLICIES 1 Focus and scope TheSloveniangeographicaljournalActageographicaSlovenica(printversion:ISSN:1581-6613,digitalver­sion: ISSN: 1581-8314) is published by the Anton Melik Geographical Institute of the Slovenian Academy of Sciences and Arts Research Center. Acta geographica Slovenica publishes original research papers from all fields of geography and related disciplines,andprovidesaforumfordiscussingnewaspectsoftheory,methods,issues,andresearchfind­ings, especially in central and southeast Europe. We accept original research papers and review papers. Paperspresentingnewdevelopmentsandinnovativemethodsingeographyarewelcome.Submissions should address current research gaps and explore state-of-the-art issues. Research based on case studies shouldhavetheaddedvalueoftransnationalcomparisonandshouldbeintegratedintoestablishedornew theoretical and conceptual frameworks. Thetargetreadershipisresearchers,policymakers,anduniversitystudentsstudyingorapplyinggeog­ raphy at various levels. Submissions are accepted in English or Slovenian. Thejournalisindexedinthefollowingbibliographicdatabases:SCIE(ScienceCitationIndexExpanded), Scopus,JCR(JournalCitationReport,ScienceEdition),ERIHPLUS,GEOBASEJournals,CurrentGeographical Publications,EBSCOhost,Geosciencee-Journals,Georef,FRANCIS,SJR(SCImagoJournal&CountryRank), OCLC WorldCat, and Google Scholar. The journal’s publisher is a member of CrossRef. 2 Types of papers Unsolicited or invited original research papers and review papers are accepted. Papers and materials or sections of them should not have been previously published or under consideration for publication else­where. The papers should cover subjects of current interest within the journal’s scope. 3 Special issues The journal also publishes special issues (thematic supplements). Special issues usually consist of invited papers and present a special topic, with an introduction by the (guest) editors. The introduction briefly presents the topic, summarizes the papers, and provides important implications. 4 Peer-review process Allpapersareexaminedbytheeditor-in-chief.Thisincludesfact-checkingthecontent,spellingandgram­mar,writingstyle,andfigures.Papersthatappeartobeplagiarized,areghost-written,havebeenpublished elsewhere, are outside the scope of journal, or are of little interest to readers of Acta geographica Slovenica mayberejected. Ifthearticleexceedsthemaximumlength,theauthor(s)mustcorrectthisbeforethearticle isreviewed.Thepaperisthensenttoresponsibleeditors,whochecktherelevance,significance,originality, clarity,andqualityofthepaper.Ifacceptedforconsideration,thepapersarethensenttopeerreviewer(s)for double-blindreview. Paperarerejectedoracceptedbasedonthepeerreviewsandeditorialboard’sdecision. 5 Publication frequency Acta geographica Slovenica is published three times a year. 6 Open-access policy This journal provides immediate free open access to its content and supports greater global exchange of knowledge by making research freely available. The papers in Acta geographica Slovenica and its prede­cessorsActageographica/GeografskizbornikandGeographicaSlovenicaareavailableonlinefreeofcharge. The author(s) receive a free print copy. The journal’s publication ethics and publication malpractice statement is available online, as well as information on subscriptions and prices for print copies. AUTHOR GUIDELINES Beforesubmittingapaper,pleasereadthedetailsonthejournal’sfocusandscope,peer-reviewprocess,pub­lication frequency, history, and open-access policy. This information is available in the editorial policies. 1 The papers Research papers must be prepared using the journal’s template and contain the following elements: – Title: this should be clear, short, and simple. – Informationaboutauthor(s):submitnames(withoutacademictitles),institutions,ande-mailaddresses through the online submission system. – Highlights:authorsmustprovide3–5highlights.Thissectionmustnotexceed400characters,including spaces. – Abstract: introduce the topic clearly so that readers can relate it to other work by presenting the back­ground, why the topic was selected, how it was studied, and what was discovered. It should contain one or two sentences about each section (introduction, methods, results, discussion, and conclusions). The maximum length is 800 characters including spaces. – Key words: include up to seven informative key words. Start with the research field and end with the place and country. – Main text: The main text must not exceed 30,000 characters, including spaces (without the title, affil-iation,abstract,keywords,highlights,referencelist,andtables).Donotusefootnotesorendnotes.Divide the paper into sections with short, clear titles marked with numbers without final dots: 1 Section title. Use only one level of subsections: 1.1Subsectiontitle. Research papers should have the following structure: • Introduction: present the background of the research problem (trends and new perspectives), state of the art (current international discussion in the field), research gap, motivation, aim, and research questions. • Methods: describe the study area, equipment, tools, models, programs, data collection, and analysis, define the variables, and justify the methods. • Results:followtheresearchquestionsaspresentedintheintroductionandbrieflypresenttheresults. • Discussion:interprettheresults,generalizefromthem,andpresentrelatedbroaderprinciplesandrela­tionshipsbetweenthestudyandpreviousresearch.Criticallyassessthemethodsandtheirlimitations, and discuss important implications of the results. Clarify unexpected results or lacking correlations. • Conclusion:presentthemainimplicationsofthefindings,yourinterpretations,andunresolvedques­ tions, offering a short take-home message. Review papers (narratives, best-practice examples, systematic approaches, etc.) should have the follow­ ing structure: • Introduction:include1)thebackground;2)theproblem:trends,newperspectives,gaps,andconflicts; and 3) the motivation/justification. • Materialandmethods:provideinformationsuchasdatasources(e.g.,bibliographicdatabases),search terms and search strategies, selection criteria (inclusion/exclusion of studies), the number of studies screened and included, and statistical methods of meta-analysis. • Literaturereview:usesubheadingstoindicatethecontentofthevarioussubsections.Possiblestruc­ture:methodological approaches,models ortheories,extent of support fora giventhesis,studies that agree with one another versus studies that disagree, chronological order, and geographical location. • Conclusions:provideimplicationsofthefindingsandyourinterpretations(separatefromfacts),iden­tify unresolved questions, summarize, and draw conclusions. – Acknowledgement:usewhenrelevant.Inthissectionauthorscanspecifythecontributionofeachauthor. – Reference list: see the guidelines below. 2 Paper submission 2.1 Open journal system Author(s) must submit their contributions through the Acta geographica Slovenica Open Journal System (OJS) using the Word document template. Enter all necessary information into the OJS. Any addition, deletion, or rearrangement of names of the author(s) in the authorship list should be made and confirmed by all coauthors before the manuscript has been accepted, and is only possible if approved by the journal editor. Tomakeanonymouspeerreviewpossible,thepapertextandfiguresshouldnotincludenamesofauthor(s). Donotusecontractionsorexcessiveabbreviations. Useplaintext,withsparinguseof bold and italics. Do not use auto-formatting, such as section or list numbering and bullets. If atext is unsatisfactory, the editorial board mayreturn it to the author(s) for professional copyediting orrejectthepaper. Seethesectiononthepeer-reviewprocessfordetails. Author(s)maysuggestreviewers when submitting a paper. 2.2 Language Papers are published in English. Papers are submitted in English or Slovenian and copyedited/translated after acceptance by a profes­sional chosen by the editorial board. The translation or copyediting costs are borne by the author(s) (translation €500, copyediting €200) and must be paid before layout editing. All papers should have English and Slovenian abstracts. 2.3 Supplementary file submission Supplementary files (figures) can be submitted to the OJS packed in one zip file not exceeding 50 MB. 2.4 Submission date Thejournalpublishesthesubmissiondateofpapers.Pleasecontacttheeditorwithanyquestions. 3 Citations Examples for citing publications are given below. Using “gray literature” is highly discouraged. 3.1 Citing papers • deKerk,G.V.,Manuel,A.R.2008:Acomprehensiveindexforasustainablesociety:TheSSI–theSustainable Society Index. Ecological Economics 66-2,3. DOI: https://doi.org/10.1016/j.ecolecon.2008.01.029 • Fridl, J., Urbanc, M., Pipan, P. 2009: The importance of teachers’ perception of space in education. Acta geographica Slovenica 49-2. DOI: https://doi.org/10.3986/AGS49205 • Gams, I. 1994a: Types of contact karst. Geografia fisica e dinamica quaternaria 17. • Gams, I. 1994b: Changes of the Triglav glacier in the 1955-94 period in the light of climatic indicators. Geografski zbornik 34. • Perko, D. 1998: The regionalization of Slovenia. Geografski zbornik 38. • van Hall, R. L., Cammeraat, L. H., Keesstra, S. D., Zorn, M. 2016: Impact of secondary vegetation suc­cessiononsoilqualityinahumidMediterraneanlandscape.Catena,Inpress.DOI:https://doi.org/10.1016/ j.catena.2016.05.021 (25.11.2016). 3.2 Citing books • Cohen, J. 1988: Statistical power analysis for the behavioral sciences. New York. • Fridl, J., Kladnik, D., Perko, D., Orožen Adamič, M. (eds.) 1998: Geografski atlas Slovenije. Ljubljana. • Luc,M.,Somorowska,U.,Szmańda,J.B.(eds.)2015:Landscapeanalysisandplanning.Heidelberg.DOI: https://doi.org/10.1007/978-3-319-13527-4 • Nared, J., Razpotnik Visković, N. (eds.) 2014: Managing cultural heritage sites in Southeastern Europe. Ljubljana. 3.3 Citing parts of books or proceedings • Gams,I.1987:AcontributiontotheknowledgeofthepatternofwallsintheMediterraneankarst:acase study on the N. island Hvar, Yugoslavia. Karst and man, Proceedings of the International Symposium on Human Influence in Karst. Ljubljana. • Hrvatin,M.,Perko,D.,Komac,B.,Zorn,M. 2006:Slovenia. SoilErosioninEurope. Chichester. DOI: https://doi.org/10.1002/0470859202.ch25 • Komac, B., Zorn, M. 2010: Statistično modeliranje plazovitosti v državnem merilu. Od razumevanja do upravljanja, Naravne nesreče 1. Ljubljana. • Zorn, M., Komac, B. 2013: Land degradation. Encyclopedia of Natural Hazards. Dordrecht. DOI: https://doi.org/10.1007/978-1-4020-4399-4_207 3.4 Citing expert reports, theses, and dissertations • BregValjavec,M.2012:Geoinformaticmethodsforthedetectionofformerwastedisposalsitesinkarstic and nonkarstic regions (case study of dolines and gravel pits). Ph.D. thesis, University of Nova Gorica. Nova Gorica. • Holmes, R. L., Adams, R. K., Fritts, H. C. 1986: Tree-ring chronologies of North America: California, Eastern Oregon and Northern Great Basin with procedures used in the chronology development work includingusermanualforcomputerprogramCOFECHAandARSTAN.ChronologySeries6.University of Arizona, Laboratory of tree-ring research. Tucson. • Hrvatin, M. 2016: Morfometrične značilnosti površja na različnih kamninah v Sloveniji. Ph.D. thesis, Univerza na Primorskem. Koper. • Šifrer, M. 1997: Površje v Sloveniji. Elaborat, Geografski inštitut Antona Melika ZRC SAZU. Ljubljana. 3.5 Citing online material with authors and titles • Bender,O.,Borsdorf,A.,Heinrich,K.2010:TheinteractivealpineinformationsystemGALPIS.Challenges formountainregions,Tacklingcomplexity.Internet:http://www.mountainresearch.at/images/Publikationen/ Sonderband/bender-borsdorf-heinrich.pdf (4.8.2014). 3.6 Citing online material without authors • Internet: http://giam.zrc-sazu.si (18.11.2016). • Internet 1: http://giam.zrc-sazu.si/ (22.7.2012). • Internet 2: http://ags.zrc-sazu.si (23.7.2012). 3.7 Citing sources without authors • Popisprebivalstva,gospodinjstev,stanovanjinkmečkihgospodarstevvRepublikiSloveniji,1991–končni podatki. Zavod Republike Slovenije za statistiko. Ljubljana, 1993. • WCED–Worldcommissiononenvironmentalanddevelopment:Ourcommonfuture–Brundtlandreport. Oxford, 1987. 3.8 Citing cartographic sources • Buser,S. 1986:OsnovnageološkakartaSFRJ1:100.000,listTolmininVidem(Udine). Saveznigeološki zavod. Beograd. • Digitalni model višin 12,5. Geodetska uprava Republike Slovenije. Ljubljana, 2005. • DržavnatopografskakartaRepublikeSlovenije1:25.000,listBrežice.GeodetskaupravaRepublikeSlovenije. Ljubljana, 1998. • FranciscejskikatasterzaKranjsko,k.o.Sv.Agata,listA02.ArhivRepublikeSlovenije.Ljubljana,1823–1869. • The vegetationmap of forest communities of Slovenia 1:400,000. Biološki inštitut Jovana Hadžija ZRC SAZU. Ljubljana, 2002. 3.9 Citing official gazettes • 1999/847/EC: Council Decision of 9 December 1999 establishing a Community action programme in the field of civil protection. Official Journal 327, 21.12.1999. • Zakon o kmetijskih zemljiščih. Uradni list Republike Slovenije 59/1996. Ljubljana. • Zakonovarstvuprednaravnimiindrugiminesrečami.UradnilistRepublikeSlovenije64/1994,33/2000, 87/2001, 41/2004, 28/2006 in 51/2006. Ljubljana. 3.10 In-text citations Please ensure that every reference cited in the text is also in the reference list (and vice versa). In-text cita­tionsshouldstatethelastnameoftheauthor(s)andtheyear,separateindividualcitationswithsemicolons, order the quotes accordingto year, and separate the page information from the name of the author(s) and yearinformationwithacomma;forexample:(Melik1955),(Melik,IlešičandVrišer1963;Kokole1974,7–8; Gams 1982a; Gams 1982b). For sources with more than three authors, list only the first followed by et al.: (Melik et al. 1956). Cite page numbers only for direct citations: Perko (2016, 25) states: »Hotspots are…« To cite online material with authors, cite the name: (Zorn 2010). To cite online material without authors, cite only Internet fol­lowed by a number: (Internet 2). 3.11 Works cited list Arrange references alphabetically and then chronologically if necessary. Identify more than one reference by the same author(s) in the same year with the letters a, b, c, etc., after the year of publication: (1999a, 1999b). Use this format for indirect citations: (Gunn 2002, cited in Matei et al. 2014). Include the Digital Object Identifier (DOI) in the reference if available. Format the DOI as follows: https://doi.org/… (for example: https://doi.org/10.3986/AGS.1812). 4 Tables and figures Number all tables in the paper uniformly with their own titles. The number and the text are separated by a colon, and the caption ends with a period. Example: Table 1: Number of inhabitants of Ljubljana. Table 2: Changes in average air temperature in Ljubljana (Velkavrh 2009). Tablesshouldcontainnoformattingandshouldnotbetoolarge;itisrecommendedthattablesnotexceed one page. Upload figures to the OJS as separate supplementary files in digital form. If the graphic supplements prepared cannot be uploaded using these programs, consult the editorial board in advance. Numberallfigures(maps,graphs,photographs)inthepaperuniformlywiththeirowntitles.Example: Figure 1: Location of measurement points along the glacier. All graphic materials must be adapted to the journal’s format. Illustrations should be exactly 134mm wide (one page) or 64mm wide (half page, one column), and the height limit is 200mm. To make anonymous peer review possible, include the name of the author(s) with the title of the illus­tration in the supplementary file metadata, but not in the paper text.MapsshouldbemadeindigitalvectorformwithCorelDraw,AdobeIllustrator,orasimilarprogram,espe­cially if they contain text. They can exceptionally be produced in digital raster form with at least 300 dpi resolution, preferably in TIFF or JPG format. For maps made with CorelDraw or Adobe Illustrator, two separate files should be prepared; the original file (.cdr or .ai format) and an image file (.jpg format). For maps made with ArcGIS with raster layers used next to vector layers (e.g., .tif of relief, airborne or satelliteimage),threefilesshouldbesubmitted:thefirstwithavectorimagewithouttransparencytogeth­er with a legend and colophon (export in .ai format), the second with a raster background (export in .tif format), and the third with all of the content (vector and raster elements) together showing the final ver­sion of the map (export in .jpg format). Do not print titles on maps; they should appear in a caption. Save colors in CMYK, not in RGB or other formats. Use Times New Roman for the legend (size 8) and colophon (size 6). List the author(s), scale, source, and copyright in the colophon. Write the colophon in English (and Slovenian, if applicable). Example: Scale: 1:1,000,000 Content by: Drago Perko Map by: Jerneja Fridl Source: Statistical Office of the Republic of Slovenia, 2002 © 2005, ZRC SAZU Anton Melik Geographical Institute Graphs should be made in digital form using Excel on separate sheets and accompanied by data. Photos must be in raster format with a resolution of 240 dots per cm or 600 dpi, preferably in .tif or .jpg formats; that is, about 3,200 dots per page width of the journal. Figures containing a screenshot should be prepared at the highest possible screen resolution (Control Panel\AllControlPanelItems\Display\ScreenResolution).ThefigureismadeusingPrintScreen,andthe capturedscreenispastedtotheselectedgraphicprogram(e.g.,Paint)andsavedas.tif.Thesizeoftheimage or its resolution must not be changed.Examples of appropriate graphic data forms: see the templates of maps in cdr and mxd files for a whole­pagemapinlandscapeviewandanexampleofcorrectfilestructureforsubmittingamapmadewithESRI ArcGIS. SUBMISSION PREPARATION CHECKLIST Aspartofthesubmissionprocess,checkyoursubmission’scompliancewiththefollowingitems.Submissions may be returned to author(s) that do not follow these guidelines. 1. The journal policies have been reviewed. 2. The submission has not been previously published and is not being considered for publication else­where (or an explanation has been provided in comments to the editor). 3. Themetadata(title,abstract,keywords,fulladdress,etc.)areprovidedinEnglishandSlovenian,when applicable. 4. The submission isin Microsoft Word format and the document template was used (single-spaced text, 12-point font, no formatting except italics and bold). 5. The manuscript has been checked for spelling and grammar. 6. Allfigurelocationsinthetextaremarked.Figuresarenotinthetextandareprovidedassupplementary files: cdr, .ai for maps and illustrations; .tif for photographs; xlsx for graphs. 7. Tables are placed in the text at the appropriate place. 8. The reference list was prepared following the guidelines. 9. All references in the reference list are cited in the text, and vice versa. 10. Where available, URLs and DOI numbers for references are provided. 11. Supplementary files are in one .zip file. 12. I agree for this article to be translated or copyedited at my expense AFTER the article is accepted for publication (see guidelines for details). 13. Permissionhasbeenobtainedfortheuseofcopyrightedmaterialfromothersources,includingonline sources; see the copyright notice below. 14. The instructions for ensuring a double-blind review have been followed. ACTA GEOGRAPHICA SLOVENICA EDITORIAL REVIEW FORM Acta geographica Slovenica editorial review form 1 This is an original scientific paper. (The paper is an original and the first presentation of research results with the focus on methods, the­oretical aspects or a case study.)* Yes No 2 The paper follows the standard IMRAD/ILRAD scheme.* Yes No 3 The paper's content is suitable for reviewing in the AGS journal. (The paper is from the field of geography or related fields of interest, the presented topic is interesting for the readers of Acta geographica Slovenica and well presented. In case of negative answer add com­ments below.)* Yes No 4 Editorial notes regarding the paper's content. 5 Thereferencelistissuitable(theauthorcitespreviouslypublishedpaperswithsimilartopicsfromother relevant geographic scientific journals).* Yes, the author cited previously published papers on a similar topic. No, the author did not cite previously published papers on a similar topic. 6 Notes to editor-in-chief regarding previously published scientific work with the focus on AGS.* 7 Is the language of the paper appropriate and understandable?* 8 Recommendation of the editor* The paper is accepted and can be sent to the review process. Reconsider after a major revision (see notes). The paper is rejected. ACTA GEOGRAPHICA SLOVENICA REVIEW FORM 1 RELEVANCE 1a) Are the findings original and the paper is therefore a significant one? Yes No Partly 1b) Is the paper suitable for the subject focus of the AGS journal? Yes No 2 SIGNIFICANCE 2a Does the paper discuss an important problem in geography or related fields? Yes No Partly 2b Does it bring relevant results for contemporary geography? Yes No Partly 2c What is the level of the novelty of research presented in the paper? High Middle Low 3 ORIGINALITY 3a Has the paper been already published or is too similar to work already published? Yes No 3b Does the paper discuss a new issue? Yes No 3c Are the methods presented sound and adequate? Yes No Partly 3d Do the presented data support the conclusions? Yes No Partly 4 CLARITY 4a Is the paper clear, logical and understandable? Yes No 4b If necessary, add comments and recommendations to improve the clarity of the title, abstract, keywords, introduction, methods or conclusion: 5 QUALITY 5a Isthepapertechnicallysound?(Ifno,theauthorshoulddiscusstechnicaleditor[rok.ciglic@zrc-sazu.si] for assistance.) Yes No 5b Does the paper take into account relevant current and past research on the topic? Yes No Propose amendments, if no is selected: 5d Is the references list the end of the paper adequate? Yes No Propose amendments, if no is selected: 5e Is the quoting in the text appropriate? Yes No Partly Propose amendments, if no is selected: 5f Which tables are not necessary? 5g Which figures are not necessary? 6 COMMENTS OF THE REVIEWER Comments of the reviewer on the contents of the paper: Comments of the reviewer on the methods used in the paper: 7 RECOMMENDATION OF THE REVIEWER TO THE EDITOR-IN-CHIEF My recommendation is: Please rate the paper from 1 [low] to 100 [high]: Personal notes of the reviewer to editor-in-chief. COPYRIGHT NOTICE TheActageographicaSlovenicaeditorialboardandthepublisher,theZRCSAZUAntonMelikGeographical Institute, are committed to ensuring ethics in publication and the quality of published books and jour­nalsbyfollowingtheActaGeographicaSlovenicaPublicationEthicsandPublicationMalpracticeStatement. Authorsmustrespectthecopyrightrulesofdataowners;forexample,therulesoftheSlovenianSurveying and Mapping Authority are available at its webpage. ForpapersenttoActageographicaSlovenica,authorsagreethatallmoralrightsoftheauthorsremainwith the authors; material rights to reproduction and distribution in Slovenia and other countries are exclusively cededtothepublisherfornofee,foralltime,forallcases,forunlimitededitions,andforallmedia;andmate­rialrightstothepaperfigures(maps,photos,graphs,etc.)arecededtothepublisheronanon-exclusivebasis. Authors allow publication of the paper or its components on the internet. Authors give permission to the publisher to modify the paper to conform to its guidelines, including the length of the paper. Authors shall provide a professional translation of papers not originally in English. The name of the translator must be reported to the editor. No honoraria are paid for papers in Acta geographica Slovenica or for the reviews. The first author of the paper shall receive one free copy of the publication. PRIVACY STATEMENT Thenamesande-mailaddressesprovidedtothisjournalsitewillbeusedexclusivelyforthestatedpurposes of this journal and will not be made available for any other purpose or to any other party. PUBLISHER Anton Melik Geographical Institute Research Center of the Slovenian Academy of Sciences and Arts PO Box 306 SI–1001 Ljubljana Slovenia SOURCES OF SUPPORT Slovenian Academy of Sciences and Arts Slovenian Research Agency JOURNAL HISTORY Acta geographica Slovenica (print version:ISSN:1581-6613, digital version: ISSN: 1581-8314) was founded in 1952. It was originally named Geografski zbornik / Acta geographica (ISSN 0373-4498). Altogether 42 volumes were published. In 2002 Geographica Slovenica (ISSN 0351-1731, founded in 1971, 35 volumes) was merged with the journal. Since2003(fromvolume43onward)thenameofthejointjournalhasbeenActageographicaSlovenica. The journal continues the numbering system of the journal Geografski zbornik / Acta geographica. Those interested in the history of the journal are invited to read the paper »The History of Acta geo­graphica Slovenica in volume 50-1.« All published issues of Acta geographica Slovenica are available free of charge at http://ags.zrc-sazu.si or http://ojs.zrc-sazu.si/ags. ISSN: 1581-6613 UDC – UDK: 91 ACTA GEOGRAPHICA SLOVENICA GEOGRAFSKI ZBORNIK 60-1 2020 © 2020, ZRC SAZU, Geografski inštitut Antona Melika Print/tisk: Present, d.o.o. Ljubljana 2020 ACTA GEOGRAPHICA SLOVENICA GEOGRAFSKI ZBORNIK 60-1 • 2020 Contents Mojca POKLAR Comparison of the sonar recording method and the aerial photography methodfor mapping seagrass meadows 7 Vanja PAVLUKOVIĆ, Uglješa STANKOV, Daniela ARSENOVIĆ Social impacts of music festivals: A comparative study of Sziget (Hungary) and Exit (Serbia) 21 Péter János KISS, Csaba TÖLGYESI, Imola BÓNI, László ERDŐS, András VOJTKÓ,István Elek MAÁK, Zoltán BÁTORI The effects of intensive logging on the capacity of karst dolines to provide potential microrefugia for cool-adapted plants 37 Radu SĂGEATĂ Commercial services and urban space reconversion in Romania (1990–2017) 49 Kristina IVANČIČ, Jernej JEŽ, Blaž MILANIČ, Špela KUMELJ, Andrej ŠMUC Application of a mass movement susceptibility model in the heterogeneous Miocene clastic successions of the Slovenj Gradec Basin, northeast Slovenia 1 Andrej GOSAR Measurements of tectonic micro-displacements within the Idrija fault zone in the Učjavalley (W Slovenia) 79 Piotr RAŹNIAK, Sławomir DOROCKI, Anna WINIARCZYK-RAŹNIAK Economic resilienceofthe command andcontrolfunctionof citiesin Centraland EasternEurope 95 Mateja FERK, Rok CIGLIČ, Blaž KOMAC, Dénes LÓCZY Management of small retention ponds and their impact on flood hazard prevention in the Slovenske Gorice Hills 107 Gregor KOVAČIČ Sediment production in flysch badlands: A case study from Slovenian Istria 127 Vesna LUKIĆ, Aleksandar TOMAŠEVIĆ Immigrant integration regimes in Europe: Incorporating the Western Balkan countries 143 Mitja DURNIK Community development: LocalImmigrationPartnershipsin Canadaand implications forSlovenia 155 ISSN 1581-6613 9 771581 661010