VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF wASTE-FILLED DOLINES VEGETACIJA KOT BIOINDIKATOR ANTROPOGENE DEGRADACIJE KRAšKE POKRAJINE: PRIMER Z ODPADKI ZAPOLNJENE VRTAČE Mateja BREG VALJAVEC1*, Daniela RIBEIRO1 & Andraž ČARNI2,3 1 Research Center of the Slovenian Academy of Sciences and Arts, Anton Melik Geographical Institute, Gosposka ulica 13, 1000 Ljubljana, Slovenia. e-mail: mateja.breg@zrc-sazu.si 2 Research Center of the Slovenian Academy of Sciences and Arts, Jovan Hadži Institute of Biology, Novi trg 5, 1000 Ljubljana, Slovenia. 3 University of Nova Gorica, Vipavska cesta 13, 5000 Nova Gorica, Slovenia Received/Prejeto: 03.10.2016 COBISS: 1.01 ACTA CARSOLOGICA 46/1, 95–110, POSTOJNA 2017 Abstract UDC 913:581.55(497.47) 581.55:551.435.82:628.472.2(497.47) Mateja Breg Valja�ec, Daniela Ribeiro & Andraž Čarni: Veg� etation as the bioindicator of human�induced degradation in karst landscape: case study of waste�filled dolines Mismanagement practices, such as unsustainable waste dis- posal, created many degraded sites. In karst landscapes, old uncontrolled landfills are often located in dolines (sinkholes). Buried waste material in dolines represents degradation of landform, habitats and a potential danger of groundwater pol- lution. Buried waste provides heterogeneous ecological con- ditions on the surface, thus plant communities or individual plant species that developed on the surface of landfills can be used as a bioindicators of waste-filled doline sites and therefore indicators of land degradation. we aimed to discover the po- tential of vegetation to detect unknown locations of old dump- sites in suffusion dolines in Logaško polje (Dinaric Karst, Slo- venia), either by plant communities or by plant species. we aimed to ascertain whether vegetation can indicate the dump- ing period by estimation of community succession stage. Loca- tions and the age of waste-filled dolines (doline-dumps) were preliminary identified by a historical landscape study. Thus, we used time series of aerial photographs and digital photogram- metry tools for 3D modelling of historical terrain. Ecological evaluation was based on sampling the floristic composition of plots (5x5 m). we analysed ecological conditions by Ellenberg bioindicator values, structure by life history traits and natural- ness by hemerobic levels of plants. we studied in detail 30 up to 50-years-old waste-filled dolines that are interspersed by dry and mesic grasslands. Ecological evaluation demonstrated that the main driver of ecological diversity at doline-dumps is the time at which the doline was backfilled and succession started. Annual and eutrophic communities dominate the youngest doline-dumps, middle aged doline-dumps are covered by ni- Izvleček UDK 913:581.55(497.47) 581.55:551.435.82:628.472.2(497.47) Mateja Breg Valja�ec, Daniela Ribeiro & Andraž Čarni: Vegetacija kot bioindikator antropogene degradacije kraške pokrajine: primer z odpadki zapolnjene �rtače Slabe prakse upravljanja, kot je netrajnostno odlaganje odpa- dkov, so povzročile nastanek številnih degradiranih območij. V kraških pokrajinah, so tak primer stara odlagališča odpadkov v vrtačah. Z odpadki zasute vrtače predstavljajo degradacijo reli- efne oblike, habitatov in potencialno nevarnost za onesnaženje podzemne vode. V vrtačah odloženi odpadki povzročajo hete- rogene ekološke razmere in niše za razvoj rastlinskih združb in posameznih rastlinskih vrst. Tako združbe kakor vrste je možno uporabiti kot bioindikatorje za odkrivanje z odpadki zasutih vr- tač in antropogene degradacije v pokrajini. Namen prispevka je določiti potencial vegetacije (rastlinskih združb ali rastlinskih vrst) za odkrivanje degradacije na primeru neznanih odlagališč odpadkov v sufozijskih vrtačah Logaškega polja (Dinarski kras, Slovenija). Ugotavljali smo, ali lahko stopnja sukcesije določi obdobje degradacije. Lokacije in starost z odpadki napolnjenih vrtač so bile najprej določene s historičnogeografsko analizo. Uporabili smo zgodovinske aerofotogrametrične posnetke, me- tode digitalne fotogrametrije in 3D modeliranje zgodovinske- ga reliefa. Ekološka ocena degradiranih vrtač je določena na osnovi floristične sestave vzorčnih ploskev (5x5 m), in sicer s pomočjo Ellenbergovih bioindikatorskih vrednosti, strukture z življenjskimi oblikami rastlin in njihovimi funkcionalnimi zna- ki ter stopnjah hemerobije (naravnosti) rastlin. Podrobneje smo preučili vzorec 30 z odpadki zapolnjenih vrtač (v zadnjih 50 le- tih). Slednje so v pokrajini nerazpoznavne, preraščene s razno- liko vegetacijo znotraj travnikov. Ekološka ocena je pokazala, da je glavno gonilo ekološke raznolikosti na degradiranih vrtačah čas v katerem je bila vrtača zapolnjena. Enoletne in evtrofne združbe prevladujejo na najmlajših odlagališčih, na srednje ACTA CARSOLOGICA 46/1 – 201796 MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI trophilous perennial forbs and, finally, communities developed towards mesic grassland. we conclude that plant communi- ties in combination with diagnostic plant species can be used as a bioindicators of doline-dumps in agricultural landscape of Logaško polje and can therefore indicate the sites of potential groundwater pollution sources but not the type of long-time buried waste. Key words: suffusion dolines, plant communities, waste dump- ing, Ellenberg indicator values, polje. starih odlagališčih smo zaznali nitrofilne trajnice in najbolj raz- vite združbe na gojenih mezofilnih travnikih, ki so zadnji sta- dij travniške sukcesije. Dokazali smo, da so rastlinske združbe v povezavi z diagnostičnimi rastlinskimi vrstami, uporabne kot bioindikatorji za določanje lokacij starih (do 50 let) in recentnih degradiranih vrtač v kmetijski pokrajini kraškega polja. Z upo- rabljeno metodo lahko posredno odkrivamo potencialne vire onesnaženja kraške vode, ni pa metoda dovolj preizkušena za opredeljevanje tipa podzemno odloženih odpadkov. Ključne besede: sufozijske vrtače, rastlinske združbe, odlaganje odpadkov, Ellenbergove indikatorske vrednosti, kraško polje. INTRODUCTION Dolines are the most common, typical and representa- tive landform (Sauro 2003), even more they are consid- ered as the diagnostic karst landform (Ford & williams 1989, 2007) of karst landscapes. According to this, in conservation policy more attention should be paid to the geodiversity potential of unique landforms (Smrekar et al. 2016), habitat biodiversity and conservation. From geographic point of view, dolines can be considered as unique and individual geotop (Giusti 2016) with micro- climate characteristics (e.g. thermal inversion) (Gams 2003) that induce specific vegetation patterns which can be called landform-conditioned vegetation pat- terns or landform-vegetation units (Baroni et al. 2007). The landform–vegetation units express consistent cor- relations among landforms, geomorphic processes and diagnostic plant communities. These communities act as phytoindicators of specific geomorphologic units de- rived from the activity of a single geomorphic process or from a combination of different geomorphic proc- esses (Baroni et al. 2007). The species composition of the herb layer of the dolines is markedly different than in surrounding grass- lands (Bátori et al. 2009), so doline could be recognized as an individual landform-vegetation unit. In Central Eu- rope dolines are considered as potential refugia for many vascular plants under upcoming global warming and are thus very valuable from a nature conservation point of view (Bátori et al. 2014). Dolines can be considered as the link between karst surface and underground, espe- cially if we consider their role in karst processes (Gams 2000) and genesis (e.g. the theory of denuded caves by Mihevc et al. 1998). The direct connection between the surface and the underlying high permeability aquifers makes karst aquifers extremely vulnerable to pollution (Gutiérrez et al. 2014). Besides, karst underground is a unique habitat for many protected and endemic species (Pipan & Culver 2013; Fišer et al. 2014) and the quality of underground abiotic and biotic sphere is the result of ecological conditions on the karst surface. Despite the high coverage of carbonate rocks and the economic importance of karst areas in Slovenia, cur- rent standards for karst protection are loosely defined (Ravbar & šebela 2015) and human-induced degrada- tion is still in progress. Human-induced filling of the do- lines is considered as degradation of typical karst land- scape from the perspective of typical geomorphology and landform. However, if we look from the perspective of the farmer or agricultural scientists, filling of dolines (if the material were natural and non-toxic, such as rocks and gravel of different size) could be considered as a positive measure to ensure more land of greater quality that enables the usage of modern machines etc. It can be compared to other agricultural land improvements like cultural terraces that are in Slovenia very common (šmid Hribar et al. 2017; Kladnik et al. 2017). Nevertheless, human activities may lead to a great variety of impacts, including degradation of the karst landscape, destruction of the epikarst and loss of karst landforms (Gutiérrez et al. 2014). Old pollution sourc- es in a form of waste-filled dolines are of great concern due to the fact that waste material may remain active for many decades and even hundreds of years, depending on the physical conditions. Vegetation development is dis- turbed depending on the type of buried waste and reme- diation used. Gutiérrez et al. (2014) pointed out that il- legal waste dumping and infiltration of liquid waste from landfills have effects on 1) water quality deterioration; 2) changes in karst landscape; 3) ecosystem contamination, and 4) extinction of rare species. Some studies on doline degradation in Dinaric Karst landscapes were done in regard to uncontrolled waste dumping and landfilling (Breg 2007; Cernatič Gregorič & Zega 2010; Kovačič & Ravbar 2013; Breg Valjavec 2014) and the impacts were evaluated from various perspectives, such as hydrology ACTA CARSOLOGICA 46/1 – 2017 97 VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... METHODS GEOGRAPHICAL AND GEOLOGICAL SETTINGS OF THE STUDy AREA The study was conducted on Logaško polje, one of the most northern of the 130 poljes in the Dinaric Karst (Gams 1978; Mihevc et al. 2010). Due to intensive affor- estation that affected Dinaric Karst in past decades this region is characterized as a landscape coldspot (Perko et al. 2017). The topography of polje is the result of natural hydro-geomorphic processes and long-term human in- terventions into the karst geomorphology and hydroge- ology. As these combined processes are going on, many dolines have already been overgrown with vegetation, some of them for a couple of years, others even for dec- ades. Apparently they are “masked” by the surround- ing agricultural landscape, mainly intensive grasslands (Fig. 1). Previous studies on doline degradation on Logaško polje (Breg 2007) pointed out many potential and existing sites of doline degradation. Logaško polje is an overflowing ponor polje. A large portion of the polje lies on dolomite, while limestone at the altitude of the plateau represents a 10 to 15 m higher rock cut terrace. The northern and western parts of the polje consist of early Triassic layered dolomite, with lay- ers from 0.5 to 1 meter thick. Permian deposits of flint sandstone and clay slate appear on the far north edge of Logaško polje. Lower Cretaceous and Cenomanian de- posits can be found on the south-east and east part of Logaško polje and consist of a range of grey and dark grey platy limestone, intermitted with grainy bituminous dolomite a few centimetres thick. On the bottom of the polje, located at an altitude between 470 and 480 meters, lie Quaternary river and stream deposits of an average thickness of 3 to 4 meters (Pleničar & Buser 1970). Dolines in alluvial (Quaternary deposits) parts of the karst polje are formed by suffusion processes and are differently called by different authors: suffusion dolines (Ford & williams 2007), cover dolines (Sauro 2003), (Kogovšek & Petrič 2013) or topography (Breg Valjavec & Zorn 2015) but never from an ecological point of view. It is evident from previous studies (Tintner & Klug 2011; Tintner et al. 2012) on Central European landfills that aboveground vegetation differs visually and measur- ably from natural or semi-natural vegetation in the sur- rounding habitat. Plant communities respond to exter- nal influences and the importance of site factors to plant species composition has been demonstrated repeatedly (Goslee & Sanderson 2010). It has been proved that veg- etation can indicate characteristics of landfill cover layer and, in some cases (organic waste), can even indicate some characteristics of the underlying waste, including landfill gases (Tintner & Klug 2011; Tintner et al. 2012). The characteristics of communities growing on land- fill surfaces depend on the type of succession, whether it was completely natural or human-induced (Baasch et al. 2012). Hemeroby, urbanity and ruderality are mea- sures of disturbance and human impact that have been used in urban ecology (Hill et al. 2002) and studies on spontaneous succession (Tropek et al. 2010).The diag- nostic power of individual species can be estimated by a fidelity measure that compares the presence of species within the community and beyond it. Species with high fidelita values can be treated as diagnostic (Peinado et al. 2013). In addition to species themselves, we can use spe- cies indicator values which act as an important tool for evaluating ecological conditions (Ellenberg et al. 1992; Diekmann 2003; Jóźwiak & Jóźwiak 2014). Even though these values have some limitations, they are widely used in applied plant ecology (Carpenter & Goodenough 2014). Plant functional traits can also be used; these describe the functional response of vegetation to land- use changes (Garnier et al. 2007). It is also important to estimate how close to nature the vegetation is; human influence on vegetation can be estimated by hemerobic levels (Klotz et al. 2002). So far, no study has been pub- lished on human-induced land degradation of dolines with regard to bioindication.The aim of this study is to discover the potential of plant species and plant commu- nities for detecting the locations of waste-filled dolines (doline-dumps) to identify unknown potential pollution sites and degraded habitats. Special diagnostic species that show the presence of doline-dumpsite in the agro- ecosystems of a polje were identified for individual class- es. Since species composition does not entirely describe community functioning, we attempted to find the linkage between floristic composition, structure and ecological conditions of the habitat. In view of the time complex- ity of doline waste dumps (from recent up to 50-year-old recession plots) and the heterogeneity of dumped waste, we also aim to ascertain whether plant communities can be used as an estimator of the period of time since doline dumping was completed. ACTA CARSOLOGICA 46/1 – 201798 alluvial dolines (Gams 2003). In the present research the term suffusion doline was selected. The presence of do- lines in agro-ecosystems hampers agricultural produc- tion; these also appear to be “suitable locations” for waste disposal. They have been mostly backfilled with mixed waste materials of industrial (e.g. organic) or household origin and recently with construction and demolition waste. Nowadays, the study area is a flattish surface cov- ered by grasslands; notwithstanding, some surfaces are recessed due to subsidence of the material that has been used to backfill the dolines (Fig. 1). HISTORICAL LANDSCAPE ANALySIS Precise locations and shapes of dolines were detected and determined by historical landscape analysis. we used ster- eo-image processing of historical aerial photographs from 1972 to generate historical 3D digital terrain model and to determine the initial state of doline landform (shape, depth, volume) (Fig. 2). In this study we used and updated the results of study on 3D stereo image processing and electrical resistivity imaging of dolines on Logaško polje published by Breg Valjavec (2014). Detailed explanation of methods and data is available in referenced article. Besides, series of historical aerial photographs (1975, 1986, 1989, 1992, 1994, 1998, 2001, 2006 and 2011) were acquired to determine the period of doline dumping for individual do- line (doline-dump age). By visual analysis the landscape history of individual doline and doline-dumpsite was de- termined and mapped. The map was used as the input data for field study of vegetation while results on the age of in- dividual doline-dumpsite enabled the control data for the age of succession at studied dolines. ECOLOGICAL EVALUATION Ecological evaluation is the basis for all forms of envi- ronmental and ecological management, including assess- ment of ecosystem structure and landscape degradation (Burger 2008). Field surveys were conducted during the optimal season for vegetation sampling (May–June) in order to make a species inventory to determine which plant species had invaded and colonized the doline-dumps. The ecological evaluation started with sampling plots of 5 x 5 m², one plot per doline-dump; which consisted of site reconnaissance and phyto-sociological relevés data collection for each plot and estimation of the spe- cies coverage according to the Braun-Blanquet approach (Braun-Blanquet 1964). Vegetation data was stored in the Turboveg database management program (Hennekens & Schaminée 2001). A data table (164 species x 39 sample sites) was built and was subjected to numerical analysis. For the purpose of numerical analysis, we transformed the cover values into percentages and square rooted. Classification was then carried out by PC-ORD (Mccune & Mefford 1999), ran in the JUICE 7.0 program (Tichý 2009; Tichý & Holt 2006). A dendrogram was prepared using ward’s method and Euclidean distance as a measure of resem- blance. we used the phi coefficient as a fidelity measure of species to an individual cluster. we also calculated Fischer's exact test and gave a zero fidelity value to spe- cies with P>0.05. The threshold phi value (multiplied by 100 in the JUICE program that was used in calculations) set for the species to be considered as characteristic was 40 (Chytrý et al. 2002). Life forms were determined following Raunkiaer (1934) and Pignatti et al. (2005); ecological indicator val- ues were used after Ellenberg et al. (1992) commonly used to describe substrate conditions (Diekmann 2003) and hemeroby according to Klotz et al. (2002). For interpreta- tion of the ecological conditions, the indicator values were passively projected onto a non-metric multidimensional Fig. 1: Approximately 30 years old waste-filled doline (marked with red line), which visually expresses different vegetation (lower grass) from the surrounding grassland and the preserved doline (marked with yellow line). MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI ACTA CARSOLOGICA 46/1 – 2017 99 RESULTS CLASSIFICATION OF PLANT COMMUNITIES 30 doline-dumpsites were selected as the sampling sites for further field assessment of plant communities. Land- scape history was studied through historical analysis of dolines geomorphology and it shows (Fig. 8, left image) that the most intensive period of doline dumping was in the 1980s and at the beginning of 1990s. Therefore 19 doline-dumps are 20−30 years old and six out of 30 doline-dumps are 30−40 years old. Four doline-dumps’ vegetation are younger than 10 years and only one doline-dump vegetation is older than 40 years. Upon preliminary field vegetation study, it can be summarized that doline-dumps inside agricultural grasslands present a circular shape, represented by a lightly concave surface topography of slightly differ- ent vegetation, mainly of more eutrophic or ruderal character (see Fig. 1). They cover small surface areas (on average 200 m²) and are randomly dispersed in the agricultural landscape of the study area. Their depth (before they were filled) was ranging from 3 to 7 m on average (which is also the thickness of the dumped material). with phytocenological assessment of study plots and statistical analysis of Ellenberg indicator values we recognized five clusters that are ecologically well in- terpretable (Fig. 3, Tab. 2, Fig. 8). The plant communi- ties, sampled at 39 sites in total, were classified into five community clusters (1, 2, 3, 4 and 5) corresponding to scaling (NMDS) diagram in order to reveal ecological relationships among clusters and indicator values; in this way, we could better explain environmental gradients un- Fig. 2: Study area of Logaško polje (left image) and extended view of two doline-dumps landscape history (L29A, L29b). derlying the main ordination axes. Differences between clusters were presented by Box-wishers diagrams in R project (R Development Core Team 2015). VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... ACTA CARSOLOGICA 46/1 – 2017100 two general groups (i and ii) differentiated by absence or presence of human-induced degradation: (i) Communities of semi-natural grasslands/ab- sence of land degradation: (1) dry grasslands, (3) mesic grasslands; and (i) Communities of doline-dumps/presence of land degradation: (2) oldest doline-dumps floristically close to mesic grasslands, (4) annual, eutrophic communities and (5) nitrophilous perennial forb communities. Fig. 3: dendrogram, Euclidean distance as a resemblance mea- sure and Ward’s method. The classification resulted in five clus- ters: (1) – dry grasslands, (2) – oldest doline-dumps floristically close to mesic grasslands; (3) – mesic grasslands, (4) – annual, eutrophic communities and (5) – perennial, nitrophilous forb communities. tab. 1: Synoptic table of five clusters showing the frequency and fidelity of species (upper case). Group No. 1 2 3 4 5 No. of relevés 4 16 5 5 9 Ranunculus bulbosus 100 96.3 6 −−− ∙ −−− ∙ −−− ∙ −−− Carex flacca 75 84.0 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Carex caryophyllea 75 84.0 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Bromopsis erecta 100 80.7 19 −−− 20 −−− ∙ −−− ∙ −−− Arenaria serpyllifolia 75 79.6 6 −−− ∙ −−− ∙ −−− ∙ −−− Festuca rupicola 75 71.4 ∙ −−− 20 −−− ∙ −−− ∙ −−− Salvia pratensis 100 71.2 25 −−− 20 −−− 20 −−− ∙ −−− Vicia lathyroides 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Galium verum 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Leucanthemum praecox 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Sedum annuum 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Thlaspi praecox 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Carex montana 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Brachypodium pinnatum 50 66.7 ∙ −−− ∙ −−− ∙ −−− ∙ −−− Leontodon hispidus 100 63.2 12 −−− 60 −−− 20 −−− ∙ −−− Trifolium campestre 50 61.3 6 −−− −−− ∙ −−− ∙ −−− Medicago lupulina 100 57.0 38 −−− 60 −−− 20 −−− ∙ −−− Thymus longicaulis 50 56.7 12 −−− ∙ −−− ∙ −−− ∙ −−− Hypericum perforatum 50 51.9 ∙ −−− ∙ −−− 20 −−− ∙ −−− Briza media 50 45.8 ∙ −−− 20 −−− ∙ −−− 11 −−− Lotus corniculatus 75 42.3 19 −−− 60 −−− 20 −−− ∙ −−− Convolvulus arvensis ∙ −−− 75 50.3 20 −−− 40 −−− 11 −−− Trisetum flavescens 50 −−− 94 35.7 60 −−− ∙ −−− 89 −−− Arrhenatherum elatius 75 −−− 100 34.0 60 −−− 40 −−− 67 −−− Prunella vulgaris ∙ −−− 38 −−− 100 70.1 20 −−− 11 −−− Holcus lanatus ∙ −−− 62 −−− 100 52.4 20 −−− 56 −−− Daucus carota 50 −−− 25 −−− 80 47.2 20 −−− ∙ −−− Cynosurus cristatus 75 −−− 56 −−− 100 47.2 ∙ −−− 33 −−− Veronica persica ∙ −−− 12 −−− ∙ −−− 80 73.1 11 −−− Capsella bursa-pastoris ∙ −−− 12 −−− 20 −−− 80 68.8 ∙ −−− Lamium purpureum ∙ −−− 19 −−− ∙ −−− 60 60.7 ∙ −−− Lactuca serriola ∙ −−− ∙ −−− ∙ −−− 40 59.0 ∙ −−− Chenopodium album ∙ −−− ∙ −−− ∙ −−− 40 59.0 ∙ −−− Cerastium brachypetalum ∙ −−− 6 −−− ∙ −−− 40 53.1 ∙ −−− Sonchus asper ∙ −−− ∙ −−− 20 −−− 40 43.1 ∙ −−− MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI ACTA CARSOLOGICA 46/1 – 2017 101 Group No. 1 2 3 4 5 No. of relevés 4 16 5 5 9 Plantago major ∙ −−− ∙ −−− 20 −−− 40 43.1 ∙ −−− Anthriscus sylvestris ∙ −−− 19 −−− 20 −−− ∙ −−− 89 72.7 Ajuga reptans ∙ −−− 44 −−− 40 −−− ∙ −−− 100 65.6 Rumex obtusifolius ∙ −−− 50 −−− 20 −−− 60 −−− 100 54.2 Heracleum sphondylium ∙ −−− 31 −−− 20 −−− 20 −−− 78 52.4 Alopecurus pratensis ∙ −−− 12 −−− ∙ −−− ∙ −−− 44 52.0 Vicia sepium ∙ −−− 62 −−− 40 −−− 20 −−− 89 47.2 Glechoma hederacea ∙ −−− 75 −−− 40 −−− 20 −−− 89 44.4 Aegopodium podagraria ∙ −−− 6 −−− ∙ −−− 20 −−− 44 43.5 Poa trivialis ∙ −−− 69 −−− 80 −−− 60 −−− 100 39.4 Colchicum autumnale ∙ −−− ∙ −−− 20 −−− ∙ −−− 33 36.7 Dactylis glomerata 100 −−− 100 −−− 100 −−− 60 −−− 100 −−− Leucanthemum vulgare 100 −−− 88 −−− 100 −−− 60 −−− 100 −−− Taraxacum officinale 25 −−− 94 −−− 100 −−− 100 −−− 100 −−− Plantago lanceolata 25 −−− 100 −−− 100 −−− 80 −−− 89 −−− Achillea millefolium 75 −−− 100 23.3 100 −−− 80 −−− 56 −−− Trifolium pratense 75 −−− 88 −−− 100 −−− 80 −−− 44 −−− Veronica chamaedrys ∙ −−− 81 −−− 100 −−− 60 −−− 100 −−− Cerastium holosteoides 50 −−− 88 −−− 100 −−− 80 −−− 33 −−− Ranunculus acris 25 −−− 88 −−− 100 −−− 20 −−− 78 −−− Knautia arvensis 75 −−− 69 −−− 100 −−− 40 −−− 67 −−− Centaurea carniolica 50 −−− 69 −−− 100 −−− 40 −−− 67 −−− Galium mollugo 25 −−− 81 −−− 40 −−− 40 −−− 89 −−− Centaurea jacea 50 −−− 69 −−− 100 −−− 20 −−− 67 −−− Festuca pratensis 25 −−− 69 −−− 100 −−− 40 −−− 67 −−− Crepis biennis ∙ −−− 69 −−− 80 −−− 40 −−− 89 −−− Rumex acetosa 25 −−− 56 −−− 100 −−− 20 −−− 89 −−− Poa pratensis 75 −−− 62 −−− 80 −−− 20 −−− 56 −−− Allium carinatum 50 −−− 69 −−− 80 −−− 20 −−− 56 −−− Veronica arvensis 50 −−− 75 −−− 40 −−− 80 −−− 22 −−− COMMUNITIES OF SEMI-NATURAL GRASSLANDS Dry grassland communities (cluster 1) are the character- istic vegetation of dry shallow rendzic laptosols on car- bonate bedrock. They appear on the edge of polje where limestone bedrock is exposed and it continues to the karst plateau. Some characteristic species of this group are Ra- nunculus bulbosus, bromus erectus and Carex caryophyllea (Tab. 1/1). Considering Ellenberg indicator values, this vegetation type appears on the driest sites with high pH (alkaline), nutrient-poor soils and highly exposed to light (Fig. 5). These conditions enable the establishment of a relatively high proportion of therophytes (Fig. 3). Among all studied vegetation are dry grasslands the most natural ones (Fig. 4) since the proportion of oligohemerobic and mesohemerobic species is the highest. Mesic grasslands (cluster 3) are another grassland diagnostic community developed on humid deep loamy soils of polje (Fig. 3). The vegetation in these grasslands is very similar to vegetation of old doline-dumps and Fig. 4: NmdS ordination diagram with passively projected Ellen- berg bioindicator values. detailed legend is specified in Fig. 3. VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... ACTA CARSOLOGICA 46/1 – 2017102 tab. 2: The results of historical landscape study and ecology study. DOLINE HISTORICAL LANDSCAPE STUDy ECOLOGy STUDy TyPE OF SITE AGE OF SUCCESSION (period since doline backfilling was completed) NUMERICAL VEGETATION ANALySIS (field survey & dendrogram) CLASSIFICATION OF COMMUNITIES T8b grassland / 1 - dry grasslands (i) Communities of semi-natural grasslands / absence of land degradation T2b grassland / 1 - dry grasslands T9b grassland / 1 - dry grasslands T6b grassland / 1 - dry grasslands L6 backfilled doline 26 2 - oldest doline-dumps (ii) communities of doline-dumps / presence of land degradation L2 backfilled doline 29 2 - oldest doline-dumps L13 backfilled doline 40 2 - oldest doline-dumps L4 backfilled doline 29 2 - oldest doline-dumps L14 backfilled doline 40 2 - oldest doline-dumps T4 a,b grassland / 2 - oldest doline-dumps L29B backfilled doline 26 2 - oldest doline-dumps L24 backfilled doline 23 2 - oldest doline-dumps L17 backfilled doline 26 2 - oldest doline-dumps L5 backfilled doline 26 2 - oldest doline-dumps L28 backfilled doline 29 2 - oldest doline-dumps L1 backfilled doline 40 2 - oldest doline-dumps L11 backfilled doline 35 2 - oldest doline-dumps L12A backfilled doline 40 2 - oldest doline-dumps L15 backfilled doline 43 2 - oldest doline-dumps L21 backfilled doline 23 2 - oldest doline-dumps T5b grassland / 3 - mesic grasslands (i) Communities of semi-natural grasslands / absence of land degradation T7b grassland / 3 - mesic grasslands T1b grassland / 3 - mesic grasslands T3b grassland / 3 - mesic grasslands L16 a backfilled doline 29 3 - mesic grasslands L29A backfilled doline 9 4 - annual, eutrophic communities (ii) communities of doline-dumps / presence of land degradation L3 backfilled doline 9 4 - annual, eutrophic communities LA1 backfilled doline 9 4 - annual, eutrophic communities LA2 backfilled doline 9 4 - annual, eutrophic communities L12B backfilled doline 40 4 - annual, eutrophic communities L25 backfilled doline 23 5 - nitrophilous perennial forb communities (ii) communities of doline-dumps / presence of land degradation L26 backfilled doline 23 5 - nitrophilous perennial forb communities L27 backfilled doline 23 5 - nitrophilous perennial forb communities L22 backfilled doline 23 5 - nitrophilous perennial forb communities L23 backfilled doline 23 5 - nitrophilous perennial forb communities L7 backfilled doline 26 5 - nitrophilous perennial forb communities L8 backfilled doline 26 5 - nitrophilous perennial forb communities L9 backfilled doline 26 5 - nitrophilous perennial forb communities L18 backfilled doline 29 5 - nitrophilous perennial forb communities a sits that were classified different by individual method. b reference sites that don’t have data on age of succession they are considered to be grasslands. MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI ACTA CARSOLOGICA 46/1 – 2017 103 some plots mediate between the two groups (Tab. 1). Diagnostic species of this group are Prunella vulgaris, holcus lanatus, daucus carota and Cynosurus cristatus (Tab. 1/3). These sites show an intermediate site condi- tion among all identified clusters. In comparison to the most similar old doline-dumps, these communities are poorer in nutrients and more light-demanding (Fig. 5). These communities form a close canopy of hemicrypto- phytes, preventing therophytic species to settle (Fig. 6). Compared to old doline-dumps, this type of vegetation is more natural, since more oligohemerobic and mesohe- merobic species are present and fewer euhemerobic and polyhemerobic ones (Fig. 7). Communities of semi natural grasslands are consid- ered as referential sites where no degradation occurred. COMMUNITIES OF DOLINE-DUMPS Oldest doline-dumps (cluster 2) are floristically close to mesic grasslands. These dolines were backfilled a long time ago, on average 29 years. Agro-meliorations took Fig. 5: Comparison of ecological conditions in plant communities estimated by Ellenberg bioindicator values. Fig. 6: Comparison of life history traits (only hemicryptophytes and therophytes are presented) and species richness. The median, 25 %–75 % of values, non-outlier range and outliers are presented here. VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... ACTA CARSOLOGICA 46/1 – 2017104 place in Logaško polje in late 1980s to improve agricul- tural land use. Dolines that were backfilled during agro- meliorations and before 1990 were covered with a fer- tile topsoil layer and these surfaces were further used as croplands and grasslands. The diagnostic species of this group are Convolvulus arvensis, trisetum flavescens and Arrhentherum elatius (Tab. 1/2). These stands show high similarity to the mesic grasslands of the cluster 3. Middle-aged doline-dumps dominated by nitrophi- lous perennial forb communities (cluster 5) are diag- nostic for dolines that were filled around 20 years ago, on average of 23 years (Fig. 3). The diagnostic species, with highest abundance, of this group are Anthriscus sylvestris, Ajuga reptans, Rumex obtusifolius, Alopecurus pratensis, heracleum sphondyllium, Vicia sepium, Gle- choma hederacea, Aegopodium podagraria, Poa trivialis and Colchicum autumnale (Tab.1/5). The communities are humid and nutrient-rich (Fig. 5). The proportion of hemicriptophytics is high and there is a relatively low proportion of therophytic species (Fig. 6). These com- munities are dominated by perennial and ruderal forb species. Annual, eutrophic communities (cluster 4) (Fig. 3), represent the youngest doline-dumps, covered by an- nual, eutrophic communities as an evidence of recent land disturbance. Considering results of historical land- scape analysis, these dolines have been recently filled, the median is 9 years. Considering socio-economic Fig. 7: Comparison of hemeroby between five groups. Legende: series 1 – oligohemerob, 2 − mesohemerob, 3 − β euhemerob, 4 − α euhemerob, 5 − p polyhemerob, for groups see Fig. 3. Fig. 8: Comparison of results of historical analysis and results of ecological evaluation. MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI ACTA CARSOLOGICA 46/1 – 2017 105 development the intensity of agricultural production on Logaško polje decreased at the beginning of millennium, due to the increase of urbanization. Therefore there was less need for arable land and the waste dumped in dolines was covered with infertile top layers usually consisting of construction and demolition waste or even remained uncovered until the aboveground vegetation developed, naturally or human-induced. The prevailing land use is extensive meadows. During extensive field work in the spring, we noted that grasslands are fertilized either by organic or/and mineral fertilizers to improve the ecolog- ical (nutrition) conditions of cover layer. This can also be evident by the presence of eutrophic communities and species at these sites. The diagnostic species of the group are Capsella-bursa pastoris, Veronica persica and Lam- ium purpureum (Tab. 1/4). The sites are nutrient-rich and warm (Fig. 5). The proportion of theropytic species is high due to open stands (Fig. 6). These stands are the least natural of all, being dominated by euhemerobic species (Fig. 7). In these communities, weed species are well represented, showing the initial, eutrophic character of the communities. Relations among clusters can be seen on an ordina- tion diagram (Fig. 4) and show that dry grasslands appear with the highest light and reaction, showing a relatively distant position in the diagram. The highest temperature and continentality (well pronounced extreme condi- tions) are in stands of annual, eutrophic communities, while the other three groups form a fairly homogenous group, from perennial nitrophilous forb communities to mesic grasslands. The highest trophic status includes an- nual, eutrophic (youngest doline dump) and nitrophil- ous forb communities (middle-aged doline dump) that are at the beginning of the successional sere. Recent waste and municipal solid waste slag result in a ruderal pioneer stage dominated by weedy species of open soil. ways of further succession from this stage cannot yet be predicted with certainty as also Tintner and Klug (2011) claimed. In the figure 6, the median, 25 %–75 % of values, non-outlier range and outliers are presented. Detailed legend is specified in Fig. 3. It is also important to estimate how close to na- ture the vegetation is; human influence on vegetation was estimated by hemerobic levels (Klotz et al. 2002). The comparison of hemeroby between all five groups (Fig. 7) shows that series 1 (dry grassland) correspond to oligohemerob, series 2 to mesohemerob (oldest doline- dumps), series 3 (mesic grassland) to euhemerob, series 4 (annual, eutrophic communities) to euhemerob and series 5 (perennial, nitrophilous forb communities) to polyhemerob. DISCUSSION GENERAL DISCUSSION ON COMMUNITIES OF DEGRADED DOLINES Degraded dolines can be detected by sampling vegeta- tion (plants species and plant communities) within cul- tivated and natural grasslands of agricultural landscapes. According to Gutiérrez et al. (2014) the illegal waste dumping on Logaško polje has affected changes in karst landforms. Besides, backfilling of dolines indirectly in- fluences on the extinction of rare species that are com- mon for dolines as ecological (climate) refugia (Batori et al. 2014). The aboveground vegetation of waste-filled dolines differs visually and measurably from natural or semi-natural vegetation in the surrounding agricultural (dry and mesic) grasslands and thus influences grassland biodiversity in total. The time and cost-demanding assessment by physi- cal or chemical parameters could be simplified by the vegetation ecology approach as Tintner and Klug (2011) defended. Plant communities of waste-filled dolines do respond to external influences and the importance of site factors to plant species composition has been demon- strated in accordance to referenced literature (Goslee & Sanderson 2010). we succeeded in confirming that indicator species of cluster 1 (dry grasslands) were sampled on carbonate bedrock (limestone, dolomite), where extensive grass- lands prevail due to shallow soils over carbonate bed- rock. These species show the highest level of natural- ness and a low level of hemeroby (Hill et al. 2002). In these conditions, the reaction is high, the availability of nutrients is low and there is fairly high species richness (Merunková & Chytrý 2012). Hemicriptophytes domi- nate in this type of vegetation, although sporadically we can also find some therophytes (Batalha et al. 2015). Dry grasslands vegetation is a typical landscape feature of the karst landscape (Kaligarič & Ivajnšič 2014). By means of our analysis, we could identify this vegetation type in the field and distinguish the natural vegetation of dry grasslands from the vegetation on doline-dumps. This vegetation can be classified within the class of grasslands on base-rich soils Festuco-brometea (Pipen- baher et al. 2011). VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... ACTA CARSOLOGICA 46/1 – 2017106 The second and third clusters can be treated as cul- tivated (fertilized) mesic grasslands (meadows) of allu- vial plain and the oldest doline-dumps, floristically close to mesic grassland. This type of vegetation can develop either over deeper, fertile soils, formed on loamy un- consolidated Holocene river sediments or over doline- dumps with a thicker cover layer of allochthonous fer- tile soil. The sites are humid and have a neutral reaction. Hemicryptophytic species are here dominant, forming dense stands and not allowing therophytes to establish (Čámská & Skálová 2012; šilc et al. 2014). The fourth cluster is dominated by annual, eutro- phic communities. These communities are characteristic of disturbed surfaces, since disturbance favours annual and ruderal species. we found communities with the highest hemeroby within this cluster. The sites are nutri- ent rich, humid, with relatively high reaction (Perronne et al. 2014; Sans et al. 2014). Since these sites are warm and have a relatively high reaction, we presume that there is carbonate gravel and rock in the upper horizons of the backfilling material. These habitats have high po- tential to be settled by invasive species (such as Parthe- nocissus or Erigeron), so management of these sites must be done with prudence (Zisenis 2015). These commu- nities indicate that the dolines have been covered with gravel (probably from construction waste). The relatively early successional stage, marked by the therophytic com- munities (Sojneková & Chytrý 2015), indicates that the backfilling has been done recently. No reference site is included in this cluster, only dumped dolines (5 of them), predominantly younger doline-dumps that were filled up to 10 years ago. This vegetation can be classi- fied within the class of annual weed communities Stel- larieatea mediae (šilc & Čarni 2007). The fifth cluster represents nitrophilous forb com- munities. These communities settle on sites with high nutrient availability. The species are mainly hemicrypto- phytic, often spread by rhizomes and build dense stands. The sites are humid and fertile (Pastor et al. 2014; Shush- pannikova 2014). These communities indicate that at least the upper horizons of the doline have been filled with organic waste, which has provided the fertility to the site. Doline-dumps (9 of them) included in this clus- ter are predominantly dolines that were filled on average 25 years ago. These communities are classified within the class of tall-herb semi-natural forb vegetation of nutrient rich sites Galio-Urticetea (Čarni 1994). THE SUCCESSION SERES In accordance to Baasch et al. (2012) results the structure of communities depends on the age of succession which is diverse from communities at undisturbed dry and mesic grasslands. The revegetation and succession time of doline-dumps is a very important variable that influences the stability of a meadow species community. we noted that after an extended succession period (20–40 years), species communities on doline-dumps became very sim- ilar to cultivated meadows and it was more difficult to differentiate them from the surrounding grasslands. This corresponds to the results obtained by Prach and Pýšek (2001), who mentioned that in smaller disturbed sites, succession towards natural vegetation is easier and more direct. In this case, we needed to use for identification of doline-dumps, indicator plant species from the group of nitrophilous forb communities, which provides link- age with previous successional phases (e.g. Urtica doica, Rumex obtusifolia etc.). The results show that upper lay- ers consist of relatively fertile soil, although we are not quite sure about the lower layers. This vegetation can be classified as molinio-Arrhenatheretea (šilc et al. 2014), a class of anthropogenic pastures and meadows on deeper, fertile soils. The results confirm that dry grasslands are beyond the successional sere. The succession begins with annual, synanthropic vegetation, develops towards perennial ni- trophilous forb communities and succession is arrested with mesic grasslands. Annual and eutrophic vegetation appears on new doline-dumps but this type of vegetation can be replaced by perennial species in a short period of time (Prach 1987, Sojneková & Chytrý 2015). In comparison with abandoned fields (non-cultivated), where this change appears in a few years, in our case it is a relatively long process, lasting up to 10 years, due to the type of top cover material, presumably construction and demolition waste or excavated materials. In the study area, the next successional stage con- sists of nitrophilous perennial forb species that dominate the sites for the next period of 10 years. These mid-suc- cessional species do not allow the further development of the succession sere (Bartha et al. 2014). Mowing and ex- porting material (nutrients) from the ecosystem (Herbst et al. 2013) has enabled the development of communities with lower nutrient demand. The next successional stage is represented by the oldest doline-dumps, which are similar to mesic grass- lands. This stage is dominated by Arrheatherum elatius, a competitive nitrophilous grass. This species loses its com- petitive power in some stands in this and further succes- sional stages, probably due to the lack of phosphorous in the soil (Prach et al. 2007). The oldest doline-dumps are floristically very close to mesic grassland and the transi- tion is sometimes difficult to detect, especially in cases in which mesic meadows are fertilized (Gustavsson et al. 2007). According to Prach et al. (2014), the succession in Central Europe would reach semi-natural vegetation in MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI ACTA CARSOLOGICA 46/1 – 2017 107 CONCLUSION The geomorphological analysis and historical records were the basis for the vegetation assessment which re- veals the spatio-temporal patterns and provides the ex- planation for ecological processes occurring at the indi- vidual doline-dump. At Logaško polje hydro-geomorphic processes are continuously forming new suffusion do- lines and reshaping existing ones. Doline landform in combination with thermal inversion enables landform- induced-vegetation which is destroyed by doline back- filling. Eventually, a unique karst landform and its role of habitat (refugia) disappeared for ever. However, vegeta- tion patterns of such degraded grasslands differ from the patterns of surrounding non-degraded grasslands. Thus, we conclude that vegetation is a reliable bio-indicator for the detection of doline-dumps in agricultural grassland landscapes. The revegetation and succession time of do- line-dumps is a very important variable that influences the stability of meadow species community. The natural succession of doline-dump begins with annual, synan- thropic vegetation, develops towards perennial nitrophi- lous forb communities and stops at mesic grasslands that are maintained by regular mowing. we noted that after an extended succession period (20–40 years), species communities on doline-dumps became very similar to cultivated meadows and thus it’s more difficult to differ- entiate them from the surrounding grasslands. we were able to assess unsustainable landscape management practices, such as inappropriate waste dis- posal, although the methods used should be further test- ed in comparable vulnerable agricultural landscapes and other doline types (e.g. solution dolines) in environments with different geological and ecological conditions. about 25 years, with the species composition changing a little later on. Our study showed similar results, as after 30 years the vegetation of the oldest doline-dumps very much resembles mesic grasslands. In well-established, mown communities, the trophic status became lower. According to Tintner and Klug (2011) sandy to loamy material with thick cover layers that is characteristic for oldest doline-dumps (covered with thick- loamy soils during agro-meliorations) vegetation develops very close to community of mesic meadows with high successional stability and the classification of these two communities is possible only by diagnostic plants. It was reported by Prach et al. (2013) that invasive species sometimes inter- fere in the succession line; however, our results showed that such species do not play an important role in the succession sere. we can confirm that spontaneous suc- cession is suitable for the creation of mesic grasslands on waste-filled dolines. ACKNOwLEDGMENTS The study was financially supported by the Ministry of Higher Education, Science and Technology of the Re- public of Slovenia (OP.13.2.1.01.0067) and the Slovenian Research Agency (P1-0236 to AČ). we would like to thank Iztok Sajko for the prepara- tion of some graphical material. REFERENCES Baasch, A., Kirmer, A. & S. Tischew, 2012: Nine years of vegetation development in a post mining site: effects of spontaneous and assisted site recovery.- Jour- nal of Applied Ecology, 49, 251–260, Oxford. DOI: http://dx.doi.org/10.1111/j.1365-2664.2011.02086.x Baroni, C., Armiraglio, S., Gentili, R. & A. Carton, 2007: Landform–vegetation units for investigating the dynamics and geomorphologic evolution of alpine composite debris cones (Valle dell'Avio, Adamello Group, Italy).- Geomorphology, 84, 59–79. DOI: http://dx.doi.org/10.1016/j.geomorph.2006.07.002 VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... ACTA CARSOLOGICA 46/1 – 2017108 Bartha, S., Szentes, S., Horváth, A., Házi, J., Zimmermann, Z., Molnár, C., Dancza, I., Margóczi, K., Pál, R.w., Purger, D., Schmidt, D., Óvári, M., Komoly, C., Suty- inski, Z., Szabó, G., Csathó, A., Juhász, M., Penksza, K. & Z. Molnár, 2014: Impact of mid-successional dominant species on the diversity and progress of succession in regenerating temperate grasslands.- Applied vegetation science, 17, 2, 201–213. Batalha, M.A., Pipenbaher, N., Bakan, B., Kaligarič,M. & S. škornik, 2015: Assessing community assembly along a successional gradient in the North Adriatic Karst with functional and phylogenetic distances.- Oecologia, 178, 4, 1205−1214. Bátori, Z., Csiky, J., Erdős, L., Morschhauser, T., Torok, P. & L. Kormoczi, 2009: Vegetation of the dolines in Mecsek mountains (south Hungary) in relation to the local plant communities.- Acta Carsologica, 38, 2–3, 237–252. DOI: http://dx.doi.org/10.3986/ ac.v38i2-3.125 Bátori, Z., Farkas, T., Erdös, L., Tölgyesi, C., Körmöczi L. & A. Vojtkó, 2014: A comparison of the vegeta- tion of forested and non-forested solution dolines in Hungary: a preliminary study.- Biologia, 69, 10: 1339–1348. Braun-Blanquet, J., 1964: Pflanzensoziologie. Grudzüge der Vegetationskunde. 3. Auflage. Springer Verlag, pp. 865, wien. Breg, M., 2007: Degradation of dolines on Logaško polje (Slovenia). Acta Carsologica, 36, 2, 223–231. DOI: http://dx.doi.org/10.3986/ac.v36i2.191 Breg Valjavec, M., 2014: Study of filled dolines by using 3D stereo image processing and electrical resistivity imaging.- International Journal of Speleology, 43, 1, 57–68. DOI: http://dx.doi.org/10.5038/1827-806- x.43.1.6 Breg, M. & M. Zorn, 2015: Degraded karst relief: waste- filled dolines.- In: Daniels, J.A.: Advances in envi- ronmental research, 40, Nova Science Publishers, pp. 77–95. New york. Burger, J., 2008: Environmental management: Integrat- ing ecological evaluation, remediation, restoration, natural resource damage assessment and long-term stewardship on contaminated lands.- Science of the Total Environment, 400, 6–19. DOI: http://dx.doi. org/10.1016/j.scitotenv.2008.06.041 Čámská, K. & H. Skálová, 2012: Effect of low-dose N ap- plication and early mowing on plant species com- position of mesophilous meadow grassland (Ar- rhenatherion) in Central Europe.- Grass and Forage Science, 67, 3, 403–410. Čarni, A., 1994: Les associations des ourlet nitrophiles dans le sud-est de la Slovénie comme indicateur des habi- tats.- Colloques phytosociologiques, 22, 467–497. Carpenter, w. & A. Goodenough, 2014: How robust are community-based plant bioindicators? Empirical testing of the relationship between Ellenberg values and direct environmental measures in woodland communities.- Community Ecology, 15, 1, 1–11. DOI: http://dx.doi.org/ 10.1556/ComEc.15.2014.1. 1 Cernatič Gregorič, A. & M. Zega, 2010: The impact of human activities on dolines (sinkholes) – typical geomorphologic features on Karst (Slovenia) and possibilities of their preservation.- Geographica Pannonica, 14, 4, 109–117. Chytrý, M., Tichý, L., Holt, J. & Z. Botta-Dukát, 2002: Determination of diagnostic species with statistical fidelity measures.- Journal of Vegetation Science, 13, 79–90. DOI: http://dx.doi.org/10.1111/j.1654- 1103.2002.tb02025.x Diekmann, M., 2003: Species indicator values as an im- portant tool in applied plant ecology – a review.- Basic and Applied Ecology, 4, 493–506. DOI: http:// dx.doi.org/10.1078/1439-1791-00185 Ellenberg, H., weber, H.E., Düll, R., wirth, V., werner, w. & D. Paulißen, 1992: Zeigerwerte von Pflanzen in Mitteleuropa.- Scripta Geobotanica 18, 1–248. Fišer, C., Pipan, T. & D.C. Culver, 2014: The vertical ex- tent of groundwater metazoans: an ecological and evolutionary perspective.- Bioscience, 64, 11, 971– 979. DOI: http://dx.doi.org/ 10.1093/biosci/biu148 Ford, D.C. & P.w. williams, 1989: Karst Geomorphology and hydrology.- Academic Division of Unwin Hy- man Ltd, pp. 601, London. Ford, D. & P.w. williams, 2007: Karst hydrogeology and geomorphology. John whiley & Sons, pp. 562, Chichester. Gams, I., 1978: The polje: the problem of definition.- Zeitschrift für Geomorphologie, 22, 2, 170–181. Gams, I., 2000: Doline morphogenetical processes from global and local viewpoints.- Acta Carsologica, 29, 123–138. Gams, I., 2003: Kras v Sloveniji v prostoru in času. ZRC Publishing, pp. 516, Ljubljana. Garnier, E., Lavorel, S., Ansquer, P., Castro, H., Cruz, P., Dolezal, J., Erikson, O., Fortunel. C., Freitas, H., Golo- dets, C., Grigulis, K., Jouany, C., Kazakou, E., Kigel, J., Kleyer, M., Lehsten, V., Lepš, J., Meier, T., Pakeman, R., Papadimitriou, M. Papanastasis, M.P., Quested, H., Quétier, F., Robson, M., Roumet, C., Rusch, G., Skarpe, C., Sternberg, M., Theau, J-P., Thébault, A., Vile, D. & M.P. Zavarovali, 2007: Assessing the effects of land-use change on plant traits, communities and ecosystem functioning in grasslands: a standardized methodology and lessons from an application to 11 European sites.- Annals of botany, 99, 5, 967–985. DOI: http://dx.doi.org/10.1093/aob/mcl215 MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI ACTA CARSOLOGICA 46/1 – 2017 109 Giusti, C., 2016: Aspects of geodiversity of palaeozoic limestones in the Black mountains of southern France.- Acta geographica Slovenica, 56, 1, 153– 170. DOI: http://dx.doi.org/10.3986/AGS.963 Gutiérrez F., Parise, M., Dewaele, J. & H. Jourde, 2014: A review on natural and human-induced geohazards and impacts in karst.- Earth-Science Reviews, 138, 61–88. DOI:10.1016/j.earscirev.2014.08.002 Goslee, S.C. & M.A. Sanderson, 2010: Landscape context and plant community composition in grazed agri- cultural systems of the Northeastern United States.- Landscape Ecology, 25, 1029–1039. DOI: http:// dx.doi.org/10.1007/s10980-010-9477-y Gustavsson, E., Lennartsson, T. & M. Emanuelsson, 2007: Land use more than 200 years ago explains current grassland plant diversity in a Swedish agricultural landscape.- Biological conservation, 138, 1, 47–59. Hennekens, S.M. & J.H.J. Schaminée, 2001: TURBOVEG, a comprehensive database management system for vegetation data.- Journal of Vegetation Science, 12, 589–591. Herbst, C., wäschke, N., Barto, E. K., Arnold, S., Geuß, D., Halboth, I., Reschke, S., Säger, J., Meiners, T. & E. Obermaier, 2013: Land use intensification in grasslands: higher trophic levels are more negative- ly affected than lower trophic levels.- Entomologia Experimentalis et Applicata, 147, 3, 269–281. Hill, O.H., Roy, D. B. & K. Thompson, 2002: Hemeroby, urbanity and ruderality: bioindicators of distur- bance and human impact.- Journal of Applied Ecol- ogy, 39, 5, 708–720. DOI: http://dx.doi.org/10.1046/ j.1365-2664.2002.00746.x Jóźwiak, M.A. & M. Jóźwiak, 2014: Bioindication as chal- lenge in modern environmental protection.- Eco- logical Chemistry and Engineering 21, 4, 577–591. DOI: http://dx.doi.org/10.1515/eces-2014-0041 Kaligarič, M. & D. Ivajnšič, 2014: Vanishing landscape of the “classic” Karst: changed landscape identity and projections for the future.- Landscape and Urban Planning, 132, 148–158. Kladnik, D., Kruse, A., B. Komac, 2017: Terraced land- scapes: an increasingly prominent cultural land- scape type. Acta geographica Slovenica, 57, 2, In Press. DOI: http://dx.doi.org/10.3986/AGS.4770 Klotz, S., Kühn, I. & w. Durka (eds.), 2002: BIOLFLOR – Eine Datenbank zu biologisch-ökologischen Me- rkmalen der Gefäßpflanzen in Deutschland. Schrif- tenreihe für Vegetationskunde 38. Bonn. Kovačič, G. & N. Ravbar, 2013: Analysis of human in- duced changes in a karst landscape: the filling of dolines in the Kras plateau, Slovenia.- Science of to- tal environment, 447, 143–151. DOI: http://dx.doi. org/10.1016/j.scitotenv.2013.01.002 Kogovšek, J. & M. Petrič, 2013: Increase of vulnerabil- ity of karst aquifers due to leakage from landfills.- Environmental Earth Sciences, 70, 901–912. DOI: http://dx.doi.org/10.1007/s12665-012-2180-3 Mccune, B. & M.J. Mefford, 1999: PC-ORD. Multivariate Analysis of Ecological Data, Version 4. MjM Soft- ware Design, Glenden Beach. Merunková, K. & M. Chytrý, 2012: Environmental con- trol of species richness and composition in upland grasslands of the southern Czech Republic.- Plant Ecology, 213, 4, 591–602. Mihevc, A., Slabe, T. & S. šebela, 1998: Denuded caves – an inherited element in the karst morphology, the case from Kras.- Acta Carsologica, 27, 1, 165–174. Mihevc, A., Prelovšek, M. & N. Zupan Hajna, 2010: In- troduction to the dinaric Karst, Inštitut za razisko- vanje krasa ZRC SAZU, pp 71, Postojna. Pastor, J., Gutiérrez-Ginés, M.J., Bartolomé, C. & A.J. Hernández, 2014: The complex nature of pollution in the capping soils of closed landfills: case study in a Mediterranean setting. In: Hernendez-Soriano (ed.) Environmental Risk Assessment of Soil Con- tamination. InTech, publisher,pp. 199–223, Rijeka. Peinado, M., Díaz, G., Ocaňa-Peinado, F.M., Aguirre, J.L., Macías, M.A., Delgadillo, J. & A. Aparicio, 2013: Statistical measure of fidelity applied to di- agnostic species in plant sociology.- Modern Ap- plied Science, 7, 6, 106−120. DOI: http://dx.doi. org/10.5539/mas.v7n6p106 Perko, D., Hrvatin, M., R. Ciglič, 2017: Determination of Landscape Hotspots in Slovenia.-Acta geographica Slovenica, 57, 1, 7‒29. Perronne, R., Le Corre, V., Bretagnolle, V. & S. Gaba, 2014: Stochastic processes and crop types shape weed community assembly in arable fields.- Journal of Vegetation Science, 26, 2, 348–359. DOI: http:// dx.doi.org/10.1111/jvs.12238 Pignatti, S., Menegoni, P. & S. Pietrosanti, 2005: Bioindi- cazione attraverso le piante vascolari. Valori di indi- cazione secondo Ellenberg per le piante della flora d’Italia.- Braun-Blanquetia, 39, 1–97. Pipan, T. & D.C. Culver, 2013: Forty years of epikarst: what biology have we learned?.- International jour- nal of speleology, 42, 3, 215–223. DOI: http://dx.doi. org/10.5038/1827-806x.42.3.5. Pipenbaher, N., Kaligarič, M. & S. škornik, 2011: Floris- tic and functional comparision of karst pastures and karst meadows from the north adriatic karst.- Acta Carsologica, 40, 3, 515–525. Pleničar, M. & S. Buser, 1970: Basic Geological map of SFRy. Interpretation of Map L33 – 77, 1 : 100 000.- Belgrade. VEGETATION AS THE BIOINDICATOR OF HUMAN-INDUCED DEGRADATION IN KARST LANDSCAPE: CASE STUDy OF ... ACTA CARSOLOGICA 46/1 – 2017110 Prach, K., 1987: Succession of Vegetation on Dumps from Strip Coal Mining, N. w. Bohemia, Czecho- slovakia.- Folia Geobotanica & Phytotaxonomica, 22, 4, 339–354. Prach, K. & P. Pyšek, 2001: Using spontaneous succes- sion for restoration of human-disturbed habitats: Experience from Central Europe.- Ecological Engi- neering, 17, 55–62. DOI: http://dx.doi.org/10.1016/ S0925-8574(00)00132-4 Prach, K., Pyšek, P. & V. Jarošík, 2007: Climate and pH as determinants of vegetation succession in Cen- tral European man-made habitats.- Journal of Veg- etation Science, 18, 5, 701–710. DOI: http://dx.doi. org/10.1111/j.1654-1103.2007.tb02584.x Prach, K., Lencová, K., Řehounková, K., Dvořáková, H., Jírová, A., Konvalinková, P., Mudrák, A., Novák, J. & R. Trnková, 2013: Spontaneous vegetation suc- cession at different central European mining sites: a comparison across seres.- Environmental Science and Pollution Research, 20, 11, 7680–7685. DOI: http://dx.doi.org/10.1007/s11356-013-1563-7 Prach, K., Rehounková, K., Lencová, K., Jírová, A., Kon- valinková, P., Mudrák, O., študent, V., Vanĕček, Z., Tichý, L., Petřik, P., šmilauer, P. & P. Pyšek, 2014: Vegetation succession in restoration of disturbed sites in Central Europe: the direction of succession and species richness across 19 seres.- Applied Veg- etation Science, 17, 193–200. DOI: http://dx.doi. org/10.1111/avsc.12064 R Development Core Team, 2015: R: A language and en- vironment for statistical computing. Vienna. Ravbar, N. & S. šebela, 2015: The effectiveness of protec- tion policies and legislative framework with special regard to karst landscapes: Insights from Slovenia.- Environmental Science and Policy, 51, 106–116. DOI: http://dx.doi.org/10.1016/j.envsci.2015.02.013 Raunkiaer, C., 1934: The life forms of plants and statisti- cal plant geography.- Charendon Press, pp. 632, Ox- ford. Sans, F.x., Armengot, L., Blanco-Moreno, J. M., Bocci, G., Carlesi, S. & P. Barberi, 2014: Effect of conser- vation practices on functional diversity and assem- bly of weed communities: a database of functional traits.- Building Organic Bridges, 3, 977–978. Sauro, U., 2003: Dolines and dolines: aspects of evolu- tion and problems of classification.- Acta Carsolo- gica, 32, 2, 41–52. Shushpannikova, G.S., 2014: Formation and degradation of meadows under the impact of hay harvesting and grazing in the Vychegda and Pechora floodplains.- Russian journal of ecology, 45, 1, 33–37. DOI: http://dx.doi.org/10.1134/S1067413614010123 Smrekar A., Zorn, M. & B. Komac, 2016: Heritage pro- tection through a geomorphologist’s eyes: from recording to awareness raising.- Acta geographi- ca Slovenica, 56, 1, 123–127. DOI: http://dx.doi. org/10.3986/AGS.3348 šilc, U. & A. Čarni, 2007: Formalized classification of the weed vegetation of arable land in Slovenia.- Preslia, 79, 283–302. šilc, U., Aćić, S., škvorc, ž., Krstonošić, D., Franjić, J. & Z. Dajić Stevanović, 2014: Grassland vegetation of the molinio-Arrhenatheretea class in the Nw Balkan Peninsula.- Applied vegetation science, 17, 3, 591– 603. DOI: http://dx.doi.org/10.1111/avsc.12094 šmid Hribar, M., Geršič, M., Pipan, P., Repolusk, P., Tiran, J., Topole, M., R. Ciglič, 2017: Cultivated terraces in Slovenian landscapes. Acta geograph- ica Slovenica, 57, 2, In Press. DOI: http://dx.doi. org/10.3986/AGS.4597 Sojneková, M. & M. Chytrý, 2015: From arable land to species-rich semi-natural grasslands: Succession in abandoned fields in a dry region of central Europe.- Ecological Engineering, 77, 373–381. DOI: http:// dx.doi.org/10.1016/j.ecoleng.2015.01.042 Tichý, L. & J. Holt, 2006: jUICE, program for manage- ment, analysis and classification of ecological data.- Masaryk University, Vegetation Science Group, pp. 98 Brno. Tichý, L., 2009: Juice, software for vegetation classifica- tion.- Journal of Vegetation Science, 13, 3, 451–453. DOI: http://dx.doi.org/10.1111/j.1654-1103.2002. tb02069.x Tropek, R., Kadlec, T., Karesova, P., Spitzer, L., Kocarek, P., Malenovsky, I., Banar, P., Tuf, I.H., Hejda, M. & M. Konvicka, 2010: Spontaneous succession in limestone quarries as an effective restoration tool for endangered arthropods and plants.- Journal of Applied Ecology, 47, 139–147. DOI: http://dx.doi. org/10.1111/j.1365-2664.2009.01746.x Tintner, J. & B. Klug, 2011: Can vegetation indicate land- fill cover features?- Flora- Morphology, Distribu- tion, Functional Ecology of Plants, 206, 6, 559–566. DOI: http://dx.doi.org/10.1016/j.flora.2011.01.005 Tintner, J., Smidt, E., Böhm, K. & L. Matiasch, 2012: Risk assessment of an old landfill regarding the potential of gaseous emissions – A case study based on bioin- dication, FT-IR spectroscopy and thermal analysis.- waste management, 32, 2418–2425. DOI: http:// dx.doi.org/10.1016/j.wasman.2012.07.022 Zisenis, M., 2015: Alien plant species: A real fear for ur- ban ecosystems in Europe?- Urban Ecosystems, 18, 355–370. DOI: http://dx.doi.org/10.1007/s11252- 014-0400-1 MATEJA BREG VALJAVEC, DANIELA RIBEIRO & ANDRAž ČARNI