97 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 original scientifi c article DOI 10.19233/ASHN.2017.12 received: 2017-11-20 POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH- EASTERN ALPINE AND DINARIC PHYTOGEOGRAPHIC REGIONS OF SLOVENIA AND CROATIA IN THE LIGHT OF CLIMATE CHANGE Aljaž KOŽUH Pševo 9, Pševo, 4000 Kranj e-mail: aljazeko@gmail.com Mitja KALIGARIČ Department of Biology, Faculty of Natural Sciences and Mathematics and Faculty of Agriculture and Life Sciences, University of Maribor, Koroška 160, Maribor, Slovenia e-mail: mitja.kaligaric@um.si Danijel IVAJNŠIČ Department of Biology, Faculty of Natural Sciences and Mathematics, University of Maribor, Koroška 160, Maribor, Slovenia e-mail: dani.ivajnsic@um.si ABSTRACT We studied the potential distribution of silver fi r (Abies alba Miller) in the Alpine and Dinaric phytogeographic regions of Slovenia and Croatia in the light of climate change. A decline of silver fi r in southern Europe due to summer droughts and heat has already been observed, along with the spread of its range towards the north-east in continental Europe due to a warmer climate with milder winters. In this study, we modelled habitat suitability for the silver fi r in regard to the most probable climate change scenarios. No major changes in habitat suitability were found in either region. Habitat suitability should slightly increase in the central and western parts of the Alpine region in more optimistic scenarios and on Pohorje and in the Dinaric region in more pessimistic scenarios. A more distinctive change of habitat suitability would probably be suppressed by weather extremes, such as summer drought and heat, a cold winter period, and extreme weather phenomena. Key words: global warming, ecological modelling, habitat suitability, RCP, silver fi r, species range change DISTRIBUZIONE POTENZIALE DELL’ABETE BIANCO (ABIES ALBA) NELLE REGIONI FITOGEOGRAFICHE ALPINA SUD-ORIENTALE E DINARICA IN SLOVENIA E CROAZIA IN RELAZIONE AI CAMBIAMENTI CLIMATICI SINTESI Gli autori hanno studiato la distribuzione potenziale dell’abete bianco (Abies alba Miller) nelle regioni fi toge- ografi che alpina e dinarica della Slovenia e della Croazia in relazione ai cambiamenti climatici. Una diminuzione dell’abete bianco nell’Europa meridionale, dovuta alla siccità e al caldo estivi, era già stata osservata, insieme all’espansione del suo areale verso nord-est nell’Europa continentale. In questo studio gli autori hanno modellato l’idoneità dell’habitat per l’abete bianco in relazione agli scenari più probabili di cambiamento climatico. Non è risultato alcun cambiamento importante nell’idoneità dell’habitat in nessuna delle due regioni. Secondo scenari più ottimistici, l’adeguatezza dell’habitat dovrebbe aumentare leggermente nelle parti centrale e occidentale della regione alpina, mentre secondo scenari più pessimistici dovrebbe ingrandirsi sul Pohorje e nella regione dinarica. Un cambiamento più distintivo dell’idoneità dell’habitat verrebbe probabilmente soppresso da condizioni meteorologi- che estreme, quali la siccità estiva e il caldo, un freddo periodo invernale e fenomeni meteorologici estremi. Parole chiave: riscaldamento globale, modellistica ecologica, idoneità dell’habitat, RCP, abete bianco, cambiamenti dell’areale della specie 98 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 INTRODUCTION Recently, the spatial distributions of many plant spe- cies, among them trees, including the silver fi r (Abies alba), have been noted to be much different from their natural distributions owing to several anthropogenic factors and infl uences. With this research we aimed to reveal the potential distribution of silver fi r without anthropogenic infl uences in the Alpine and Dinaric phytogeographic regions of Slovenia and Croatia. Furthermore, we also aimed to determine differences between the two areas. Primarily, we tested the present natural habitat suitability for the silver fi r of those two study areas by considering environmental variables and ecological modelling techniques. Finally, we examined future potential spatial distributions of the silver fi r based on the four most likely future climate scenarios grounded on four representative concentration pathways (RCP) of greenhouse gases. The silver fi r (Abies alba) grows up to 50 m high and 2.5 m thick, an evergreen tree with coniform to oviform crown, fl at branches, fl at needles with white lines and upright cones. It blossoms from April to June (Brus & Robič, 2002). Its main growth period is around 50 to 60 days from May to July (Aussenac, 2002). The silver fi r is mainly a European tree species (Brus & Robič, 2002). Its natural habitat is located mostly in the mountain regions of eastern, western, southern and central Europe (Anić et al., 2009). It grows in the Alps, Vosges and Jura, on the Balkan Peninsula and in the Carpathians. There are also some isolated ranges on the Apennine Peninsula, Corsica, in the Massif Central and Pyrenees (Brus & Robič, 2002). Its range spreads between 40° and 52° in latitude (between Poland and northern Greece) and between 5° and 27° in longitude (between the western Alps and the Carpathians) (Anić et al., 2009). In the Alpine and Dinaric regions it prospers between 400 and 1200 meters above sea level (Brus & Robič, 2002) in humid habitats with more than 1000 mm annual rainfall, and in the Mediterranean with average annual temperatures between 7 and 13 °C (Aussenac, 2002). It prefers fresh, deep and nutrient rich soils and is not sensitive to geological bedrock: it grows on carbonate or non-carbonate substrates (Brus & Robič, 2002) despite water accessibility being lower on carbonate (Ficko et al., 2011). The silver fi r grows at late succession phases, mostly in a community with the common beech (Fagus sylvatica) and the spruce (Picea abies) (Brus & Robič, 2002). However, it is a rather weak competitor and as such prospers only in a narrow gradient of environmen- tal conditions. In most of the favourable areas for the silver fi r, the beech is more successful in less and spruce in more extreme conditions (Ellenberg, 1988). Neverthe- less, the silver fi r is more competitive in shady forests with slower growth during the spring compared to the beech (Diaci et al., 2010). It does not tolerate extreme winter cold and summer drought and heath (Gazol et al., 2015; Koprowski, 2013). In the Mediterranean, its growth is limited mostly by low precipitation and water accessibility in spring and summer, while in central Europe its growth is limited due to low temperatures in late winter and early spring (Gazol et al., 2015). Forest managers gave preference to coniferous rather than to deciduous trees (Ellenberg, 1988; Ficko et al., 2011). The silver fi r is more common and widespread in the Dinaric region than in the Alpine (Slovenian Forest Service, 2010). Young specimens of fi r are frequently consumed and damaged by deer (Brus & Robič, 2002). It is therefore no surprise that the silver fi r population size is negatively correlated with deer population size (Diaci et al., 2010). The global increase of greenhouse gas concentration and mean temperature is currently well documented (Ogrin, 2004). The CO2 concentration has risen from 280 to over 400 ppm since 1750 (Anić et al., 2009). Without anthropogenic emissions, it only rose by 20 ppm between the years 8000 and 2000 B.C. (Anić et al., 2009). In the study area, the increase of average annual temperature and decrease of annual rainfall was record- ed during the 20th century (Ogrin, 2004; ARSO, 2016). Similar trends are expected in the 21st century. Average annual temperature increased by 1 to 1.5 °C over the 20th century (Ogrin, 2004; Gazol et al., 2015) and it should increase additionally by 1.5 to 6 °C according to different scenarios during the 21st century (Ogrin, 2004). The variability in precipitation patterns are even higher (Ogrin, 2004). During the summer, the precipitation amount is expected to decrease by 20% followed then by a 30% increase during the winter (Kutner & Kobler, 2011). However, on the annual scale, a 10% decrease of precipitation is expected (Anić et al., 2009). It should be emphasized that extreme weather events (heat and cold waves, droughts, fi res, irregular precipitation, etc.) are more and more frequent and intensive (Kutner & Kobler, 2011; ARSO, 2016) and signifi cantly affect the silver fi r populations. The silver fi r population in the Mediterranean is expected to decrease due to more intensive summer droughts, heat waves and fi res (Gazol et al., 2015). The beech-fi r forests in the Dinaric region are expected to be gradually replaced by thermophile forests (Kutner & Kobler, 2011). In central Europe, the silver fi r range extension towards the northeast is expected because of less extreme cold conditions in late winter and early spring, as well as its present range stability despite the possible competition with thermophile tree species (Ruosch et al., 2016). We presumed that altitude, annual mean air tem- perature and precipitation variables represent the most important natural determinants of the silver fi r´s spatial distribution. Climate change should affect silver fi r pop- ulations and its distribution in the study area especially in the lowlands. On the other hand, it is not expected that silver fi r will spread to altitudes higher than its up- 99 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 per limit, because winter cold is still too severe in the high mountains (ARSO, 2016; Ogrin, 2004). From that perspective, we focused on three hypotheses: 1. there are more potentially suitable silver fi r habitats in the Dinaric than in the Alpine region; 2. its range will move towards higher altitudes and will be narrower along this gradient; 3. Potential silver fi r range will decrease especially in the Dinaric region because it also occupies mountain peaks. MATERIALS AND METHODS Study area The Alpine phytogeographic region of Slovenia and the Dinaric phytogeographic region of Slovenia and Croatia were chosen as the study area (Fig. 1). The Alpine region of Slovenia contains Alpine geographic region of Slovenia with Julian Alps, Kamnik-Savinja Alps and the Karawanks (they also contain several mountain plateaus – Pokljuka, Jelovica, Menina, Velika Planina and Dobrovlje), which are a part of the Southern Lime- stone Alps; and Pohorje and Kozjak as part of the Central Alps. The Dinaric region of Slovenia contains mostly the Dinaric geographic region of Slovenia with poljes and Dinaric plateaus (Senegačnik, 2012). The north-western border are the plateaus of Banjšice and Trnovski gozd. At the border with Croatia it stretches from Snežnik at the west across Kočevski Rog to the mountain edge of Bela Krajina at the east. In Croatia we marked off the Dinaric region from the Slovenian border with Gorski Kotar at the north across the Velebit and its continental hinterland with poljes to the southern edge of Velebit at the south. Alpine valleys stretch between 500 and 1000 m a.s.l., relief plateaus from 1000 to 1600 m a.s.l.; whereas the tree line extends up to 1900 m a.s.l., with the highest peaks reaching up to 1000 m above it. In the Dinaric region, poljes are distributed between 400 and 800 m a.s.l., plateaus between 800 and 1500 m a.s.l., and the highest peaks up to 1800 m a.s.l., thus stretching just above the tree line (Požar & Novak, 2005; Senegačnik, 2012). The average annual temperature of alpine plateaus is 2 to 6 °C, whereas on the Dinaric plateaus average temperatures are signifi cantly higher and range from 4 to 7 °C. Annual rainfall on both considered regions reaches 1500 to 3000 mm (Zaninović et al., 2008; ARSO, 2016). Collection of spatial data Initially, silver fi r spatial distribution data for the two phytogeographical regions in Slovenia and Croatia were gathered from the Slovenia Forest Service (Slo- venia Forest Service, 2010; url: http://www.zgs.si/slo/ gozdovi_slovenije/o_gozdovih_slovenije/karte/index. html) and the Flora Croatica database (Nikolić, 2015), which was established by the Faculty of Science in Zagreb (Url: https://hirc.botanic.hr/fcd/). Thereafter, Fig. 1: The chosen study area. Sl. 1: Izbrano območje raziskave. Tab. 1: Bioclimatic variables from the Worldclim data- base. Tab. 1: Bioklimatske spremenljivke podatkovne baze Worldclim. Symbol Description BIO1 Annual mean temperature BIO2 Mean Diurnal Range (mean of monthly (max temp – min temp)) BIO3 Isothermality (BIO2/BIO7) (*100) BIO4 Temperature Seasonality (standard deviation *100) BIO5 Max Temperature of Warmest Month BIO6 Min Temperature of Coldest Month BIO7 Temperature Annual Range (BIO5-BIO6) BIO8 Mean Temperature of Wettest Quarter BIO9 Mean Temperature of Driest Quarter BIO10 Mean Temperature of Warmest Quarter BIO11 Mean Temperature of Coldest Quarter BIO12 Annual Precipitation BIO13 Precipitation of Wettest Month BIO14 Precipitation of Driest Month BIO15 Precipitation Seasonality (Coeffi cient of Variation) BIO16 Precipitation of Wettest Quarter BIO17 Precipitation of Driest Quarter BIO18 Precipitation of Warmest Quarter BIO19 Precipitation of Coldest Quarter 100 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 bioclimatic (Worldclim 1.4; Hijmans et al., 2005) and elevation data were considered as major contributions to potential future distribution of the silver fi r under the selected climate change model (CCSM4) and four representative concentration pathways greenhouse gas scenarios (RCP2.6, RCP4.5, RCP6.0, RCP8.5 [Tab. 2]) until the end of the century (2070) (Tab. 1). The horizon- tal resolution of these geospatial datasets corresponds to 30 arc seconds (approximately 1 km² in mid latitudes). The considered greenhouse gas (GHG) scenarios are named after possible changes of radiative forcing in the year 2100 relative to the preindustrial age (Meinshausen et al., 2011). Scenario RCP2.6 anticipates a recent peak of emissions of GHG (between years 2010 and 2020); scenario RCP4.5 anticipates the GHG peak around 2040; scenario RCP6.0 around 2080; and fi nally, sce- nario RCP8.5 a continuous increase of GHG emissions until the end of the 21st century (Weyant et al., 2009). Consequently, a global annual temperature increase is inevitable (Stocker et al., 2013) (Tab. 2). Spatial data processing and ecological modelling The acquired spatial databases of silver fi r distribu- tion in the study area were unifi ed (by leaning on the WGS84 coordinate system) and spatially fi ltered with ArcGIS software (ESRI, 2016). Selected environmental variables (bioclimatic and altitude) were extracted by using a background bias fi le (Barbet-Massin et al., 2014). Additionally, all 19 bioclimatic variables were PCA transformed in order to avoid possible correla- tion of explanatory variables. The resulting fi rst three components (BioPCA), explaining 87.5% of variability, together with altitude were considered using the habitat suitability modelling procedure. In that light, the Mahalonobis Typicality species dis- tribution modelling (SDM) approach within Idrisi Selva software (Clark Labs, 2015) was selected. This method is less sensitive to spatially auto-correlated occurrence data and is frequently being used to model plant distribution from the climate change perspective (Clark Labs, 2015). After completing the present scenario, the accuracy and reliability of the produced habitat suitability map was verifi ed with ROC analysis and the resulting AUC value. The fi nal processing of future environmental conditions, captured in future BioPCA components, gave us fi ve habitat maps (present, and four future, RCP scenarios) for the silver fi r in the study area. However, the continuous maps were simplifi ed for easier interpretation into four suitability maps by applying the following thresholds: 1 = 0 – 25%, 2 = 25 – 50%, 3 = 50 – 75% and 4 = 75 – 100%. Finally, a comparative table summarizing the proportions of each suitability class within both phytogeographic regions was produced (Tab. 3) RESULTS The ROC analysis results and the corresponding AUC value for the silver fi r suitability in the study area by applying the Mahalonobis typicality model are shown Tab. 2: The considered future climate scenarios (RCP2.6, RCP4.5, RCP6.0, RCP8.5) according to CCSM global climate model. Tab. 2: Upoštevani podnebni scenariji (RCP2.6, RCP4.5, RCP6.0, RCP8.5) po globalnem podnebnem modelu CCSM. Scenario Solar radiation change (W/m²) Increase of global annual temperature by year 2100 (°C) (variability) RCP2.6 2.6 1.0 (0.3 to 1.7) RCP4.5 4.5 1.8 (1.1 to 2.6) RCP6.0 6.0 2.2 (1.4 to 3.1) RCP8.5 8.5 3.7 (2.6 to 4.8) Table 3: The proportion of potential habitat area for the silver fi r in each phytogeographic region by considering four suitability thresholds and climate scenarios. Tabela 3: Delež potencialnega habitata jelke po upoštevanih razredih ustreznosti in podnebnih napovedih na obravnavanih fi togeografskih območjih. Model Alpine phytogeographic region Dinaric phytogeographic region Both regions together Habitat suitability (%) Habitat suitability (%) Habitat suitability (%) 0-25 25-50 50-75 75-100 0-25 25-50 50-75 75-100 0-25 25-50 50-75 75-100 Present 40 26 22 11 53 24 13 10 49 25 16 10 RCP2.6 40 23 24 13 54 25 15 7 50 24 18 9 RCP4.5 41 27 17 14 49 24 19 7 47 25 19 9 RCP6.0 39 26 21 14 40 27 26 8 39 27 24 9 RCP8.5 36 25 26 13 71 6 12 11 60 12 16 12 101 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 in Fig. 2. The curve for the model is steep and fl attens quickly (Fig. 2), the AUC value is close to 1 (0.954) thus indicating a satisfactory level of agreement between oc- currence data and predicted suitability. Figure 3 shows the present silver fi r´s (Abies alba) habitat suitability in the chosen Alpine and Dinaric phytogeographic regions, and fi gure 4 its future (year 2070) potential suitability according to considered GHG scenarios (RCP 2.6, 4.5, 6.0, 8.5) and the CCSM4 global climate model. The present habitat suitability is the highest in Dinaric plateaus of Slovenia and Alpine plateaus and the middle mountain zone (Kamnik-Savinja Alps and the surround- ing plateaus) (Fig. 3). Low suitability is detected in the western Julian Alps and in the Dinaric region of Croatia (Fig. 3). Quite distinctive contrasts are evident on both sides of the border (Fig. 3). The lowest suitability can be identifi ed in the high mountains. The alpine valleys, the peak of Pohorje and several surrounding poljes exhibit low suitability as well (Fig. 3). All future scenarios show common spatial features of the silver fi r’s potential habitat but differ in a few details (Fig. 4). They all show lower habitat suitability in the western Julian Alps and the southern part of the Dinaric region in Croatia (Fig. 4). The fi rst scenario RCP2.6 is similar to the present one. The difference is noticeable in a somewhat lower habitat suitability on the Dinaric plateaus of Slovenia and higher on western Alpine plateaus of Slovenia (Pokljuka, Jelovica). The RCP4.5 scenario is the most similar to the present one. The only difference is a bit higher habitat suitability in the central Karawanks and a bit lower in the Trnovski gozd area. In comparison with RCP2.6 there is a bit higher silver fi r habitat suitability on the Dinaric plateaus of Slovenia and a bit lower on Pokljuka and Jelovica (Fig. 4). The RCP6.0 scenario is similar to the present one and to RCP4.5. There is a bit lower suitability in the central part of the Dinaric region of Slovenia, whereas in Trnovski gozd it is similar to the present one. A notice- able difference of silver fi r habitat suitability is in the eastern part of the Alpine region of Slovenia (eastern Karawanks, Pohorje) compared to other scenarios. There is a bit higher habitat suitability in the central part of the Dinaric region in Croatia as well (Fig. 4). The RCP8.5 scenario predicts the best conditions by the end of the century for the considered species mostly in Slovenia except in the western part of the Alpine region (Pokljuka, Jelovica, central Karawanks, western Kamnik-Savinja Alps) resulting in the highest habitat suitability. On the other hand, it is simultaneously the worst scenario for the Dinaric region of Croatia where there is very low habitat suitability almost throughout the whole region. This scenario assumes the highest contrast on both sides of the border (Fig. 4). Future scenarios also show a little tendency of potential habitat optimum shift from west to east in the Alpine region of Slovenia from less (RCP2.6) to the warmest scenario (RCP8.5) (Fig. 4). The proportions of potential silver fi r habitat area for each of the applied thresholds and considered regions separately and together are shown in Tab. 3. In the Alpine region, the proportions of the suitable area are similar considering all thresholds and RCP scenarios. However, in the 4th quartile of the potential habitat suitability a clear positive trend towards warmer climate conditions can be identifi ed. In the Dinaric Region the largest potential habitat in the fourth class is predicted in the case of scenario RCP8.5. If climate scenarios RCP2.6, 4.5 or 6.0 are realized, the Dinaric region could be Fig. 3: Recent habitat suitability for the silver fi r (Abies alba) in the chosen Alpine and Dinaric phytogeographic regions. Sl 3: Aktualna primernost jelke (Abies alba) v izbrani alpski in dinarski fi togeografski regiji. Fig. 2: ROC analysis curve and the corresponding AUC value. Sl. 2: Krivulja ROC analize in pripadajoča AUC vred- nost. 102 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 occupied with less silver fi r. Overall, high variability in silver fi r potential habitat in the study area is assured in both - more optimistic and more pessimistic - thresholds considered. DISCUSSION The infl uence of climate change on silver fi r popula- tions across Europe was already studied by Gazol et al. (2015). They outlined that in southwestern Europe, silver fi r populations could decrease owing to increased aridity, but increase in the Continental temperate zone of central Europe due to climate warming. Ruosch et al. (2016) draw similar conclusions, predicting that silver fi r range should decrease in southern Europe and spread northeast toward central Europe in future. They also predict that the present range should remain stable despite possible competition with thermophilous tree species. Kutner & Kobler (2011) tried to predict the change of forest vegetation in Slovenia by considering different climate scenarios with the use of ecological modelling. They calculated that the share of beech-fi r forests will substantially decrease by the year 2100 and could be mostly replaced by thermophile forests. The coniferous forests with prevalent spruce and fi r are expected to be replaced mostly by broadleaf forests. Koprowski (2013) tried to determine the response of silver fi r growing outside its natural range concerning spring extreme weather phenomena in Poland. The higher March temperatures should stimulate silver fi r growth especially in the western part of the study area, at the edge of continental plains with less spring frost and where colder winter periods are less pronounced. Anić et al. (2009) tried to reveal the infl uence of climate change on silver fi r´s ecological niche in Croatia and proposed that the niche will gradually decrease in the 21st century because of temperature rise. Ficko et al. (2011) found that silver fi r’s range in Slovenia shifted Fig. 4: Potential habitat suitability for silver fi r (Abies alba) for the year 2070 by four future climate scenarios (RCP2.6 [A], RCP4.5 [B], RCP6.0 [C], and RCP8.5 [D]) in the chosen Alpine and Dinaric phytogeographic regions. Sl. 4: Primernost habitata jelke (Abies alba) leta 2070 po štirih prihodnjih klimatskih scenarijih (RCP2.6 [A], RCP4.5 [B], RCP6.0 [C], and RCP8.5 [D]) v izbrani alpski in dinarski fi togeografski regiji. 103 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 towards cooler and more humid habitats over the last 40 years and slightly expanded. Based on our results, none of the hypotheses can be completely proven. The fi rst hypothesis about higher portion of suitable habitats in the Dinaric region can be at least partly proven, because the model, especially in Slovenia, shows more optimal potential habitats in the Dinaric compared to the Alpine region by apply- ing the 4th quartile suitability threshold (75-100%) for the Mahalonobis probability distribution. Surprisingly, though the environment in Gorski Kotar is very similar to that on the Slovenian side of the border (Čavlović et al., 2006; Kutnar & Kobler, 2011), the results show there much lower habitat suitability for the silver fi r. However, the spatial pattern of the silver fi r is hetero- geneous; the tree grows in places like the Velebit and in its surroundings despite less favourable condition (Nikolić, 2015). The second hypothesis can neither be proven nor rejected because the model does not show any distinctive change in the Alpine region or the results are not distinctive enough to draw proper conclusions. In this case, a more accurate scale would be needed to adequately test this research question. The third hypothesis can be completely rejected, because the silver fi r´s range will probably not decrease in the Dinaric region and could even increase, especially in the most pessimistic scenario, RCP8.5. The results confi rm the fact that silver fi r is most common in the middle altitude (mountain) zone (Brus & Robič, 2002). Lower habitat suitability in the west- ern and central Julian Alps might be the consequence of Mediterranean infl uence in the Soča valley (Ogrin, 2004). Thermophilic vegetation is also present there and could displace the silver fi r (Kutner & Kobler, 2011), although all references do not confi rm that (Ruosch et al., 2016). However, climate conditions there are more variable; even though there are higher rates of precipita- tion and longer dry periods (ARSO, 2016). In the Dinaric region of Croatia the results show mostly low habitat suitability, but some scenarios (especially RCP6.0) still indicate better potential habitat suitability in the northern and central parts of the region. However, in the southern part all scenarios show mostly low potential habitat suitability. We could conclude that especially in Gorski Kotar, where the environment is currently similar to that on the Slovenian side of the border, po- tential habitat suitability is also similarly high. On the Velebit, especially its southern part and its continental hinterland, the habitats might actually be less suitable for silver fi r today and still might be in the future. Maybe also the Mediterranean effect of summer droughts, heat and fi res is and will be more distinctive there. Future scenarios predict a lower share of optimal habitats in the Dinaric region in optimistic scenarios (RCP2.6 and RCP4.5) and higher in the most pessimistic (RCP8.5). In the Alpine region, all scenarios are simi- lar; however, most of habitats of greater suitability are also shown by considering the RCP8.5 scenario. The Dinaric region could be placed in southern Europe and the Mediterranean, where the silver fi r´s range should mostly decrease, especially because of hotter summers and more severe droughts (Aussenac, 2002; Kutnar & Kobler, 2011; Gazol et al., 2015; Ruosch et al., 2016). Such climate change consequences have already been spotted there (Anić et al., 2009; Čavlović et al., 2012; ARSO, 2016); but, surprisingly, our results do not con- fi rm such a response of the silver fi r. However, better potential habitat suitability for the silver fi r in central Eu- rope, where the Alpine region can be placed (Ruosch et al., 2016), is confi rmed but without signifi cant change. That could be the consequence of a milder climate in the Alps, where more distinctive summer heat and droughts are not present yet (ARSO; 2016). Finally, some restrictions and limitations regarding the research should be pointed out. We are aware that the bioclimatic variables of the Worldclim database are uncertain in some mountain areas, especially on geographically heterogeneous landscapes; this is why the results should be treated with some caution. Ow- ing to more accurate data in Slovenia, the forecast of habitat suitability is probably much more representative. The exaggerated difference on both sides of the border, despite similar environmental conditions, is certainly not a representative result, but likely the consequence of unbiased spatial data or a highly variable spatial pattern of the considered species concerning the considered environmental predictors. CONCLUSIONS The expected climate change could not have any distinctive infl uence on the silver fi r distribution range. In western and central parts of the Alpine region, the optimistic future climatic scenarios predict somewhat more favourable conditions for the silver fi r; in contrast, in the eastern part of its current range, in the Dinaric region, the pessimistic climate scenario (RCP 8.5) results in a more potentially suitable habitat area. Such results could be the consequence of higher mean air tem- peratures, but their favourable effect should probably be partly suppressed by more common and intensive weather extremes. Because of some research restrictions and limitations, the results can deviate from the actual expected state in the future. ACKNOWLEGMENTS We thank the Croatian Botanical Society and prof. Toni Nikolić in particular, for allowing the use of spatial distribution data of silver fi r in Croatia. The research for this paper was partly funded by the P1-0164 grant, provided by the Slovenian Research Agency. 104 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 POTENCIALNA RAZŠIRJENOST JELKE (ABIES ALBA) V JUGOVZHODNO-ALPSKEM IN DINARSKEM FITOGEOGRAFSKEM OBMOČJU SLOVENIJE IN HRVAŠKE V LUČI KLIMATSKIH SPREMEMB Aljaž KOŽUH Pševo 9, Pševo, 4000 Kranj e-mail: aljazeko@gmail.com Mitja KALIGARIČ Department of Biology, Faculty of Natural Sciences and Mathematics and Faculty of Agriculture and Life Sciences, University of Maribor, Koroška 160, Maribor, Slovenia e-mail: mitja.kaligaric@um.si Danijel IVAJNŠIČ Department of Biology, Faculty of Natural Sciences and Mathematics, University of Maribor, Koroška 160, Maribor, Slovenia e-mail: dani.ivajnsic@um.si POVZETEK Zaradi vse bolj izrazitega vpliva klimatskih sprememb na vegetacijo smo s to raziskavo želeli ugotoviti njihov vpliv na potencialno razširjenost jelke (Abies alba Miller) v alpski in dinarski fi togeografski regiji na območju Slove- nije in Hrvaške. Že danes je opazno krčenje areala jelke na južnem območju razširjenosti zaradi vse intenzivnejših poletnih suš in vročine v Sredozemlju ter širjenje areala proti severovzhodu zaradi toplejše klime in milejših zim kontinentalne Evrope. Preverjali smo primernost habitata za jelko s pomočjo ekološkega modeliranja za sedanje stanje in štiri najbolj verjetne prihodnje scenarije. Rezultati niso pokazali večjih sprememb v primernosti habitata v obeh regijah. Primernost habitata naj bi se nekoliko povečala, v osrednjem in zahodnem delu alpske regije ob bolj optimističnih scenarijih, na Pohorju in v Dinarski regiji pa ob bolj pesimističnih scenarijih. Izrazitejše izboljšanje primernosti habitata pa bodo najbrž vseeno zavrli vse intenzivnejši vremenski ekstremi, kot so poletna suša in vročina, zimski mraz in vremenske ujme. Ključne besede: ekološko modeliranje, jelka, globalno segrevanje, primernost habitata, RCP, sprememba areala 105 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 REFERENCES Agencija Republike Slovenije za Okolje (ARSO) (Slovenian Environment Agency) (2016): Podnebje Slovenije – preglednice, karte in izredni dogodki (The climate of Slovenia – tables, maps and extreme weather phenomena). Vremenski portal meteo.si – Uradna vremenska napoved za Slovenijo, »Javne informacije Slovenije« ARSO – met (Weather portal meteo.si – The offi cial weather forecast for Slovenia, »Public informa- tion of Slovenia« ARSO - met). Acquired from: http:// meteo.arso.gov.si/met/sl/climate/. Anić, I., J. Vukelić, S. Mikac, M. Bakšić & D. Ugarković (2009): Utječaj globalnih klimatskih prom- jena na ekološku nišu obične jele (Abies alba Mill.) u Hrvatskoj (Effects of global climate change on the eco- logical niche of silver fi r (Abies alba Mill.) in Croatia). Šumarski list br. 3-4, CXXXIII, 135-144. Aussenac, G. (2002): Ecology and ecophysiology of circum-Mediterranean fi rs in the context of climate change. Annual Forest Science, 59, 823-832. Barbet-Massin, M., C.H. Albert & W. Thuiller (2012): Selecting pseudoabsences for species distribution mod- els: how, where and how many. Methods in Ecology and Evolution, 3, 327-338. Brus, R. & D. Robič (2002): Sprehod po gozdu: najpogostejša slovenska drevesa in grmi (A walk through the forest: the most common Slovenian tree and bush species). DZS, Ljubljana. Clark Labs (2015): Idrisi 17.0 The Selva Edition. Clark University 950 Main St., Worcester MA 01610 USA. Clark Labs (2015): Species Distribution Modeling with TerrSet´s Habitat and Biodiversity Modeler. Clark University 950 Main St., Worcester MA 01610 USA. Ac- quried from: https://clarklabs.org/species-distribution- modeling-in-terrsets-land-change-modeler/ Čavlović, J., A. Bončina, M. Božić, E. Goršić, T. Simončič & K. Teslak (2015): Depression and growth recovery of silver fi r in uneven-aged Dinaric forests in Croatia from 1901 to 2001. Forestry, 88, 586-598. Diaci, J., D. Rozenbergar & A. Boncina (2010): Stand dynamics of Dinaric old-growth forest in Slovenia: Are indirect human infl uences relevant? Plant Biosystems, 144(1), 194-201. Ellenberg, H. (1988): Vegetation Ecology of Central Europe, fourth edition. Cambridge University Press, Cambridge, pp. 197-207. ESRI (2016): ArcGIS Desktop: Release 10. Redlands, CA: Environmental Systems Research Institute. Ficko, A., A. Poljanec & A. Boncina (2011): Do changes in spatial distribution, structure and abundance of silver fi r (Abies alba Mill.) indicate its decline? Forest Ecology and Management, 261, 844-85. Gazol, A., J. J. Camarero, E. Gutierrez, I. Popa, L. Andreu-Hayles, R. Motta, P. Nola, M. Ribas, G. Sangüe- sa-Barreda, C. Urbinati & M. Carrer (2015): Distinct effects of climate warming on populations of silver fi r (Abies alba) across Europe. Journal of Biogeography, 42, 1150 – 1162. Hijmans, R. J., S. Cameron & J. Parra (2005): World- Clim – Global Climate Data. Museum of Vertebrate Zo- ology, University of California, Berkeley, ZDA. Acquired from: http://www.worldclim.org/ IPCC AR5 WG1 (2013): Climate Change 2013: The Physical Science Basis. In: Stocker, T.F., D. Qin, G. K. Plattner, M. Tignor, S. K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex & P. M. Midgley (eds): Working Group 1 (WG1) Contribution to tke Intergovernmental Panel on Climate Change (IPCC) 5th Assessement Report (AR5). Cambridge University Press, 1523 pp. Koprowski, M. (2013): Reaction of silver fi r (Abies alba) growing outside its natural range to extreme weather events and a long-term increase in march tem- perature. Tree-ring research, 69(2), 49-61. Kutnar, L. & A. Kobler (2011): Prediction of forest vegetation shift due to different climate-change scenarios in Slovenia (Prognoza promjena šumske vegetacije zbog različitih scenarija klimatskih promjena u Sloveniji). Šumarski list 3-4, CXXXV, 113-126. Meinshausen, M., S.J. Smith, K. Calvin, J. S. Daniel, M. L. T. Kainuma, J-F. Lamarque, K. Matsumoto, S. A. Montzka, S. C. B. Raper, K. Riahi, A. Thomson, G. J. M. Velders, & D. P. P. van Vuuren (2011): The RCP greenhouse gas concentrations and their extensions from 1765 to 2300. Climatic Change, 109, 213-241. doi:10.1007/s10584-011-0156-z. Nikolić, T. (2015): Flora Croatica baza podataka (http://hirc.botanic.hr/fcd). Prirodoslovno-matematički fakultet, Sveučilište u Zagrebu. Ogrin, D. (2004): Vreme in Podnebje (Weather and climate). In Bat, M., F. Lovrenčak, R. Pavlovec & D. Ogrin (eds.): Narava Slovenije, Mladinska knjiga, Ljubljana, pp. 72-101. Požar, S. & F. Novak (2005): ATLAS Slovenije (Atlas of Slovenia). Geodetski Zavod Slovenije and Mladinska knjiga Založba, d.d., Ljubljana, 487 pp. Ruosch, M., R. Spahni, F. Joos, P. D. Henne, W. O. van der Knaap & W. Tinner (2016): Past and future evolution of Abies alba forests in Europe – comparison of a dynamic vegetation model with palaeo data and observations. Global Change Biology, 22, 727-740. Senegačnik, J. (2012): Slovenija in njene pokrajine (Slovenia and its landscapes). Modrijan, Ljubljana, 472 pp. Weyant, J., C. Azar, M. Kainuma, J. Kejun, N. Naki- cenovic, P. R. Shukla, E. La Rovere & G. Yohe (2009): Future IPCC Activities – New Scenarios: Report of 2.6 Versus 2.9 Watts/m2 RCPP Evaluation Panel. Intergov- ernmental Panel on Climate Change, Thirtieth session, Antalaya. IPCC Secretariat, Geneva, Switzerland, 81 pp. 106 ANNALES · Ser. hist. nat. · 27 · 2017 · 2 Aljaž KOŽUH et al.: POTENTIAL DISTRIBUTION OF SILVER FIR (ABIES ALBA) IN SOUTH-EASTERN ALPINE AND DINARIC ..., 97–106 Zaninović, K., M. Gajić-Čapka, M. Perčec Tadić, M. Vučetić, J. Milković, A. Bajić, K. Cindrić, L. Cvitan, Z. Katušin, D. Kaučić, T. Likso, E. Lončar, Ž. Lončar, D. Mihajlović, K. Pandžić, M.Patarčić, L. Srnec, V. Vučetić (2008): Klimatski atlas Hrvatske / Climate atlas of Croa- tia 1961-1990., 1971-2000. Državni hidrometeorološki zavod, Zagreb, 172 pp. Zavod za gozdove Slovenije (ZGS) (Slovenian Forest Service) (2010): Karte Drevesnih vrst: Jelka (Tree spe- cies maps: silver fi r). Oddelek za gozdnogospodarsko načrtovanje (Department for forestmanagement plan- ning), Rok Pisek, marec 2010. Acquired from: http:// www.zgs.si/slo/gozdovi_slovenije/o_gozdovih_sloveni- je/karte/index.html.