Responses of carabid beetles to urbanization in Transylvania (Romania)

Our reference: LAND 2010 P-authorquery-v9 AUTHOR QUERY FORM Journal: LAND Please e-mail or fax your responses and any corrections to: E-mail: [email protected] Article Number: 2010 Fax: +353 6170 9272 Dear Author, Please check your proof carefully and mark all corrections at the appropriate place in the proof (e.g., by using on-screen annotation in the PDF file) or compile them in a separate list. To ensure fast publication of your paper please return your corrections within 48 hours. For correction or revision of any artwork, please consult http://www.elsevier.com/artworkinstructions. Any queries or remarks that have arisen during the processing of your manuscript are listed below and highlighted by flags in the proof. Click on the ‘Q’ link to go to the location in the proof. Location in Query / Remark: click on the Q link to go article Please insert your reply or correction at the corresponding line in the proof Reference(s) given here were noted in the reference list but are missing from the text – please position each reference in the text or delete it from the list. Q1 Uncited reference: This section comprises references that occur in the reference list but not in the body of the text. Please cite each reference in the text or, alternatively, delete it. Any reference not dealt with will be retained in this section. Q2 Please check the insertion of part label ‘B’ in the captions of Figs. 3–5. Thank you for your assistance. Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 Landscape and Urban Planning xxx (2011) xxx–xxx 1 Contents lists available at ScienceDirect Landscape and Urban Planning journal homepage: www.elsevier.com/locate/landurbplan Responses of carabid beetles to urbanization in Transylvania (Romania)1 Béla Tóthmérésza,∗, István Máthéb, Enik˝ o Balázsc, Tibor Magurad 2 aDepartment of Ecology, University of Debrecen, POB 71, H-4010 Debrecen, Hungary3 bDepartment of Technology and Life Sciences, Sapientia University, RO-530104 Miercurea Ciuc, Piat¸a Libert˘ at¸ii 1, Romania4 cRO-537250 Miercurea Ciuc, Piat¸a Majláth Gusztáv Károly, 4A/24, Romania5 dHortobágy National Park Directorate, POB 216, H-4002 Debrecen, Hungary6 7 article info8 9 Article history:10 Received 29 October 200911 Received in revised form 30 September 2010 12 13 Accepted 7 October 201014 Available online xxx 15 Keywords:16 Globenet 17 Urban18 Arthropods19 Gradient 20 Species richness21 Forest specialist species22 abstract To investigate the impact of urbanization on carabid beetles samples were taken over two years using pitfalltraps along a rural–urbanforest gradient representing increasinghuman disturbance in andnearby the city of Sfântu Gheorghe (Romania). We predicted that total number of species should decrease, whereas number of opportunistic and matrix species should increase towards the urban end of the gradient. Both the overall species richness and the number of individuals were significantly the highest in the suburban area followed by the rural area and the lowest in the urban area. These findings contradicted the increasing disturbance hypothesis; the number of species did not decrease by the increasing disturbance. The proportion of the forest specialist individuals and species significantly decreased from the rural towards the urban area, supporting the habitat specialist hypothesis. An opposite pattern was observed in species richness of the generalist carabids, supporting the opportunistic species hypothesis. Both the proportion of matrix species and their density were significantly higher in the urban area, supporting the matrix species hypothesis. Our findings also highlighted that overall diversity is not an appropriate indicator; species with different habitat affinities should be analysed separately to evaluate the real effect of urbanization. © 2011 Elsevier B.V. All rights reserved. 1. Introduction23 Urbanization is a conversion of lands to urban or other built-up24 areas (Pickett, Cadenasso, & Grove, 2001; Xu et al., 2007). These25 areas account only for a few percentage of the earth’s land sur26 face. However, their influence on the functioning and services of27 ecosystems are rather large (Alberti, 2005; Berling-Wolff & Wu,28 2004; Grimm, Grove, Pickett, & Redman, 2000). Urbanization is an29 increasinglyimportantforceshapingthelandscapeviahabitatfrag30 mentation and loss (Gibb & Hochuli, 2002; Miyashita, Shinkai, &31 Chida, 1998)and the alteration of habitat structure (Antrop, 2000;32 Fernandez-Juricic, 2004; Shochat, Stefanov, & Whitehouse, 2004). 33 All these modifications affect species richness and community 34 structure in urban areas. They create opportunities for generalist35 speciesfavouringurbanenvironments,andfacilitatetheinvasionof36 alien and/or invasive species (Godefroid & Koedam, 2007; Honnay, 37 Piessens, Van Landuyt, Hermy, & Gulinck, 2003). Understanding38 the relationship between urbanization and ecological processes is a39 major objective of urban ecology (Breuste, Feldmann, & Uhlmann, 40 ∗Corresponding author. Tel.: +36 52 512900; fax: +36 52 431148. E-mail addresses: [email protected] (B. Tóthmérész), [email protected] (I. Máthé), [email protected] (E. Balázs), [email protected] (T. Magura). 1998; Wu & David, 2002), and in itself is a key research topic in 41 landscape ecology (Wu & Hobbs, 2002). 42 A way to estimate the effects of urbanization on nature is 43 to study the structure and function of ecological systems along 44 rural–urban gradients (McDonnell & Pickett, 1990; Niemelä, Kotze, 45 & Ashworth, 2000). Along these gradients, the original, native habi46 tat(ruralareaand/orwildland)isfirstbrokenupbynon-continuous 47 development and habitation with moderate disturbance (subur48 ban area). The remaining habitat fragments in urban areas are 49 influenced by the densely populated, built-up and often highly 50 disturbed city centres and they are more affected, managed, and 51 fragmented than their suburban and rural complements. In 1998, 52 an international research project called Globenet (Global Net53 work for Monitoring Landscape Change) was initiated to assess 54 and compare the impact of urbanization on biodiversity (Niemelä 55 et al., 2000). This project applies the rural–suburban–urban gra56 dient approach (Pickett et al., 2001) in forested habitats using a 57 common, standardized methodology (pitfall trapping) and eval58 uating the responses of common invertebrates to urbanization. 59 Until now, the majority of the published papers in the frame of 60 the Globenet project investigated carabid beetles (Elek & Lövei, 61 2007; Ishitani, Kotze, & Niemelä, 2003; Magura, Tóthmérész, & 62 Molnár, 2004; Magura, Lövei, & Tóthmérész, 2008; Niemelä et al., 63 2002; Sadler, Small, Fiszpan, Telfer, & Niemelä, 2006; Venn, Kotze, 64 0169-2046/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.landurbplan.2011.02.038 Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 2B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx & Niemelä, 2003). Studies analysing other target invertebrates are65 ratherlimited (for spiders: Alaruikka,Kotze, Matveinen, & Niemelä,66 2002;Magura, Tóthmérész, Hornung, &Horváth, 2008; for isopods:67 Hornung, Tóthmérész, Magura, & Vilisics, 2007; Magura, Hornung,68 & Tóthmérész, 2008; Vilisics, Elek, & Lövei, 2007).69 The aim of the present study was to investigate the effects of70 urbanization on carabid beetles along a rural–urban gradient rep-71 resenting increasing human disturbance. Several hypotheses were72 formulated to explain the effects of disturbance on biotic com-73 munities. We tested the following hypotheses: (i) According to74 the increasing disturbance hypothesis formulated by Gray (1989),75 an increase in disturbance would monotonously decrease diver-76 sity. Thus, diversity should decrease from a high value in rural77 area to a low one in the heavily disturbed urban area. (ii) Frequent78 and/or severe disturbance would affect sensitive species; it primar-79 ily affects the habitat specialist (here the forest specialist) species.80 Thus, the habitat specialist hypothesis predicts that diversity of for-81 est specialist species should decrease from the less disturbed rural82 forest towards the more disturbed urban area (Magura et al., 2004).83 (iii)Species that are able tocope with disturbance may benefitfrom84 the disturbance caused by urbanization, and they should gain dom85 inance in the disturbed suburban and heavily disturbed urban area;86 opportunistic species hypothesis (Gray,1989).(iv)Thestudiedforests87 are surrounded by a matrix (open habitats). Urbanization changes88 considerable the structure of forested habitats, and it makes them89 vulnerable to the invasion of the matrix species. Species penetrat-90 ing from the surrounding matrix (here the open-habitat species)91 may benefit from the habitat alteration. We are mentioning this92 new hypothesis as matrix species hypothesis.93 2. Materials and methods94 2.1. Study area95 The study areas were in and around the city of Sfântu Gheo-96 rghe (Sepsiszentgyörgy, Western-Transylvania, Romania; 45◦51N;97 25◦47E). The distance between sampling areas (rural, suburban,98 urban) was 3–10km and all studied sites covered an area of 99 greater than 10ha. It has been stressed recently that a forest patch 100 needs to have a minimum size to maintain an intact, habitat-101 specific carabid assemblage; it is estimated to be at least tens of102 hectares (Niemelä, 2001). Therefore, our site selection fulfilled this103 criterion.104 Rural sites were in a 90-year-old oak-hornbeam-beech forest105 on north-western slope at 630–719m elevation. Percentage cover 106 of the canopy layer was 70–80%; frequent species in the canopy:107 Fagus sylvatica,Quercus petraea,and Carpinus betulus. There were108 denseshrublayer (cover was 20%) and arelatively sparseherb layer109 (cover was 5–10%). Suburban sites were selected in a 60-year-old110 oak-hornbeam-beech forest on western slope at 600–700m ele-111 vation. The same species were frequent in the canopy as in the112 rural forest. Percentage cover of the canopy layer was 80–90% 113 with moderate shrub layer (percentage cover was 10%). Cover of114 herb layer was 10–15%. These suburban sites were popular for115 recreation by the local population. There were numerous path-116 ways and trampling intensity was high. Dead trees were harvested, 117 and fallen trees were also removed. Urban sites were in a castle118 park with moderately closed canopy (70–80% percentage cover)119 with sparse shrub layer (percentage cover was 5%) and dense herb 120 layer (percentage cover was 30–40%). Besides the native species (C. 121 betulus,Fraxinus exelsior,Quercus robur,Acer campestre,F. sylvatica, 122 Picea abies,Abies alba,Pinus nigra,Pinus strobus,Tilia cordata,Tilia123 platyphyllos,Aesculus hippocastanum), several non-native, exotic124 species were also present: Liriodendron tulipifera,Magnolia acumi125 nata,Tsuga canadensis,Caragana arborescens,F. sylvatica subsp.126 atropurpurea,andThuja plicata.Intheparkfallentreesandbranches 127 were removed. Shrub layer was strongly thinned. Grass was reg128 ularly moved, and the mowed grass and leaf litter were taken 129 away. There were several paved and asphalt-covered paths in the 130 park. 131 2.2. Sampling design 132 Sampling design followed the Globenet protocol (Niemelä et al., 133 2002). Forested sampling areas were selected along a rural–urban 134 gradient within the city, and in the surrounding forest, as required 135 by the Globenet protocol. Four sites, at least 100m apart were 136 selected within each sampling area. Carabid beetles were collected 137 by randomly placing ten pitfall traps at least 10m apart at each site. 138 This resulted in a total of 120 traps along the rural–urban gradient 139 (3 areas×4 sites×10 traps). Pitfall traps consisted of plastic cups 140 (diameter 65mm, volume 250ml) containing 75% ethylene glycol 141 as a killing-preserving solution. The traps were covered with bark 142 pieces to protect them from litter and rain. Trapped beetles were 143 collected fortnightly from the end of April to the end of September 144 in both 2004 and 2005. Traps were placed at the same locations in 145 both years. Carabids were identified to species using keys in H˚ urka 146 (1996).147 2.3. Data analyses 148 The carabid assemblages along the rural–urban gradient was 149 displayed by multidimensional scaling (MDS) using the Man150 hattan distance of the relative abundance of carabid species 151 (Legendre & Legendre, 1998). Nested analyses of variance with 152 repeated measures (using General Linear Models) were performed 153 to test differences in the overall carabid density, species richness, 154 standardized species richness, the ratio of forest, generalist and 155 open-habitat species in the assemblages among the three sampling 156 areas (rural, suburban, urban), among the 12 sites, and between the 157 two years (2004 and 2005). Data from the individual traps were 158 used. Sites were nested within the sampling areas and years were 159 concernedasrepeating(Sokal&Rohlf,1995).Toeliminatetheeffect 160 of sample size, species richness was standardized for every trap 161 using species rarefaction or expected species richness (Heck, van 162 Belle, & Simberloff, 1975; Niemelä & Kotze, 2009). The minimum 163 variance, unbiased estimates of the expected number of species 164 was used (Smith & Grassle, 1977): 165 ES(m)=ST − ST  i=1 N−ni m N n,166 where ES(m) is the expected number of species in a subsample con167 taining mindividuals; ST is the total number of species, niis the 168 abundance of the ith species and Nis the total number of indi169 viduals. We choose m=10 individuals (the lowest catch in a trap). 170 Calculations were performed by the DivOrd package (Tóthmérész, 171 1993). 172 Carabid beetles were categorised into forest, generalist and 173 open-habitat species according to the information in H˚ urka (1996).174 The distribution of data used in the ANOVA model was normal 175 (tested by the Kolmogorov–Smirnov test, Sokal & Rohlf, 1995). 176 When ANOVA revealed a significant difference between the means, 177 a Tukey test was performed for multiple comparisons among 178 means. 179 Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx 3 3. Results 180 3.1. Carabid assemblages along the gradient181 The total carabid catch consisted of 6971 individuals repre-182 senting 50 species (3651 individuals of 39 species in 2004, and183 3320 individuals of 41 species in 2005; Table 1). In the rural area184 20 species and 2076 individuals were caught (999 individuals 19185 species in 2004, 1077 individuals 15 species in 2005); 26 species186 and 4194 individuals were captured in the suburban area (2352187 individuals22speciesin2004,1842individuals21speciesin 2005),188 and 701 individuals belonging to 36 species were captured in the189 urban area (300 individuals 26 species in 2004, 401 individuals190 29 species in 2005). The most numerous species was Pterostichus191 oblongopunctatus in both years, and in total, made up 26.7% of the192 total catch. However, it occurred rarely in the urban area. In the193 rural forest, P. oblongopunctatus,Carabus glabratus,Abax parallelus 194 andMolops piceus were themostabundantin both years.Inthe sub195 urbanarea,P. oblongopunctatus,Pterostichus hungaricus,C. glabratus 196 and Carabus violaceus were the most numerous. In the urban area A. 197 parallelus,Pseudoophonus rufipes,Abax carinatus and Harpalus latus 198 were the most common (Table 1). 199 Urban carabid assemblages differed from suburban and rural 200 assemblages; MDS ordination revealed a clear separation between 201 them (Fig. 1). The assemblages of suburban and rural areas were 202 very similar to each other. The carabid assemblages in the urban 203 sites were separated from the others along the first axis. The size 204 of the convex hull on the ordination scatterplot was the highest in 205 the case of urban area, indicating a high heterogeneity, that is the 206 composition of the trapped carabids changed considerably from 207 trap to trap (Fig. 1). 208 Table 1 The numbers and habitat preference of carabid beetle species captured in pitfall traps in and around the city of Sfântu Gheorghe, Transylvania (Romania), in 2004 and 2005. Species sequence is according to the biannual total (most common first). F=forest specialist species, G=habitat generalist species, O =open-habitat species. Species Habitat affinity 2004 2005 Total Rural Sub-urban Urban Rural Sub-urban Urban Pterostichus oblongopunctatus F 361 797 6 281 412 2 1859 Carabus glabratus F 266 396 0 227 369 0 1258 Abax parallelus F 117 112 119 219 144 255 966 Pterostichus hungaricus G 9 492 0 2 305 0 808 Carabus violaceus G 30 236 6 49 313 9 643 Molops piceus F 94 78 0 91 43 0 306 Carabus coriaceus F 28 80 1 16 34 1 160 Pterostichus niger G 26 8 0 77 41 0 152 Abax parallelepipedus F 2 41 0 20 81 0 144 Cychrus semigranosus F 23 35 0 19 15 0 92 Pseudoophonus rufipes G 0 10 45 0 1 25 81 Platyderus rufus G 0 41 6 0 31 1 79 Carabus auronitens F18402610058 Abax carinatus G 0025622154 Abax schueppeli F 4 8 0 23 18 0 53 Leistus rufomarginatus F 528814643 Harpalus latus G 0013001932 Carabus intricatus F 110 0130 024 Leistus piceus G101100921 Harpalus quadripunctatus F00800917 Licinus depressus O001200416 Laemostenus terricola G10900515 Harpalus progrediens O00600612 Trechus quadristriatus G10300711 Notiophilus rufipes G02402210 Carabus arvensis G1500208 Badister bullatus O0020057 Panagaeus bipustulatus G0050016 Amara convexior G0020013 Notiophilus biguttatus G0100023 Platynus assimilis G0120003 Synuchus vivalis G0000213 Amara familiaris G0020002 Carabus convexus G1100002 Cymindis humeralis O0000202 Leistus ferrugineus G0000022 Poecilus cupreus O0110002 Stomis pumicatus G0000022 Amara montivaga O0010001 Amara similata O0010001 Anchomenus dorsalis O0000011 Anysodactylus binotatus O0000011 Calathus melanocephalus O0000011 Harpalus distinguendus O0100001 Loricera pilicornis G0000101 Notiophilus palustris G0000011 Ophonus affinis O0000011 Ophonus cordatus O0000011 Pterostichus macer O0010001 Pterostichus melanarius G0010001 Number of individuals 999 2352 300 1077 1842 401 6971 Number of species 19 22 26 15 21 29 50 Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 4B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx Fig. 1. Ordination (non-metric multidimensional scaling using the Manhattan distance of the relative frequency of the species) of the carabid assemblages along the studied Romanian urbanization gradient based on the catches of pitfall traps in 2004 and 2005. Stress of the two-dimensional configuration was 22.86%. Open symbols denote data from 2004, while filled ones data from 2005. 3.2. Carabid diversity along the gradient 209 The total number of individuals was significantly the highest in210 the suburban area followed by the rural area and it was the low-211 est in the urban area (Fig. 2aand Table 2). The total number of212 carabid species was also significantly the highest in the suburban213 area followed by the rural area and was the lowest in the urban214 area (Fig. 2b and Table 2). After standardizing the sample size by 215 species rarefaction, the species richness was significantly higher 216 20 40 60 d d c b a Number of individuals 2004 2005 a 4 6 8 c c b b a Number of species a Rural Suburban Urban 3 4 5 C B b b a a a Estimated number of species a A Fig. 2. Mean (±SE) values of the total number of carabid individuals (A), the total numberofcarabidspecies(B) andtheestimatednumberofspeciesfor10 individuals (C) along the studied urbanization gradient calculated for the pitfall traps. Different letters indicate significant differences by Tukey test. in the rural and suburban areas than in the urban one. There was 217 no statistically significant difference between the rural and sub218 urban areas (Fig. 2c and Table 2). These findings contradicted the 219 increasing disturbance hypothesis. 220 Both the ratio of forest specialist carabid species and the ratio 221 of their abundance decreased significantly from the rural area 222 towards the urban one (Fig. 3 and Table 2) supporting the habitat 223 specialist hypothesis. An opposite tendency was observed for gen224 eralists. The share of both the generalist species and individuals 225 increased significantly from rural to urban area, albeit difference in 226 the ratio of generalist individuals were not statistically significant 227 betweenthesuburbanandurbanareas(Fig.4andTable2).Ourfind228 ings partially supported the opportunistic species hypothesis. Both 229 the ratio of the open-habitat individuals and species were signifi230 cantly higher in the urban area compared to the rural or suburban 231 ones (Fig. 5 and Table 2), supporting the matrix species hypothesis. 232 4. Discussion 233 The disturbance gradient from rural to urban is a gradient of 234 a number of disturbance events, such as trampling, management, 235 and perhaps pollution. We found that both the species richness and 236 the number of individuals were the highest in the suburban area 237 followed by the rural area and the lowest in the urban area, con238 tradicting the increasing disturbance hypothesis. Proportion of the 239 forest specialists decreased from the rural towards the urban area, 240 supporting the habitat specialist hypothesis. Generalist carabids 241 showed the opposite pattern, supporting the opportunistic species 242 hypothesis. Both the proportion of matrix species and their density 243 were significantly higher in the urban area, supporting the matrix 244 species hypothesis. 245 4.1. Ratios vs. totals 246 Analysing total number of individuals and species richness as an 247 indicator of the impacts of urbanization on invertebrates was not 248 an entirely suitable parameter because given groups of species may 249 suffer (e.g. habitat specialists), while other groups may benefit (e.g. 250 generalists and/or matrix species) from the disturbance and habi251 tatalteration caused by urbanization. Species withdifferent habitat 252 affinities (forest specialists, generalists, matrix species) should be 253 considered separately to detect accurately the diversity pattern 254 along the urbanization gradient (McIntyre, 2000; Magura et al., 255 2004; Magura, Tóthmérész, & Molnár, 2008). The overall impact 256 of urbanization is different on different species, so a more articu257 lated interpretations is not possible using the summary diversity 258 descriptors. These limitations could be resolved by considering the 259 ratios (vs. total numbers) of species with different habitat affinities 260 in an assemblage. 261 4.2. Increasing disturbance hypothesis 262 Increasing disturbance hypothesis predicts that increasing dis263 turbancewould monotonously decrease diversity(Gray,1989).Our 264 results, however, did not support this prediction as the total num265 ber of carabid species was significantly the highest in the suburban 266 area followed by the rural area and was the lowest in the urban 267 area. Some papers published in the frame of the Globenet project 268 also contradicted this hypothesis (Alaruikka et al., 2002; results 269 from Bulgaria in Elek & Lövei, 2007; Magura et al., 2004; Niemelä 270 et al., 2002), whereas others supported it (results from Canada 271 and Finland in Gaublomme, Hendrickx, Dhuyvetter, & Desender, 272 2008; Ishitani et al., 2003; Niemelä et al., 2002; Sadler et al., 2006; 273 Venn et al., 2003). As there is a significant relationship between 274 the trapped number of individuals and the collected number of 275 Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx 5 Table 2 Nested ANOVA with repeated measures showing differences in total number of individuals and species, in estimated species richness and in proportion of forest specialist, generalist and open-habitat individuals and species along the rural–urban gradient and among the 12 sites. Year=the effect of study year (2004 and 2005). Variable Source df MS F p Total number of individuals Between-subjects effects Gradient 2 38703.30 214.66 <0.001 Sites 9 180.30 1.14 ns Error 108 158.70 Within-subjects effects Year 1 456.50 4.68 <0.05 Year×Gradient 2 1499.20 15.37 <0.001 Year×Sites 9 403.70 4.14 <0.001 Error 108 97.50 Total number of species Between-subjects effects Gradient 2 444.87 85.27 <0.001 Sites 9 5.22 2.35 <0.05 Error 108 2.22 Within-subjects effects Year 1 0.42 0.17 ns Year×Gradient 2 6.02 2.48 ns Year×Sites 9 4.87 2.01 <0.05 Error 108 2.42 Estimated number of species richness Between-subjects effects Gradient 2 48.37 8.96 <0.01 Sites 9 5.40 6.11 <0.001 Error 108 0.88 Within-subjects effects Year 1 1.12 0.92 ns Year×Gradient 2 4.72 3.85 <0.05 Year×Sites 9 1.39 1.14 ns Error 108 1.23 Proportion of forest individuals Between-subjects effects Gradient 2 2.57 12.81 <0.01 Sites 9 0.20 7.16 <0.001 Error 108 0.03 Within-subjects effects Year 1 0.05 2.74 ns Year×Gradient 2 0.31 17.10 <0.001 Year×Sites 9 0.05 2.65 <0.01 Error 108 0.02 Proportion of forest species Between-subjects effects Gradient 2 3.50 45.98 <0.001 Sites 9 0.08 2.97 <0.01 Error 108 0.03 Within-subjects effects Year 1 0.01 0.60 ns Year×Gradient 2 0.08 3.52 <0.05 Year×Sites 9 0.06 2.78 <0.01 Error 108 0.02 Proportion of generalist individuals Between-subjects effects Gradient 2 1.95 15.80 <0.01 Sites 9 0.12 4.31 <0.001 Error 108 0.03 Within-subjects effects Year 1 0.03 2.17 ns Year×Gradient 2 0.26 16.74 <0.001 Year×Sites 9 0.05 3.36 <0.01 Error 108 0.02 Proportion of generalist species Between-subjects effects Gradient 2 2.06 38.28 <0.001 Sites 9 0.05 2.12 <0.05 Error 108 0.03 Within-subjects effects Year 1 0.01 0.38 ns Year×Gradient 2 0.06 2.92 ns Year×Sites 9 0.06 2.71 <0.01 Error 108 0.02 Proportion of open-habitat individuals Between-subjects effects Gradient 2 0.11 6.37 <0.05 Sites 9 0.02 6.37 <0.001 Error 108 0.002 Within-subjects effects Year 1 0.001 0.51 ns Year×Gradient 2 0.002 0.67 ns Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 6B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx Table 2 (Continued) Variable Source df MS F p Year×Sites 9 0.002 0.38 ns Error 108 0.003 Proportion of open-habitat species Between-subjects effects Gradient 2 0.22 7.59 <0.05 Sites 9 0.03 4.07 <0.01 Error 108 0.01 Within-subjects effects Year 1 0.001 0.15 ns Year×Gradient 2 0.001 0.21 ns Year×Sites 9 0.0002 0.06 ns Error 108 0.004 species, a possible reason for the inconsistent results is the differ-276 ence in the number of carabid individuals captured by pitfall traps.277 Usingrarefaction, the prediction from the decreasingdiversitywith278 increasing disturbance was not supported: the (rarified) number279 of species was significantly higher in the rural and suburban areas280 than in the urban one. One possible reason of this failure is that the 281 rural–urban gradient is a complex system where many environ-282 mental factors (temperature, moisture, edaphic conditions, acidity,283 pollution, decomposition, etc.) interact (Niemelä, 1999). These fac-284 tors are likely to be different in the studied countries, which could285 lead to variation in responses of carabids along the gradients286 (Ishitani et al., 2003). Moreover, in the modified suburban and/or287 urban areas with increasing edge or edge-like habitats the species288 pattern may be strongly modified (Lövei, Magura, Tóthmérész, &289 Ködöböcz, 2006). A more obvious reason is the diverse responses of290 carabids with different habitat affinities to disturbance. Forest spe291 cialists may suffer, while generalists and species penetrating from292 thesurroundingmatrixmay benefit from the disturbance and habi-293 tatalterationcausedbyurbanization.Forthatreason,itislikelythat294 the overall diversity is not the most appropriate indicator for dis295 turbance. Therefore, species with different habitat affinities should296 be analysed separately to evaluate the real effect of urbanization297 (Magura et al., 2004; Magura, Hornung, et al., 2008).298 4.3. Habitat specialist hypothesis 299 In accordance with the habitat specialist hypothesis, both the300 proportion of individuals and the species of forest specialist cara-301 bids decreased significantly from the rural area towards the urban302 one. All the Globenet papers, which studied forest species sep-303 arately, demonstrated that urbanization caused a pronounced304 change in the assemblages with the strongest effect upon the for-305 est specialist species (Magura, Lövei, & Tóthmérész, 2010; Niemelä 306 & Kotze, 2009). Forest specialist species require microsites with a307 particular kind of environmental heterogeneity, such as favourable308 microclimate, presence of dead and decaying trees, significant309 cover of leaf litter, shrubs and herbs, together forming an undis-310 turbed forest habitat (Desender, Ervynck, & Tack, 1999). Habitat311 alteration caused by urbanization appears to eliminate favourable312 microsites for forest specialists and contributes to the decline of 313 forest specialists’ proportion in the assemblage. Along the stud314 ied gradient, disturbance was the highest in the urban area (paved315 paths, thinned shrub layer), it was moderate in the suburban area316 (dead trees harvested, and fallen trees and branches removed),317 and lowest in the rural area. This decreasing disturbance was also318 expressed by the increased abundance and species richness of for-319 est specialist carabid species. 320 4.4. Opportunistic species hypothesis 321 Opportunistic species hypothesis predicts that species that 322 are able to cope with disturbance would increase their dom-323 0.5 0.6 0.7 0.8 0.9 b c b b a Proportion of forest individuals 2004 2005 a Rural Suburban Urban 0.3 0.4 0.5 0.6 0.7 0.8 Ba a c c b b Proportion of forest species A Fig. 3. Mean (±SE) proportions of the forest specialist individuals (A) and the forest Q2 specialist species (B) along the studied urbanization gradient for the pitfall traps. Different letters indicate significant differences by Tukey test. inance (Gray, 1989). Our results did support this hypothesis, 324 as the proportion of both the individuals and species in gen325 eralists were significantly the highest in the heavily disturbed 326 urban area compared to the other moderately or lightly dis327 turbed suburban and rural areas. Data from Canada (Niemelä 328 et al., 2002), Denmark (Elek & Lövei, 2007), Finland (Niemeläet 329 al., 2002; Venn et al., 2003) and Hungary (Magura et al., 2004)330 supported this prediction, as opportunistic species were domi331 nant; the generalist species were frequent, or their proportion 332 was the highest in the urban areas. There was no difference in 333 the number of generalist individuals along the rural–urban gradi334 ent in Belgium (Gaublomme et al., 2008) or Japan (Ishitani et al., 335 2003), and none of the species gained clear dominance in the 336 urban area in Bulgaria (Niemelä et al., 2002). A surprising pat337 tern was found in Finland where more generalist individuals were 338 collected from rural areas than either urban or suburban ones 339 (Alaruikka et al., 2002). 340 Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx 7 0.1 0.2 0.3 0.4 0.5 B a a bc bc c Proportion of generalist individuals 2004 2005 b A Rural Suburban Urban 0.2 0.3 0.4 0.5 c c b b a Proportion of generalist species a Fig. 4. Mean (±SE) proportions of the generalist individuals (A) and the generalist species (B) along the studied urbanization gradient for the pitfall traps. Different letters indicate significant differences by Tukey test. 0.00 0.03 0.06 0.09 b b a a a Prop. of open-habitat individuals 2004 2005 a Rural Suburban Urban 0.00 0.04 0.08 0.12 B b b a a a Proportion of open-habitat species a A Fig. 5. Mean (±SE) ratios of the open-habitat individuals (A) and the open-habitat species (B) along the studied urbanization gradient for the pitfall traps. Different letters indicate significant differences by Tukey test. 4.5. Matrix species hypothesis 341 Ourresultsdidsupportthishypothesis,astheproportionofboth 342 the individuals and species of open-habitat carabids were signifi343 cantly the highest in the heavily disturbed urban area compared to 344 the other moderately or lightly disturbed suburban and rural areas. 345 The significant alteration of the original habitats in the urban area 346 was reflected by the high number of matrix species in the species 347 pool;stilltheirproportionwerelowcomparedtogeneralistandfor348 estspecies in caseoftraps. In theurbanarea,the forest patcheswith 349 closed canopy and moderate closure because of the walking paths 350 and thinned shrubs allows the colonisation and survival of open351 habitat species. Results concerning the matrix species are reported 352 from Finland and Hungary; open-habitat species were more abun353 dant in the urban area in Finland (Venn et al., 2003) and in Hungary 354 (Magura et al., 2004). Profound changes in habitat quality during 355 urbanization (Gilbert, 1989; Niemelä, 1999) provide possibility to 356 the matrix species to invade the altered urban habitats. Koivula and 357 Niemelä (2003) also pointed out that matrix species can invade dis358 turbedforesthabitatsbecauseofthealterationofabioticfactorsand 359 biotic interactions. 360 4.6. Summary and recommendations 361 Themodifications caused by urbanizationchangedconsiderably 362 the structure of forested habitats. They affected species richness 363 andcommunity structure in urban areas. Diversity of forest special364 ist species adapted to the forest habitats decreased considerably 365 by the increasing urbanization. Regarding the total number of 366 species, this decrease was compensated by the invasion of gen367 eralist and open-habitat species. In the urban area there were open 368 patches produced by walking paths, thinned shrubs and lawn; the 369 open patches allowed the colonisation and survival of open-habitat 370 species and supported generalist species. In the modified suburban 371 and/or urban areas there was an increasing edge or edge-like habi372 tats which also may have a contribution to the increased species 373 richness of these areas. Forest specialist species require microsites 374 with a particular kind of environmental heterogeneity. Thus, it 375 is vital to increase the patchiness of the urban parks and create 376 closed-canopy forest patches with fallen tree trunks, shrubs, herbs 377 and thick litter layer. It is also important to minimize the open 378 patches created by wide paths and/or roads; the asphalt-covered 379 paths/roads are barriers for the carabids and many other compo380 nents of the soil fauna, thus they are especially harmful and paved 381 paths are preferred. 382 Uncited reference Q1 383 Lövei and Sunderland (1996).384 Acknowledgements 385 We are thankful for Gabor Lövei and Johan Kotze for the 386 helpful comments and/or proposal regarding our manuscript. 387 We are grateful for the Sapientia Foundation Inst. Research Pro388 grammes (Romania) supporting the field work (research grant no. 389 1357/2004) and TÁMOP 4.2. IM is grateful for the Domus Hun390 garica Foundation (Hungary) supporting his visit in Debrecen. TM 391 thanks the Bolyai Research Fellowship of the Hungarian Academy 392 of Sciences for supporting this research. 393 References394 Alaruikka, D., Kotze, D. J., Matveinen, K., & Niemelä, J. (2002). Carabid beetle and 395 spider assemblages along a forested urban–rural gradient in Southern Finland. 396 Journal of Insect Conservation,6, 195–206. 397 Please cite this article in press as: Tóthmérész, B., et al. Responses of carabid beetles to urbanization in Transylvania (Romania). Landscape Urban Plan. (2011), doi:10.1016/j.landurbplan.2011.02.038 ARTICLE IN PRESS GModel LAND20101–8 8B. Tóthmérész et al. / Landscape and Urban Planning xxx (2011) xxx–xxx Alberti, M. (2005). The effects of urban patterns on ecosystem function. International 398 Regional Science Review,28, 168–192.399 Antrop, M. (2000). Changing patterns in the urbanized countryside of Western400 Europe. Landscape Ecology,15, 257–270. 401 Berling-Wolff, S., & Wu, J. (2004). Modeling urban landscape dynamics: A review.402 Ecological Research,19, 119–129.403 Breuste, J., Feldmann, H., & Uhlmann, O. (1998). Urban ecology. Berlin: Springer.404 Desender, K., Ervynck, A., & Tack, G. (1999). Beetle diversity and historical ecology405 of woodlands in Flanders. Belgian Journal of Zoology,129, 139–156.406 Elek, Z., & Lövei, G. L. (2007). Patterns in ground beetle (Coleoptera: Carabidae)407 assemblages along an urbanisation gradient in Denmark. Acta Oecologica,32,408 104–111.409 Fernandez-Juricic, E. (2004). Spatial and temporal analysis of the distribution of410 forest specialists in an urban-fragmented landscape (Madrid, Spain) implica411 tions for local and regional bird conservation. Landscape and Urban Planning,69,412 17–32.413 Gaublomme, E., Hendrickx, F., Dhuyvetter, H., & Desender, K. (2008). The effects of414 forest patch size and matrix type on changes in carabid beetle assemblages in415 an urbanized landscape. Biological Conservation,141, 2585–2596.416 Gibb, H., & Hochuli, D. F. (2002). Habitat fragmentation in an urban environment417 large and small fragments support different arthropod assemblages. Biological418 Conservation,106, 91–100.419 Gilbert, O. L. (1989). The ecology of urban habitats. London UK: Chapman and Hall.420 Godefroid, S., & Koedam, N. (2007). Urban plant species patterns are highly driven 421 by density and function of built-up areas. Landscape Ecology,22, 1227–1239.422 Gray, J. S. (1989). Effects of environmental stress on species rich assemblages. Bio-423 logical Journal of the Linnean Society,37, 19–32. 424 Grimm, N. B., Grove, J. M., Pickett, S. T. A., & Redman, C. L. (2000). Integrated425 approaches to long-term studies of urban ecological systems. Bioscience,50,426 571–584.427 Heck,K. L.,van Belle,G., &Simberloff, D.(1975). Explicitcalculationoftherarefaction 428 diversity measurement and the determination of sufficient sample size. Ecology, 429 56, 1459–1461.430 Honnay, O., Piessens, K., Van Landuyt, W., Hermy, M., & Gulinck, H. (2003). Satellite431 based land use and landscape complexity indices as predictors for regional plant432 species diversity. Landscape and Urban Planning,63, 241–250.433 Hornung, E., Tóthmérész, B., Magura, T., & Vilisics, F. (2007). Changes of isopod 434 assemblages along an urban–suburban–rural gradient in Hungary. European435 Journal of Soil Biology,43, 158–165.436 H˚ urka, K. (1996). Carabidae of the Czech and Slovak Republics. Zlin: Kabourek.437 Ishitani, M., Kotze, D. J., & Niemelä, J. (2003). Changes in carabid beetle assemblages438 across an urban–rural gradient in Japan. Ecography,26, 481–489.439 Koivula, M., & Niemelä, J. (2003). Gap felling as a forest harvesting method in boreal 440 forests: Responses of carabid beetles (Coleoptera, Carabidae). Ecography,26,441 179–187.442 Legendre, P., & Legendre, L. (1998). Numerical ecology. Amsterdam: Elsevier Science.443 Lövei,G. L.,Magura, T.,Tóthmérész,B.,&Ködöböcz,V.(2006). Theinfluence ofmatrix444 andedgesonspeciesrichness patternsofground beetles(Coleoptera,Carabidae)445 in habitat islands. Global Ecology and Biogeography,15, 283–289.446 Lövei, G. L., & Sunderland, K. D. (1996). Ecology and behavior of ground beetles447 (Coleoptera Carabidae). Annual Review of Entomology,41, 231–256. 448 Magura, T., Hornung, E., & Tóthmérész, B. (2008). Abundance patterns of terrestrial449 isopods along an urbanization gradient. Community Ecology,9, 115–120.450 Magura, T., Lövei, G. L., & Tóthmérész, B. (2008). Time-consistent rearrangement451 of carabid beetle assemblages by an urbanisation gradient in Hungary. Acta 452 Oecologica,34, 233–243. 453 Magura, T., Tóthmérész, B., Hornung, E., & Horváth, R. (2008). Urbanisation and454 ground-dwelling invertebrates. In L. N. Wagner (Ed.), Urbanization 21st century455 issues and challenges (pp. 213–225). Nova Science Publishers. 456 Magura, T., Tóthmérész, B., & Molnár, T. (2008). A species-level comparison of occur-457 rence patterns in carabids along an urbanisation gradient. Landscape and Urban458 Planning,86, 134–140.459 Magura,T., Lövei,G.L.,&Tóthmérész, B.(2010). Doesurbanisationdecreasediversity 460 in ground beetle (Carabidae) assemblages? Global Ecology and Biogeography,19,461 16–26. Magura, T., Tóthmérész, B., & Molnár, T. (2004). Changes in carabid assem462 blages along an urbanisation gradient. Landscape Ecology,19, 747– 463 759. 464 McDonnell, M. J., & Pickett, S. T. A. (1990). Ecosystem structure and function along 465 urban–rural gradients an unexploited opportunity for ecology. Ecology,71,466 1232–1237. 467 McIntyre, N. E. (2000). Ecology of urban arthropods: A review and a call to action. 468 Annals of the Entomological Society of America,93, 825–835. 469 Miyashita, T., Shinkai, A., & Chida, T. (1998). The effects of forest fragmentation 470 on web spider communities in urban areas. Biological Conservation,86, 357– 471 364. 472 Niemelä, J. (1999). Ecology and urban planning. Biodiversity and Conservation,8,473 119–131. 474 Niemelä, J. (2001). Carabid beetles (Coleoptera: Carabidae) and habitat fragmenta475 tion: A review. European Journal of Entomology,98, 127–132. 476 Niemelä, J., Kotze, J., & Ashworth, A. (2000). The search for common anthropogenic 477 impacts on biodiversity: A global network. Journal of Insect Conservation,4, 3–9. 478 Niemelä, J., & Kotze, J. D. (2009). Carabid beetle assemblages along urban to rural 479 gradients: A review. Landscape and Urban Planning,92, 65–71. 480 Niemelä,J., Kotze,J. D.,Venn, S.,Penev,L.,Stoyanov,I.,Spence,J.,etal. (2002).Carabid 481 beetle assemblages (Coleoptera, Carabidae) across urban–rural gradients: An 482 international comparison. Landscape Ecology,17, 387–401. 483 Pickett, S. T. A., Cadenasso, M. L., & Grove, J. M. (2001). Urban ecological sys484 tems: Linking terrestrial ecological, physical, and socioeconomic components 485 of metropolitan areas. Annual Review of Ecology, Evolution, and Systematics,32,486 127–157. 487 Sadler, J. P., Small, E. C., Fiszpan, H., Telfer, M. G., & Niemelä, J. (2006). Investigat488 ing environmental variation and landscape characteristics of an urban–rural 489 gradient using woodland carabid assemblages. Journal of Biogeography,33,490 1126–1138. 491 Shochat, E., Stefanov, W. L., & Whitehouse, M. E. A. (2004). Urbanization and spider 492 diversity: Influences of human modification of habitat structure and productiv493 ity. Ecological Applications,14, 268–280. 494 Smith, W., & Grassle, F. J. (1977). Sampling properties of a family of diversity mea495 sures. Biometrics,33, 283–292. 496 Sokal, R. R., & Rohlf, F. J. (1995). Biometry. New York: Freeman. 497 Tóthmérész, B. (1993). NuCoSA 1.0: Number cruncher for community studies and 498 other ecological applications. Abstracta Botanica,17, 283–287. 499 Venn, S. J., Kotze, D. J., & Niemelä, J. (2003). Urbanization effects on carabid diversity 500 in boreal forests. European Journal of Entomology,100, 73–80. 501 Vilisics, F., Elek, Z., & Lövei, G. L. (2007). Composition of terrestrial isopod assem502 blages along an urbanisation gradient in Denmark. Pedobiologia,51, 45–53. 503 Wu, J., & David, J. (2002). A spatially explicit hierarchical approach to modeling 504 complex ecological systems: Theory and applications. Ecological Modelling,153,505 7–26. 506 Wu, J., & Hobbs, R. (2002). Key issues and research priorities in landscape ecology: 507 An idiosyncratic synthesis. Landscape Ecology,17, 355–365. 508 Xu, C., Liu, M., Zhang, C., An, S., Yu, W., & Chen, J. M. (2007). The spatiotemporal 509 dynamics of rapid urban growth in the Nanjing metropolitan region of China. 510 Landscape Ecology,22, 925–937. 511 Béla Tóthmérész is professor of ecology at the Ecological Institute, University of 512 Debrecen.Hisresearch interestsinclude the theoryof diversity,communityecology, 513 and urbanization. 514 István Máthé is assistant professor of the Department of Technical and Natural Sci515 ences at the Sapientia Hungarian University of Transylvania, Romania. His research 516 interests include carabid’s ecology and urbanization. 517 Enik˝ o Balázs is a master of sciences student at the Babes¸-Bolyai University, Cluj 518 Napoca, Romania. Her research interest is the effect of urbanization on Carabids. 519 Tibor Magura is field biologist at the Hortobágy National Park Directorate, and 520 leader of the Carabidology Research Group at the University of Debrecen. His main 521 research interests include the distribution, biogeography and ecology of ground 522 beetles. 523