LANDSCAPE MACROTYPOLOGIES AND MICROTYPOLOGIES OF SLOVENIA Drago Perko, Rok Ciglic, Mauro Hrvatin There are four groups of landscape types in Slovenia: Alpine landscapes (top left), Pannonian landscapes (top right), Dinaric landscapes (bottom right), and Mediterranean landscapes (bottom left). These have a decisive influence on the microtypification of Slovenia and its parts. V Sloveniji so štiri skupine pokrajinskih tipov: alpske pokrajine (levo zgoraj), panonske pokrajine (desno zgoraj), dinarske pokrajine (desno spodaj) in sredozemske pokrajine (levo spodaj), ki odlocilno vplivajo na mikrotipizacijo Slovenije in njenih delov. DOI: https://doi.org/10.3986/AGS.10384 UDC: 911.5(497.4) COBISS: 1.02 Drago Perko1, Rok Ciglic1, Mauro Hrvatin1 Landscape macrotypologies and microtypologies of Slovenia ABSTRACT: This work discusses the historical development of landscape typologies of Slovenia, focus-ingonmethodology,terminology,criteriaforthedivisionofterritory,andlandscapetypehierarchy.Itpresents allfivemacrotypologiesofSloveniacreatedbetween1946and2013,Slovenia’sclassificationinnineselect-ed macrotypologies of Europe produced between 1995 and 2016, and eight examples of microtypologies of smaller areas of Slovenia made between 1985 and 2020. It compares and evaluates similar typologies. If,inadditiontothelandscapetypology,ageographicalregionalizationwasalsoproduced,commonpoints are sought between the two. The macrotypologies and microtypologies of Slovenia are accompanied by an original and updated map. KEY WORDS: regional geography, landscape, landscape type, landscape typification, landscape typolo­gy, landscape diversity, Slovenia, Europe Pokrajinske makrotipologije in mikrotipologije Slovenije POVZETEK: Obravnavamo zgodovinski razvoj pokrajinskih tipologij Slovenije predvsem glede na metodologijo,terminologijo,kriterijedelitevozemljainhierarhijopokrajinskihtipov.Predstavljamovseh petobstojecihmakrotipologijSlovenije,izdelanihmedletoma1946in2013,uvrstitevSlovenijev9izbranih makrotipologijEvrope,izdelanihmedletoma1995in2016,in8primerovmikrotipologijmanjšihobmocij Slovenije, izdelanih med letoma 1985 in 2020. Istovrstne tipologije med sabo primerjamo in vrednotimo. Cejeavtorhkratispokrajinskotipologijoizdelaltudigeografskoregionalizacijo,išcemonjuneskupnetocke. Makrotipologije in mikrotipologije Slovenije so opremljene z izvirnim in posodobljenim zemljevidom. KLJUCNE BESEDE: regionalna geografija, pokrajina, pokrajinski tip, pokrajinska tipizacija, pokrajinska tipologija, pokrajinska raznolikost, Slovenija, Evropa The paper was submitted for publication on August 23rd, 2021. Uredništvo je prejelo prispevek 23. avgusta 2021. 1 Research Centre of the Slovenian Academy of Sciences and Arts, Anton Melik Geographical Institute, Ljubljana, Slovenia drago.perko@zrc-sazu.si (ORCID: https://orcid.org/0000-0002-2568-9268), rok.ciglic@zrc-sazu.si (ORCID: https://orcid.org/0000-0003-3517-3780), mauro.hrvatin@zrc-sazu.si (ORCID: https://orcid.org/0000-0002-6021-8736) 1 Introduction The Anton Melik Geographical Institute at the Slovenian Academy of Sciences and Arts Research Center (hereinafter:theinstitute)hasbeendealingwithSlovenianlandscapes,theirtypes,typifications,andtypolo-gieseversinceitsfoundationin1946,largelyaspartofitsRegionalGeographyDepartmentandwithstrong support from the Geographic Information Systems and Thematic Cartography departments. It was espe­cially intensively involved in this area after Slovenia’s independence in 1991, when it produced seminal geographical volumes on Slovenia with government financial support in the 1990s. Landscapes and their types feature prominently in these works. Thisarticle,whichfocusesonasystematicoverviewoflandscapetypologiesofSlovenia,coincideswith theseventy-fifthanniversaryoftheinstitute,thethirtiethanniversaryofSlovenia’sindependence,and,what ismost relevant to this article, the production of the first landscape typology of Slovenia. It was published seventy-five years ago by the academy member Anton Melik (Melik 1946), who also initiated the insti­tute’s foundation and was its first director. The introductory section primarily describes the development of landscape studies in Slovenia and especially at the institute, where focused research in this area has been conducted in Slovenia ever since its establishment. The second section presents and compares all five geographical macrotypologies for all ofSloveniaproducedbySloveniangeographersbetween1946and2013.Thethirdsectionestablisheswhich landscape types of the nine European macrotypologies Slovenia belongs to and how these macrotypolo­gies overlap with Slovenian ones in Slovenian territory. The fourth section presents and compares eight examples of landscape microtypologies (i.e., typologies of small parts of Slovenia), which the institute’s researchers produced between 1985 and 2020. The conclusion provides an example showing how current landscape macrotypologies could be improved using state-of-the-art methods. Certain older findings were updated and expanded, and some are published here for the first time. The reference list at the end is also very important because it provides an extensive overview of publica­tions on geographical landscape typologies in Slovenia and abroad. In countries as diverse as Slovenia, landscape typifications present a great challenge for researchers. Incountrieswheretheyformpartofofficialdocumentsthatspecifyfinancialobligations,theyareimpor­ tant for most citizens. This also includes Slovenia. 1.1 Terminology Themaintechnicaltermsinlandscapeclassificationincludelandscapetypification,landscapetypology,and landscape type (from Lat. typus ‘bas-relief, image, figure’, Gr. týpos ‘example, model’; Snoj 2016). They are combinedhereintoageneraldefinition:landscapetypificationistheprocessofclassifyinglandscapesaccord­ing to their characteristics into landscape types, and its result or final classification is a landscape typology. Closely related to these are the terms region and regionalization (from Lat. regio ‘region, place, bor­ der’;Snoj2016).Theycanbecombinedintothefollowingdefinition:regionalizationistheprocessofdividing territory into regions and their hierarchical classification. Increasingly important is also the term landscape diversity (Perko, Hrvatin, and Ciglic 2017), which revealshowconcentratedtheinterchangeisbetweenlandscapesorlandscapeelementswithinagivenspa­tial unit. It can also be expressed numerically using a landscape diversity index, which can be defined as theaverage(arithmeticmean)oftheratiosbetweenthenumberoftypesfound(classes)andalltypes(class­es) of landscape elements considered within a given spatial unit. Its formula is: Sn ni 1 ni s LDI = 1. = ( +…+ ) Si=1 Ni SNi Ns LDI = Landscape Diversity Index S = number of landscape elements considered N = number of all types (classes) of a given landscape element n = number of types found (classes) of the same given landscape element InSlovenian,thetermspokrajina‘region;landscape’andregija‘region’areusuallysynonymous,although sometimes the term pokrajina is used to refer to more natural or physical geographical spatial units, and the term regija is associated with more social or political geographical spatial units (Perko 1998a; Kladnik and Perko 1998). Therelationshipbetweenalandscapeorregionandalandscapetypeisasfollows:everyregionbelongs to a specific landscape type, and several regions can be part of the same landscape type. Landscape typi­ficationandlandscapetypesarecharacterizedbytheprincipleofsimilarity,andgeographicalregionalization and regions are characterized by the principle of individuality. Every landscape type can occur several times and separately, and so it is treated as a common noun phrase and hence not capitalized. In turn, every region occurs only once and is hence capitalized (Perko 1998b; Perko, Hrvatin, and Ciglic 2015). For example, a landscape type can be a mountain range, of which there are several in Slovenia and many more elsewhere around the globe, but there is only one mountain region in the world called Julian Alps (in Slovenia) and only one called Andes (in South America). Whendividingandcombiningregionsandlandscapetypesaspartofgeographicalregionalizationand landscapetypification,theyarehierarchicallyclassifiedintoseverallevels.Everyregioncanbepartofalarg­erregionanddividedintosmallerregions.Similarappliestolandscapetypesandothergeographicalunits. Hence, a series of terms have developed: macrounits (e.g., macroregion) and submacrounits at the highest levels,mesounits(e.g.,landscapemesotype)andsubmesounitsatthemiddlelevels,andmicrounitsandsub­microunits (e.g., submicroregion) at the lowest levels. These hierarchical terms are used by most typology authors presented in the second, third, and fourth sections below. 1.2 Landscape typologies around the world Elsewhere around the world, landscape typologies and other landscape classifications are also related to several levels: global, regional, national, and local. Described in the articles are mainly the classification methods and their usage, and problems associated with data (e.g., Udvardy 1975; Belbin and McDonald 1993; Meeus 1995; Bailey 1996; Burrough et al. 2001; Olson et al. 2001; Bohn et al. 2003; Mücher et al. 2003, 2006, 2010; Leathwick, Overton, and McLeod 2003; Zhou et al. 2003; Wolock et al. 2004; Žiberna, Natek,andOgrin2004;Rivas-Marínez,Penas,andDíaz2004a;2004b;HargroveandHoffman2005;Metzger etal.2005;Wascher2005;Bryan2006;Jongmanetal.2006;Owenetal.2006;Renetzederetal.2008;Romportl 2009;VanEetveldeandAntrop2009;Castillo-Rodríguezetal.2010;SotoandPintó2010;Izakovicová2014; Grondinetal.2014;RomportlandCerny2014;Cullumetal.2016;Mezosi2016;Kozjek,Dolinar,andSkok 2017; see also Drozg et al. 2004; Simensen, Halvorsen, and Erikstad 2018; and Stupariu et al. 2021). Landscapeclassificationsaroundtheglobeareusedforresearch,educational,economic,planning,and other purposes. Europe has adopted several measures for raising residents’ awareness about the impor­tance of landscapes (Wascher 2005). As early as 1996, the Council of Europe adopted the Pan-European BiologicalandLandscapeDiversityStrategy(CouncilofEurope1996),followedbytheEuropeanLandscape Convention in 2000, which encourages member states to identify their landscapes (Van Eetvelde 2009; Dempsey and Wilbrand 2017). Based on this, Slovenia updated its 1998 classification of landscape types (Marušic,Ogrin, and Jancic 1998),entitlingit Regionalna razdelitevkrajinskih tipov (Regional Distribution of Landscape Types; Bratina Jurkovic 2008; Council of Europe 2007). Several countries also use landscape classifications for official purposes at the national level; in addi­tiontoSlovenia(Perko,Hrvatin,andCiglic2015),thisisalsodoneintheUK(WarnockandGriffiths2015) and Norway (Strand 2011), which also produced a map of landscape types at the local level for Nordland County(Erikstad,Uttakleiv,andHalvorsen2015).TheuseofthemapofEurope’sbiogeographicalregions for the purposes of the Natura 2000 network is an example of official use of a landscape classification at the European level (Biogeographical regions…2016). With regard to landscape classifications that are used for official purposes, it is especially important that they be as accurate and objective as possible and that in producing them subjective definition of cri­teriaanddigitizationandinterpolationerrorsbeavoidedasmuchaspossible(McMahonetal.2004;Ellison 2010). However, many researchers are aware that subjectivity is inevitable (Loveland and Merchant 2004;Leathwick,Overton,andMcLeod2003;NatekandŽiberna2004;Owenetal.2006)becauselandscapeclas­sificationisanabstraction(Bernertetal.1997)thatnevercapturesalllandscapefactors(Zonneveld1994). Sometimes the boundaries in a landscape are clear (Bailey 1996), but often they are defined arbitrarily (Leathwick,Overton,andMcLeod2003)andsubjectivelyregardlessofthequalityoftheinputdata(Fnukalová and Romportl 2014). A common weakness of existing landscape classifications is also a poor description of how classification units are defined and classified (Loveland and Merchant 2004), even though a good descriptionisimperative(Mücheretal.2003;Bailey2004;Erikstadetal.2015;Lauschetal.2015).Another common weakness observed in recent classifications is the omission of field mapping, resulting from the increasinglyhigherqualityofdigitaldataandmodernsoftware(WieczorekandMigon2014).Classifications based on landscape stereotypes are considered especially poor (McMahon et al. 2004). How landscape classifications should be tackled depends on concrete cases (Lu and Weng 2007), and some see the future primarily in enhanced evaluation of current landscape classifications with new quan­titativemethodsthatwilleitherfullyconfirmalandscapetypologyproducedoreliminateanyweaknesses and improve it (Congalton 1991; Ciglic and Perko 2015). We see the future of landscape typologies moving in two directions: toward evaluating, testing, and improvingexistinglandscapetypologiesusingmoderntools,andtowardproducingcompletelynewland-scape typologies, while taking into account criteria similar to those for improving the existing typologies. An example of improving an existing landscape typology using modern tools and for a specific purpose is presented in the conclusion. 1.3 Studying Slovenian landscapes at the institute ProfessionalsoutsideSloveniahavenotdealtwithlandscapetypificationsofSlovenia.Theyhaveonlystud­ied Slovenia at the level of Europe (Ciglic 2009) – that is, in a very generalized manner, which is discussed inthethirdsection.MoststudiesofSlovenianlandscapesandthelandscapetypologiesproducedhavebeen concentrated at the geographical institute. The institute’s primary organizational unit specializing in Slovenian landscapes is the Regional Geography Department. It was formally founded on October 14th, 1994, which was relatively late con-sideringthatitwasalreadymentionedintheinstitute’sfirstcharter,whichitsfounder,thegeneralassembly of the Slovenian Academy of Sciences and Arts, adopted on November 6th, 1948. According to the char­ter, the institute was to be divided into departments. Its Articles 5 and 6 specify that it can be divided into physical,human,andregionalgeographysections,aswellasintofurthersectionsandsubsectionsifneed­ed.Inaddition,thecharterdefinesresearchonSlovenianlandscapesasoneoftheinstitute’smainpriorities (Natek and Perko 1999). The first such section or unit, the Institute of Cartography, was established in 1952 and then renamed the Thematic Cartography Department on October 14th, 1994, when the institute was last reorganized. TheGeographicInformationSystemsDepartmentwasalsoestablishedaspartofthisreorganization.Without both of these departments, it is practically impossible to imagine contemporary research on landscapes andthepublicationoffindingsinthisarea.Theinstitutecurrentlyhassevendepartments,aphysicalgeog­raphy laboratory, a library, and a museum, which adds up to a total of ten units: • The Physical Geography Department (since October 14th, 1994); • The Human Geography Department (since October 14th, 1994); • The Regional Geography Department (since October 14th, 1994); • The Natural Hazards Department (since October 14th, 1994); • The Environmental Protection Department (since October 14th, 1994); • The Geographic Information Systems Department (since October 14th, 1994); • The Thematic Cartography Department (since February 7th, 1952); • The Physical Geography Laboratory (since April 18th, 2018); • The Geographical Museum (since May 7th, 1946); and • The Geographical Library (since September 1st, 1964). The institute was named after Anton Melik in 1976, ten years after his death. Since 1981, it has been operating as part of the Slovenian Academy of Sciences and Arts’ Research Center. Its Regional Geography Department studies Slovenia as a whole and as part of Europe or the world and,mostimportantly,itexploresSlovenianregions.TheprojecthasalsocarriedoutthelargestSlovenian geographical project to date: it produced a volume on Slovenia’s regional geography, which practically all geographers from Slovenian universities and research institutes contributed to. Work on the volume was completed as part of the first geographical project in Slovenia since 1993, when research in the country began to largely be funded through projects and programs approved by the Slovenian Research Agency. The funding for the project was approved in 1993 and it lasted three years, which even now continues to be the average duration of agency-funded projects. More general and long-term programs normally last six years. Tworesearch programs that alsofocusedon landscapetypifications havebeen carriedout at theinsti­tute to date: • The Regional Geography of Slovenia, which lasted from January 1st, 1999, to December 31st, 2003, and was headed by Drago Perko, focused entirely on studying Slovenian regions and landscapes; • The Geography of Slovenia, which has been taking place since January 1st, 2004, and is headed by Blaž Komac, dedicates two of its five thematic sections to Slovenian landscapes: one in full and one in part. Twelve research projects dealing with landscapes have been carried out so far: • CompletionoftheRegionalGeographicalVolumeonSlovenia,whichwascarriedoutfromJanuary1st, 1993,toDecember31st,1995,andheadedbyDragoPerko,concludedwiththepublicationofnineschol­arly volumes and the book Slovenija: pokrajine in ljudje (Slovenia: Landscapes and People; Perko and Orožen Adamic 1998) for general audiences (Figure 1); • The National Atlas of Slovenia, which was carried out from January 1st, 1993, to December 31st, 1995, andheadedbyAndrejCerne,concludedwiththepublicationofGeografskiAtlasSlovenije(Geographical Atlas of Slovenia; Figure 2) and Nacionalni atlas Slovenije (National Atlas of Slovenia), in both of which Slovenian landscapes play an important role; • Cultural Landscapes of Slovenia, which lasted from November 1st, 1994, to December 31st, 1999, and washeadedbyDragoPerko,focusedonculturallandscapeelements;thetwopublicationswiththegreat-est impact were the multimedia KulturAtlas Slowenien (Culture Atlas of Slovenia) published as part of the German series KulturAtlas Europa (Aimée et al. 1996) and the scholarly volume Kulturne pokrajine v Sloveniji (Cultural Landscapes in Slovenia; Urbanc 2002); • Geographical Microregionalization of Slovenia, which took place from January 1st, 1996, to December 31st, 2000, and was headed by Drago Perko, focused on state-of-the-art methods used in geographical regionalizations and typifications of small spatial units; • Common Lands in Slovenia: Cultural Landscape between the Past and Future was carried out from September1st,2005,toAugust31st,2007;itwasheadedbyDragoPerkoanditexploredcommonlands as an important landscape element of Slovenian regions; • DeterminingNaturalLandscapeTypesofSloveniaUsingaGeographicInformationSystem,carriedout from May 1st, 2010, to April 30th, 2013, and headed by Drago Perko, established the effectiveness of modern geoinformation methods in landscape typifications; • Terraced Landscapes in Slovenia as Cultural Values, carried out from July 1st, 2011, to June 30th, 2014, and headed by Drago Kladnik, dealt with the differences between cultural terraces in various Slovenia’s landscapetypes,anditconcludedwiththebookTerasiranepokrajineinSlovenian(TerracedLandscapes; Perko,Ciglic,andGeršic2016),whichlateralsoappearedinanabridgedEnglishedition(Perkoetal.2017); • TextbooksasToolsforShapingtheGeographicalImaginationofSlovenianLandscapes,carriedoutfrom July1st,2011,toJune30th,2014,undertheleadershipofMimiUrbanc,determinedtheroleofSlovenian landscapes in textbooks and concluded with the publication of the book Oblikovanje predstav o sloven-skihpokrajinahvizobraževalnemprocesu(ShapingtheGeographicalImaginationofSlovenianLandscapes in Education; Urbanc et al. 2016); • Landscape Diversity and Hotspots of Slovenia, carried out from October 1st, 2014, to September 30th, 2017, and headed by Drago Perko, focused on determining landscape diversity as one of the elements shaping Slovenian identity (Perko and Ciglic 2015; Perko, Hrvatin, and Ciglic 2017); • CulturalLandscapesCaughtbetweenPublicGood,PrivateInterests,andPolitics,carriedfromOctober 1st, 2014, to September 30th, 2017, and headed by Mimi Urbanc, elucidated cultural landscapes from aspectsthathadbeenpoorlystudiedinSloveniangeographyuntilthenandgavethemnewdimensions; • Advancement of ComputationallyIntensive Methods for Efficient Modern General-Purpose Statistical Analysis and Inference, carried out from March 1st, 2016, to February 28th, 2019, and headed by ErikŠtrumbelj, explored the use of artificial intelligence in landscape typifications. Asalreadymentionedabove,thefirstresearchprojectapprovedwasactuallytheconcludingactofmore than ten years of efforts to produce a regional geographical volume on Slovenia, which officially began by accepting the volume as part of the institute’s program of work in 1981. The volume’s thematic concept envisagedalandscapetypologicalpresentationofSloveniaintheintroduction,whichwastoformthebasis for the geographical regionalization of Slovenia and its presentation by regions (Ilešic 1981). Until fund­ing for the project was approved in 1993, work on the volume had proceeded very slowly but, once the funding was secured, researchers made a commitment to complete their work in three years (i.e., by the end of 1995), which they also did. They divided Slovenian into nine macrounits and sixty-one mesounits, which they described in nine volumes published in 1996: • The macrounit alpski svet ‘Alpine region’ with four mesounits (Pak and Perko 1996b); • The macrounit zahodni predalpski svet ‘western Prealpine region’ with four mesounits (Pak and Perko 1996c); • The macrounit Ljubljanska kotlina ‘Ljubljana Basin’ with six mesounits (Pak and Perko 1996a); • Themacrounitsubmediteranskisvet‘sub-Mediterraneanregion’withsevenmesounits(KladnikandPerko 1996a); • The macrounit visoki kraški svet ‘high karst region’ with eight mesounits (Kladnik and Perko 1996b); • The macrounit nizki kraški svet ‘low karst region’ with seven mesounits (Perko 1996a); • Themacrounitsrednjiinjužnisubpanonskisvet‘centralandsouthernsub-Pannonianregion’withseven mesounits (Perko 1996b); • Themacrounitvzhodniinseverovzhodnipredalpskisvet‘easternandnortheasternPrealpineregion’with eight mesounits (Drozg and Perko 1996a); and • Themacrounitsevernisubpanonskisvet‘northernsub-Pannonianregion’withtwelvemesounits(Drozg and Perko 1996b). In 1997, the institute also produced a more popular version of the regional geographical volume of Sloveniawithasimplifieddivisionofthecountryintofourmacrounitsandforty-sixmesounits.Itwaspub­lished a year later in a single volume (Perko and Orožen Adamic 1998), together with a popular-science filmaboutSlovenianlandscapesreleasedonavideotape.ThefilmwasspecialbecauseitpresentedSlovenian landscapes for the first time together with their natural sounds (Perko and Križnar 1998). The volume (Figure 1)wastheresultofworkcontributedbythirty-twoauthors,sixcartographers,fortyphotographers, and thirty-four other professionals. It contains 736 pages, 455 photos, 118 maps, and 426 graphs (Natek and Perko 1999). Beforetheprojectslistedabove,whichwerefundedbytheagency,theinstitutealsocarriedoutanoth-er large project associated with Slovenian landscapes: Krajevni leksikon Slovenije (Gazetteer of Slovenia; Figure 3). The institute managed to have most Slovenian geographers employed as researchers at univer­sitiesandinstitutesworkontheproject,alongsidemanygeographyteachers.Theyallperformedfieldwork, wrote texts, and took photos, and most processed the settlements within one municipality. Work began in1993andwascompletedintwoyears.Thebookwaspublishedin1995,presentingall5,981settlements (i.e., villages and towns) of that time with descriptions, photos, and tables (Orožen Adamic, Perko, and Kladnik 1995). It contains 640 pages, 492 color photos, thirty-nine 1:100,000 color thematic maps with settlements, which were an outstanding technological achievement of the institute’s digital cartography at that time, and several other thematic maps (Natek and Perko 1999). Among them is also a geographical regionalization of Slovenia with six macrounits and sixty-two mesounits, which in places are divided into further subunits. An interactive CD, the first of its kind in Slovenia, was released together with the book, andPrirocnikrajevnileksikonSlovenije(PocketGazetteerofSlovenia;OroženAdamic,Perko,andKladnik 1996), about half the size of the original gazetteer in terms of both its format and content, was published a year later. The last major project associated with landscapes was carried out at the institute between 2017 and 2019aspartoftheprogramGeografijaSlovenije(GeographyofSlovenia).ItsfinaloutcomewasTheGeography of Slovenia(Perko, Ciglic, and Zorn 2020a), abook published in 2020 as part of Springer’s World Regional Geography Book Series, which was the first extensive and detailed geographical presentation of Slovenia inEnglish(Figure4).The360-pageworkbeginswithachapteraboutSlovenia’sexceptionallandscapediver­sity, followed by five thematic sections comprised of twenty-two chapters, including one on the landscape typificationofSloveniaandoneonitsgeographicalregionalization.Thetextiscomplementedby222color photos,includingmanythematicmaps.ThebookcontributesgreatlytopromotingSlovenianresearchand enhancingSlovenia’sinternationalprofile,anditssubtitleSmallbutDiversehighlightsthefactthat,despite being a small country, Slovenia is characterized by above-average geographical diversity and hence italso offers numerous research challenges in studying landscapes (Ciglic and Perko 2013a). Figure 1: Cover of Krajevni leksikon Slovenije, which also introduced Figure 2: Cover of Geografski atlasSlovenije, which also contains a chap-a new geographical regionalization of Slovenia (Orožen Adamic, Perko, ter on the geographical regionalization and landscape typology of and Kladnik 1995). Slovenia (Fridl et al. 1998a). Because a quarter of a century has already passed since the publication of the gazetteer, national atlas, and regional geographical volume on Slovenia, and because the country has experienced great changes in practically all areas, updating all these seminal geographical works on Slovenia should be an important priorityinthefuture.ThiswilldemandcooperationbetweenallgeographicalresearchinstitutionsinSlovenia (the geographical institute and the three geography departments at the universities of Ljubljana, Maribor, and Primorska) and substantial funding, which is needed to carry out all the projects, from research to book publication. As already mentioned, all these major geographical works on Slovenia include landscape typologies andgeographicalregionalizations,buttheyarelargelylimitedtoonetypificationandregionalizationonly. Thefollowingsectionprovides,forthefirsttimeinSloveniangeography,asystematicoverviewofallland-scape macrotypologies by Slovenian authors and compares them. Added to this is a section on landscape microtypificationsbecausesomemacrotypologieshadanimportantinfluenceontheproductionofmicroty­pologies. 2 Macrotypologies of Slovenia SloveniangeographersandsimilarprofessionalshaveproducedvariousclassificationsofSlovenia(Kladnik 1996;KladnikandPerko1998;PerkoandCiglic2020a,2020b).Mostdivisionsarebasedononelandscape element or only a few elements. They include several regionalizations, but only five classifications can be defined as landscape typologies. Three were produced by the geographical institute’s directors, and two were also co-created by other institute employees. Nearly seventy years passed between the first one pro­duced in 1946 and the last one produced in 2013, and during this period people’s attitude toward nature andlandscapechangedsignificantly. However,withthedevelopmentofmoderntechnology,geographical methodology changed even more. This also influenced the design and characteristics of these five land-scapetypologiesofSlovenia,whicharepresentedbelow.TheywereproducedfortheterritoryofallofSlovenia at the highest spatial levels and can therefore be defined as macrotypologies. 2.1 Melik’s typology (1946) ThefirstlandscapetypologytocoverallofSloveniawasproducedbyAntonMelik(1890–1966),thegreat­estSloveniangeographer,whoalsoinitiatedtheestablishmentofthegeographicalinstitutein1946(Figure 7).Hereceivedabachelor’sdegreeingeographyandhistoryinViennain1916andadoctorateinLjubljana in 1927. That same year, he began teaching at the University of Ljubljana. In 1940, he became a member of the Slovenian Academy of Sciences and Arts, and in 1946 he was appointed director of the geograph­icalinstitute,whichwasnamedafterhimin1976,tenyearsafterhisdeath.Hismostimportantworksinclude the first general geographical work on Slovenia in two volumes (Melik 1935–1936) and the first regional geographical work on Slovenia in four volumes (Melik 1954, 1957, 1959, and 1960). The first major work that he tackled as the institute’s director was a landscape typology of Slovenia (Melik 1946), which covered Slovenia and ethnic Slovenian areas outside Slovenia. Without using any special methodology and relying primarily on his expertise and fieldwork experience, he divided the territory in terms of geomorphology, rocks, and climate. According to Melik (1946), the purpose of his typologywas not to divideSlovenia only into naturallandscape units, but primarily into such homoge­nousunitsthatofferedsimilar opportunitiesfortheeconomyanditsdevelopmentforpeopleandsociety in general. He published a map with boundaries between the types and described their natural and social char­acteristics in the main text, also listing the most important (micro)regions within each type. HeentitledthemappresentingthistypologyPrirodnogospodarskasestavaSlovenije(NaturalEconomic Structure of Slovenia) and referred to its units as prirodni sestavni deli ‘natural component parts’. Figure 5: Original map of Melik’s landscape typology of Slovenia (Melik 1946) with legend: I High Alpine mountains, II Low Alpine mountains, III Hills, IV Low hills, V Plains and wide valleys, VI Dinaric high plateaus, VII Karst hills with low plateaus and lowlands, and VIII Karst poljes. p p. 16 Figure 6: Updated map of Melik’s landscape typology of Slovenia. p p. 17 Gornja Radgona Radlje ob Dravi Murska Sobota Lenart v Slovenskih Goricah Dravograd Lendava Slovenj Gradec Ravne na Koroškem MARIBOR Ruše Ljutomer Kranjska Gora Crna na Koroškem Jesenice Ptuj Ormož Slovenska Bistrica Slovenske Konjice Bled Bovec Velenje Tržic D Mozirje Acta geographica Slovenica, 61-3, 2021 Radovljica Bohinjska Bistrica Kranj Kamnik Žalec CeljeRogaška Slatina Tolmin Železniki ŠentjurCerkno Škoa Loka Mengeš TrbovljeHrastnik Zagorje ob Savi Landscape types Domžale Laško High Alpine mountains Žiri LJUBLJANA Litija Low Alpine mountains Sevnica Idrija Hills Nova Gorica VrhnikaGrosuplje Krško Low hills Brežice Ajdovšcina Logatec Trebnje Plains and wide valleys Dinaric high plateaus Novo Mesto Karst poljes Karst hills with low Cerknica Postojna plateaus and lowlands Ribnica Sežana Metlika Kocevje Adriatic Sea Crnomelj Ilirska Bistrica Piran Koper Izola 0 10203040 km Map by: Manca Volk BahunSource: RS, Surveying and Mapping Authority© 2021, ZRC SAZU Anton Melik Geographical Institute Figure7: AntonMelik(1890–1966),author ofthefirst landscapetypol-Figure 8: Svetozar Ilešic (1907–1985), author of the second landscape ogy of Slovenia. typology of Slovenia. Hedividedtheterritoryintoeightmacrounits,whosenamesrepresentedtypicallandscapetypes(e.g., hribovje ‘hills’). He then divided these further into three submacrounits and thirteen mesounits, which he partly named as a landscape type (e.g., the mesounit širše doline ‘wide valleys’), a region (e.g., the sub-macrounitBelakrajina‘WhiteCarniola’),orsomethinginbetween(e.g.,themesoregiondolinaobzgornji Krki ‘valley along the upper Krka River’). The basic types in Melik’s division of Slovenia were as follows (Figures 5 and 6): • visokogorski alpski predel ‘high Alpine mountains’; • predel alpskega sredogorja ‘low Alpine mountains’; • hribovje ‘hills’ as the largest type, covering 23.8% of Slovenia; • predel goric ‘low hills’; • ravnine in vecje doline ‘plains and wide valleys’; • dinarske visoke planote ‘Dinaric high plateaus’; • kraško hribovje z nižjimi planotami in podolji ‘karst hills with low plateaus and lowlands’; and • kraška polja ‘karst poljes’ as the smallest type, covering only 1.3% of Slovenia. It is interesting that, compared to Melik’s regionalization (Melik 1954, 1957, 1959, and 1960), which was slightly deficient in certain aspects, his typology (Melik 1946) met with significantly less response in theSlovenianprofessionalcommunity,eventhoughitwasbasedonmorethoroughresearch(Perko1998a). 2.2 Ilešic’s typology (1958) ThesecondlandscapetypologyofallofSloveniawasproducedbySvetozarIlešic(1907–1985),whoreceived abachelor’sdegreeingeographyandhistoryinLjubljanain1930,followedbyadoctoratein1933.In1967, he became a member of the Slovenian Academy of Sciences and Arts, and he served as the second direc­tor of the geographical institute from 1967 to 1981 (Figure 8). VariousclassificationsofSlovenianterritoryformedanimportantpartofhisworkasaresearcher.Even though he advocated a uniform and complex geography, he believed that nature and society differ from one another to the extent that two basic types of territorial geographical division are necessary: a land-scape-physiognomic one, which falls into the realm of physical geography, and an economic-functional one, which is the subject of human geography (Ilešic 1958). Amongthedivisionsheproduced,hislandscapetypologyofSloveniastandsout.Hepublisheditthree times with minor modifications. First, he only published the names and more non-technical descriptions of individual units (Ilešic 1956), which he called pokrajine ‘landscapes’. Two years later (Ilešic 1958), he published the typology again as part of a theoretical discussion on issues connected with the geographi­caldivisionofSlovenia.HeentitleditPokrajinsko-fiziognomicneregijeSlovenije(Landscape-Physiognomic RegionsofSlovenia)inSlovenianandsimplyRégionsphysionomiquesdelaSlovénie(PhysiognomicRegions ofSlovenia)inFrench.Thetypologywasthesameasthepreviousone,butheaddedamapwithunitbound­ariesandimprovedtheunits’hierarchy.Hereplacedthetermpokrajina‘landscape’withregija‘region’(e.g., makroregija ‘macroregion’), but he retained it with concrete names (e.g., the macroregion Alpske pokra­jine ‘Alpine landscapes’). He referred to the typology using two names: a landscape-type division and an ecological division. The units of the former were pokrajinski tipi ‘landscape types’ and those of the latter were ekološka obmocja ‘ecological areas’. On the 1972 map (Ilešic 1972), he then combined both names intothetitlePokrajinsko-ekološkarazclenjenostSlovenije(Landscape-EcologicalDivisionofSlovenia),which he himself translated into English as Landscape Types and Ecological Areas of Slovenia. Just like with Melik, his typology covered Slovenia and its nearby surroundings, and with regard to methodologyheonlypresentedthecriterionofdivision– thatis,auniform,homogenouslandscapephys­iognomy or external landscape image, which is largely affected by natural landscape elements and the anthropogenic elements strongly associated with them, such as land use, settlement type, and agriculture. Ilešic divided the territory into five macroregions and ten mesoregions (two macroregions had three mesoregions,twomacroregionshadtwomesoregions,andonemacroregionhadnomesoregions),which representedtypicallandscapetypes(e.g.,themacroregionpredalpskepokrajine‘sub-Alpinelandscapes’or themesoregionsubmediteransko-subalpskepokrajine‘sub-Mediterranean–sub-Alpinelandscapes’).Hethen dividedthesefurtherintothirty-sevensubmesoregionsandtenmicroregions, nearlyallofwhichwereactu-ally regions and not types (e.g., the submesoregion Julijske Alpe ‘Julian Alps’ or its microregion Zgornja Soška dolina ‘Upper Soca Valley’). Thebasictypes(tenmesoregionswithinthemacroregionsandonemacroregionwithnomesoregions) in Ilešic’s division of Slovenia were as follows (Figures 9 and 10): • zahodne alpske pokrajine ‘western Alpine landscapes’; • vzhodne alpske pokrajine ‘eastern Alpine landscapes’; • zahodne predalpske pokrajine ‘western sub-Alpine landscapes’; • vzhodne predalpske pokrajine ‘eastern sub-Alpine landscapes’; • severovzhodne predalpske pokrajine ‘northeastern sub-Alpine landscapes’; • pravesubpanonskepokrajine‘sub-Pannonianlandscapes’asthelargesttype,covering22.0%ofSlovenia; • subpanonsko-subdinarske pokrajine ‘sub-Pannonian–sub-Dinaric landscapes’; • kraške pokrajine notranje Slovenije ‘karst landscapes of inner Slovenia’; • submediteransko-subalpske pokrajine ‘sub-Mediterranean–sub-Alpine landscapes’ as the smallest type, covering 0.5% of Slovenia; • submediteranske pokrajine ‘sub-Mediterranean landscapes’; and • submediteransko-dinarske pokrajine ‘sub-Mediterranean–Dinaric landscapes’. Ilešic’stypologywasmetwithawiderpublicresponsethanthatofMelik.ThiswaspartlybecauseIlešic promotedhistypologywithseveralpublicationsandpartlybecausehisdivisionofSloveniacombinedtypol­ogy at the higher levels with regionalization at the lower levels. Figure 9: Original map of Ilešic’s landscape typology of Slovenia (Ilešic 1958) with legend: IA Western Alpine landscapes, IB Eastern Alpine landscapes, IIA Western sub-Alpine landscapes, IIB Eastern sub-Alpine landscapes, IIC Northeastern sub-Alpine landscapes, IIIA Sub-Pannonian landscapes, IIIB Sub-Pannonian–sub-Dinariclandscapes,IV Karstlandscapesof innerSlovenia, VASub-Mediterranean–sub-Alpine landscapes,VBSub-Mediterranean landscapes, and VC Sub-Mediterranean–Dinaric landscapes. p p. 20 Figure 10: Updated map of Ilešic’s landscape typology of Slovenia. p p. 21 Murska Sobota Gornja Radgona Radlje ob Dravi Lenart v Slovenskih Goricah Dravograd Lendava Ravne na Koroškem MARIBOR Ruše Ljutomer Kranjska Gora Slovenj Gradec Crna na Koroškem Jesenice Ptuj Ormož Slovenska Bistrica Bled Bovec D Tržic Mozirje Velenje Slovenske Konjice Radovljica Bohinjska Bistrica Žalec CeljeRogaška Slatina Tolmin Železniki Kranj Kamnik ŠentjurCerkno Škoa Loka Mengeš TrbovljeHrastnik Zagorje ob Savi Laško Domžale Landscape types Western Alpine landscapesEastern Alpine landscapes Žiri LJUBLJANA Litija Sevnica Western sub–Alpine landscapes Krško Eastern sub–Alpine landscapes Idrija Nova Gorica VrhnikaGrosuplje Logatec Brežice Northeastern sub–Alpine landscapes Trebnje Ajdovšcina Sub-Pannonian landscapes Novo Mesto Sub-Pannonian–sub–Dinaric landscapes Cerknica Karst landscapes of inner Slovenia Postojna Sub-Mediterranean–sub–Alpine landscapes Ribnica Sežana Mediterranean landscapes Metlika Sub–Mediterranean–Dinaric landscapes Kocevje Adriatic Sea Crnomelj Ilirska Bistrica Piran Koper Izola 0 10203040 km Map by: Manca Volk BahunSource: RS, Surveying and Mapping Authority© 2021, ZRC SAZU Anton Melik Geographical Institute 2.3 Perko’s typology (1996) ThethirdlandscapetypologyofallofSlovenianterritorywasproducedhalfacenturyafterMelikbyDrago Perko while preparing a volume on Slovenia’s regional geography (Perko and Orožen Adamic 1998) and the national atlas of Slovenia (Fridl et al. 1998a, 2001). This was the first partly computer-based landscape typology of Slovenia. Its research bases were first published in 1998, together with the natural geograph­ical regionalization of Slovenia (Perko 1998a). Perkostartedtheprocessoftypificationin1995usingageographicinformationsystem(Idrisi,nowTerrset). He entered four data layers: the surface elevation and inclination, and lithology and vegetation types. The inclinationandelevationdatawerebasedona100-meterdigitalelevationmodel,andthelithologyandveg­etationdatawereobtainedthroughdigitizationofa1:250,000lithologicalmapwiththirty-sevenbasicunits and a vegetation map with sixty-two basic units, which were both rasterized to a 100-meter grid. All four layers were then generalized and simplified into seven classes. Perko covered and joined all four layers. Altogether 2,401 different combinations were theoretically possible (Perko and Ciglic 2020a). To reduce the number of combinations, he filtered all four layers and the combined layer three times using the modus inside of a moving 11×11 cell square window, obtaining forty-eight larger and spatially separatehomogenouscoreswiththesamecombinationofelevation,inclination,lithology,andvegetation. Heprintedthesecoresona1:250,000scalemapand,withthehelpofexpertsforindividualpartsofSlovenia, manuallyplottedtheboundaries,mostlyinmorphologicalboundariesandlargerwatercourses.Intheend, he combined these forty-eight manually delineated landscape units into nine landscape types, which he combined further into four landscape type groups (Perko and Ciglic 2020a). The nine landscape types were (Figures 11 and 12): • alpska gorovja ‘Alpine mountains’; • alpska hribovja ‘Alpine hills’ as the largest type, covering 23.0% of Slovenia: • alpske ravnine ‘Alpine plains’; • panonska gricevja ‘Pannonian low hills’; • panonske ravnine ‘Pannonian plains’; • dinarske planote ‘Dinaric plateaus’; • dinarska podolja in ravniki ‘Dinaric lowlands’; • sredozemska gricevja ‘Mediterranean low hills’; and • sredozemske planote ‘Mediterranean plateaus’ as the smallest part, covering 3.3% of Slovenia. The four landscape type groups were: • alpske pokrajine ‘Alpine landscapes’; • panonske pokrajine ‘Pannonian landscapes’; • dinarske pokrajine ‘Dinaric landscapes’; and • sredozemske pokrajine ‘Mediterranean landscapes’. Based on this landscape typology, Perko et al. also produced a naturalgeographical regionalization of Slovenia, which was first published in 1996 (Kladnik 1996). The four groups of landscape types present­ed in the typology were replaced by macroregions, and the forty-eight landscape cores turned into mesoregions(Kladnik1996;Perko1998a;Perko2001;Perko2007a;Perko,Hrvatin,andCiglic2015;Perko, Hrvatin, and Ciglic 2017). In addition, the Bay of Trieste, which the Slovenian territorial waters extend to, was defined as a special region (Perko and Kladnik 1998). This is the combined or connecting scheme of the landscape typology and regionalization described above: • The macroregion Alpe ‘Alps’ (corresponding to the landscape type group Alpine landscapes in terms of boundariesandareascovered)includeselevenmesoregions,fourofwhich(e.g.,theJulianAlps)fallunder thelandscapetypeAlpinemountains,five(e.g.,theSavaHills)belongtothelandscapetypeAlpinehills, and two (e.g., the Sava Plain) fall under the landscape type Alpine plains; Figure11:OriginalmapofPerko’slandscapetypologyofSloveniafrom1996withlegend:1Alpinemountains,2Alpinehills,3Alpineplains,4Pannonian low hills, 5 Pannonian plains, 6 Dinaric plateaus, 7 Dinaric lowlands, 8 Mediterranean low hills, and 9 Mediterranean plateaus. p p. 23 Figure 12: Updated map of Perko’s landscape typology of Slovenia. p p. 24 Gornja Radgona Radlje ob Dravi Murska Sobota Lenart v Slovenskih Goricah Dravograd Lendava MARIBOR Ravne na Koroškem Ruše Ljutomer Kranjska Gora Slovenj Gradec Crna na Koroškem Jesenice Ptuj Ormož Slovenska Bistrica Slovenske Konjice Radovljica Bled Tržic D Bovec Mozirje Velenje Bohinjska Bistrica Kranj Kamnik Žalec CeljeRogaška Slatina Tolmin Železniki ŠentjurCerkno Ško"a Loka Zagorje ob Savi Mengeš TrbovljeHrastnik Laško Landscape types Domžale Alpine mountains Alpine hills LJUBLJANA Žiri Litija Alpine plains Sevnica Idrija Vrhnika Grosuplje Krško Pannonian low hills Nova Gorica Pannonian plains Logatec Trebnje Brežice Ajdovšcina Dinaric plateausDinaric lowlands Novo Mesto Cerknica Mediterranean low hills Postojna Mediterranean plateaus Ribnica Sežana Metlika Kocevje Adriatic Sea Ilirska Bistrica Crnomelj Piran Koper Izola 0 10203040 km Map by: Manca Volk BahunSource: RS, Surveying and Mapping Authority© 2021, ZRC SAZU Anton Melik Geographical Institute • The macroregion Panonska kotlina ‘Pannonian Basin’ (corresponding to the landscape type group Pannonian landscapes) has twelve mesoregions, nine of which (e.g., the Lendava Hills) fall under the landscape type Pannonian low hills and three (e.g., the Drava Plain) under the type Pannonian plains; • The macroregion Dinarsko gorovje ‘Dinaric Alps’ (corresponding to the landscape type group Dinaric landscapes) includes nineteen mesoregions, eleven of which (e.g., the Idrija Hills) fall under the land­scape type Dinaric plateaus and eight (e.g., the Ljubljana Marsh) under the landscape type Dinaric lowlands); • ThemacroregionSredozemlje‘Mediterranean’(correspondingtothelandscapetypegroupMediterranean landscapes) includes six mesoregions, four of which (e.g., the Koper Hills) fall under the landscapetype Mediterraneanlowhillsandtwo(e.g.,theKarstPlateau)underthelandscapetypeMediterraneanplateaus). Among all five typologies presented, the one produced by Perko has become the most established. It hasbeenincorporatedinallmajorgeographicalworksonSloveniapublishedafterSlovenia’sindependence: the eleventh volume of Enciklopedija Slovenije(Encyclopedia of Slovenia, 1997), Geografski atlas Slovenije (GeographicalAtlasofSlovenia;Fridletal.1998a),thebookonSlovenia’sregionalgeographytitledSlovenija: Pokrajineinljudje(Slovenia:RegionsandPeople;PerkoandOroženAdamic1998),NacionalniatlasSlovenije (National Atlas of Slovenia; Fridl et al. 2001), Popisni atlas Slovenije 2002 (Census Atlas of Slovenia, 2002; Dolencetal.2007),theatlasSloveniainFocus(Fridletal2008),Terasiranepokrajine(TerracedLandscapes; Perko,Ciglic,andGeršic2016;Perkoetal.2017),andTheGeographyofSlovenia:SmallbutDiverse(Perko, Ciglic,andZorn2020a).IthasbeenintroducedtoprimaryschoolstudentsviaŠolskakartaSlovenijevmer­ilu 1 : 500.000 (1:500,000 School Map of Slovenia; Perko 1997), Geografski atlas za osnovno šolo (GeographicalAtlasforPrimarySchools;Fridletal.1998b),andthepopularsciencemagazineGeografski obzornik (Geographical Horizon; Perko 1998c). In 2008, it was also incorporated into Slovenian legisla­tion, being used as the basis for assessing land quality (Perko, Hrvatin, and Ciglic 2015). 2.4 Špes et al.’s typology (2002) This typology was created as part of producing the methodology for studying environmental vulnerabil­ityatthenationalandlocallevelsasstipulatedbythe2003Slovenianenvironmentalprotectionlegislation to ensure a more quality sustainable development of Slovenia. A detailed description of the methodolo­gy and data sources used is provided in Špes et al. (2002). The researchers first produced a three-level landscape ecological regionalization based on the domi­nant landscape elements. Thecriteria of division at the highest level were primarily elevation differences, the share of carbonate rock, and average annual temperatures and precipitation, and the main criterion at the second level was the alternation of major concave and convex landforms (e.g., hills and valleys or plateausandlowlands).Atthefirstlevel,theydefinedfivemacroregions(AlpineSlovenia,PrealpineSlovenia, Pannonian Slovenia, Dinaric–Karst Slovenia, and Mediterranean Slovenia), referring to them as pokra­jinskoekološke makroenote ‘landscape ecological macrounits’. At the second level, they defined sixty mesoregions called pokrajinskoekološke mezoenote ‘landscape ecological mesounits’, and at the third level they defined 223 microregions called pokrajinskoekološke podenote ‘landscape ecological subunits’. Then the researchers combined similar landscape ecological units to produce a landscape typology with fourteen landscape ecological types: thirteen for land and one for sea. Their land types included the following (Figures 13 and 14): • visokogorski svet ‘mountains’; • širše recne doline v visokogorju, hribovju in na krasu ‘wide valleys in mountains, hills, and karst’; • visokekraškeplanoteinhribovjavkarbonatnihkamninah‘highkarstplateausandhillsofcarbonaterocks’; • hribovja v pretežno nekarbonatnih kamninah ‘hills of non-carbonate rocks’ as the largest type, covering 22.4% of Slovenia; Figure 13: Original map of the Špes et al. landscape typology of Slovenia (Špes et al. 2002) with legend: 1 Mountains, 2 Wide valleys in mountains, hills, and karst, 3 High karst plateaus and hills of carbonate rocks, 4 Hills of non-carbonate rocks, 5 Inter-mountain basins, 6 Low hills in inner Slovenia, 7 Plains and wide valleys in low hills in inner Slovenia, 8 Karst poljes and lowlands, 9 Low karst of Inner Carniola and Lower Carniola, 10 Low karst of White Carniola, 11 Karst and Podgorje Karst plateaus, 12 Low hills in the Littoral, and 13 Wide valleys and plains in the Littoral. p p. 26 Figure 14: Updated map of the Špes et al. landscape typology of Slovenia. p p. 27 Gornja Radgona Radlje ob Dravi Murska Sobota Lenart v Slovenskih Goricah Dravograd MARIBOR Lendava Ravne na Koroškem Ruše Slovenj Gradec Ljutomer Kranjska Gora Crna na Koroškem Jesenice Ptuj Ormož Slovenska Bistrica Radovljica Slovenske Konjice Bled Bovec Velenje Tržic D Mozirje Bohinjska Bistrica Žalec CeljeRogaška Slatina Kranj Kamnik Tolmin Železniki Mengeš TrbovljeHrastnik Šentjur Landscape types Cerkno Ško#a Loka Zagorje ob Savi Laško Domžale Žiri LJUBLJANA Litija MountainsWide valleys in mountains, hills, and karst Sevnica High karst plateaus and hillsof carbonate rocks Idrija Nova Gorica Vrhnika Grosuplje Krško Hills of non–carbonate rocks Ajdovšcina Brežice Logatec Inter–mountain basins Trebnje Novo Mesto Low hills in inner Slovenia Plains and wide valleys in lowhills in inner Slovenia Cerknica Postojna Karst poljes and lowlands Ribnica Low karst of Inner and Lower Carniola Low karst of White Carniola Metlika Kocevje Karst Plateau and Podgorje Karst Plateau Sežana Crnomelj Adriatic Sea Low hills in the Littoral Ilirska Bistrica Piran Koper Izola Wide valleys and plains in the Littoral 0 10203040km Map by: Manca Volk BahunSource: RS, Surveying and Mapping Authority© 2021, ZRC SAZU Anton Melik Geographical Institute • medgorske kotline ‘inter-mountain basins’; • gricevje v notranjem delu Slovenije ‘low hills in inner Slovenia’; • ravnine in širše doline v gricevju notranjega dela Slovenije ‘plains and wide valleys in low hills in inner Slovenia’; • kraška polja in podolja ‘karst poljes and lowlands’; • nizki kras Notranjske in Dolenjske ‘low karst of Inner Carniola and Lower Carniola’; • nizki kras Bele krajine ‘low karst of White Carniola’; • Kras in Podgorski kras ‘Karst and Podgorje Karst plateaus’; • gricevje v primorskem delu Slovenije ‘low hills in the Littoral’; and • širše doline in obalne ravnice v primorskem delu Slovenije ‘wide valleys and plains in the Littoral’ as the smallest type, covering 1.2% of Slovenia. Theyaddedageographicaldefinitiontosomeofthetypes(e.g.,innerSloveniaorLowerCarniola),and with one type they simply combined the names of two microregions or subunits (i.e., Karst and Podgorje Karst). Both Ilešic and Špes et al. combined the types and regions to some extent. Ilešic divided the landscapetypes into smaller regions,and Špes et al. combined smallerregions into landscape ecological types,treat­ing them as relatively homogenous areas that respond similarly to anthropogenic impacts and, as such, are extremely important for environmental vulnerability analyses. 2.5 Perko, Hrvatin, and Ciglic’s typology (2013) Thelandscapetypesofallfourtypologiespresentedsofarrepresentedcontiguousandmoreorlessround­ed-off polygons on the map. An individual landscape type may have represented only one polygon in one placeoritmayhaveappearedinseveralplaces,intheformofseveralspatiallyseparatedpolygons.However, in the landscape typology produced by Drago Perko, Mauro Hrvatin, and Rok Ciglic in 2013, the land­scape types were represented by cells that either touched one another, forming a polygon (i.e., they were contiguous), or were located far from one another (i.e., they were independent). Basedonthespatialoverlapof195landformunits,938lithologicalunits,andsixty-fivevegetationunits, theresearchers used the geographical information system to produce a typification at multiple levels with a different number of types, which makes it applicable to various areas and for various purposes. At the most detailed level, as many as 11,889,150 unit combinations were theoretically possible and at the most generallevel,whichreliedonthespatialoverlapoffourlandformunits,sevenlithologicalunits,andseven vegetationunits,thenumberofpossiblecombinationswas196,whichwasmanageable.Thecombinations formedthebasisfordefininglandscapetypes.Adetaileddescriptionofthemethodanddatasourcesused is provided in the article where this typology was first published (Perko, Hrvatin, and Ciglic 2015). Here, only the most generalized typology, which includes twenty-four landscape types, is provided for illustra­tion. The researchers used a 25-meter digital elevation model with 32,436,693 cells as the basic GIS layer. Atthisresolution,all threevectorlayerswerethenconvertedintorasterformat.Thisway,eachsquarecell with a 25m baseline and an area of 6.25 ares was furnished with information on the landform unit, litho-logical unit, and vegetation unit it belonged to. Throughgeneralization,theresearchersgraduallyreducedandlogicallycombinedthenumberofsmall­ er units. They produced intermediate typologies of various precision, only retaining four landform units, seven lithological units, and seven vegetation units for the simplest typology at the last level. The final landform units included: • Plains; • Low hills; • High hills; and • Mountains. The combined lithological units included: • Non-carbonate sediments (clay, silt, sand, silicate gravel); • Carbonate sediments (carbonate gravel, rubble, till); • Fine-grained clastic rocks (claystone, siltstone, marl); • Flysch (sandstone, marl); • Coarse-grained clastic rocks (carbonate conglomerate, silicate sandstone, conglomerate); • Carbonate rocks (limestone, dolomite, carbonate clastic rocks); and • Metamorphic and igneous rocks with tuffs (metamorphic rocks, igneous rock, tuffs, tuffites). The outcome of the final stage of generalizing vegetation units was as follows: • Downy oak and sessile oak (occasional European hophornbeam); • European hornbeam, oak, and red pine (occasional black alder, elm, and fir); • Beech; • Beech and fir; • Beech and European hophornbeam; • Beech, chestnut, and various oaks; and • Fir, spruce, and highland vegetation. By statistically analyzing the spatial overlap of all three unit types, their combination was determined for each cell. For example, if a cell ranked as low hills in terms of landform unit, as flysch in terms of litho-logical unit, and as downy oak and sessile oak in terms of vegetation unit, its combination was »low hills + flysch + downy oak, sessile oak.« This means that part of the landscape, which in the GIS virtual reali­ty was represented by this cell, was composed of low hills made of flysch and covered in downy oak and sessile oak. Combinationswiththehighestfrequencyandthusareaweredefinedasimportantandusedfordeter­miningthelandscapetypes.Ofthe196unitcombinationstheoreticallypossible,175wereactuallyidentified. Thetwentymostfrequentunitcombinationscoveringnearlytwo-thirdsofSloveniaweredefinedasland-scape types, and the rest were logically categorized under four additional types, in which the landform unit was the key factor. Thelandscapetypes identifiedwereas follows (Figure15;thenumberofthetypeusedinthemap leg­end is provided in parentheses): • ravnina+nekarbonatneusedline+beligaber,dob,rdecibor‘plains+non-carbonatesediments+European hornbeam, oak, red pine’ (1); • ravnina+nekarbonatneusedline+bukev,kostanj,hrasti‘plains+non-carbonatesediments+beech,chest­nut, various oaks’ (2); • ravnina+karbonatneusedline+beligaber,dob,rdecibor‘plains+carbonatesediments+Europeanhorn- beam, oak, red pine’ (3); • ostale ravnine ‘other plains’ (4); • gricevje+nekarbonatneusedline+beligaber,dob,rdecibor‘lowhills+non-carbonatesediments+European hornbeam, oak, red pine’ (5) as the smallest type, covering 1.6% of Slovenia; • gricevje + nekarbonatne usedline + bukev, kostanj, hrasti ‘low hills + non-carbonate sediments + beech, chestnut, various oaks’ (6); • gricevje + drobnozrnate klasticne kamnine + bukev, kostanj, hrasti ‘low hills + fine clastic rocks + beech, chestnut, various oaks’ (7); • gricevje + fliš + puhasti hrast, graden ‘low hills + flysch + downy oak, sessile oak’ (8); • gricevje + fliš + bukev, kostanj, hrasti ‘low hills + flysch + beech, chestnut, various oaks’ (9); • gricevje + karbonatne kamnine + puhasti hrast, graden ‘low hills + carbonate rocks + downy oak, ses­sile oak’ (10); • gricevje + karbonatne kamnine + beli gaber, dob, rdeci bor‘low hills + carbonate rocks + European horn­beam, oak, red pine’ (11); • gricevje + karbonatne kamnine + bukev ‘low hills + carbonate rocks + beech’ (12); • gricevje + karbonatne kamnine + bukev, jelka ‘low hills + carbonate rocks + beech, fir’ (13); • gricevje + karbonatne kamnine + bukev, kostanj, hrasti ‘low hills + carbonate rocks + beech, chestnut, various oaks’ (14); • ostala gricevja ‘other low hills’ (15); • hribovje + grobozrnate klasticne kamnine + bukev ‘high hills + coarse clastic rocks + beech’ (16); • hribovje + karbonatne kamnine + bukev ‘high hills + carbonate rocks + beech’ (17); • hribovje + karbonatne kamnine + bukev, jelka ‘high hills + carbonate rocks + beech, fir’ (18); Figure 15: Map of the Perko, Hrvatin, and Ciglic landscape typology of Slovenia from 2013 (the type number in the legend corresponds to the type number from the list of types in the text; Perko, Hrvatin, and Ciglic 2013). p p. 30 Gornja Radgona Radlje ob Dravi Murska Sobota Dravograd Lendava Lenart v Slovenskih Goricah MARIBOR Ravne na Koroškem Ruše Ljutomer Slovenj Gradec Kranjska Gora Crna na Koroškem Jesenice Ptuj Ormož Slovenska Bistrica Slovenske Konjice Bled Bovec Velenje Tržic D Mozirje Radovljica Bohinjska Bistrica Tolmin Železniki Kranj Kamnik Žalec CeljeRogaška Slatina Šentjur 30 Mengeš Trbovlje Cerkno Škofja Loka Landscape types Zagorje ob Savi Hrastnik Laško Domžale 1 13 2 14 Žiri LJUBLJANA Litija Sevnica Idrija Vrhnika 3 15 Nova Gorica Grosuplje Krško 4 16 5 17 Logatec Brežice a Trebnje Ajdovšcina 6 18 7 19 Novo Mesto Postojna Cerknica 8 20 9 21 Ribnica Adriatic Sea Sežana 10 22 Metlika 11 23 Kocevje 12 24 Ilirska Bistrica Piran Koper Izola Crnomelj 0 10203040km Map by: Manca Volk BahunSource: RS, Surveying and Mapping Authority© 2021, ZRC SAZU Anton Melik Geographical Institute • hribovje+karbonatnekamnine+bukev,gabrovec‘highhills+carbonaterocks+beech,Europeanhophorn-beam’ (19); • hribovje + metamorfne in magmatske kamnine s tufi + bukev, kostanj, hrasti ‘high hills + metamorphic and igneous rocks with tuffs + beech, chestnut, various oaks’ (20); • ostala hribovja ‘other high hills’ (21) as the largest type, covering 14.9% of Slovenia; • gorovje + karbonatne kamnine + bukev ‘mountains + carbonate rocks + beech’ (22); • gorovje + karbonatne kamnine + jelka, smreka, visokogorsko rastje ‘mountains + carbonate rocks + fir, spruce, highland vegetation’ (23); and • ostala gorovja ‘other mountains’ (24). Because the typology was completely computerized, an important part of the typification process was dedicatedtotestingitontheground.Theresearchersfirstcheckedafewtestareasbythemselvesandthen, basedontheirexperience,selectedfivetestcasesonwoodedlandandfiveonneighboring,alreadycleared land for each of the twenty-four landscape types. They also developed a suitable procedure and informa­tionformforthegeographicalinstitute’sresearchersthatvisitedandcheckedthetestareas.Fieldworkwas documented through written reports, photos, drawings, and other material. Theresearchersconceivethelandscapetypesdefinedandtestedinthemannerdescribedaboveasrel- ativelyhomogenousnaturalspatialunitsthathaveasimilarimpactonsociallandscapeelements,respond similarly to human-induced changes, and require similar protection. AccordingtoPerko,Hrvatin,andCiglic,themainadvantageoftheirtypologyisitsflexibilitybecause, on the one hand, organizing data in GIS allows constant updates, additions, and expansions, and, on the other, relatively fast and simple creation of landscape typologies at various levels, with various precision, a different number of types, and for various purposes. 2.6 Comparing the macrotypologies A great deal of similarity can be established between the macrotypologies of Slovenia presented, but also several important differences (Table 1). Melik and Ilešic only presented tentative criteria for dividing the territory into types, whereas in the other three typologies the criteria were significantly more refined and other methodological approaches to typification were also comprehensibly described. The authors of the four oldest landscape typologies manually drew the boundaries of the landscape types presented as polygons, whereas in the most recent typology the landscape types presented as a set of cells were defined by the computer itself based on the criteria selected by the researchers. There are also differences in aspects related to the purpose of the typology. Špes et al. advocated the ecologicalaspectoflandscapetypesbecausetheywereinterestedintheirvulnerability.Ilešicadoptedasim­ilarapproach,whereasMelikemphasizedtheeconomicaspectbecauseheviewedlandscapetypesthrough the prism of economic development opportunities. These aspects are also evident from the types’ names (Table 1). Three approaches to hierarchizing and combining types and regions can be observed: Ilešic dividedthetypesatthelowestlevelintoregions,Špesetal.combinedsimilarsmallerregionsintotypes,andPerko combined the types and regions at the same levels. In addition, the researchers combined the elements of regionalization with their typologies in other ways. Melik almost never mixed types and regions in the typologyitself,butinhisdescriptionsofindividualtypesheconsistentlylistedthemicroregionsintowhich he divided the types. In turn, Ilešic practically augmented his typology with regionalization: his units at the first and second levels are types and his units at the lower levels are regions. The fewest types were defined by Melik (eight) and the most by Perko, Hrvatin, and Ciglic (twenty-four, or three times as many). The greatest imbalance in size between the largest and smallest types can be found in Ilešic’s typology, where the largest type is over forty times larger than the smallest one. The smallest difference can be observed in Perko’s typology, where the largest type is only seven times larger than the smallest one. The three most recent typologies were partly or fully produced using a geographic information sys­tem. Perko and Špes et al. converted the data into raster format using a 100-meter digital elevation model, and the authors of the most recent typology used 25-meter digital elevation model. To establish similarities between the typologies (Figures 16 to 24), ten contingency tables were creat­ed to compare each typology with all the others (ten tables was the maximum number possible for five typologies). The frequency (area) of all types within one typology was determined across all types of theother four typologies and vice versa. Thus, for example, the landscape ecological type mountains in Špes et al.’s typology can be categorized in full under the landscape type Alpine mountains in Perko’s typology and, vice versa, the landscape type Alpine mountains in Perko’s typology can be categorized under fivelandscape ecological types in Špes et al.’s typology (about six-tenths under the type mountains and one-tenth under four other types). Table 1: Comparison of macrotypologies of Slovenia. Authors Year Typology Terminology No. of basic Average basic Ratio between Resolution described as used for type types type size (km˛) largest and smallest (m) basic types Melik 1946 Natural- natural 8 2,534 19:1 – -economic component part Ilešic 1958 Landscape- landscape type 11 1,843 41:1 – -typological ecological area and ecological Perko 1996 Landscape landscape type 9 2,252 7:1 100 Špes et al. 2002 Landscape landscape 13 1,559 18:1 100 ecological ecological type Perko, Hrvatin, 2013 Landscape landscape type 24 845 9:1 25 and Ciglic Figure16: Accordingto the1996 geographical regionalizationofSlovenia producedby Perkoet al.,theSavaHillsconstitutethelargestSlovenian region (Perko1998a;Perko and Ciglic2020b). Melik largely includedthe SavaHillsunderhills (Melik 1946), Ilešic largelyunder easternsub-Alpine landscapes(Ilešic 1972), Perko in full under Alpine hills (Perko and Ciglic 2020a), and Špes et al. mostly under hills of non-carbonate rocks (Špes et al. 2002). MATEVŽ LENARCIC, GIAM ZRC SAZU ARCHIVE IVAN KEBE, SHUTTERSTOCK.COM Correlation coefficients were calculated. The higher they are, the more similar two typologies are and to a lesser extent the types of one typology are scattered across the types of the other. Thegreatestsimilarity wasestablishedbetweenIlešic’sandPerko’stypologies,witha correlationcoef­ficientof0.8302,andthesmallestsimilaritywasdeterminedbetweenIlešic’sandPerko,Hrvatin,andCiglic’s typologies (0.4978). AllthetypologieshavethehighestcorrelationcoefficientswithPerko’stypology:thecoefficientestab­lished for Melik’s typology was 0.7622, for Ilešic’s typology it was 0.8302, for Špes et al. it was 0.7393, and for Perko, Hrvatin, and Ciglic it was 0.6819. Therefore, to present Slovenia’s basic landscape characteris­tics, an overview of landscape types in Perko’s typology is added below (Table 2). On average, Perko’s typology has the highest correlation coefficient (0.7534), followed by that of Špes etal.(0.6812),Ilešic(0.6796),andMelik(0.6596).Perko,Hrvatin,andCiglic’stypologyhasthelowestcoef­ficient(0.5782),whichmakessensebecauseitusesadifferentmethodologythanthefouroldertypologies. The correlation coefficients between individualmacrotypologies are not low, which suggests that the macrotypologies of Slovenia differ from one another in finer details but are similar in general. Their sim­ilarity to the macrotypologies of Slovenia produced by researchers from abroad is discussed in the next section. Table 2: Some basic characteristics of the landscape types in Perko’s typology (updated based on Perko and Ciglic 2020a). p p. 38–41 Variable Landscape type group Alpine landscapes Pannonian landscapes Area (km˛) 8,540.95 4,291.51 % of area 42.13 21.17 Mean elevation (m) 731.90 260.61 Mean inclination (°) 20.18 7.48 Most frequent rock type limestone 17.17%; dolomite 13.40%; clay and silt 30.66%; marl 20.96%; carbonate gravel, rubble, and till 13.34%; silicate gravel 19.13%; sand 14.65% silicate sandstone and conglomerate 11.03% Most frequent vegetation type beech 42.53%; beech, chestnut, beech, chestnut, and various oaks 65.53%; and various oaks 19.17%; beech hornbeam and pedunculate oak 16.37%; and hophornbeam 10.82% hornbeam 10.03% Insolation (MJ/m˛) 3,876.93 4,145.64 % of fields 4.23 30.61 % of vineyards 0.09 2.16 % of orchards 1.50 2.83 % of meadows and pastures 18.44 19.02 % of forests 66.88 36.72 % of built-up areas 4,68 7,43 Average landscape diversity 0.182028 0.151146 % of hotspot area 13.68 6.00 % of coldspot area 3.35 7.78 Population in 1931 561,090 439,367 % of population in 1931 40.15 31.44 Population in 1961 696,945 497,312 % of population in 1961 43.79 31.25 Population growth index, 1931–1961 124.21 113.19 Population in 1991 924,174 547,787 % of population in 1991 47.01 27.86 Population growth index, 1931–1991 164.83 124.52 Population growth index, 1961–1991 132.70 110.01 Population in 2011 959,695 540,148 % of population in 2011 46.81 26.35 Population density in 2011 112,36 125,86 (people per km˛) Population growth index, 1931–2011 171.04 122.94 Population growth index, 1991–2011 103.77 98.73 Number of settlements in 2011 2,237 1,636 Settlement density in 2011 26.19 38.12 (number per 100km˛) Average size of settlement in 2011 429.01 330.16 (people per settlement) Population in 2017 by houses 1,032,122 574,203 Landscape type group Dinaric landscapes Mediterranean landscapes Slovenia 5,706.25 1,734.29 20,273.30 28.15 8.55 100.00 579.91 351.74 556.83 12.70 11.22 14.62 limestone 54.66%; dolomite 21.17%; flysch 45.95%; limestone 37.83% limestone 26.67%; clay and silt 13.13%; clay and silt 11.07% dolomite 12.39% beech 35.36%; beech and fir 29.78%; downy oak and hophornbeam 31.20%; beech 29.53%; beech, chestnut, hornbeam and fir 12.80% downy oak 16.82%; beech, chestnut, and various oaks 24.94% and various oaks 16.04%; sessile oak 15.07%; beech 14.57% 4,006.00 4,376.02 4,012.84 3.84 4.37 9.71 0.30 4.16 0.94 0.83 3.46 1.76 17.91 19.08 18.47 73.33 62.10 61.90 3,31 6,17 5,01 0.147703 0.141510 0.162362 3.86 2.48 8.33 14.89 19.34 8.90 238,274 158,919 1,397,650 17.05 11.37 100.00 253,909 143,357 1,591,523 15.95 9.01 100.00 106.56 90.21 113.87 309,492 184,533 1,965,986 15.74 9.39 100.00 129.89 116.12 140.66 121.89 128.72 123.53 352,941 197,405 2,050,189 17.22 9.63 100.00 61,85 113,82 101,13 148.12 124.22 146.69 114.04 106.98 104.28 1,642 515 6,030 28.78 29.70 29.74 214.95 383.31 340.00 381,404 212,550 2,200,279 Variable Landscape type Alpine mountains Alpine hills Alpine plains Pannonian low hills Area (km˛) 3,061.77 4,659.97 819.21 2,994.52 % of area 15.10 22.99 4.04 14.77 Mean elevation (m) 1,055.45 582.38 373.09 288.77 Mean inclination (°) 26.24 18.94 4.56 10.29 Most frequent rock type limestone 37.22%; silicate sandstone and carbonate gravel, rubble, clay and silt 31.10%; carbonate gravel, rubble, conglomerate 17.26%; and till 51.70%; clay marl 29.71%; and till 19.67%; metamorphic rocks 16.57%; and silt 21.62%; carbonate sand 20.26% dolomite 14.38% dolomite 14.87% conglomerate 15.40% Most frequent beech 49.42%; dwarf pine beech 41.85%; beech, hornbeam 65.33%; beech, chestnut, vegetation type and other highland chestnut, and various beech 20.69%; and various oaks 88.18% vegetation 12.92%; beech oaks 31.37%; beech and red pine 10.51% and fir 12.10%; beech hophornbeam 12.25% and hophornbeam 11.48% Insolation (MJ/m˛) 3,705.84 3,953.62 4,080.05 4,131.55 % of fields 0.53 3.11 24.43 21.23 % of vineyards 0.00 0.17 0.01 3.00 % of orchards 0.49 2.18 1.40 3.55 % of meadows and pastures 12.68 21.74 21.17 22.97 % of forests 74.38 68.12 31.79 43.07 % of built-up areas 1.64 4.03 19.74 5.56 Average landscape diversity 0.183556 0.182367 0.174393 0.154415 % of hotspot area 12.46 14.10 15.84 6.31 % of coldspot area 1.21 3.35 11.37 3.80 Population in 1931 82,906 263,809 210,332 258,404 % of population in 1931 5.93 18.88 15.05 18.49 Population in 1961 89,621 292,985 314,339 256,737 % of population in 1961 5.63 18.41 19.75 16.13 Population growth index, 108.79 112.32 144.28 100.08 1931–1961 Population in 1991 92,224 338,344 493,606 252,074 % of population in 1991 4.69 17.21 25.11 12.82 Population growth index, 111.95 129.98 226.56 96.04 1931–1991 Population growth index, 102.90 115.71 157.03 95.96 1961–1991 Population in 2011 89,581 354,244 515,870 251,208 % of population in 2011 4.37 17.28 25.16 12.25 Population density in 2011 29.26 76.02 629.72 83.89 (people per km˛) Population growth index, 108.74 135.81 236.77 97.92 1931–2011 Population growth index, 97.13 104.49 104.51 101.96 1991–2011 Number of settlements in 2011 303 1528 406 1252 Settlement density in 2011 9.90 32.79 49.56 41.81 (number per 100km˛) Average size of settlement in 295.65 231.84 1270.62 200.65 2011 (people per settlement) Population in 2017 by houses 95,134 398,339 538,649 282,718 Landscape type Pannonian plains Dinaric plateaus Dinaric lowlands Mediterranean low hills Mediterranean plateaus 1,296.99 6.40 195.59 1.00 silicate gravel 59.62%; clay and silt 20.65% 3,809.32 18.79 668.09 15.24 limestone 59.13%; dolomite 24.29% 1,896.93 9.36 402.83 7.61 limestone 45.69%; clay and silt 28.29%; dolomite 14.90% 1,061.02 5.23 305.11 12.66 flysch 71.56%; clay and silt 13.75%; limestone 10.10% 673.27 3.32 425.22 8.94 limestone 82.52%; dolomite 11.22% hornbeam and pedunculate oak 44.44%; hornbeam 24.22%, pedunculate oak 17.40%; beech, chestnut, and various oaks 13.23% 4,178.22 52.26 0.22 1.16 9.89 22.06 11.75 0.143597 5.31 16.98 182,780 13.08 240,575 15.12 131.59 beech 39.01%; beech and fir 38.65%; beech and hophornbeam 10.02% 3,947.64 1.26 0.24 0.61 12.87 83.21 1.47 0.144251 3.20 16.37 102,324 7.32 77,143 4.85 75.18 hornbeam and fir 32.42%; beech 28.03%; hornbeam and pedunculate oak 13.65%; beech and fir 11.97% 4,123.22 9.02 0.40 1.26 28.02 53.51 7.02 0.154634 5.17 11.90 137,160 9.81 176,766 11.11 130.30 downy oak 26.45%; beech, chestnut, and various oaks 25.66%, sessile oak 20.77%; beech 10.72% 4,372.86 6.18 6.03 5.37 1.81 57.89 7.80 0.149312 3.86 14.18 129,568 9.27 120,570 7.58 93.65 downy oak and hophornbeam 69.80%; beech 20.64% 4,381.01 1.51 1.21 0.46 24.23 68.74 3.61 0.129215 0.29 27.46 30,367 2.17 22,787 1.43 75.52 295,713 15.04 164.49 63,940 3.25 62.36 245,552 12.49 180.97 160,791 8.18 124.89 23,742 1.21 78.68 125.01 82.95 138.88 133.36 104.19 288,940 14.09 222.78 65,565 3.20 17.21 287,376 14.02 151.50 172,369 8.41 162.46 25,036 1.22 37.19 158.04 63.90 211.83 133.88 82.97 96.08 102.46 117.06 107.20 105.45 384 29.61 809 21.24 833 43.91 380 35.81 135 20.05 752.45 81.04 344.99 453.60 185.45 291,485 77,076 304,328 183,488 29,062 3 Slovenia in European macrotypologies Alltheauthorsofthelandscapetypologiesdiscussed(Melik1946;Ilešic1958;Perko1998a;Špesetal.2002;Perko, Hrvatin, and Ciglic 2015) and other Slovenian geographers highlight the landscape diversity of Slovenia’ssmallterritory,resultingfromitslocation attheintersectionoffourmajorEuropeangeographical units: the Alps, the Pannonian Basin, the Dinaric Alps, and the Mediterranean (Melik 1935; Gams 1998; Kladnik and Perko 1998; Plut 1999; Perko, Hrvatin, and Ciglic 2017; Perko, Ciglic, and Zorn 2020b). Thelandscape and similar divisions of Europe that also include Slovenia reveal how the country is viewed by professionals outside Slovenia. Nine such classifications of Europe were reviewed to examine which macrounits Slovenia is catego­rizedunderandtowhatextentthesedivisionsaresimilartoSlovenianlandscapetypologies.Theyinclude: 1. TheEnvironmentalStratificationofEurope(Mücheretal.2003;Metzgeretal.2005;Jongmanetal.2006); 2. The European Landscape Classification (LANMAP2; Mücher et al. 2003; Mücher et al. 2006; Mücher et al. 2010); 3. The Digital Map of European Ecological Regions (DMEER 2003); 4. Biogeographical Regions (2016); 5. ThePhysical-GeographicalClassificationofEurope(Germ.Physisch-geographischeGliederungEuropas; Bohn et al. 2003); 6. The Pan-European Landscape Types (Meeus 1995); 7. The Terrestrial Ecoregions of the World (Olson et al. 2001); 8. The Biogeographical Provinces of Europe (European Environment Agency 1995); and9. TheBiogeographicMapofEurope(Rivas-Marínez,PenasandDíaz2004b)andBioclimaticMapofEurope (Rivas-Marínez, Penas and Díaz 2004a). OlderclassificationsofEuropewerestillproducedinamoretraditionalandsubjectivemanner(Mücheret al. 2003), whereas more recent ones (e.g., the Environmental Stratification of Europe and the European Landscape Classification) were produced using geographic information systems and digital data layers. The size of the basic spatialdata unit (cell) of some is only 1km˛. A detailed description ofall nine classifi­cations is provided in Ciglic and Perko (2012). For easier comparison between the typologies, Slovenia’sterritory is displayed at the same scale on all the maps (Figures 25 to 34). 3.1 Environmental Stratification of Europe A group of researchers (Mücher et al. 2003; Metzger et al. 2005; Jongman et al. 2006) used quantitativemethods and digital data on natural factors (elevation, slope, proximity of the ocean, latitude, and vari­ous climatic factors) to produce an environmental stratification of Europe with a resolution of 1km˛. Theydefinedeighty-fourenvironmentalclassesandcombinedthemintothirteenenvironmentalzones, which were further combined into six biogeographical regions. Including islands in the Atlantic Ocean,the stratification comprises fourteen zones and seven regions. Slovenia is categorized under three of the six environmental regions: Alpine, Mediterranean, and Continental(Figure25),andunderfiveofthethirteenenvironmentalzones:Alpinesouth,Mediterraneanmountains, Mediterranean north, Pannonian, and Continental. At the lowest level, twelve of the eighty-four classes can be identified in Slovenia (Metzger et al. 2005). 3.2 European Landscape Classification This classification was produced based on climate, elevation, soil, and land-use data. Major urban areas, water surfaces, and tide areas were specified separately (Mücher et al. 2006). The authors defined eight types at the first level, thirty-one at the second level, and seventy-six at the third level, and they specified 350 landscape types at the lowest (i.e., fourth) level (Mücher et al. 2006). Atthefirstlevel,Sloveniaisclassifiedunderthreetypes:Mediterranean,Continental,andAlpine(Figure 26).PracticallytheentirecountryliesintheMediterraneantypebut,interestingly,theGoriziaHills,atyp-icalMediterraneanlandscape,arepartoftheAlpinetype.Atthesecondlevel,Sloveniaincludeseighttypes (withoutseparatelydefinedurbanareasofLjubljanaandMaribor).Thethirdlevelincludestwelveofatotal of seventy-six types, and the final, fourth level includes nineteen of a total of 350. 3.3 Digital Map of European Ecological Regions Thisdigitalmapwasproducedbasedonclimate,topography,andgeobotanicaldata.Theauthors’goalwas to present areas with homogenous ecological conditions. This typology relies heavily on vegetation, with ecological regions largely named after the type of vegetation (Mücher et al. 2003; DMEER 2003). It includes sixty-eight European ecological regions(DMEER 2003). On the map (Figure 27), Slovenia liesinfourunitsandbordersone:Dinaricmountainsmixedforests,Illyriandeciduousforests,Alpineconif­ erous and mixed forests, Pannonian mixed forests, and Po river Basin mixed forests. 3.4 Biogeographical Regions BiogeographicalregionswerespecifiedforthepurposesoftheNatura2000network,asstipulatedinCouncil Directive 92/43/EEC. This isthe first time that non-administrative borders were defined in an official EU document (Mücher žet al. 2003). The last version is from 2016. Earlier versions were based on combining naturalvegetationinthememberstatesoftheEuropeanCommunityandtheCouncilofEurope(Noirfalise 1987).Forestcommunitieswerecombinedintobiogeographicalregions,andthemapwasgeneralized.Later versions also used the Map of Potential Vegetation produced by the German Federal Agency for Nature Conservation (The indicative map…1996). The2016classificationincludeselevenbiogeographicalregions.Itreliesonnaturalvegetation,butsome boundaries also run along administrative (national) borders, which is why it deviates from a completely natural division. According to the 2016 division, Slovenia is part of the Continental and Alpine regions, bordering the Pannonian region to the northeast and the Mediterranean region to the southwest (Figure 28). 3.5 Physical-Geographical Classification of Europe TheGermanFederalAgencyforNatureConservationproducedaphysical-geographicalclassificationofEurope (Germ.Physisch-geographischeGliederungEuropas)basedonclimate,rocks,andsoil,whichformedthebasis for the 1:2,500,000 Map of Natural Vegetation of Europe (Germ. Karte der natürlichen Vegetation Europas). Europewasdividedintofoursubcontinents(Germ.Subkontinent):NorthernEurope(Nordeuropa),Western andCentralEurope(West-undMitteleuropa),SouthernEurope(Südeuropa),andEasternEurope(Osteuropa). These were thendividedinto nine majorareas(Großraum)and forty-seven physical-geographical regions (physisch-geographischeRegion),whichwerefurtherdividedintosubunitsorsubregions(Bohnetal. 2003). Slovenia is part of two subcontinents (i.e., Western and Central Europe, and Southern Europe), three majorareas(Alpine,Carpathian,andMediterranean;Figure29),andfourregions(theAlps,theNorthItalian Plain, the Pannonian Basin, and the Balkan Peninsula). Figure 27: European ecological regions (DMEER 2003). 3.6 Pan-European Landscape Types Meeus (1995) produced a map of pan-European landscape types based on landforms (as a consequence of the rock base and climate), economic land-use potential, sustainability of human activity, nature con­servation, settlement pattern, field pattern, visual impression, and the quality of the view. The typology includes thirty landscape types combined into nine groups. Worth mentioning among them is the group Regional landscapes, which comprises types that only appear in one or several places thanks to their exceptional natural or cultural features (Meeus 1995). NearlyallofSloveniafallsunderasinglelandscapetype:Mediterraneansemi-bocage(Figure30).The French word bocage denotes a landscape of mixed farmland and forests or trees. This type is character-izedbyaMediterraneanclimate,diverselanduse,andthepredominanceofruralsettlements(Meeus1995). Three other types can be found on the edges of Slovenia and its immediate vicinity: the Collective open field in the east, the Delta type in the west, and the Mountain type in the north (Meeus 1995). 3.7 Terrestrial Ecoregions of the World This map was produced based on the biogeographical characteristics of landscapes around the world. Ecoregions are characterized by special combinations of natural communities and species. The borders between these areas match the natural conditions prior to major changes introduced by man. The Digital MapofEuropeanEcologicalRegions(seeSection3.3)wasusedtoidentifyunitsinEurope.Inareasforwhich nobiogeographicaldivisionswerefound,theauthorsreliedonlandformsandvegetation(Olsonetal. 2001). Theterrestrialpartoftheworldwasdividedintoeightgeographicalrealmsandfourteenbiomes.Within these, the authors further defined 867 ecoregions (Olson et al. 2001). At the level of biomes, of the fourteen in the world and eight in Europe, three are present in Slovenia: Mediterraneanforests,woodlandandscrub,temperatebroadleafandmixedforests,andtemperateconif­erousforests(Figure31).EcoregionsinSloveniaoritsimmediatevicinityincludePannonianmixedforests, Dinaric mountains mixed forests, Illyrian deciduous forests, Alpine coniferous and mixed forests, and Po basin mixed forests. 3.8 Biogeographical Provinces of Europe The Europe’s Environment report of 1995 used biogeographical provinces of the world, which were based on Udvardy’s Classification of the Biogeographical Provinces of the World (1975). Europe is divided into nineteen provinces. The classification also covers northern Africa, Turkey, and theareaeastoftheCaspianSea.InadditiontoEurope’sbiogeographicalprovinces,thereportalsoshowed four EU biogeographical zones covering the territory of the EU at that time. Nearly all units are named after geographical names and occur only once, which makes this more of a regionalization than a typifi­cation (European Environment Agency 1995). Sloveniaispartoffourprovinces:Balkanhighlands,Continental,Mediterraneansclerophyll,andCentral European highlands (Figure 32). 3.9 The Biogeographic Map of Europe and the Bioclimatic Map of Europe Rivas-Martínez, Penas, and Díaz (2004b) produced an extensive biogeographical map of Europe, which coverstheareafromtheArabianPeninsulaandtheCaspianSeatotheCanaryIslands,Spitsbergen,Novaya Zemlya, and Franz Josef Land. Itincludesfiveregions:Circumarctic,Eurosiberian,Mediterranean,Irano-Turanian,andSaharo-Arabian. Thefirsttworegionsarefurtherdividedintothreesubregionseach.Thethirdlevelincludesthirtyprovinces, and the last (i.e., fourth) level includes seventy-two sectors. The higher levels can be considered more as a typification and the lower levels more as a regionalization. All of Slovenia is part of the Eurosiberian region and the Alpino-Caucasian subregion, which is fur­ther divided into the Alpine province and the Apennino-Balkan province (Figure 33). At the lowest level, it belongs to the Eastern Alpine, Illyrian, and Padanian sectors. Rivas-Martínez, Penas, and Díaz (2004a) also produced the Bioclimatic Map of Europe, which con­tains three levels and is based on climatic features. Europe is divided into four macrobioclimates, which are further divided into sixteen bioclimates with certain variants for some. WesternSloveniaispartoftheTemperateOceanicbioclimateandeasternSloveniaispartoftheTemperate Continentalbioclimate(Figure34).TheborderbetweenthemrunsintheDinaricdirectionfromthenorth­west toward the southeast. Taking into account both maps provides a better idea of the natural conditions in Slovenia. 3.10 Comparing the European divisions In terms of the levels of the divisions presented on the maps (Figures 25 to 33), Slovenia is most often (i.e., infourdivisions)foundinthreetypes(classes),inthreedivisionsitisfoundinfourtypes,andintwodivi­sions it is found in two and five types, respectively. In an individual typology, the highest share of all types in Slovenia appears in the Environmental Stratification of Europe (Metzger et al. 2005; Jongman et al. 2006): three of a total of six types (i.e., half), whichtheoreticallycorrespondsto148typespermillionkm˛ofthecountry’sarea.Estonia,whichistwice the size of Slovenia, has only one type, which corresponds to twenty-two types per million km˛ of its ter­ritoryornearlyseventimesfewertypesthanSlovenia.Italy,whichisfifteentimeslargerthanSlovenia,has twotypes,whichmeansonlyseventypespermillionkm˛orthatSloveniahastwenty-onetimesmoretypes. ThesmallestshareofalltypesinSloveniaisfoundontheBiogeographicMapofEurope(Rivas-Martínez, Penas,andDíaz2004b) –thatis,onlytwoofatotalofthirtytypes,whichcorrespondstoninety-eighttypes per million km˛ or less than a tenth of all types in this classification. Nonetheless, here, too, Slovenia is more diverse than Estonia and Italy. The former has two types, corresponding to forty-four per million km˛ or half the number of types as in Slovenia, and the latter has five types, corresponding to seventeen per million km˛ or one-sixth the number of types as in Slovenia. Similar ratios between Slovenia and other countries also apply to other divisions. In general, in terms of the number of types extending into its territory alone, Slovenia is comparable to the largest European countries,anditevenoutranksthemintermsoftypedensity.ThisconfirmsSlovenia’sabove-averageland­scapediversityanddemonstratesthatprofessionalsoutsideSloveniaalsoperceiveitthatway–thatis,similar to Slovenian geographers (Perko, Ciglic, and Zorn 2020b). ThemapshowinghowthebordersofthetypesfeaturedintheEuropeandivisionsexaminedareinter­connected (Figure 35) shows the extent of similarity between these divisions in Slovenia’s territory and howtheyoverlapwiththemostestablishedlandscapetypologyofSlovenia(Perko2007a),whichisbelieved to display Slovenia’s actual geographical features relatively well. Becausethesedivisionsuseddifferentmethods,differentlandscapeelements,anddataofdifferentquality (Ciglic and Perko 2012), there are clear differences between them (Figure 36), but they roughly all come close to the landscape typology of Slovenia (Perko 2007a), indicating its position at the intersection of the main European landscape types (i.e., the Alpine, Pannonian, Dinaric, and Mediterranean), even though they may sometimes be referred to with different names (e.g., Illyrian or Carpathian). 4 Examples of microtypifications of parts of Slovenia In addition to landscape typologies covering all of Slovenia, the institute has also produced a significant numberoflandscapetypologiesofsmallerpartsofSlovenia,ormicrotypologies,forvariouspurposesand using various methods. Geographers have also produced many classifications of parts of Slovenia outside the institute, among whichclassificationsbasedonasinglelandscapeelementandmicroregionalizationsdistinctlypredominate. Morecomplexclassificationsandlandscapetypificationsarerare,andtheylargelyrelyonterrainandrocks. For example, Ivan Gams divided southeastern Carinthia into eight landform types (Gams 1970), part oftheUpperSocaValleyintofivelandscape-ecologicalunits(Gams1975),andtheMariborregionintofive landscape-ecologicalunits(Gams1979).Inaddition,togetherwithotherresearchers,hedividedtheVoglajna andUpperSotladrainagebasinsineasternSloveniaintoeightnaturalgeographicalunits(Gamsetal.1974), the area surrounding the village of Breginj in Slovenia’s extreme west into five landscape-ecological units (Gams,Lovrencak,andPlut1978),andtheareaaroundthevillageofKamenintofourlandscape-ecological units(Gams,Lovrencak,andPlut1978).Cerne,Klemencic,andPlut(1981)definedtwelvelandscape-eco­logical units in the Municipality of Tržic, and Plut (1981) identified sixteen microchores, Ferreira (2006) dividednorthernUpperCarniolaintofivelandscape-ecologicaltypes,andTopole(1992)dividedtheMirna drainagebasinintofourmorphologicaltypes.WorthhighlightinghereisalsoastudyofsoilinthePohorje Mountains,inwhichRepe(2017)usedquantitativemethodsandgeographicinformationsystemstodeter­mine soil types based on pedogenetic factors, using many layers with natural landscape elements, which could also provide a basis for determining natural landscape types. A historical overview of selected examples of microtypologies produced at the institute between 1985 and 2020 (Figure 37) shows the development of the methodology for determining landscape types, espe­ciallyhowtraditionalgeographyusingpaperthematicmapswasenrichedbygeographicinformationsystems withdigitalthematiclayersthatrelyonincreasinglyaccuratedata.Themethodologiesusedforsomeland-scapemicrotypificationsrelatetothemethodologiesoflandscapemacrotypificationsofSlovenia,andothers are completely unique. 4.1 The Kokra Valley Thefirstmicrotypificationexampleoriginatesfromthemid-1980s,wheninitialstepswerestillbeingmade to use computers more widely and frequently in geography and elsewhere. The Kokra Valley was select­edbecauseitextendsfromawideplaintothehighmountainsandhenceprovidesarelativelygoodexample for illustrating Slovenia’s geographical characteristics (Perko, Hrvatin, and Ciglic 2017). TheKokraValleycovers222km˛,or1%ofSlovenia.TheKokrais54kmlong.ItrisesintheKarawanks, makesitswaythroughtheKamnik–SavinjaAlpsintotheLjubljanaBasin,andeventuallyflowsintotheSava at Kranj, 25km northwest of Ljubljana, through a conglomeratecanyon, which is a protected natural site. A landscape microtypology of the Kokra Valley was produced in 1985 as part of a study examining theconnectionbetweenthenaturalandculturallandscapeinthebasin.Adetaileddescriptionofthemethod and data sources used is provided in a 1986 article (Perko 1986). This typology was still produced the traditional, analog way, although conceptually already along the linesofageographicinformationsystem.Allcalculationsweremadeonacomputer,whereasthemapswere stillmanuallyproduced.Allareasweremeasuredbycountingthesquaresontransparentgraphpaperplaced overthe maps. Inthisway,atype ofasimplegeographic informationsystemwascreated, with rasterlayers Figure 37: Distribution of the microtypologies in Slovenia presented. p Gornja Radgona Radlje ob Dravi Murska Sobota Lenart v Slovenskih Goricah Dravograd MARIBOR Lendava Ravne na Koroškem Ruše Slovenj Gradec Ljutomer Kranjska Gora Crna na Koroškem Jesenice Ptuj Ormož Slovenska Bistrica Bled D Bovec Radovljica Tržic Mozirje Velenje Slovenske Konjice Bohinjska Bistrica Acta geographica Slovenica, 61-3, 2021 Žalec CeljeRogaška Slatina Kranj Kamnik Tolmin Železniki Šentjur Mengeš Trbovlje Hrastnik Examples of microtypologies Cerkno Škoa Loka Domžale Zagorje ob Savi Laško Kokra ValleyEastern Krka Basin Žiri LJUBLJANA Litija Nova Gorica Idrija Vrhnika Grosuplje Sevnica Volcji Potok Arboretum Krško Logatec Brežice Trebnje Lower Drava Valley City Municipality of Ljubljana Ajdovšcina Municipality of Idrija Novo Mesto Cerknica Municipality of Kocevje Postojna Ribnica City Municipality of Velenje Sežana Metlika Kocevje Adriatic Sea Crnomelj Ilirska Bistrica Piran Koper Izola 0 10203040 km Map by: Manca Volk BahunSource: RS, Surveying and Mapping Authority© 2021, ZRC SAZU Anton Melik Geographical Institute with a 420×440 square resolution and a 50m baseline. Before the microtypology was published (Perko 1986), the measurements were also checked through manual planimeter surveying – that is, by measur­ingtheareasusingaplanimeter.TheareaoftheKokraValleymeasuredwithaplanimeterwas22,330hectares and the number of squares counted was 89,376, corresponding to 22,344 hectares. The total area covered by the squares was larger than that measured with the planimeter because all the squares into which the basinatleastpartiallyextendedwereaddedtothearea. Itisinterestingthatthedifferencebetweenthetwo areas established this way and today’s accurate digital measurement of 22,191 hectares is only 0.6 or 0.7%. Figure38:OriginalmapoflandscapetypesintheKokra Valley (Perko1992)withlegend:1.Recentgravelterraces,2. Oldgravelterraces, 3.Conglomerate terraces, 4. Tertiary low hills, 5. Thermal belt, 6. Hills and uplands, 7. Small mountain basins, and 8. High mountains. Thisstudydidnotyetapplythetermlandscapetype;instead,itusedthetermtopetorefertoanalmost completelyhomogenousarea(landscapeunit)intermsofoneorseverallandscapeelements,andtheterm topiccomplexortopicfamily,whichcombinestopesthatarestillsufficientlysimilartoconstitutearelatively homogenous whole. Basedonvariousdatasourcesandmaps,fivesyntheticthematicmapsweredrawnatascaleof1:50,000: • Amapofmorphologicaltopiccomplexeswiththreeunitsormorphocomplexes:plainsandbasins(21.7% of the area), low hills (12.8%), and hills and mountains (65.5%); • Amapoflithologicaltopiccomplexeswitheightunitsorlithocomplexes:igneousrock(3.0%),slate(7.5%), limestoneanddolomite(31.8%),marl(14.2%),moraine(1.0%),scree(16.5%),conglomerate(11.3%),and gravel (14.0%); • Amapofclimatictopiccomplexeswithfiveunitsorclimaticcomplexes:basinsandplains(21.7%),ther­malbelt(12.8%),hillsanduplands(49.7%),smallmountainbasins(2.9%),andhighmountains(12.9%); • Amapofpedologicaltopiccomplexeswithsixunitsorpedocomplexes:lithosol(10.0%),rendzina(48.8%), ranker (9.9%), fluvial soil (6.1%), brown rendzina (9.6%), and acidic brown soil (15.6%); and • A map of phytological topic complexes with seven units or phytocomplexes: alpine vegetation (10.0%), beech on acidic soil (20.6%), beech on alkaline soil (37.5%), beech and fir (6.4%), pine on acidic soil (11.3%), sessile oak and hornbeam (8.2%), and alder and willow (6.0%). By overlapping these maps, it was established that the borders of the topic complexes overlapped rel­atively well, and therefore it was easy to produce a map of ecological topic complexes, again at a scale of 1:50,000, with the following eight units or eco-complexes (Figure 38): • mlade (holocenske) prodne terase ‘recent (Holocene) gravel terraces’ (3.7%); • stare (würmske) prodne terase ‘old (Würm) gravel terraces’ (7.4%); • konglomeratne terase ‘conglomerate terraces’ (10.6%); • terciarno gricevje ‘Tertiary low hills’ (6.1%); • topli pas ‘thermal belt’ (6.7%); • hribovje in sredogorje ‘hills and uplands’ (49.7%); • gorska kotlinica ‘small mountain basins’ (2.9%); and • visokogorje ‘high mountains’ (12.9%). Later on, the term ecological topic complex was replaced by landscape type (Perko 1992). The names of the eight landscape types are as follows (Figure 39): • mlajše prodne terase ‘recent gravel terraces’; • starejše prodne terase ‘old gravel terraces’; • konglomeratne terase ‘conglomerate terraces’; • terciarno gricevje ‘Tertiary low hills’; • topli pas ‘thermal belt’; • hribovje in sredogorje ‘hills and uplands’; • gorska kotlinica ‘small mountain basins’; and • visokogorje ‘high mountains’. 4.2 Eastern Krka Basin TheEasternKrkaBasinanditsedgescover733km˛,or3.6%ofSlovenia.Itformstheeasternpartofalong tectonic basin at the intersection of the Alps, the Dinaric Alps, and the Pannonian Plain along the Slovenian–Croatian border. Three large rivers deposit their material in the basin. The alpine Sava, which crosses it diagonally from the northwest to the southeast, deposits mostly gravel and sand. In turn, its two tributaries,thekarstKrka,whichflowsalongthebasin’ssouthernedgeinanorth–southdirection,andthe PannonianSotla,whichmeandersalongthebasin’seasternedgefromnorthtosouth,depositmaterialthat is sandier and more clay-likein nature. It is the watercourses thatcreate the mainlandscapedifferences in thispartofSloveniabydepositingandtransportingmaterialofvariousgranulationandfloodingeveryyear. A landscape microtypology of the Eastern Krka Basin was produced in 1989 as part of research on the connection between natural and social landscape elements, especially the impact of natural elements on settlement. A detailed description of the method and data sources used is provided in an article pub­lished that same year (Perko 1989). Maps at various scales and 1:10,000 aerial photos, mostly taken in 1986, constituted the main source of data on natural landscape elements that this microtypology is based on. First, all the maps and aerial photos were enlarged or reduced to the 1:50,000 scale and then overlapped with a 2×2cm square grid, which at the 1:50,000 scale corresponds to 1km˛. By entering data on landscape elements by square into a computer, a geographic information system with raster layers consisting of 41×33 squares with a 1km baseline was created. Of the total of 1,353 squares, 733 covered the Eastern Krka Basin. Because the features on the maps and aerial photos were presented either as points, lines, or polygons, data were primarily entered into the square grid in three ways: • By counting point features (e.g., dolines), measuring the length of line features (e.g., watercourses), and measuring the area covered by features (e.g., flood plains) in a square; • By determining the lowest and highest values of features (e.g., lowest and highest elevation) in a square; and • By identifying the predominant feature (e.g., the predominant type of soil) in a square. Seventeen maps and layers were prepared, containing more analytical natural landscape elements or variables (e.g., ridge lengths), which were then combined into five further synthetic maps and layers (e.g., rock types or lithocomplexes). Each of the twenty-three variables had its own map drawn on a transparent sheet and a correspond­ing layer in the geographic information system acquired by reading data by squares in the kilometer grid placed over the map. Incombiningmapswithanalyticalvariablesintomapswithsyntheticvariablesandthefinallandscape type map, the maps were overlapped on a transparent desk with light shining through all the transparent sheets from below. This study already applied the terms landscape and landscape types, even though the term topic com-plexorsimplycomplexwasstilllargelyusedforthesyntheticlayers,andeco-complexwasstillusedinstead of landscape type. The following five synthetic maps formed the basis for producing the microtypology: • Morphocomplexes with three units: plains (40% of the area), low hills (41%), and hills (19%); • Lithocomplexes with eight units: clay, loam, and sand (37%), sand and gravel (15%), marl (11%), marl and limestone (14%), limestone (5%), limestone and dolomite (7%), dolomite (11%), and slate (1%); • Hydrocomplexes with six units: floodplains (19%), areas with impeded drainage on plains and in the valleys(16%),areaswithnormaldrainageonplainsandinthevalleys(10%),areaswithimpededdrainage in the low hills and hills (20%), areas with normal drainage in the low hills and hills (21%), and areas with karst and partially karst drainage (14%); • Pedocomplexes with seven units: gley (20%), pseudo-gley (7%), fluvial soil (10%), acidic brown soil on clayandloam(10%),brownsoilongravel(8%),brownsoilonmarl(10%),andrendzina,chromiccam­bisol, or brown skeletal soil (35%); and • Phytocomplexeswithnineunits:willow,alder,andpoplar(22%),pedunculateoakwithEuropeanhorn- beam (5%), common hornbeam on acidic soil (13%), common hornbeam on alkaline soil (8%), oak on acidicsoil(6%),beechonacidicsoil(27%),beechonalkalinesoil(11%),beechwithhophornbeam(4%), and beech with common hornbeam (4%). An overview of complex values by square showed that many squares had the same or similar values – that is, that they were similar or even the same. In this sense, two squares are considered the same if they have completely the same combination of complex values. Intheory,thenumberofpossiblecombinationsequaledtheproductofpossibledifferentcombinations forindividualcomplexes. Becausethemorphocomplexeshadthreeunitsorpossiblevalues,thelithocom­plexeshadeight,thehydrocomplexeshadsix,thepedocomplexeshadseven,andthephytocomplexeshad nine possible values, 9,072 combinations were theoretically possible (3×8×6×7×9), but in reality 140 combinations were identified (1.5%), with only twenty-three combinations (0.3%) appearing at least ten times. The most frequent was the combination of the morphocomplex plains, the lithocomplex sand and gravel,thehydrocomplexnormaldrainageonplainsandinthevalleys,thepedocomplexbrownsoilongrav-el, and the phytocomplex common hornbeam on alkaline soil. This combination was found in forty-seven squares, which means it was typical of 47km˛ of the basin. The map of eco-complexes was produced by overlapping the maps of partial complexes and identify­ing the most common combinations of square values in the layers of these same complexes. Figure 42: Updated map of landscape types in the Eastern Krka Basin. p p. 57 The eco-complex map (Figures 40 and 42) and layer (Figure 41) include nine types: • ravninaizholocenskegline,iloviceinpeska‘plainmadeofHoloceneclay,loam,andsand’(105km˛,14.3% of the area); • ravnina iz holocenskega proda in peska ‘plain made of Holocene gravel and sand’ (49km˛, 6.7%); • ravninaizpleistocenskegline,iloviceinpeska‘plainmadeofPleistoceneclay,loam,andsand’(85km˛,11.6%); • ravnina iz pleistocenskega peska in proda ‘plain made of Pleistocene sand and gravel’ (59km˛, 8.1%); • gricevje iz holocenske gline in ilovice ‘low hills made of Holocene clay and loam’ (34km˛, 4.6%); • gricevje iz pleistocenskein pliocenske gline in ilovice‘low hills made of Pleistocene and Pliocene clay and loam’ (70km˛, 9.5%); • gricevje iz miocenskega laporja ‘low hills made of Miocene marl’ (88km˛, 12.0%); • gricevje iz mezozojskega laporja, apnenca in dolomita ‘low hills made of Mesozoic marl, limestone, and dolomite’ (107km˛, 14.6); and • hribovje ‘hills’ (136km˛, 18.6%). 4.3 Volcji Potok Arboretum The Volcji Potok Arboretum is located approximately 15km north of Ljubljana, between Domžale and Kamnik,eastoftheKamnikBistricaRiver.ItsmostprominentpartisVolcjihrib(literallyWolfHill),which rises to an elevation of 398m and where there are the ruins of a medieval castle. The arboretum covers just under 1km˛, and the square study area with the arboretum measures 2.25km˛, which is by far the smallest example amongthe microtypologies presented. Despite its small size, the area has a diverse land­scape, ranging from the most recent Holocene deposits via older Pleistocene deposits to Mesozoic and Paleozoic rocks, which differences in soil, vegetation, landforms, and other natural features are associat­ed with. This is where, with its first low hills, the bottom of the Ljubljana Basin begins to ascend to the extensive Sava Hills, which extend all the way to the Croatian border to the east. A landscape microtypology of this area was produced in 1992 as a geographical basis for protecting thearboretum,which was declareda cultural siteof national importancein 1999 duetoitsimportant gar­den architecture heritage. This microtypology is primarily interesting because a digital elevation model was applied for the first timeforthetypificationandbecauseitcoversaverysmallareawithevensmallerlandscapetypes.Adetailed description of the method and data sources used is provided in a 1993 article (Perko 1993). Atthattime,thewholeofSloveniawasonlycoveredbya100-meterdigitalelevationmodel,andsoamore accurateonewas producedinhouse. First, the mostly one-meter contour lines from the basic 1:5,000 topo­graphicmapweredigitized,andinsomeplacesintermediatepointswereaddedifneeded.Theresultingvector layer was converted into a 5×5m raster grid, through which a five-meter 1,500×1,500m digital elevation model was obtained, in which every cell or square measured 25m˛ (a quarter of an are). This digital eleva­tionmodelwasusedtodrawthree-dimensionalspatialimagesofthearea(Figure43),calculatetheelevations, slopes,andsurfaceaspect,andidentifyfourlandformunits(lowhills,asmallvalley,analluvialfan,andaplain). Because of the small size of the area, the cartographic bases were able to be extensively updated and then digitized. Data were processed using the IDRISI software. The entire square area was processed sep­arately from the are within the arboretum, which at that time covered 8,252 ares. The study used the terms landscape type and geoecological unit as synonyms. The main landscape differences were related to the age of fluvial sediments (Šifrer 1961) and surface aspect,whichiswhythesetwolandscapeelementswerecrucialinthearea’smicrotypificationandmicrore­gionalization. Fifteen microregions were identified: Volcji hrib, Kopasti hrib, Dolina jezerc, Dolina rododendronov, Hribarica,Rusula,Jelovadraga,Jamce,Vhribih,Dobrava,Volcjivršaj,Grašcinsko,Zalše,Spodnjepolje,and Srednjepolje.Theywerelargelynamedafteroldmicrotoponyms(Figure43).Inaddition,thefollowingfive main landscape types or geological units were defined (Figures 44 and 45): • aluvialnaravninazdrobnozrnatimiusedlinami‘Alluvialplainwithfine-grainedsediments’(characteris-tics:locatedimmediatelyalongcreeks,predominanceofsilicateparticles,thinandacidicsoil,floodplain; microregions: Dolina jezerc, Dolina rododendronov, and Jelova draga); • aluvialna ravnina z debelozrnatimi usedlinami ‘Alluvial plain with coarse-grained sediments’ (charac­teristics: mix of silicate and carbonate particles, thinner, less acidic, and more permeable soil; microregions: Spodnje polje, Grašcinsko, and Zalše); • periglacialnivršaj‘Periglacialalluvialfan’(characteristics:fine-graineddepositsareolderandmoreinclined than those in the alluvial plains at lower elevations; microregion: Volcji vršaj); • konglomeratnaterasa‘conglomerateterrace’(characteristics:predominanceofcarbonategravel,thicker, more acidic, and less permeable soil; microregions: Srednje polje, Hribarica, Rusula, and Dobrava); and • hribovje ‘hills’ (characteristics: predominance of Triassic dolomite and Permo-Carboniferous claystone and sandstone, oak and hornbeam are common on the sunny slopes (the southern slopes of Volcji hrib and Kopasti hrib), and Scots pine and beech on the shady sides; microregions: Volcji hrib, Kopasti hrib, Jamce, and V hribih). 4.4 Lower Drava Valley The Lower Drava Valley was processed within the confines of today’s twenty-three municipalities: Cirkulane, Destrnik, Dornava, Gorišnica, Hajdina, Juršinci, Kidricevo, Križevci, Ljutomer, Majšperk, Markovci,Ormož,Podlehnik,Ptuj,Razkrižje,SredišceobDravi,SvetiAndražvSlovenskihGoricah,SvetiTomaž,TrnovskaVas,Veržej,Videm,Zavrc,andŽetale.Togethertheycover1,034.5km˛,or5.1%ofSlovenia. The valley lies southeast of Maribor along the Mura and Drava rivers, before they enter Croatia. It is com­posed of a plain along the Mura to the northeast, low hills in the center, a plain along the Drava with the oldest Slovenian town of Ptuj to the south, and low rugged hills to the southwest. Thearea’slandscapemicrotypologywasproducedin1996aspartofastudyoftheconnectionbetween natural landscape elements and natural hazards because within the landscape types as relatively homoge­nousunitstheriskofnaturalhazardsandprotectionagainstthemaresimilar.Agoodknowledgeoflandscape typesandthedifferencesbetweenthemhelpspreventnaturalhazards,reducetheirnumber,oratleastdecrease their impacts. A detailed description of the method and data sources used is provided in an article pub­lished the same year (Fridl et al. 1996). The data were based on a 100-meter 59×42km digital elevation model. The researchers had the ele­vation,slope,surfaceaspect,andinsolationrasterlayersattheirdisposal,alongwiththerocktype,soiltype, potential vegetation type, real vegetation type, and groundwater vector layers. In determining the corre­lationbetweenthem,itwasestablishedthatrocks,potentialvegetation,andslopewerethemostimportant factors, which is why these three layers were also taken into account in identifying the landscape types. Becausethe rocks were combined into five classes, the vegetation into four classes, and the slopes into six classes, 120 different combinations of these three natural elements were theoretically possible. In real-ity,halfofthem(i.e.,sixty)wereestablished.Themicrotypificationonlytookintoaccountthecombinations covering over 1,000 hectares, and combinations covering a smaller area were added to their most similar types. Hence, the following nineteen combinations (Figures 46 and 47) or landscape types (rock + vege­tation + slope in degrees) remained: • silikatni prod + beli gaber in dob + 0 do 2o ‘silicate gravel + hornbeam and pedunculate oak + 0 to 2°’ (208.8km˛, 20.2% of the terrain); • silikatniprod+brestindob+0do2o ‘silicategravel+elmandpedunculateoak+0to2°’(19.6km˛,1.9%); • silikatni prod + beli gaber + 0 do 2o ‘silicate gravel + hornbeam + 0 to 2°’(21.7km˛, 2.1%); • silikatni prod + bukev, kostanj in hrasti + 0 do 2o ‘silicate gravel + beech, chestnut, and various oaks + 0 to 2°’ (44.4km˛, 4.3%); • silikatni prod + bukev, kostanj in hrasti + 2 do 6o ‘silicate gravel + beech, chestnut, and various oaks + 2 to 6°’(31.0km˛, 3.0%); • silikatni prod + bukev, kostanj in hrasti + 6 do 12o ‘silicate gravel + beech, chestnut, and various oaks + 6 to 12°’ (14.5km˛, 1.4%); • pesek + bukev, kostanj in hrasti + 0 do 2o ‘sand + beech, chestnut, and various oaks + 0 to 2°’ (15.5km˛, 1.5%); • pesek + bukev, kostanj in hrasti + 2 do 6o ‘sand + beech, chestnut, and various oaks + 2 to 6°’ (45.5km˛, 4.4%); • pesek+bukev,kostanjinhrasti+6do12o ‘sand+beech,chestnut,andvariousoaks+6to12°’(71.3km˛, 6.9%); • pesek + bukev, kostanj in hrasti + 12.0 do 19.9o ‘sand + beech, chestnut, and various oaks + 12 to 19.9°’ (40.3km˛, 3.9%); • glinainmelj+beligaberindob+0do2o ‘clayandsilt+hornbeamandpedunculateoak+0to2°’(95.1km˛, 9.2%); • glina in melj + bukev, kostanj in hrasti + 0 do 2o ‘clay and silt + beech, chestnut, and various oaks + 0 to 2°’ (51.7km˛, 5.0%); • glina in melj + bukev, kostanj in hrasti + 2 do 6o ‘clay and silt + beech, chestnut, and various oaks + 2 to 6°’ (52.7km˛, 5.1%); • glina in melj + bukev, kostanj in hrasti + 6 do 12o ‘clay and silt + beech, chestnut, and various oaks + 6 to 12°’ (50.6km˛, 4.9%); • glina in melj + bukev, kostanj in hrasti + 12 do 20o ‘clay and silt + beech, chestnut, and various oaks + 12 to 20°’ (33.1km˛, 3.2%); 1. Silicate gravel + hornbeam and pedunculate oak + 0 to 2o; 2. Silicate gravel + elm and pedunculate oak + 0 to 2o; 3. Silicate gravel + hornbeam + 0 to 2o; 4. Silicate gravel + beech, chestnut, and various oaks + 0 to 2o; 5. Silicate gravel + beech, chestnut, and various oaks + 2 to 6o; 6. Silicate gravel + beech, chestnut, and various oaks + 6 to 12o; 7. Sand + beech, chestnut, and various oaks + 0 to 2o; 8. Sand + beech, chestnut, and various oaks + 2 to 6o; 9. Sand + beech, chestnut, and various oaks + 6 to 12o; 10. Sand + beech, chestnut, and various oaks + 12 to 20o; 11. Clay and silt + hornbeam and pedunculate oak + 0 to 2o; 12. Clay and silt + beech, chestnut, and various oaks + 0 to 2o; 13. Clay and silt + beech, chestnut, and various oaks + 2 to 6o; 14. Clay and silt + beech, chestnut, and various oaks + 6 to 12o; 15. Clay and silt + beech, chestnut, and various oaks + 12 to 20o; 16. Marl + beech, chestnut, and various oaks + 2 to 6o; 17. Marl + beech, chestnut, and various oaks + 6 to 12o; 18. Marl + beech, chestnut, and various oaks + 12 to 20o; and 19. Sandstone + beech, chestnut, and various oaks + 20 to 30o. Figure 47: Updated map of landscape types in the Lower Drava Valley and Prlekija region. p • lapor + bukev, kostanj in hrasti + 2 do 6o ‘marl + beech, chestnut, and various oaks + 2 to 6°’ (39.3km˛, 3.8%); • lapor + bukev, kostanj in hrasti + 6 do 12o ‘marl + beech, chestnut and various oaks + 6 to 12°’ (75.4km˛, 7.3%); • lapor+bukev,kostanjinhrasti+12do20o ‘marl+beech,chestnutandvariousoaks+12to20°’(74.4km˛, 7.2%); and • pešcenjak+bukev,kostanjinhrasti+20do30o ‘sandstone+beech,chestnutandvariousoaks+20to30°’ (48.6km˛, 4.7%). 4.5 City Municipality of Ljubljana The City of Ljubljana covers 274.9km˛, or 1.4% of Slovenia. It lies along the Sava River and its tributary, the Ljubljanica, where the hills nearly cut the bottom of the Ljubljana Basin into two parts: the Sava Plain to the north and the Ljubljana Marsh to the south. This makes the basin’s landscape relatively diverse. A landscape microtypology of this area was produced, together with its microregionalization, in 2000 for the 18th Conference of Slovenian Geographers, which took place in Ljubljana that year and covered the municipality from most geographical angles. A detailed description of the method and data sources used is provided in an article published the same year (Hrvatin and Perko 2000). The geographic information system was based on a 100-meter 29.3×21.7km digital elevation model. This time around, the typification used the rock type and potential vegetation type vector layers and, for the first time, a relief coefficient raster layer. The relief coefficient or the surface roughness coefficient (Perko 2000) is a more complex indicator compared to elevation and slope because it combines both. Conceptually it is based on spatial changes in surface elevation and slope, and methodologically it relies on their coefficient of variation. It is the geo­metric mean of the elevation coefficient (i.e., the ratio between the standard deviation in the elevations of a square cell and its eight neighbors, multiplied by 100) and the slope coefficient (i.e., the ratio between thestandarddeviationintheslopesofasquarecellanditseightneighbors,multipliedby100;Perko2000). In calculating the relief coefficient for all of Slovenia, it was determined that the results yield a better approximationofSlovenia’sactualsurfaceroughnessiftheaverageofallcells(theentirearea)isusedinstead ofthelocalaverageofninecells.Thereliefcoefficientcalculatedusingthelocalninecellswaslaterrenamed thelocalreliefcoefficient,andtheonecalculatedusingallcellswasrenamedtheregionalreliefcoefficient. Atthesametime,incalculatingthecoefficient,theslopewasreplacedbysurfaceaspect,sothattheresult­ing coefficient combined the vertical and horizontal spatial variability (i.e., elevations and aspects; Perko 2007b). Usingthereliefcoefficientmakesitpossibletocomputehomogenousareasofthesameorsimilarsur­ face roughness or morphological units or morphological surface types. The lowest value of the regional relief coefficient for all of Slovenia was 0, the highest was 111.5, and the average was 9.3 (Perko 2000)). AfterreviewingthefrequencydistributionofregionalreliefcoefficientvaluesinthemosttypicalSlovenian plains, low hills, hills, and mountains, the coefficients were logically combined into four basic morpho­logical classes (Perko 2000): • Flat surface or plain with values ranging between 0 and 1 (223,843, or 11.0% of cells); • Slightly rough surface or low hills with values ranging between 1 and 10 (974,279, or 48.1% of cells); • Very rough surface or hills with values ranging between 10 and 20 (701,095, or 34.6% of cells); and • Extremely rough surface or mountains with values over 20 (127,981, or 6.3% of cells). In all cases, the values at the lower limit of a class are included in that class and the values at the upper limit of a class are included in the next higher class. Because there are only a few cells in the City Municipality of Ljubljana with a relief coefficient over 20, the third and fourth classes were combined and thus only three classes remained: • Flat surface (39.3% of cells); • Slightly rough surface (26.9% of cells); and • Very rough surface (33.8% of cells). Before the typification, rocks were combined into four classes: • Late Paleozoic non-carbonate rocks (Carboniferous and Permian shale, quartz sandstone, and quartz conglomerate) are the oldest and also most widespread (33.6% of the area); • Mesozoic carbonate rocks (limestone and dolomite), in some places including layers of marl, shale, silt­stone, sandstone, tuff, and tuffite (10.3%); • Pleistocenesandgravel,insomeplacesformingaconglomerate,andHolocenegravelandsand(37.2%); and • Quaternary clay, silt, sand, and peat (18.9%). Potentialvegetation,whichcouldgrowundertoday’secologicalconditions(climate,bedrock,soil,and so on) without human and animal intervention, were also combined into four classes: • Communities of beech forests cover over a third of the municipality and mostly grow in the hills; beech andhard-fernforests(Blechno-Fagetum)primarilygrowonsilicaterockandpre-Dinaricbeech;andbroad-leaved sanicle forests grow on carbonate rock (Hacquetio-Fagetum; 36.8%); • Hornbeam and sessile oak forests (Querco-Carpinetum) grow on Holocene gravel plains and the Würm gravel terrace of the Ljubljana Plain (31.2%); • Hornbeamandpedunculateoakforests(Quercoroboris-Carpinetum)primarilygrowinareasinfluenced by a high groundwater level, especially in the Ljubljana Marsh (20.8%); and • Scotspineandblueberryforests(Vacciniomyrtilli-Pinetum)growinshallow,nutrient-deficientsoil;Scots pine is often mixed with spruce (11.2%). Based on the spatial overlapping of these three natural landscape elements, the landscape types were then identified. Values were determined for every cell or hectare of the City Municipality of Ljubljana in terms of the surfaceroughness,rock,andvegetationclasses,andtheircombinationsweredefined.Becausesurfacerough­ness included three classes, and rocks and vegetation included four, forty-eight combinations were theoreticallypossible.Thirty-ninewereactuallyidentified.Onlysevencombinationsappearedoverathou-sand times, together covering 76.7% or over three-quarters of the municipality. However, not only the absolute frequency of a specific combination is important in typification, but alsoitsrelativefrequency.Therefore,thetheoreticalprobabilityofoccurrenceortheoreticalfrequencywas calculatedforeachcombinationbymultiplyingitspartialprobabilities,alongwithitsactualfrequencyand the ratio between the actual and theoretical frequency. For example, the combination of flat surface, grav­el, and hornbeam and sessile oak forests was typical of 5,254 cells. Its actual frequency was 0.19 (the ratio between5,254cellswiththiscombinationandthenumberofallcells,27,489)anditstheoreticalfrequency was 0.046, which was obtained by multiplying the theoretical probabilities of the occurrence of flat sur­face (the ratio between 10,787 cells and the number of all cells), gravel (the ratio between the 10,216 cells withgravelandthenumberofallcells),andhornbeamandsessileoakforests(theratiobetweenthe8,583 cellswiththeseforestsandthenumberofallcells).Theratiobetweentheactualandtheoreticalfrequency was 4.20, which meant that the combination described was over four times more frequent than theoreti­cally expected. Allcombinationswithratiosbetweentheactualandtheoreticalfrequencyhigherthan0.5weredefined as important to the area. Fifteen such combinations were established, covering as much as 91.6% of the area. The remaining combinations, which were partly also the result of inaccurate digitization of vector layers, were logically added to these fifteen combinations. Thefifteencombinationsselectedwerereferredtoasnaturallandscapetypes(Figure48 and49).They included: • pokrajina z nerazgibanim površjem, glino ter gozdom belega gabra in doba ‘landscape with aflat surface, clay, and hornbeam and pedunculate oak forests’ (4,255.3km˛, 15.5% of the area); • pokrajinazmocnorazgibanimpovršjem,karbonatnimikamninamitergozdombukve‘landscapewithavery rough surface, carbonate rocks, and beech forests’ (1,990.2km˛, 7.2%); • pokrajina z mocno razgibanim površjem, nekarbonatnimi kamninami ter gozdom bukve ‘landscape with a very rough surface, non-carbonate rocks, and beech forests’ (5,090.9km˛, 18.5%); • pokrajinaznerazgibanimpovršjem,prodomtergozdombelegagabraingradna‘landscapewithaflatsur-face, gravel, and hornbeam and sessile oak forests’ (5,442.8km˛, 19.8%); • pokrajina z rahlo razgibanim površjem, nekarbonatnimi kamninami ter gozdom rdecega bora ‘landscape with a slightly rough surface, non-carbonate rocks, and Scots pine forests’ (1,314.0km˛, 4.8%); • pokrajinazmocnorazgibanimpovršjem,nekarbonatnimikamninamitergozdomrdecegabora‘landscape with a very rough surface, non-carbonate rocks, and Scots pine forests’ (1,586.1km˛, 5.8%); • pokrajina z rahlo razgibanim površjem, karbonatnimi kamninami ter gozdom bukve ‘landscape with a slightly rough surface, carbonate rocks, and beech forests’ (791.7km˛, 2.9%); • pokrajina z rahlo razgibanim površjem, prodom ter gozdom belega gabra in gradna ‘landscape with a slightly rough surface, gravel, and hornbeam and sessile oak forests’ (2,504.2km˛, 9.1%); 1. Landscapewithaflatsurface,clay,andhornbeamandpedunculateoak 8. Landscape with a slightly rough surface, gravel, and hornbeam and forests; sessile oak forests; 2.Landscapewithaveryroughsurface,carbonaterocks,andbeechforests; 9. Landscape with a slightly rough surface,non-carbonate rocks, and 3. Landscapewithaveryroughsurface,non-carbonaterocks,andbeech beech forests; forests; 10. Landscape with a slightly rough surface, clay, and red pine forests; 4. Landscape with a flat surface, gravel, and hornbeam and sessile oak 11. Landscape with a slightly rough surface, gravel, and red pine forests; forests; 12. Landscape with a slightly rough surface, clay, and hornbeam and 5. Landscapewithaslightlyroughsurface,non-carbonaterocks,andred pedunculate oak forests; pine forests; 13. Landscape with a flat surface, clay, and red pine forests; 6. Landscapewithaveryroughsurface,non-carbonaterocks,andredpine 14. Landscapewithaflatsurface,gravel,andhornbeamandpedunculate forests; oak forests; and 7. Landscape with a slightly rough surface, carbonate rocks, and beech 15. Landscape with a slightly rough surface, gravel, and beech forests. forests; Figure 49: Updated map of landscape types in the City Municipality of Ljubljana. p • pokrajina z rahlo razgibanim površjem, nekarbonatnimi kamninami ter gozdom bukve ‘landscape with a slightly rough surface, non-carbonate rocks, and beech forests’ (1,347.0km˛, 4.9%); • pokrajina z rahlo razgibanim površjem, glino ter gozdom rdecega bora ‘landscape with a slightly rough surface, clay, and Scots pine forests’ (195.2km˛, 0.7%); • pokrajinazrahlorazgibanimpovršjem,prodomtergozdomrdecegabora‘landscapewithaslightlyrough surface, gravel, and Scots pine forests’ (505.8km˛, 1.8%); • pokrajina z rahlo razgibanim površjem, glino ter gozdom belega gabra in doba ‘landscape with a slightly rough surface, clay, and hornbeam and pedunculate oak forests’ (357.4km˛, 1.3%); • pokrajinaznerazgibanimpovršjem,glinotergozdomrdecegabora‘landscapewithaflatsurface,clay,and Scots pine forests’ (206.2km˛, 0.8%); • pokrajina z nerazgibanim površjem, prodom ter gozdom belega gabra in doba ‘landscape with a flat sur­face, gravel, and hornbeam and pedunculate oak forests’ (887.9km˛, 3.2%); and • pokrajina z rahlo razgibanim površjem, prodom ter gozdom bukve ‘landscape with a slightly rough sur­face, gravel, and beech forests’ (1,019.8km˛, 3.7%). 4.6 Municipality of Idrija The Municipality of Idrija covers 293.7km˛, or 1.4% of Slovenia. It lies on very rough terrain at the inter­section of the Alps and the Dinarides, approximately 50km west of Ljubljana. The town of Idrija is well known for its now abandoned mercury mine. A landscape microtypology of the area, together with its microregionalization, was produced in 2010 as part of the geographical expert bases for the municipality’s development strategy. A detailed descrip­tion of the method and data sources used is provided in an article published the same year (Perko and Hrvatin 2010). For the first time, the geographic information system was based on a 25-meter 14.0×14.4km digital elevation model. Available were the raster layers for elevation, slope, surface aspect, and insolation, and vector layers for rock types, soil types, potential vegetation types, and land use types. However, the val­ues for all layers changed so quickly from one location to another that the researchers ended up with an uncontrollable number of combinations, of which most covered only a negligible area. Therefore, the microtypification was ultimately carried out by only taking into account the relief indicators (i.e., surface slope and elevation, relative elevation difference, and morphological surface type) or terrain, which con­tributes the most to the external appearance of this area. Seventeen microregions (Figure 50) and five landscape types (Figure 51) were identified: • kotlinica‘smallbasin’(2.0km˛,0.7%,overtwo-thirdsoftheareawithelevationsbetween300and400m, and slopes ranging between 0 and 20°; microregions: the Idrija Basin, the Spodnja Idrija Basin); • dolina ‘valley’ (159.5km˛, 54.3%, over two-thirds of the area with elevations between 400 and 800m, and slopes between 12 and 45°; microregions: the Kanomlja Valley, the Sovra Valley, the Lower Idrija Valley, the Zala Valley, the Upper Idrija Valley); • podolje‘lowland’(9.2km˛,3.1%,overtwo-thirdsoftheareawithelevationsbetween500and700m,and slopes between 0 and 12°; microregion: the Godovic Lowland); • srednje visoka planota ‘medium-high plateau’ (41.8km˛, 14.2%, over two-thirds of the area with eleva­tions between 600 and 800m, and slopes between 2 and 12°; microregions: the Crni Vrh Plateau, the Ravne Plateau, the Vrsnik Plateau, the Zavratec Plateau); and • visoka planota ‘high plateau’ (81.2km˛, 27.7%, over two-thirds of the area with elevations between 700 and1,000m, and slopes between 6 and 30°; microregions: the Dole Plateau, the Trnovo Forest Plateau, the Krnice Plateau, the Ledine Plateau, the Vojsko Plateau). Figure 50: Map of microregions in the Municipality of Idrija (Perko and Hrvatin 2010). p Figure 51: Map of landscape types in the Municipality of Idrija (Perko and Hrvatin 2010). p p. 70 70 4.7 Municipality of Kocevje TheMunicipalityofKocevjecovers555.6km˛,or2.7%ofSlovenia.Itliesabout50kmsoutheastofLjubljana. The area is dominated by Dinaric lowlands and plateaus (Perko 1998a) characterized by a karst surface, a sparse network of surface streams, and a high percentage of forest cover. Themicrotypology oftheMunicipality ofKocevje was oneof the four microtypologiesselectedas part ofa2013–2015researchprojectthatfocusedondevelopingamethodforidentifyingnaturallandscapetypes atthelocallevelandmappingthesetypesforselectedSlovenianmunicipalities.Amongotherthings,thepur­poseofthatresearchwastochecktheapplicabilityofgeographicinformationtoolsanddigitaldataindesigning landscapetypologiesatthelocallevel.Placedattheforefrontwasthusanattempttodevelopasuitablemethod-ologythatcouldbeusedasamodelforproducinglandscapetypologiesatthelocallevel.Themicrotypification wasprimarilybasedonusingsegmentationandsupervisedandunsupervisedclassification.Adetaileddescrip­tion of the methodand data sources used isprovided in a reportpublished in 2015 (Ciglic2015). Based on the findings of research conducted as part of this project and previous research (e.g., Ciglic 2014), the following general steps were defined for producing a natural landscape typology: • Collecting and preparing data layers; • Assessing data layers (their interconnection and significance at various spatial scales by calculating the average moderate coefficient of variation; for more details, see Ciglic and Perko 2017); • Converting raster data into vector data using segmentation (Acharya and Ray 2005; Pratt 2007; Lotufo et al. 2008; Eastman 2012) because the basic unit used from there on was segments or polygons; • Assigning values to polygons for the cells they covered (e.g., the minimum, maximum, and average val­ues in terms of numerical data and the most frequent category in terms of nominal data); • Classifyingpolygonsusingvariousunsupervisedclassificationmodelsandestimatingthenumberoftypes; • Reviewing the unsupervised classification data and results, and conducting a field inspection; • Selecting test polygons and validation polygons of envisaged landscape types to determine the super­vised classification model; and • Selecting the most suitable model, classifying the entire area, and reviewing andmanually revising any discrepancies with the real landscape. Thedatalayerswereselectedbasedontheirapplicabilityforanindividualmunicipality.Theyincluded: • A 12.5m and 25m digital elevation model; • Average monthly and annual precipitation and temperature; • Watercourses; • Soil types; • Solar radiation (Gabrovec 1996); • Solar radiation (Zakšek 2005); • Cave sites; • Landslide sites; • Landsat 8 images; and • Rock types. Based on these data layers, several other layers were also prepared. The final dataset included the fol-lowing:elevation,slope,elevationvariation,slopevariation,totalslopecurvature,verticalslopecurvature, horizontal slope curvature, depression sites, density of dolines according to the 12.5m and 25m DEM, annual precipitation, Mediterranean character index, average annual temperature, annual temperature difference, April and October temperature difference, rock types, permeability, annual solar radiation (Gabrovec 1996, Zakšek 2005), network of major perennial and non-perennial streams, stream density, soil types, cave sites, cave density, and normalized difference vegetation index (NDVI). Four test areas were selected: one in the Slovenian alpine region, one in thePannonian region, one in theDinaricregion,andoneintheMediterraneanregion.TheMunicipalityofKocevjewastheDinaricarea selected. Several models were produced for each area, and the most suitable one was then selected. First, the average moderate coefficient of variation and the Pearson correlation coefficients were ana­lyzed to evaluate the data layers, after which several attempts of segmentation at a 12.5m resolution were made. Ultimately, a segmentation with the following settings was selected: Figure 52: Map of modeled landscape types in the Municipality of Kocevje. p p. 72 72 • Data layers (with weights): stream density (weight: 1/3), slope (weight: 1/3), NDVI (weight: 1/3); • Window width: 9; • Weight mean factor: 0.6; • Weight variance factor: 0.4; and • Similarity tolerances: 0, 1, 2, 3, 4, 5. After segmentation, several unsupervised polygon classifications were conducted. Based on the qual­ityofdata,theircorrelation,andtheanalysisoftheaveragemoderatecoefficientofvariation,forthepurposes of unsupervised hierarchical classification, which shows changes in the homogeneity of types according to their number, the minimum and mean polygon values were selected for stream density, and the mean polygon value was selected for the density of dolines (12.5m DMV), slope, annual solar radiation, pre­cipitation index, permeability, and elevation. Ward’s method with squared Euclidean distance was used for the classification. Aspartofunsupervisedclassificationexperiments,differentsetsofinputdatalayersweretested,pri­marily to establish how the data characteristics change by level. Up to the level of four or five types, the decrease in the average moderate coefficients of variation of all layers was 0.2, and then up to the level of fifteen types it was smaller than 0.2. A review of decrease in the average moderate coefficient of vari­ation for individual layers showed that most of them recorded smaller decreases at least at the level of seven to nine types. Therefore, at least four and a maximum of nine types had to be found for the typ­ification. Theresultsoftheanalysisofhierarchicalclassificationsandthearea’sgeographicalcharacteristics(rocks, terrain and insolation, climate, and settlement) were then taken into account in the supervised classifi­cation. Ultimately, the following seven landscape types were identified: • kraško polje ‘karst polje’; • kraško hribovje ‘karst hills’; • kraški ravnik ‘karst plain’; • naplavna ravnica ‘Alluvial plain’; • nekraško hribovje ‘karst plain with sediments’; • nekraško hribovje ‘non-karst hills’; and • dolinsko pobocje ‘valley slope’. After that, 298 segments or polygons were selected as landscape type case samples, corresponding to 2% of all polygons. A decision tree with the Classification and Regression Tree (Lin, Noe, and He 2006) algorithmwasusedforthesupervisedclassification.Theinputdatalayersincludedrocktype,averagerock permeability, average elevation, average slope, average stream density, and average density of dolines. In conclusion, several minor manual corrections were made to the polygon classification (Figure 52). The typology produced is a test case of using various geographic information tools for landscape typ­ification. The entire procedure includes several steps that can be combined as follows: • Assessing data; • Converting raster data into vector data (polygons); • Checking unsupervised classifications; and • Performing a supervised classification. 4.8 City Municipality of Velenje TheCityMunicipalityofVelenjecovers8.5km˛,or0.4%ofSlovenia. Itliesonthesoutheasternedgeofthe Alps, where the mountains descend toward the Pannonian Basin, roughly 60km northeast of Ljubljana. This is a landscape that has been significantly transformed by years of mining, with several lakes having been formed due to subsidence. A microtypology of this area was produced, together with its microregionalization, in 2020 as part of the geographical recommendations for the municipality’s further development. Inthiscase,alandscapemicrotypologywasfirstproducedbasedonlandscapediversity,whichbecame very important in the twenty-firstcentury, and both Slovenia and the EU have been incorporating it into various strategies and similar documents (Ciglic and Perko 2013b). A detailed description of the method and data sources used is provided in an article published in 2020 (Perko and Hrvatin 2020). Municipalities can effectively use their above-average landscape diversity as a development advan­tageandthusincreasetheireconomiccompetitivenessandraisethequalityoflifeofindividualsandsocial groups. TheCityMunicipalityofVelenjeisoneofthem.Itischaracterizedbyhighlydiversenaturallandscape elements and includes five microregions: the Plešivica Hills, Paka Hills, Velenje Basin, Pirešica Lowland, andLožnicaLowHills.Asaformermining-industrymunicipality,itisbecomingincreasinglyinnovative, creative, and friendly to its residents and the environment, to which landscape diversity can contribute significantly, especially in terms of accelerating the development of tourism and recreational and sports activities, and improving the quality of life of both its residents and visitors. This will be facilitated even further by the planned improvement of its transport accessibility or transport connections. The geographic information system used was based on a 25-meter 13.4×14.6km digital elevation model. Landscape diversity was determinedbasedon surfaceroughness andvariation inrocks and veg-etation(Perko,Hrvatin,andCiglic2017).Thedatawascombinedintosevenrelief,fifteenrock,andfifteen vegetationtypes. Theratiobetweenthenumberofrelief,rock,andvegetationtypeswithina1kmradius and the numberof allrelief, rock, and vegetationtypes was calculated foreach cell, using a moving win­dow.Thelowestpossibleratiowas0.092ifonlyonerelief,rock,andvegetationtypeoccurredwithina1km radius((1/7+1/15+1/15)/3),andthehighestpossibleratiowas1.00ifallsevenrelieftypes,fifteenrock types, and fifteen vegetation types occurred within a 1km radius at the same time (Perko, Ciglic, and Hrvatin 2015). The average landscape diversity of the City Municipality of Velenje was 0.2186, which is 34.6% high­er than that of Slovenia as the European landscape hotspot or the EU country with the highest average landscape diversity, or 20.1% higher than that of the Slovenian part of the Alps. The average landscape diversity of Slovenia is 0.1624 and that of the Slovenian Alps microregion is 0.1820 (Perko, Hrvatin and Ciglic2017).Thelowestvalueinthemunicipalitywas0.0920recordedforthePlešivicaHills,andthehigh-est value was 0.3905 recorded for the Pirešica Lowland (Figure 54). The municipality is characterized by two diagonal belts of above-average landscape diversity extend­ing from the northwest to the southeast and from the northeast to the southwest, and intersecting east of Velenje, roughly in the center of the municipality. Thebeltsalsoincludetwolandscapehotspotswiththegreatestconcentrationofdifferentnaturalland­scape elements. The larger one is at the intersection of both belts east of Velenje and the smaller one is in theextremenortheasterncornerofthemunicipalityalongthePakaValley,wherethereisalsothecellwith thegreatestlandscapediversityinthemunicipality.ItislocatedattheintersectionofthePakaHills,Velenje Basin, and Pirešica Lowland. The most distinct areas of below-average landscape diversity, with the lowest concentration of differ­ent natural landscape elements, can be found on the southwestern and eastern edges of the municipality. ThePirešicaLowlandhasthegreatestlandscapediversityamongthemicroregions(Figure53):itsland-scape diversity is 27.9% greater than that of the municipality’s average and 47.4% greater than that of the LožnicaLow Hills, the microregion with the lowest landscape diversity in the municipality – which, how­ever, is still 16.8% higher than the Slovenian average. Five landscape diversity types (Figure 54) were identified: • obmocjezelonizkepokrajinskeraznolikosti‘averylowlandscapediversityarea’(coefficientrangingfrom 0.0000 to 0.1499); • obmocje nizke pokrajinske raznolikosti ‘a low landscape diversity area’ (coefficient ranging from 0.1500 to 0.1999); • obmocje srednje pokrajinske raznolikosti ‘a medium landscape diversity area’ (coefficient ranging from 0.2000 to 0.2499); • obmocje visoke pokrajinske raznolikosti ‘a high landscape diversity area’ (coefficient ranging from 0.2500 to 0.2999); and • obmocjezelovisokepokrajinskeraznolikosti‘averyhighlandscapediversityarea’(coefficientrangingfrom 0.3000 to 0.3999). Figure 53: Map of microregions in the City Municipality of Velenje (Perko and Hrvatin 2020). p Figure 54: Map of landscape types in the City Municipality of Velenje (Perko and Hrvatin 2020). p p. 76 4.9 Comparing the microtypifications The microtypification examples presented primarily differ in term of the location in Slovenia (Figure 37), the size of the area covered, the number and size (resolution) of square cells, the year produced (Table 3), and the methodology used, and they tend to differ less in terms of the data layers used or the landscape elements observed. In terms of the landscape typology of Slovenia produced in 1996 (Perko 1998a; Perko, Hrvatin, and Ciglic 2015), all the examples are relatively diverse, with not even a single area including only one land­scape type. Two landscape types are included in the Kokra Valley (Alpine mountains and Alpine plains), theVolcjiPotokArboretum(AlpineplainsandAlpinehills),theLowerDravaValley(Pannonianlowhills and Pannonian plains), the Municipality of Kocevje (Dinaric plateaus and Dinaric lowlands), and the Municipality of Velenje (Alpine mountains and Alpine hills), and three landscape types can be found in theEasternKrkaBasin(Pannonianlowhills,Pannonianplains,andDinaricplateaus),theCityMunicipality ofLjubljana(Alpineplains,Alpinehills,andDinariclowlands),andtheMunicipalityofIdrija(Alpinehills, Dinaric plateaus, and Dinaric lowlands). IntermsofSlovenia’sregionalizationproducedin1996,extendingovertworegionsaretheKokraValley (the Kamnik–Savinja Alps and the Sava Plain) and the Volcji Potok Arboretum (the Sava Plain and the Sava Hills), extending over three regions are the Municipality of Kocevje (the Ribnica–Kocevje Lowland; the Big Mountains, Stojna Plateau, and Gotenica Mountains; the Little Mountains, Kocevje Rog Plateau, andPoljaneMountains)andtheCityMunicipalityofVelenje(theEasternKarawanks;theVelenjeandKonjice Hills; the Ložnica and Hudinja Low Hills), and extending across four regions are the Eastern Krka Basin (the Sava Hills; the Krško, Senovo, and Bizeljsko Low Hills; the Krka Plain; and the Gorjanci Hills), the Lower Drava Valley (the Mura Plain, the Slovenian Hills, the Drava Plain, and the Haloze Hills), the CityMunicipality of Ljubljana (the Sava Plain; the Cerkno, Škofja Loka, Polhov Gradec, and Rovte Hills; theSava Hills; and the Ljubljana Marsh), and the Municipality of Idrija (the Idrija Hills; the Cerkno, Škofja Loka, Polhov Gradec, and Rovte Hills; the Trnovo Forest Plateau, Mount Nanos, and the Hrušica Plateau; and the Inner Carniolan Lowland). Theaveragesizeofaquadrantis848.51km˛andoftheareatypifiedinsidethisquadrantis396.24km˛. The smallest example is the Volcji Potok Arboretum (quadrant: 2.25km˛, area typified: 0.83km˛) and the largestoneistheLowerDravaValley(quadrant:2,478.00km˛,areatypified:1,034.45km˛).Theratiobetween thesmallest,average,andlargestquadrantsis1:377:1,101,andtheratiobetweenthesmallest,average,and largest areas typified is 1:477:1,246. The Eastern Krka Basin has the smallest number of cells (i.e., 1,763 squares) and the Municipality of Kocevje has the largest number (i.e., 6,718,400 squares); the ratio between them is 1:3,811. The Eastern Krka Basin has the lowest cell resolution (i.e., 1,000m) and the Volcji Potok Arboretum has the highest (i.e., 5m), the ratio being 200:1. A comparison of the typification methods used in the examples presented shows gradual changes or developmentofmethodsoverthepastfewdecades.InthecaseoftheKokraValleyin1985,firstmapswere preparedforallthelandscapeelementsobserved,afterwhichtheywereoverlappedtocreatefivenewmaps showing the types of relief, rocks, climate, soil, and vegetation. By overlappingthe synthesized maps, ulti­mately a final map of the Kokra Valley landscape typology with eight types was produced. All areas on themapswerefirstmeasuredwithasquaregridontransparentpaperplacedoverthemaps,throughwhich asimplegeographicinformationsystemwascreated.Thesizeoftheareaswaslatercheckedwithaplanime­ter. All statistical and other calculations were already performed on a computer (Perko 1987). In the case of the Eastern Krka Basin (Perko 1989) four years later, maps were also produced for all the landscape elements. By overlapping these maps, synthetic maps were created and, based on this, the finallandscapetypologymapwithninetypeswasproduced.However,thistimethesquaregridwithwhich the maps were covered was used not only to measure the areas, but also to read data from the maps, enter them in the computer, and develop a proper geographic information system, albeit with a low resolution. InthecaseoftheVolcjiPotokArboretum(Perko1993),alldatawerealreadyenteredinthegeographic information system through digitization, including the digital elevation model, which was produced by digitizing the contour lines from a 1:5,000 map. Its five-meter resolution was a great achievement back in 1992. A special feature of this typology was that, due to the small size of the area covered, the data and the final landscape typology with five types were able to be updated and checked on the ground. Withthethreetypologiesmentionedabove,thetypeswerespatiallycompleteunits(i.e.,irregularpoly­gons), whereas the landscape typology of the Lower Drava Valley was the first one that was produced at the level of a square cell. In determining the spatial and statistical correlation between individualdata lay-ers,itwasestablishedthatrocks,slope,andvegetationwerethemostimportantlandscapeelements.Therefore, thelandscapetypesweredefinedbasedonthecombinationsoftypesofrockandvegetationandslopeclass­es within an individual cell. The nineteen most frequent combinations were selected as landscape types, and the combinations that were rarer were added to their most similar types (Fridl et al. 1996). Fouryearslater,thelandscapetypologyfortheCityMunicipalityofLjubljana(HrvatinandPerko2000) wasproducedinasimilarway,onlythatsurfaceroughnessunitswereusedinsteadofsurfaceslopes.These unitsweredefinedbasedonthereliefcoefficient(Perko2000,2007),whichsimultaneouslytakesintoaccount thespatialchangesinslopesandelevations.Thefinallandscapetypologyproducedincludedfifteentypes. The previous two landscape typologies relied on a 100-meter digital elevation model, but in 2010 the landscape types in the Municipality of Idrija were identified based on a 25-meter digital elevation model. Examiningthecorrelationbetweenthelayersrevealedthatpracticallyalllandscapedifferenceswereasso­ciatedwithsurfaceroughness,andthereforeamicrotypologywithfivetypeswasproducedusingonlyslope, elevation, difference in elevation, and morphological surface type (Perko and Hrvatin 2000). AmethodologicalinnovationinthecaseofthemicrotypificationoftheMunicipalityofKocevje(Ciglic 2015) was modeling with segmentation and supervised and unsupervised classification. In the end, seven landscapetypeswereidentified.A12.5-meterdigitalelevationmodelandsomeartificialintelligencealgo­rithms were used for the first time (Ciglic et al. 2019; Perko, Ciglic, and Hrvatin 2019). Thelastexamplepresented,the2020landscapemicrotypologyoftheCityMunicipalityofVelenje,land­scape diversity based on the concentration of different landscape elements was used to define the types for the first time(Perko and Hrvatin 2020). This time, a 25-meter digital elevation model was used again. Landscape diversity was defined based on surface roughness and variation in rocks and vegetation. The final microtypology included five types. A comparison of all eight microtypifications presented also shows the development of the basic typ­ification term landscape type. With the Kokra Valley , priority was given tothe term eco-complexand with the Eastern Krka Basin both terms were used fairly equally. With the Volcji Potok Arboretum, the term geologicalunitwasalsousedwiththesamemeaning,whereasinlatermicrotypologiesonlythetermland­scape type was used. Table 3: Comparing the microtypologies of parts of Slovenia. Name Quadrant size Quadrant size Quadrant size Resolution in Size of typified No. of landscape Year produced in km˛ in km˛ in squares meters landscape in km˛ types (current digital measurement) Kokra Valley 21.0 × 22.0 462.00 420 × 440 50 221.91 8 1987 Eastern Krka Basin 41.0 × 43.0 1,763.00 41 × 43 1,000 704.54 9 1989 Volcji Potok 1.5 × 1.5 2.25 300 × 300 5 0.83 5 1992 Arboretum Lower Drava 59.0 × 42.0 2,478.00 590 × 420 100 1,034.45 19 1996 Valley City Municipality 29.3 × 21.7 635.81 293 × 217 100 275.03 15 2000 of Ljubljana Municipality 14.0 × 14.4 201.60 560 × 576 25 293.47 5 2010 of Idrija Municipality 32.3 × 32.5 1,049.75 2,584 × 2,600 12.5 555.64 7 2015 of Kocevje City Municipality 13.4 × 14.6 195.64 536 × 584 25 84.08 5 2020 of Velenje 5 The future of landscape typologies of Slovenia To our knowledge, this is the first article that provides a systematic overview of all landscape typologies at various levels for a specific country in the world. ItpresentsallfivelandscapemacrotypologiesofSloveniaproducedbySlovenianauthors(Melik1946;Ilešic 1958; Perko 1998a; Špes et al. 2002; Perko, Hrvatin, and Ciglic 2015) and compares them with the landscape macrotypologies of Europe produced by authors outside Slovenia (Meeus 1995; European EnvironmentAgency1995;Olsonetal.2001;Bohnetal.2003;Mücheretal.2003,2006,2010;Rivas-Marínez, Penas, and Díaz 2004a; 2004b; Jongman et al. 2006; Biogeographical regions 2016; DMEER 2003) and Slovenia’s position within them. In addition, it presents several examples of microtypologies of individ­ualpartsofSlovenia(Perko1986,1989,1993;Fridletal.1996;HrvatinandPerko2000;PerkoandHrvatin 2010, 2020; Ciglic 2015), some of which were created under the influence of macrotypologies or repre­senttheirfurtherdevelopmentatthemicrolevel.Forexample,themicrotypologyandmicroregionalization of the Municipality of Idrija (Perko and Hrvatin 2010) are a development of Perko’s landscape macroty­pology (Perko 1998a). In the article, all Slovenian landscape macrotypologies and microtypologies are presentedwiththeoriginalmapandanupdatedoneatthesamescale,whichfacilitatescomparisonbetween them and especially between the landscape types. In general, the comparison between landscape macroty­pologiesprimarilyshowsdevelopmentinthemethodologyofclassifyinglandscapes,movingfromadistinctly subjective method using manually drawn borders between the landscape types as seen in the macroty­pologyproducedbyMelik(1946)andIlešic(1958)toamoreautomaticone,whichisnonethelessstilltested on the ground, as seen in Perko, Hrvatin, and Ciglic (2015). Melik’s (1946) and Ilešic’s (1958) landscape macrotypologies are the work of a single author,Perko (1998a) was assisted by experts in his detailed def-initionofborders,andthetwomostrecentlandscapetypologies(Špesetal.2002;Perko,Hrvatin,andCiglic 2015) are the work of a group of researchers. What stands out in the microtypologies is the great diversi­ty of the methodologies used to achieve the same result. The terminology developed from various expressions for spatial units in the oldest macrotypologies and microtypologies to the predominant term pokrajinski tip ‘landscape type’ in the ones produced after 2000.Thedevelopmentofconcretenamesoflandscapetypesproceededfrommixingthenamesofregions and landscape types in older typologies to distinguishing more consistently between the names of regions as geographical regionalization units (e.g., regija Ljubljanskakotlina‘Ljubljana Basin region’) and types as landscape typology units (e.g., kotlina ‘basin’). Distinguishing between proper nouns referring to regions and common nouns referring to types is the more complex because, except for Perko, Hrvatin, and Ciglic (2015), all authors link Slovenia’s macrotypologies with its regionalizations, especially at the microlevel, which means that landscape types are common as units at the macrolevel and mesolevel, and regions are common at the microlevel. However,whatstandsoutthemostishowgeographicinformationsystemsandrelateddigitalthematic cartography have almost completely supplanted other landscape typification methods at all levels, from the macrolevel to the microlevel. However, this is also typical of other areas of geography. 5.1 The importance of landscape typologies Research directly and indirectly influences people in various ways, which is why it will continue to occupy animportantplaceinthefuture.Becauselandscapeclassifications,includinglandscapetypologies,canaffect thelifeofeveryindividual,theynotonlyposearesearchchallenge,butalsohaveapracticalorappliedvalue. Defining landscape types de facto means defining homogenous areas, and familiarity with homoge-nous spatial units is important in a wide range of areas. For example, in natural disasters areas of the same landscape type, even though they may be far apart, are similarly threatened and they demand similar pro-tectionagainstnaturaldisastersandsimilarmeasures(Perko1992a;Fridletal.1996).Areaswiththesame landscape type allow for similar spatial and regional planning (Perko 2009), similar economic use (e.g., in agriculture and tourism; Ciglic and Perko 2013b), similar protection of natural and cultural heritage (Ticar, Perko, and Volk Bahun 2018; Perko and Ticar 2020), and similar military defense, and some land-scapetypesareevenassociatedwithgeographicalnames(Geršic,Ciglic,andPerko2018).Thedevelopment of the entiresociety can also be more successfully directed based on a good knowledge of landscape types (Nared et al. 2015). In this century, Europe has been promoting awareness of the importance of landscape diversity (Van Eetvelde 2009; Dempsey and Wilbrand 2017), and in countries where landscape diversity is an important feature(e.g.,inSlovenia)researchonlandscapetypesalsostrengthensnationalidentity(GeršicandPerko 2020). Landscape typologies and geographical regionalizations are often part of geography textbooks at all levels of education, which means that practically everyone gets to know them (Urbanc et al. 2016). Themethodologyandapproachesdevelopedwhileproducinglandscapetypologiescanalsobeapplied toothergeographicalareasandsimilardisciplines(e.g.,thematiccartography).Inaddition,landscapetypes canalsobestudiedinlessdirectlyrelatedresearchareas,suchasmachinelearningorartificialintelligence (Ciglic et al. 2019). Thefollowingdescriptionorproposedapproachtoimprovingexistinglandscapetypologies,usingthe example of the 1996 landscape macrotypology of Slovenia (Perko 1998a), illustrates how important land-scapetypologycanbe,howitmayconcretelyimpactthelifeofindividuals,orhowanimprovedlandscape typology may contribute to a fairer society. 5.2 Improving the landscape typology of Slovenia for a fairer society Many researchers see the future of classifying landscapes in evaluation and correction of existing land-scapeclassificationswithmodernquantitativemethods(Congalton1991;CiglicandPerko2015).Therefore, the authors of this article suggest several methodological steps for improving a landscape typology, illus­trating them with the example of Slovenia. A landscape typology is a part of Slovenia’s legislation defining financial obligations, and therefore it is important for a large part of the populace. Thus, for example, the location of agricultural land in terms of landscape typology can decisively influence the amount of taxes for a particular farm. This means that a defective landscape typology can lead to unfair distribution of taxes, and thus unfairness or even dis­crimination between people. Since 2008 in Slovenia, an important element for determining the quality of agricultural land and its taxation has been the location of individual parcels in relation to the landscape type, within which rela­tivelysimilarnaturalconditionsexist.Slovenianlegislationfrom2008takesintoaccountPerko’slandscape macrotypologyfrom1996(Perko1998a),whichdividesSloveniaintoninelandscapetypesandfourland­scape type groups. Perko produced this landscape macrotypology at a scale of 1:250,000 in 1996, and in 2008 he also prepared it as a digital data layer for the relevant ministry. When it was established a quarter of a century ago, the accuracy and diversity of landscape data were much lower. The landscape macroty­pologywas produced for publication in the volume Geographical Atlas of Slovenia (Fridl et al. 1998a), but not as a basis for taxes. Due to the availability of statistical data, the authors drew up the borders between the landscape types in such a way that they did not divide the nuclei of settlements, and so in some cases thelinesdonotfullyfollownaturalconditions(Perko1998a;PerkoandCiglic2020a).Becauseitwasdesigned on a relatively rough scale, many parcels of agricultural land, especially at the borders between individ­ual landscape types, are classified in the wrong landscape type and farmers are therefore unfairly taxed. This means that a defective landscape classification can lead to an unfair distribution of taxes, and thus unfairness or even discrimination between people. To solve the problem, we propose some methods of machinelearningandgeographicinformationsystemsforevaluatingexistinglandscapetypologyanddeter-miningtheirerrors.Basedontheresults,itwouldbepossibletocorrectthelandscapetypology,thuscontributing to better and more effective taxation, and thus to a fairer society. There are two main ways to tackle this problem: by creating a completely new landscape typology, or by updating the 1996 landscape typology. The application of a new typology would require changing leg-islativeactsandalengthyandunpredictablelegalprocedure.Incontrast,implementinganupdatedexisting typologybythesamegroupofresearcherswouldonlyrequirereplacingthedatalayerattherelevantmin­istry. This is considered data maintenance, and it allows the immediate use and implementation of an improved and revised typology, thereby eliminating the unfair taxation of some farmers and providing greater social justice. In 2012, the Surveying and Mapping Authority of the Republic of Slovenia and the Geodetic Institute of Slovenia proposed that using the nine landscape types instead of four landscape type groups as provid­edbythe1996landscapetypologywouldservebetter.Farmersandothershavealsosuggestedimprovements to the input data for determining land quality (Berk et al. 2012). However, both proposals have remained unfulfilled. Thedevelopmentoftechnologymakesitpossibletoobtainandprocessanincreasingamountofmore accurate data. For example, the digital elevation model had a resolution of 100m in 1996, and now it has a resolution of 1m (Hrvatin 2016). Other landscape data have also been improved, making it possible to respond to these demands and improve the landscape typology. Researchersaroundtheworldalsopointouttheimportanceofevaluatingandcorrectingexistingclas­sifications. Some of them have drawn attention to the insufficient evaluation of the results obtained and have called for more transparent and replicable procedures to help scientifically verify the classifications produced (McMahon et al. 2004; Hargrove and Hoffman 2005). Therefore, different evaluations of land­scape classifications have been presented; for example, in the United Kingdom (Warnock and Griffiths 2015).Congalton(1991)collectedandpresentedsomebasicmethodsforaccuracyassessment(errormatrix, user’s and producer’s accuracy, and kappa). Dragut and Eisank (2012) evaluated topographical units withmean intra-object variance, spatial auto-correlation, and an online survey; Breskvar Žaucer and Marušic (2006)usedaneuralnetworktoconfirmexistinglandscapeclassifications;Strand(2011)usedbinarylogis­ticregressiontoevaluatetheexistinglandscapemapofNorway;Mezosi(2016)evaluatedHungarianlandscape unitdelimitationwithmultidimensionalscalingandagglomerativehierarchicalclustering;Wieczorekand Migon (2014) compared automatic classification of landforms with morphographic landform classifica­tion mapping in the field; Vannametee et al. (2014) modeled a map of landforms with multiple-point geostatistics and compared it to the referenced map using a kappa coefficient and landform similarity; Jasiewicz et al. (2014) analyzed similarity between two landscapes calculated using their signatures (his­togramsoffeatures)andamodifiedWaveHedgessimilarityfunction;andZawadzkaetal.(2015)compared the classification of landforms and classification of soil categories with the use of Cramér’s V and multidi­mensionalscaling. Klingseisen et al. (2008) compared a map of landforms made with fuzzy set algorithm andamapgeneratedbyexpertvisualphoto-interpretation.BrownandBrabyn(2012,2012b)analyzedhow landscapevalues associate with landscape characters using a chi-square residual. Some authors have used severaldifferentmethodstoevaluateexistingclassifications(e.g.,CiglicandPerko2015;Ribeiroetal.2016). Lausch et al. (2015) provided a review of methods with which one can evaluate existing classifications. With modern methods and data sets now available, we presume it is possible to objectively quantita­tivelyevaluateexistingclassificationsandcorrectthemifnecessary.Namely,therearestillmanyopenissues inthelandscapeclassificationprocess,inparticulartheobjectivityandtransparencyofclassification.With the (re)modeling of an existing landscape classification and an explanation of its structure, it is possible to add missing pieces of information to its original description. An excellent tool for achieving this goal is machine learning, a subfield of artificial intelligence, which isbasedontheideathatmachinesareabletoautomaticallylearnandadaptthroughexperience.Suchatool developsalgorithmsfordataanalysisandknowledgediscoveryindatabases(datamining)andconstructs computer models for better decisions in real life. Machine learning can produce a better future through past experiences. The most widely used groups of machine learning methods are classification, regression, logical rela­tions, equations, and clustering (Ciglic et al. 2019). The authors of this article have already dealt with the evaluation of landscape typology, data layers, anddifferentmethodsatthetheoreticallevel(Perko1998a;Ciglic2012,2018;CiglicandPerko2012,2015, 2017;Perko,Hrvatin,andCiglic2015;Perko,Ciglic,andHrvatin2019;Ciglicetal.2019;PerkoandCiglic 2020a), but now, armed with more precise data, the newer methodological steps are even more feasible. We seek to apply and add to knowledge that we mainly acquired through two basic projects already mentioned in the first section of this article. In the project Determination of Natural Landscape Types of Slovenia Using a Geographic Information System, in which we produced several new landscape typolo­gies at various levels and with a different number of classes (see Section 2.5), we already addressed the problem of landscape classification with the help of modern databases and produced new classifications atdifferentlevels(Perko,Hrvatin,andCiglic2015).IntheprojectAdvancementofComputationallyIntensive MethodsforEfficientModernGeneral-PurposeStatisticalAnalysisandInference,wetestedmodernmath­ematical and statistical methods in geography in collaboration with researchers from the Department of Machine Learning and Artificial Intelligence at the University of Ljubljana’s Faculty of Computer and Information Science (Ciglic et al. 2019). Both projects offer much fundamental knowledge and guidance on how to improve existing landscape classifications more effectively based on more detailed data. In a methodological sense, the proposed landscape macrotypology improvement is roughly divided into three sections: • Evaluating and modeling the existing landscape macrotypology of Slovenia from 1996 with machine learning algorithms; • Checking the results in the field; and • Creating a revised landscape macrotypology of Slovenia. The first section includes the following: • Selecting and testing different machine learning algorithms and methods; and • ExperimentalmodelingofthelandscapemacrotypologiesofSloveniawithvariousmannersofsampling learningexamplesandthemosteffectivemachinelearningalgorithmsforsupervisedclassification(e.g., decision tree or k-nearest neighbors), and analysis of results. The second section includes the following: • Checking the results of the modeled landscape macrotypologies in the field in selected areas; and • Field study of the differences between the existing and modeled landscape macrotypologies of Slovenia in disputed areas. The third section includes the following: • Final modeling of the landscape macrotypology of Slovenia with multiple repetition and validation (Figure 55); • Determining the number of inhabitants, houses, and hectares for individual land-use types that do not belong to the same landscape type in relation to the existing and revised landscape macrotypology of Slovenia and that are wrongly and therefore unfairly taxed; • Producing a revised landscape macrotypology of Slovenia; and • Preparing a revised landscape macrotypology data layer for the relevant ministry to eliminate poten­tially unfair taxation of some farmers and to provide greater social justice. An important step (in addition to the use of modern classification models and the latest data) should betheuseofthesegmentationmethod,wherebywecanswitchfromraster-basedmodelingtoobject-based vectormodeling,thussignificantlyreducingunnecessarydataanddeterminingthesmallesthomogeneous units that encompass the most similar raster cells. The proposed methodological steps are not only useful for the area of Slovenia, but also suitable for other countries or regions. Improvinglandscapemacrotypologieswouldbebeneficialfordevelopingfundamentalresearchingeog­raphy as well as for greater application of research findings in agriculture, regional planning, preserving the countryside, settlement, planning proper land use, landscape ecology, and preserving Slovenia’s nat­ural and cultural heritage. With this conclusion of the article, we want to demonstrate how science – albeit at a more theoretical level, such as the creation of landscape typologies – can solve the problems of ordinary people through modern methods. After all, helping people should be one of the basic tasks of science. The authors would like our text on macrotypologies and microtypologies to serve as a small contribution to this. ACKNOWLEDGEMENT:TheauthorsacknowledgefinancialsupportfromtheSlovenianResearchAgency (program no. P6-0101: Geography of Slovenia, project no. L6-3643: Determination of NaturalLandscape Types of Slovenia Using a Geographic Information System, project no. L6-6852: Landscape Diversity and Hotspots of Slovenia, and project no. L1-7542: Advancement of Computationally Intensive Methods for EfficientModernGeneral-PurposeStatisticalAnalysisandInference),theSlovenianAcademyofSciences andArts(projectno.L6-3643,projectno.L6-6852),andtheMinistryofEducation,Science,andSport(from theEuropeanSocialFund;projectno.OP13.2.1.8.01.0068:DefinitionofaMethodologyforDetermination of Natural Landscape Types at the Local Level and Their Cartographic Display for Selected Slovenian Municipalities). 6 References Acharya, T., Ray, A. K. 2005: Image processing. Principles and applications. Hoboken. Aimée,E.,Fridl,J.,OroženAdamic,M.,Pavšek,M.,Perko,D.1996(eds.):KulturAtlasSlowenien.KulturAtlas Europa 6. Gotha, Stuttgart. Bailey, R. G. 1996: Ecosystem geography. New York. Bailey, R. G. 2004: Identifying Ecoregion Boundaries. Environmental Management 34 (S1). DOI: https://doi.org/10.1007/s00267-003-0163-6 Belbin, L., McDonald, C. 1993: Comparing three classification strategies for use in ecology. Journal of Vegetation Science 4-3. Berk, S., Kete, P., Žagar, T., Pegan Žvokelj, B., Košir., J. 2012: Ocena proizvodne sposobnosti zemljišc iz razpoložljivih prostorskih podatkov. Geografski informacijski sistemi v Sloveniji 2011–2012. GIS v Sloveniji 11. Ljubljana. Bernert, J. A., Eilers, J. M., Freemark, K. E., Ribic, C. 1997: A quantitative method for delineating regions: an example for the western corn belt plains ecoregion of the USA. Environmental Management 21-3. Biogeographicalregions.Europeanenvironmentalagency.Copenhagen,2016.URL:https://www.eea.europa.eu/ data-and-maps/data/biogeographical-regions-europe-3 (1. 12. 2021). Bohn, U., Neuhäusl, R., Gollub, G., Hettwer, C., Neuhäuslová, Z., Raus, Th., Schlüter, H., Weber, H. 2003: Karte der natürlichen Vegetation Europas / Map of the Natural Vegetation of Europe. Münster. Bratina Jurkovic, N. 2008: Evropska konvencija o krajini: izvajanje v Sloveniji. Ljubljana. BreskvarŽaucer,L.,Marušic,J.2006:Analizakrajinskihtipovzuporaboumetnihnevronskihmrež.Geodetski vestnik 50-2. Brown,G.,Brabyn,L.2012:Ananalysisoftherelationshipsbetweenmultiplevaluesandphysicallandscapes ataregionalscaleusingpublicparticipationGISandlandscapecharacterclassification.Landscapeand Urban Planning 107-3. DOI: https://doi.org/10.1016/j.landurbplan.2012.06.007 Bryan,B.A.2006:Synergistictechniquesforbetterunderstandingandclassifyingtheenvironmentalstruc­tureoflandscapes.EnvironmentalManagement37-1.DOI:https://doi.org/10.1007/s00267-004-0058-1 Burrough,P.A.,Wilson,J.P.,vanGaans,P.F.M.,Hansen,A.J.2001:Fuzzyk-meansclassificationoftopo-climatic data as an aid to forest mapping in the Greater Yellowstone Area, USA. Landscape ecology 16-6. DOI: https://doi.org/10.1023/A:1013167712622 Castillo-Rodríguez,M.,López-Blanco,J.,Muńoz-Salinas,E.2010:AgeomorphologicGIS-multivariateanaly-sis approach to delineate environmental units, a case study of La Malinche volcano (central México). Applied Geography 30-4. DOI: https://doi.org/10.1016/j.apgeog.2010.01.003 Council of Europe 2007: Synoptic presentation of the status of landscape policies pursued by the mem­ber states of the Council of Europe 2006. Information Document, Strasbourg. Ciglic, R. 2009: Slovenija v naravnogeografskih clenitvah Evrope. Geografski vestnik 81-2. Ciglic, R. 2012: Evaluation of digital data layers for establishing natural landscape types in Slovenia. Landscapes:Perception,Knowledge,Awareness,andAction:ProceedingsoftheFG-SHUInternational Symposium on Geography. New York. Ciglic, R. 2015: Izdelava metode dolocanja naravnih pokrajinskih tipov na lokalni ravni in njihovega kar­tografskega prikaza za izbrane slovenske obcine. Final report, ZRC SAZU. Ljubljana. Ciglic, R. 2018: Assessing the impact of input data incongruity in selected quantitative methods for mod-elling natural landscape typologies. Geografski vestnik 90-1. DOI: https://doi.org/10.3986/GV90107 Ciglic,R.,Perko,D.2012:SloveniaingeographicaltypificationsandregionalizationsofEurope.Geografski vestnik 84-1. Ciglic, R. Perko, D. 2017: A method for evaluating raster data layers according to landscape classification scale. Ecological Informatics 39. DOI: https://doi.org/10.1016/j.ecoinf.2017.03.004 Ciglic, R., Perko, D. 2013a: Europe’s hotspots. Acta geographica Slovenica 53-1. DOI: https://doi.org/ 10.3986/AGS53106 Ciglic, R., Perko, D. 2013b: Pokrajinska raznolikost Slovenije kot razvojna priložnost za turizem. Nove razvojne perspektive, Regionalni razvoj 4. Ljubljana. Ciglic, R., Perko, D. 2015: Modelling as a method for evaluating natural landscape typology: The case of Slovenia. Landscape Analysis and Planning. Cham. Ciglic,R.,Štrumbelj,E.,Cešnovar,R.,Hrvatin,M.,Perko,D.2019:Evaluatingexistingmanuallyconstructed naturallandscapeclassificationwithamachinelearning-basedapproach.Journalofspatialinformation science 18. DOI: https://doi.org/10.5311/JOSIS.2019.18.464 Congalton,R.G.1991:Areviewofassessingtheaccuracyofclassificationsofremotelysenseddata.Remote Sensing of Environment 37-1. DOI: https://doi.org/10.1016/0034-4257(91)90048-B Cullum,C.,Rogers,K.H.,Brierley,G.,Witkowski,E.T.2016:Ecologicalclassificationandmappingforland­scapemanagementandscience:Foundationsforthedescriptionofpatternsandprocesses.Progressin Physical Geography: Earth and Environment 40-1. DOI: https://doi.org/10.1177/0309133315611573 Council of Europe 1996: Pan-European biological and landscape diversity strategy. Nature and environ­ment 74. Strasbourg. Cerne,A.,Klemencic,M.,Plut,D.1981:Metodologijaopredeljevanjapokrajinskihenotnaprimerutržiške obcine. Gorenjska. Ljubljana. Dempsey,K.E.Wilbrand,S.M.2017:TheroleoftheregionintheEuropeanLandscapeConvention.Regional Studies 51-6. DOI: https://doi.org/10.1080/00343404.2016.1144923 DMEER:DigitalmapofEuropeanecologicalregions.Europeanenvironmentalagency.Copenhagen,2003. URL: https://www.eea.europa.eu/data-and-maps/data/digital-map-of-european-ecological-regions (1. 12. 2021) Dolenc,D.,Fridl,J.,Kladnik,D.,Perko,D.,Repolusk,P.(eds.)2007:PopisniatlasSlovenije2002.Ljubljana. Dragut, L., Eisank, C. 2012: Automated object-based classification of topography from SRTM data. Geomorphology 141–142. DOI: https://doi.org/10.1016/j.geomorph.2011.12.001 Drozg,V.,Kolnik,K.,Klemencic,M.M.,Lovrencak,F.,Horvat,U.(eds.)2004:Teorijainpraksaregionalizacije. Maribor. Drozg,V.,Perko,D.(eds.)1996a:Severnisubpanonskisvet.RegionalnogeografskamonografijaSlovenije. Ljubljana. Drozg,V.,Perko,D.(eds.)1996b:Vzhodniinseverovzhodnipredalpskisvet.Regionalnogeografskamono­grafija Slovenije. Ljubljana. Eastman, J. R. 2009: IDRISI Taiga. Guide to GIS and Image Processing. Worcester. Ellison,A.M.2010:Repeatabilityandtransparencyinecologicalresearch.Ecology91-9.DOI:https://doi.org/ 10.1890/09-0032.1 Erikstad, L., Uttakleiv, L. A., Halvorsen, R. 2015: Characterisation and mapping of landscape types, a case study from Norway. Belgeo 3. DOI: https://doi.org/10.4000/belgeo.17412 European Environment Agency 1995: Europe’s Environment - The Dobris Assessment. State of the envi­ ronment report 1/1995. Copenhagen. Ferreira, A. 2006: Pokrajinskoekološka clenitev Zgornje Gorenjske. Dela 26. Fnukalová,E.,Romportl,D.2014:AtypologyofnaturallandscapesofCentralEurope.AUCGeographica 49-2. DOI: https://doi.org/10.14712/23361980.2014.15 Fridl,J.,Gabrovec,M.,Hrvatin,M.,OroženAdamic,M.,Pavšek,M.,Perko,D.1996:Tipipokrajininnaravne nesrece. Spodnje Podravje s Prlekijo. Ljubljana. Fridl,J.,Kladnik,D.,OroženAdamic,M.,Pavšek,M.,Perko,D.,Repolusk,P.,Urbanc,M.(eds.)2007:Slovenia in Focus. Ljubljana. Fridl, J., Kladnik, D., Orožen Adamic, M., Perko, D. (eds.) 1998a: Geografski atlas Slovenije. Ljubljana. Fridl, J., Kladnik, D., Orožen Adamic, M., Perko, D., Zupancic, J. (eds.) 2001: National Atlas of Slovenia. Ljubljana. Fridl,J.,OroženAdamic,M.,Perko,D.,Urbanc,M.(eds.)1998b:Geografskiatlaszaosnovnošolo.Ljubljana. Gabrovec, M. 1996: Solar radiation and the diverse relief of Slovenia. Geografski zbornik 36. Gams,I.1970:GeomorfološkeinklimatskerazmerevJugovzhodniKoroški.JugovzhodnaKoroška.Ljubljana. Gams, I. 1975: Pokrajinska ekologija soseske Soca. Zgornje Posocje. Ljubljana. Gams, I. 1979: K pokrajinsko ekološki razclenitvi mariborske regije. Mariborsko Podravje. Maribor. Gams, I. 1998: Lega Slovenije v Evropi in med njenimi makroregijami. Geografija Slovenije. Ljubljana. Gams, I., Kunaver, J., Lovrencak, F., Radinja, D. 1974: Prispevek k prirodnogeografski tipologiji pokrajine v porecju Voglajne in zgornje Sotle. Voglajnsko-Sotelska Slovenija. Ljubljana. Gams, I., Lovrencak, F., Plut, D. 1978: Soca, Breginj in Kamno v pokrajinsko-ekološki primerjavi. Zgornje Posocje. Ljubljana. Geršic,M.,Ciglic,R.,Perko,D.2018:PokrajinskainzemljepisnoimenskaraznolikostvSloveniji.Pokrajina v visoki locljivosti, GIS v Sloveniji 14. Ljubljana. Geršic, M., Perko, D. 2020: Pokrajinska identiteta v Sloveniji. Pokrajine v Sloveniji, Lex localis. Maribor. Grondin, P., Gauthier, S., Borcard, D., Bergeron, Y., Noël, J. 2014: A new approach to ecological land clas­ sification for the Canadian boreal forest that integrates disturbances. Landscape Ecology 29. DOI: https://doi.org/10.1007/s10980-013-9961-2 Hargrove, W. W., Hoffman, F. M. 2005: Potential of multivariate quantitative methods for delineation and visualization of ecoregions. Environmental Management 34-S1. DOI: https://doi.org/10.1007/ s00267-003-1084-0 Hrvatin, M. 2016: Morfometricne znacilnosti površja na razlicnih kamninah v Sloveniji. Koper. Hrvatin,M.Perko,D.2000:RegionalizacijaintipizacijaMestneobcineLjubljana.Ljubljana:geografijamesta. Ljubljana. Ilešic, S. 1956: Slovenske pokrajine. Geografski obzornik 3-2. Ilešic, S. 1958: Problemi geografske rajonizacije ob primeru Slovenije. Geografski vestnik 29–30. Ilešic, S. 1972: Slovenske pokrajine. Geografski vestnik 44. Ilešic,S.1981:Redakcijskazasnovainstrukturapredvideneregionalno-geografskemonografijeSlovenije. Geografski vestnik 53. TheindicativemapofEuropeanBiogeographicalRegions,methodologyanddevelopment.Europeantopic centre on biological diversity. Paris, 1996. Izakovicevá, Z. 2014: Landscape typology of the Slovak Republic, IALE-Europe Thematic Workshop »Advances in Spatial Typologies: How to move from concepts to practice?«. Lisbon. Jasiewicz,J.,Netzel,P.,Stepinski,T.F.2014:Landscapesimilarity,retrieval,andmachinemappingofphys­iographic units. Geomorphology 221. DOI: https://doi.org/10.1016/j.geomorph.2014.06.011 Jongman, R. H. G., Bunce, R. G. H., Metzger, M. J., Mücher, C. A., Howard, D. C., Mateus, V. L. 2006: Objectives and applications of a statistical environmental stratification of Europe. Landscape Ecology 21-3. DOI: https://doi.org/10.1007/s10980-005-6428-0 Kladnik, D. 1996: Naravnogeografske clenitve Slovenije. Geografski vestnik 68. Ljubljana. Kladnik, D., Perko, D. (eds.) 1996a: Submediteranski svet. Regionalnogeografska monografija Slovenije. Ljubljana. Kladnik,D.,Perko,D.(eds.)1996b:Visokikraškisvet.RegionalnogeografskamonografijaSlovenije.Ljubljana. Kladnik, D., Perko, D. 1998: Zgodovina regionalizacij Slovenije. Slovenija: pokrajine in ljudje. Ljubljana. Klingseisen,B.,Metternicht,G.,Paulus,G.2008:Geomorphometriclandscapeanalysisusingasemi-auto­mated GIS-approach. Environmental Modelling & Software 23-1. DOI: https://doi.org/10.1016/ j.envsoft.2007.05.007 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 Lausch,A.,Blaschke,T.,Haase,D.,Herzog,F.,Syrbe,R.-U.,Tischendorf,L.,Walz,U.2015:Understanding and quantifying landscape structure – A review on relevant process characteristics, data models and landscape metrics. Ecological Modelling 295. DOI: https://doi.org/10.1016/j.ecolmodel.2014.08.018 Leathwick,J.R.,Overton,J.M.,McLeod,M.2003:AnenvironmentaldomainclassificationofNewZealand and its use as a tool for biodiversity management. Conservation Biology 17-6. Lotufo,R.A.,Audigier,R.,Saúde,A.V.,Machado,R.C.2008:Morphologicalimageprocessing.Microscope Image Processing, Burlington, San Diego, London. Loveland, T. R., Merchant, J. M. 2004: Ecoregions and ecoregionalization: geographical and ecological perspectives. Environmental Management 34-S1. DOI: https://doi.org/10.1007/s00267-003-5181-x Lu, D., Weng, Q. 2007: A survey of image classification methods and techniques for improving classifi­cation performance. International Journal of Remote Sensing 28-5. DOI: https://doi.org/10.1080/ 01431160600746456 Marušic, J., Ogrin, D., Jancic, M. 1998: Methodological bases. Ljubljana. McMahon,G.,Wiken,E.B.,Gauthier,D.A.2004:Towardascientificallyrigorousbasisfordevelopingmapped ecologicalregions. EnvironmentalManagement34-1. DOI:https://doi.org/10.1007/s00267-004-0170-2 Meeus,J.H.A.,1995:Pan-Europeanlandscapes.LandscapeandUrbanPlanning31.DOI:https://doi.org/ 10.1016/0169-2046(94)01036-8 Melik, A. 1935: Slovenija: geografski opis. Ljubljana. Melik, A. 1946: Prirodnogospodarska sestava Slovenije. Geografski vestnik 18. Melik, A. 1954: Slovenski alpski svet. Ljubljana. Melik, A. 1957: Štajerska s Prekmurjem in Mežiško dolino. Ljubljana. Melik, A. 1959: Posavska Slovenija. Ljubljana. Melik, A. 1960: Slovensko Primorje. Ljubljana. Metzger, M. J., Bunce, R. G. H., Jongman, R. H. G., Mücher, C. A., Watkins, J. W. 2005: A climatic strati­fication of the environment of Europe: A climatic stratification of the European environment. Global Ecology and Biogeography 14-6. DOI: https://doi.org/10.1111/j.1466-822X.2005.00190.x Meze, D. 1974: Porecje Kokre v pleistocenu. Geografski zbornik 10. Mezosi,G.2016:Similarityassessmentofnaturallandscapesbasedontaxonomicdistance.AppliedEcology and Environmental Research 14-3. DOI: https://doi.org/10.15666/aeer/1403_679693 Mücher, C. A., Bunce, R. G. H., Jongman, R. H. G., Klijn, J. A., Koomen, A. J. M., Metzger, M. J., Wascher, D.M.2003:IdentificationandcharacterisationofenvironmentsandlandscapesinEurope.Wageningen. Mücher, C. A., Klijn, J. A., Wascher, D. M., Schaminée, J. H. J. 2010: A new European landscape classifi­cation (LANMAP): a transparent, flexible and user-oriented methodology to distinguish landscapes. Ecological Indicators 10-1. DOI: https://doi.org/10.1016/j.ecolind.2009.03.018 Mücher,C.A.,Wascher,D.M.,Klijn,J.A.,Koomen,A.J.M.,Jongman,R.H.G.2006:AnewEuropeanland­scapemap as an integrative framework for landscape character assessment. Landscape Ecology in the Mediterranean,InsideandOutsideApproaches,ProceedingsoftheEuropeanIALEConference.Faro. Nared, J., Razpotnik Viskovic,N., Cremer-Schulte, D., Brozzi, R., Cortines Garcia,F.2015:Achieving sus­tainable spatial development in the Alps through participatory planning. Acta geographica Slovenica 55-2. DOI: https://doi.org/10.3986/AGS.1631 Natek, K., Žiberna, I. 2004: Naravnogeografske regionalizacije Slovenije. Teorija in praksa regionalizacije Slovenije. Maribor. Natek, M., Perko, D. 1999: 50 let Geografskega inštituta Antona Melika ZRC SAZU. Geografija Slovenije 1. Ljubljana. DOI: https://doi.org/10.3986/9789612544942 Noirfalise, A. 1987: Map of the Natural Vegetation of the member countries of the European Community and of the Council of Europe. Luxembourg. Ogrin, D. 2007: Climate, 1:1.000.000. Slovenia in Focus. Ljubljana. Olson, D. M., Dinerstein, E., Wikramanayake, E. D., Burgess, N. D., Powell, G. V. N., Underwood, E. C., D’Amico, J. A., Itoua, I., Strand, H. E., Morrison, J. C., Loucks, C. J., Allnut, T. F., Ricketts, T. H., Kura, Y.,Lamoreux,J.F.,Wettengel,W.W.,Hedao,P.,Kassem,K.R.2001:Terrestrialecoregionsoftheworld: A new map of life on Earth. BioScience 51-11. DOI: https://doi.org/10.1641/0006-3568(2001)051 [0933:TEOTWA]2.0.CO;2 Orožen Adamic, M., Perko, D., Kladnik, D. (eds.) 1995: Krajevni leksikon Slovenije. Ljubljana. Orožen Adamic, M., Perko, D., Kladnik, D. (eds.) 1996: Prirocni krajevni leksikon Slovenije. Ljubljana. Owen,S.M.,MacKenzie,A.R.,Bunce,R.G.H.,Stewart,H.E.,Donovan,R.G.,Stark,G.,Hewitt,C.N.2006: Urbanlandclassificationanditsuncertaintiesusingprincipalcomponentandclusteranalyses:Acase study for the UK West Midlands. Landscape and Urban Planning 78-4. DOI: https://doi.org/10.1016/ j.landurbplan.2005.11.002 Pak, M., Perko, D. (eds.) 1996a: Ljubljanska kotlina. Regionalnogeografska monografija Slovenije. Ljubljana. Pak,M.,Perko,D.(eds.)1996b:Visokogorskialpskisvet.RegionalnogeografskamonografijaSlovenije.Ljubljana. Pak, M., Perko, D. (eds.) 1996c: Zahodni predalpski svet. Regionalnogeografska monografija Slovenije. Ljubljana. Perko, D.,Hrvatin, M., Ciglic, R. 2017: Determination of landscape hotspots of Slovenia. Acta geographica Slovenica 57-1. DOI: https://doi.org/10.3986/AGS.4618 Perko, D. (ed.) 1996a: Nizki kraški svet. Regionalnogeografska monografija Slovenije. Ljubljana. Perko,D.(ed.)1996b:Srednjiinjužnisubpanonskisvet.RegionalnogeografskamonografijaSlovenije.Ljubljana. Perko, D. 1987: Pokrajina in raba tal v Pokokrju. Geografski zbornik 27. Perko, D. 1989: Vzhodna Krška kotlina – pokrajinska sestava in prebivalstvo. Geografski zbornik 29. Perko, D. 1992a: Ogroženost Slovenije zaradi naravnih nesrec. Ujma 6. Perko, D. 1992b: Tipi pokrajin v porecju Kokre. Geografski obzornik 39-2. Perko, D. 1993: Arboretum Volcji potok. Geografski obzornik 40-3. Perko, D. 1997: Slovenske pokrajine. Šolska karta Slovenije v merilu 1 : 500.000. Ljubljana. Perko, D. 1998a: The Regionalization of Slovenia. Geografski zbornik 38. Perko, D. 1998b: Geografija, regija in regionalizacija. Slovenija: pokrajine in ljudje. Ljubljana. Perko, D. 1998c: Tipizacija in regionalizacija Slovenije. Geografski obzornik 45-1. Perko,D.2001:AnalizapovršjaSlovenijesstometrskimdigitalnimmodelomreliefa.GeografijaSlovenije 3. Ljubljana. Perko, D. 2007a: Landscapes. Slovenia in Focus. Ljubljana. Perko, D. 2007b: Morfometrija površja Slovenije. Georitem 3. Ljubljana. Perko, D. 2009: Tipi naravne pokrajine kot dejavnik regionalnega razvoja in regionalnih razlik v Sloveniji. Razvojni izzivi Slovenije, Regionalni razvoj 2. Ljubljana. Perko, D., Ciglic, R. 2015: Slovenia as a European landscape hotspot. Acta Geobalcanica 1-2. DOI: https://doi.org/10.18509/AGB.2015.05 Perko, D., Ciglic, R. 2020a: Slovenia’s landscapes. The Geography of Slovenia: Small but Diverse. Cham. DOI: https://doi.org/10.1007/978-3-030-14066-3_14 Perko,D.,Ciglic,R.2020b:Slovenia’sregions.TheGeography ofSlovenia:Small butDiverse.Cham.DOI: https://doi.org/10.1007/978-3-030-14066-3_15 Perko, D., Ciglic, R., Geršic, M. (eds.) 2016: Terasirane pokrajine. Ljubljana. Perko, D., Ciglic, R., Geršic, M., Kladnik, D. (eds.) 2017: Terraced Landscapes. Ljubljana. Perko, D., Ciglic, R., Hrvatin, M. 2019: The usefulness of unsupervised classification methods for land-scapetypification:thecaseofSlovenia.ActageographicaSlovenica59-2.DOI:https://doi.org/10.3986/ AGS.7377 Perko, D., Ciglic, R., Zorn, M. 2020a (eds.): The geography of Slovenia: Small but diverse. World Regional Geography Book Series. Cham. DOI: https://doi.org/10.1007/978-3-030-14066-3 Perko, D., Ciglic, R., Zorn, M. 2020b: Slovenia: A European landscape hotspot. The Geography of Slovenia:SmallbutDiverse.Worldregionalgeographybookseries.Cham.DOI:https://doi.org/10.1007/ 978-3-030-14066-3_1 Perko, D., Hrvatin, M. 2010: Mikroregije obcine Idrija. Na prelomnici: razvojna vprašanja obcine Idrija. CAPACities 1. Ljubljana. Perko, D., Hrvatin, M. 2020: Pokrajinska raznolikost Mestne obcine Velenje. Velenje, industrijsko mesto v preobrazbi, CAPACities 4. Ljubljana. DOI: https//doi.org/10.3986/9789610502623_04 Perko,D.,Hrvatin,M.,Ciglic,R.2015:AmethodologyfornaturallandscapetypificationofSlovenia.Acta geographica Slovenica 55-2. DOI: https://doi.org/10.3986/AGS.1938 Perko, D., Hrvatin, M., Ciglic, R. 2017:Determinationoflandscape hotspots inSlovenia. Actageographica Slovenica 57-1. DOI: https://doi.org/10.3986/AGS.4618 Perko, D., Kladnik, D. 1998: Nova regionalizacija Slovenije. Slovenija: pokrajine in ljudje. Ljubljana. Perko, D., Križnar, N. 1998: Pokrajine v Sloveniji. Videoposnetek. Ljubljana. Perko, D., Orožen Adamic, M. (eds.) 1998: Slovenija: pokrajine in ljudje. Ljubljana. Perko, D., Ticar, J. 2020: Geomorfološke vrednote ter reliefni in pokrajinski tipi Slovenije. Modeliranje pokrajine, GIS v Sloveniji 15. Ljubljana. DOI: https://doi.org/10.3986/9789610504696_03 Plut,D.1981:PokrajinskaekologijaBelekrajine.RaziskovalnodeloInštitutazageografijoUniverzeEdvarda Kardelja v Ljubljani. Ljubljana. Plut, D. 1999: Regionalizacija Slovenije po sonaravnih kriterijih. Geografski vestnik 71. Pratt, W. K. 2007: Digital image processing. Los Altos. Renetzeder, C., van Eupen, M.,Mücher, C. A., Wrbka, T. 2008: A spatial regional reference framework for sustainability assessment in Europe. Sustainability impact assessment of land use changes. Berlin, Heidelberg. Repe, B. 2017: Pohorje – slovenska pedološka posebnost. Geografije Podravja. Maribor. Ribeiro,D.,Somodi,I.,Carni,A.2016:TransferabilityofapredictiveRobiniapseudacaciadistributionmodel in northeast Slovenia. Acta geographica Slovenica 56-1. DOI: https://doi.org/10.3986/AGS.772 Rivas-Marínez,S.,Penas,A.,Díaz,T.E.2004a:BioclimaticMapofEurope,Bioclimates.CartographicService, University of León. León. Rivas-Marínez,S.,Penas,A.,Díaz,T.E.2004b:BiogeographicMapofEurope.CartographicService,University of León. León. Romportl,D.2009:LandscapetypologyoftheCzechRepublic(TypologiekrajinyCeskérepubliky).Prague. Romportl, D., Cerny, K. 2014: Landscape typology as a spatial framework for phytopathological studies, IALE-Europe Thematic Workshop »Advances in Spatial Typologies: How to move from concepts to practice?«. Lisbon. Simensen,T.,Halvorsen,R.,Erikstad,L.2018:Methodsforlandscapecharacterisationandmapping:Asys­ tematic review. Land Use Policy 75. DOI: https://doi.org/10.1016/j.landusepol.2018.04.022 Snoj, M. 2016: Slovenski etimološki slovar. Ljubljana. Soto, S., Pintó, J. 2010: Delineation of natural landscape units for Puerto Rico. Applied Geography 30-4. DOI: https://doi.org/10.1016/j.apgeog.2010.01.010 Strand, G.-H. 2011: Uncertainty in classification and delineation of landscapes: A probabilistic approach to landscape modeling. Environmental Modelling & Software 26-9. DOI: https://doi.org/10.1016/ j.envsoft.2011.03.005 Stupariu, M.-S., Cushman, S. A., Plesoianu, A.-I., Patru-Stupariu, I., Fürst, C. 2021: Machine learning in landscapeecologicalanalysis:Areviewofrecentapproaches.LandscapeEcology.DOI:https://doi.org/ 10.1007/s10980-021-01366-9 Šifrer, M. 1961: Porecje Kamniške Bistrice v pleistocenu. Dela SAZU 12. Ljubljana. Šifrer, M. 1969: Kvartarni razvoj dobrav na Gorenjskem. Geografski zbornik 9. Špes,M.,Cigale,D.,Lampic,B.,Natek,K.,Plut,D.,Smrekar,A.,2002:Študijaranljivostiokolja.Geographica Slovenica 35, 1-2. Ticar,J.,Perko,D.,VolkBahun,M.2018:GeodedišcinainpokrajinskaraznolikostSlovenije.Pokrajinavvisoki locljivosti, GIS v Sloveniji 14. Ljubljana. Topole, M. 1992: Tipi pokrajin v porecju Mirne. Geografski obzornik 39-4. Udvardy, M. D. F. 1975: A Classification of the biogeographical provinces of the World. IUCN Occasional paper 18. Morges. Urbanc, M. 2002: Kulturne pokrajine v Slovenije. Geografija Slovenije 5. Ljubljana.Urbanc, M., Ferk, M., Fridl, J., Gašperic, P., Ilc Klun, M., Pipan, P., Resnik Planinc, T., Šmid Hribar, M. 2016: Oblikovanje predstav o slovenskih pokrajinah v izobraževalnem procesu. Geografija Slovenije 34. Ljubljana. VanEetvelde,V.,Antrop,M. 2009:Astepwisemulti-scaletypologyandcharacterisationfortrans-region­al integration, applied on the federal state of Belgium. Landscape and urban planning 91-3. DOI: https://doi.org/10.1016/j.landurbplan.2008.12.008 Vannametee, E., Babel, L. V., Hendriks, M. R., Schuur, J., de Jong, S. M., Bierkens, M. F. P., Karssenberg, D. 2014: Semi-automated mapping of landforms using multiple point geostatistics. Geomorphology 221. DOI: https://doi.org/10.1016/j.geomorph.2014.05.032 Warnock, S., Griffiths, G. 2015: Landscape characterisation: The living landscapes approach in the UK. Landscape Research 40-3. DOI: https://doi.org/10.1080/01426397.2013.870541 Wascher, D. M. 2005: Landscape character: Linking space and function. European Landscape Character Areas.Typologies,cartographyandindicatorsfortheassessmentofsustainablelandscapes.Wageningen. Wieczorek,M.,Migon,P.2014:Automaticreliefclassificationversusexpertandfieldbasedlandformclas­sificationforthemedium-altitudemountainrange,theSudetes,SWPoland.Geomorphology206.DOI: https://doi.org/10.1016/j.geomorph.2013.10.005 Wolock,D.M.,Winter,T.C.,McMahon,G.2004:Delineationandevaluationofhydrologic-landscaperegions in the United States using geographic information system tools and multivariate statistical analyses. Environmental Management 34-S1. DOI: https://doi.org/10.1007/s00267-003-5077-9 Zakšek, K., Podobnikar, T., Oštir, K. 2005: Solar radiation modelling. Computers & Geosciences 31. DOI: https://doi.org/10.1016/j.cageo.2004.09.018 Zawadzka, J., Mayr, T., Bellamy, P., Corstanje, R. 2015: Comparing physiographic maps with different cat-egorisations. Geomorphology 231. DOI: https://doi.org/10.1016/j.geomorph.2014.12.006 Zhou, Y., Narumalani, S., Waltman, W. J., Waltman, S. W., Palecki, M. A. 2003: A GIS-based spatial pat-ternanalysismodelforeco-regionmappingandcharacterization.Internationaljournalofgeographical information science 17-5. DOI: https://doi.org/10.1080/1365881031000086983 Zonneveld, I. S. 1994: Basic principles of classification. Ecosystem classification for environmental management. Dordrecht.Žiberna, I., Natek, K., Ogrin. D. 2004: Naravnoegografska regionalizacija Slovenije za potrebe pouka geografije v osnovni šoli. Teorija in praksa regionalizacije Slovenije. Maribor.