SEARCH

SEARCH BY CITATION

Keywords:

  • Beta diversity;
  • edge effects;
  • fragmentation;
  • meta-community;
  • species turnover

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  1. Local extinction of specialist species due to fragmentation is one of the major causes of biodiversity loss. Increased extinction rates in smaller fragments are expected to result from both smaller local population sizes, which increase the effect of environmental or demographic stochasticity, and increased edge effects. The relative effect sizes of these two factors are still poorly investigated, however.
  2. We attempt to disentangle these effects on ground beetle communities of temperate broadleaved woodland fragments situated in one of the most urbanised regions in Belgium. Assemblages were sampled along transects that extended from 30 m outside to 100 m inside both small and large historical forest fragments.
  3. Although species assemblages within the forest were highly distinct compared to those sampled outside the forest, species turnover along these transects was less pronounced within forest fragments indicating only weak edge effects. The magnitude of edge effects did not differ significantly between large and small fragments. Nevertheless, larger differences in species composition were observed with respect to fragment size, wherein highly specialised species persisted only in the largest fragment.
  4. In summary, increased local extinction processes in smaller fragments, which led to a strong reduction in specialised and wingless forest species, appeared to be the most important factor that drives changes in species composition in this historical and fragmented woodland complex.

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Different forms of anthropogenic land use result in an ever increasing fragmentation and isolation rate of the original habitat (Vitousek et al., 1997) and strongly affect species assemblages and global biodiversity (Barnosky et al., 2011). Fragmentation is presumed to alter species composition of the focal patch by two different processes: increased rates of local extinction by demographic and environmental stochasticity (Hanski, 1998) and strong edge effects resulting from an increase in the proportion of edge to interior habitat (Fagan et al., 1999; Ries et al., 2004; Ewers et al., 2007). This is expected to result in the invasion of generalist species with good dispersal abilities that replace forest specialist species, rendering small patches less suitable for the preservation of forest interior species (Ås, 1999; Magura et al., 2001; Summerville & Crist, 2004; Lövei et al., 2006; Hendrickx et al., 2007, 2009). When edge effects penetrate deeply into the interior habitat, or when patches are very small, there may even be no interior habitat left (Laurance & Yensen, 1991; Tscharntke et al., 2002; Ewers & Didham, 2006).

Given that both area and edge effects have a strong and even interacting effect on ecological dynamics in fragmented landscapes (Tscharntke et al., 2002; Ewers et al., 2007), it is crucial to understand their relative role in determining species composition to effectively implement conservation strategies (Fletcher et al., 2007). Unfortunately, very few studies effectively separate area from edge effects in fragmented forest ecosystems (Didham et al., 1998a,b; Matthews et al., 1999; Galetti et al., 2003; Ewers et al., 2007; Fletcher et al., 2007; Banks-Leite et al., 2010). The majority of these studies on the effect of habitat edges focused on species richness, which may be misleading as disturbances may favour widespread and abundant species, leading to an increase in species richness while providing no information on the species composition (Margules et al., 1994; Davies & Margules, 1998; Gaublomme et al., 2008).

Temperate broadleaved forests in Western Europe provide a good example of a former relatively continuous habitat that suffered from dramatic levels of habitat fragmentation. In Belgium, forests have been under human pressure for about 7000 years (Tack et al., 1993). Their history is therefore one of destruction, fragmentation and degradation, although there have also been periods of forest recovery. Halfway into the 19th century, woodland area reached its minimum (Tack et al., 1993). Brussels – a highly urbanised region in Belgium – occupies 160 km² holding a large population of around 1 million inhabitants. Despite this strong urbanisation, ancient forest fragments still account for 10% of the total area (Gryseels, 1998). With the exception of a few larger fragments, most forest patches are smaller than 50 ha, which poses a serious threat to true forest specialist species (Hermy et al., 1999; Gaublomme et al., 2008). While most research in this region has been conducted on plants (Honnay et al. 1999; Honnay et al., 2002), the distribution of arthropods is less well-known and a more detailed study is required to realistically predict how future changes in the spatial configuration of forest remnants will affect species composition.

In this study, we make use of ground beetles assemblages to investigate how habitat edges and patch area shape beetle assemblages within this forest complex. Ground beetles have repeatedly been shown to be highly valuable to investigate the effects of changes in habitat quality and configuration for a range of habitats (Lövei & Sunderland, 1996; Niemelä et al., 2002; Gaublomme et al., 2008; Hendrickx et al., 2009). We attempt to disentangle both area and edge effects by sampling carabid beetles along transects which were spaced perpendicular to the forest edge, in both small and large fragments of a formerly contiguous forest area. More specifically, we address the following questions: (i) What is the relative effect of edge habitat and forest size in shaping carabid communities of forest fragments in this urbanised area? and (ii) Is the degree of species turnover from forest edge to forest interior related to forest size?

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Sampling

The study was conducted by comparing species composition in 13 transects distributed over ten different forest fragments, situated around the capital Brussels (50°74′–50°89′N, 4°29′–4°41′E) in Belgium (Fig 1; Table 1). Three transects were located in the northern part of the study area, while the remaining ten transects were situated in fragments that historically formed part of the large medieval Kolenwoud forest. While the majority of these fragments were very small (between 5.3 and 43.1 ha), one fragment, that is, the Soignes forest, still has a total area of more than 4000 ha and was sampled with two transects (ZS and ZR; Table 1). Two other transects were situated in the north-western part (ZU and ZX), which was separated by a highway infrastructure about fifty years ago. The sites selected within these forests all have a similar soil type (loam), are dominated with beech (Fagus sylvatica), are at least 230 years old (since the maps of de Ferraris, 1775) and show no evidence of recent severe anthropogenic disturbances. For all studied transects, the edge perimeter constitutes an abrupt boundary between forest vegetation and surrounding matrix, the latter ranging from arable field, meadow, park, garden to pavement. Fragment area was determined from digitalised maps in ArcView GIS 3.0 (Table 1).

Table 1. Site description of the 13 forest fragments
CodeFragmentArea (ha)% Urban areaTransectn Traps
BRBrugman park5.390.32−30/0/30/6012
DUDuden park21.375.620/30/60/10012
DIDielegem park15.274.990/30/60/10012
ZUZoniën urban88.259.310/30/609
POPoelbos7.556.76−30/0/30/60/10015
RORondebos5.456.09−30/0/30/6012
VEVerrewinkel13.943.99−30/0/30/60/10015
KLKleetbos43.142.11−30/0/30/60/10015
ZXZoniën extra88.235.030/30/609
ZSZoniën suburban4383.328.8−30/0/30/60/10015
GAGasthuisbos38.922.32−30/0/30/60/10015
LALaarbeekbos37.915.52−30/0/30/60/10015
ZRZoniën ruraal4383.313.45−30/0/30/60/10015
image

Figure 1. Sampling locations in and around Brussels (inset map shows the location of the sampling area in central Belgium). The dark areas represent forest habitat.

Download figure to PowerPoint

Transects were oriented perpendicular to the forest edge and a total of five plots per transect, each plot consisting of three different pitfall traps (mouth diameter = 95 mm, 90 mm deep), were sampled per site. Plots were situated 30 m outside the forest, at the forest border (0 m), and at 30, 60 and 100 m into the forest interior. A distance of 100 m could not be achieved in all fragments due to size restrictions in the smallest fragment. For four transects located in a highly urbanised environment, no plot could be installed outside the forest (Table 1). Pitfall traps were placed parallel to the forest edge and 5 m apart. All edges had a similar southwest orientation. All 171 pitfall traps were filled with a 4% formaldehyde solution to kill and preserve the collected arthropods. Traps were operational from 1st of March 2002 until 19 November 2002 and emptied fortnightly. Beetles captured at each plot were pooled to obtain a single sample. All adults were identified to species level with the species identification key of Boeken et al. (2002).

Data analyses

Both unconstrained correspondence analysis (CA) and constrained canonical correspondence analysis (CCA) were performed to investigate the role of fragment area and distance along the transect. Given that rare species tend to have a large influence on the ordination results, we excluded all singleton species. None of these singletons represented true forest specialists. Species abundances were square-root transformed to down-weigh the effect of species whose abundances differ strongly among plots. Two species data sets were constructed, that is, one including and one excluding all plots outside the forest habitat. The latter was constructed to obtain a more detailed picture of the differences within forest fragments as species assemblages of plots outside the forest differed remarkably from the interior assemblages (see 'Results').

First, a CA was conducted to identify the main gradients that determine the variation in species composition. Next, we performed a CCA to relate the factors distance to forest edge, log transformed fragment area and their interaction to species composition and tested their significance by means of backward stepwise tests based on 10 000 permutations. Ordination analyses and statistical tests were constructed with the vegan package (Oksanen et al., 2012) in R 2.15.0 (R Development Core Team, 2012).

Since these tests only allow to detect consistent differences in species composition that are shared among all gradients with respect to distance, a second analysis was performed to estimate and test the rate of species turnover along the transects. Here, we calculated pairwise Jaccard dissimilarities between the plot located outside the forest, which is used as reference, and all interior plots for each gradient separately (Anderson et al., 2010). Given that there are five plots per gradient, this resulted in four independent measures of species turnover per gradient, that is, at 0, 30, 60 and 100 m into the forest. Subsequently, we related these measures to the factors distance, fragment area (log transformed), degree of urbanisation of the matrix and all respective interactions by means of a General Linear Mixed model with the mixed procedure in SAS v. 9.1.3 (SAS Institute, Cary, NC, USA). As species turnover along each gradient was expected to increase non-linearly with distance, we also included a quadratic effect of distance. Non-significant model terms were removed in a backward stepwise manner, starting with the highest order interaction terms. Transect was included as random effect, and fixed effects estimates of distance along the transects therefore estimate the rate of species turnover conditional on transects.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

A total of 52 198 adult beetles were collected yielding a total of 99 species, which represents about one quarter of all known ground beetle species in Belgium (Desender et al., 2008).

CA indicates that the most important source of the total variation in species composition, depicted by variation along CA1 (27% explained variance), is the separation of traps located outside the forest (−30 m) compared to those inside the forest habitat (0–100 m; Fig. 2a). Samples located at the edge of the forest habitat (0 m) tend to be situated much closer to the plots located in the forest interior compared to those outside the forest. Hence, changes in species composition between forest interior compared to forest exterior appeared to be fairly abrupt. Along the second axis, which explained 14% of the total variance, a distinct separation is observed between the interior plots of two transects of the large contiguous Soignes forest (ZR and ZS) compared to those of the smaller forest fragments. The latter group also includes the forest interior plots of the two transects in the Soignes fragment that became separated half a century ago by urban infrastructure (ZX and ZU; Fig. 1). Species that were found almost uniquely in the forest interior plots of the large continuous Soignes fragment are Abax ovalis, Carabus problematicus, Carabus auronitens, Pterostichus cristatus and Cychrus attenuatus (Fig. 2b).

image

Figure 2. (a) CA sample plot with symbols indicating the distance of the plot along the gradient perpendicular to the forest edge. Transect abbreviations refer to the forests listed in Table 1. (b) CA species plot. Full names corresponding to the species abbreviations, and species scores of all species are listed in the Supporting Information Table S1.

Download figure to PowerPoint

The importance of fragment area and distance along the transect in structuring species composition was further confirmed by CCA with transect distance, fragment area and their interaction as constraining variables (Fig. 3). These three variables explained 29% of the total variation in species composition (Table 2). Permutation tests revealed a highly significant effect of the main effects fragment area and distance along the transect (Table 2). The effect of distance along the transect differed significantly among fragments of different size and indicates a synergistic effect of distance and area (Fig. 3a).

Table 2. Results of the canonical correspondence analysis (CCA), with distance along the transect (distance) and fragment area (area; logarithmic transformed) and their interaction as explanatory variables, on all sampled plots
Proportion explained variance
 InertiaProportionRank
Total1.99651.0 
Constrained0.58270.2919 3
Unconstrained1.41380.708152
Eigenvalues for constrained axes
CCA 1CCA 2CCA 3 
0.308270.230360.04408 
Permutation test
 d.f.χ2 F P
Distance 10.303611.16750.001
Area 10.19857.20250.001
Distance × area 10.08323.06190.010
Residual521.4138  
image

Figure 3. Canonical correspondence analysis plot with samples, species and environmental variables (fragment area, distance along the transect and their interaction). (a) species data set with all plots included and (b) species data set with plots located outside the forest excluded.

Download figure to PowerPoint

Given that the total variation in species composition is mainly driven by the large assembly differences of plots outside the forest, a second analysis was performed on a restricted data set with the exterior plots removed, allowing us to better quantify information on edge effects within forest fragments. CCA on this restricted data set revealed that only fragment area had a significant effect on species composition, but not the location of the plot along the transect or the interaction between distance along the transect and fragment area (Table 3; Fig 3). Hence, no consistent shifts in species assemblages were observed that discriminate plots situated closer to the edge compared to plots situated at the interior of the forest fragment.

Table 3. Results of the canonical correspondence analysis (CCA), with distance along the transect (distance) and fragment area (area; logarithmic transformed) and their interaction as explanatory variables, on a restricted data set that excluded all plots located outside the forest habitat
Proportion explained variance
 InertiaProportionRank
Total1.82651.0 
Constrained0.36070.1975 3
Unconstrained1.45690.802544
Eigenvalues for constrained axes
CCA 1CCA 2CCA 3 
0.26510.060540.03504 
Permutation test
 d.f.χ2 F P
Distance 10.05361.60910.120
Area 10.26037.81420.001
Distance × area 10.04671.40320.188
Residual441.4659  

The full model relating species turnover to all explanatory variables and their interactions revealed no significant interaction effect of fragment area with distance along the transect, indicating that species turnover along the transects did not differ significantly between larger and smaller fragments (Table 4). Also matrix habitat, expressed as degree of urbanisation, and forest area did not have any effect on the average degree of species dissimilarity between the forest exterior and interior. After removing these effects in a backwards stepwise manner, a final model was constructed with only the main effects distance, a quadratic effect of distance and fragment area. Based on this model, only the effect of distance along the transect appeared to be significant, and the rate of species turnover decreased non-linearly from 0.77 at the forest edge to 0.87 at a distance of 100 m inside the forest (Fig 4).

Table 4. Results of the General Linear Mixed model relating species turnover (Jaccard dissimilarities) to the explanatory variables distance along the transect (distance), fragment area (area, logarithmic transformed), degree of urbanisation of the matrix (urbanisation) and a quadratic effect of distance. The interaction effects [distance × area] and [distance × distance × area] had no significant effect and were removed from in the final model
 d.f. F P EstimateSE
Intercept   0.780.022
Distance23.115.780.0060.00210.0005
Urbanisation7.250.40.70.00030.0007
Area7.020.010.9−9.99E-079.48E-06
Distance × distance23.28.150.009−0.000015.06E-06
[Distance × area]22.20.720.4  
[Distance × distance × area]21.20.30.6  
image

Figure 4. Species turnover along each transect perpendicular to the forest edge. Pairwise Jaccard dissimilarity values calculated between the plots outside the forest (as reference) and all other forest plots along each transect. The bold line represents the predicted values as estimated by means of a General Linear Mixed model.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Our results show that fragment area, rather than edge effects, is the most important variable shaping carabid assemblages within our studied forests. Changes in species composition in fragmented patches are most often explained by the replacement of habitat specialists that are bound to the interior of the patch with species residing in the matrix habitat (Halme & Niëmela, 1993; Ås, 1999; Magura et al., 2001, 2010; Summerville & Crist, 2004; Lövei et al., 2006; Didham et al., 2007; Ewers et al., 2007; Hendrickx et al., 2009). Here, we explicitly tested this phenomenon by comparing differences in species composition between forest interior and matrix habitat for forests of different size. Our results, however, do not corroborate such an effect. Rather, the beetle community found at the edge of the forest was more similar to that of the forest interior than to assemblages typifying the surrounding matrix. Moreover, a similar beetle community was found from close to the forest boundary to the interior with only slight, although significant, edge effects. A similar restriction of edge effects to only the forest border was found in recent study by Kotze et al. (2012). Notwithstanding, assemblages sampled in plots situated outside the forest appeared to be highly distinct from the species assemblages found in both larger as well as smaller forest fragments as reflected by average Jaccard dissimilarities ranging between 0.78 and 0.88. As edge effects were significantly non-linear and decreased monotonically with distance, it is unlikely that edge effects would be more prominent at distances larger than 100 m into the forest.

This general pattern is in accordance with several studies on arthropod communities in forests, where a typical forest fauna is observed even at a close distance to the edge (Heliola et al., 2001; Martin & Major, 2001; Taboada et al., 2004; Basset et al., 2008). As indicated by the ordination plots, the small differences in species composition between forest interior plots and those situated at the edge are most likely due to matrix species residing in the forest border rather than true edge preferring species (Magura et al., 2001; Molnar et al., 2001; Lövei et al., 2006).

As edge effects appeared to be restricted to plots situated at the forest edge, it is remarkable that beetle communities from small forest fragments lack forest specialist species such as A. ovalis, C. problematicus, C. auronitens, P. cristatus and C. attenuates. Moreover, forest area appeared to be the most important variable explaining differences in species composition among all plots located within the forest fragments. Given that all studied fragments are very similar in habitat, soil and forest structure, and that edge effects appear to be restricted to the forest edge, local extinction of true forest specialist due to demographic and environmental stochastic effects are the most likely cause of the strong association between forest area and species composition. The observed relationship is also unlikely to be related to the degree of urbanisation outside the forest fragment, as larger forests did not show a significantly lower degree of urbanisation in the adjacent matrix (Table 1; = −0.48; P = 0.09). Unfortunately, the Soignes forest is the only large fragment that persisted in this region rendering it difficult to generalise the persistence of these forest specialists in larger fragments. Nevertheless, a previous study conducted across 50 fragments scattered throughout the northern part of Flanders revealed that these forest specialist species were restricted to larger fragments (Desender et al., 2002). Besides being strictly bound to core forest habitat, all these species lack functional wings (Desender, 1989; Turin, 2000) rendering them very unlikely to re-colonise vacant forest fragments through the matrix after local extinction events (Hanski, 1998; Roland et al., 2000; Keller & Largiader, 2003). Generalist species are on the other hand expected to better resist habitat fragmentation than forest specialists (Didham et al., 1996). Populations from large and continuous forests are consequently less prone to local extinction as demographic and environmental fluctuations are less likely to result in local population extinction.

The importance of local extinction followed by a lack of re-colonisation was also confirmed in a previous population genetic study conducted on the forest specialist C. problematicus where we showed not only strong effects of forest isolation on population isolation but also a significant lower genetic diversity indicating larger stochastic effects in smaller populations (Gaublomme et al., 2012).

That smaller forests suffer more from edge effects because of the higher ratio of edge per unit area is another widely accepted phenomenon (Didham et al., 1998a,b; Barbosa & Marquet, 2002; Ewers et al., 2007). Our CCA analysis with only plots from inside the forest demonstrates the strength of edge effects to be independent from increasing habitat area, however.

Implications for conservation

We demonstrated that fragmentation in the forests around Brussels causes species extinctions mainly by decreasing the amount of viable core habitat area and as such population sizes, rather than due to the often suggested increasing edge effects (cf. Ewers et al., 2007).

Despite the urban character of the matrix surrounding the large ancient Soignes forest, it still contains typical forest species even close to the edge of the forest. Hence, this suggests that the most important measure to preserve this typical fauna is to retain or enlarge forest fragments as large as possible such that they can serve as a sustainable refuge for specialised and unique species. In an urban environment, there is a clear segregation between forest and surrounding matrix. But, since these sharp edges had an equal influence on beetle communities in small and large fragments, even smaller forest stands may still be valuable and important for conservation, in particular when corridors among these fragments can be developed.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Special thanks to Konjev Desender, who designed the set-up of the experiment and identified a major part of the beetles. He unfortunately passed away during the analysis and writing process of this manuscript. We thank A. Drumont and L. Gaublomme for their assistance during field work. This study was funded by a Ph.D. grant to EG of the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen) and the Belgian Science Policy (BELSPO, MO/36/014). This study was partly conducted within the framework of the Interuniversity Attraction Poles programme IAP (SPEEDY) – Belgian Science Policy. We also thank three anonymous reviewers for their constructive comments on an earlier version of this manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Anderson, M.J., Crist, T.O., Chase, J.M., Vellend, M., Inouye, B.D., Freestone, A.L., Sanders, N.J., Cornell, H.V., Comita, L.S., Davies, K.F., Harrison, S.P., Kraft, N.J.B., Stegen, J.C. & Swenson, N.G. (2010) Navigating the multiple meanings of beta diversity: a roadmap for the practicing ecologist. Ecology Letters, 14, 1928.
  • Ås, S. (1999) Invasion of matrix species in small habitat patches. Conservation Ecology, 3, 1.
  • Banks-Leite, C., Ewers, R. & Metzger, J. (2010) Edge effects as the principal cause of area effects on birds in fragmented secondary forest. Oikos, 119, 918926.
  • Barbosa, O. & Marquet, P. (2002) Effects of forest fragmentaio non the beetle assemblage at the relict forest of Fray Jorge Chile. Oecologia, 132, 296306.
  • Barnosky, A.D., Matzke, N., Tomiya, S., Wogan, G.O.U., Swartz, B., Quental, T.B., Marshall, C., McGuire, J.L., Lindsey, E.L., Maguire, K.C., Mersey, B. & Ferrer, E.A. (2011) Has the Earth's sixth mass extinction already arrived? Nature, 471, 5157.
  • Basset, Y., Missa, O., Alonso, A., Miller, S., Curletti, G., De Meyer, M., Eardley, C., Lewis, O., Mansell, M., Novotny, V. & Wagner, T. (2008) Changes in arthropod assemblages along a wide gradient of disturbance in Gabon. Conservation Biology, 22, 15521563.
  • Boeken, M., Desender, K., Drost, B., Van Gijzen, T., Koese, B., Muilwijk, J., Turin, H. & Vermeulen, R. (2002) De Loopkevers van Nederland & Vlaanderen (Coleoptera: Carabidae). Stichting Jeugdbondsuitgeverij, Utrecht, the Netherlands.
  • Davies, K. & Margules, C. (1998) Effects of habitat fragmentation on carabid beetles: experimental evidence. Journal of Animal Ecology, 67, 460471.
  • Desender, K. (1989) Dispersal and Ecology of Carabid Beetles in Belgium: An Evolutionary Approach. Studiedocumenten van het KBIN 54. Royal Belgian Institute of Natural Sciences, Brussels, Belgium.
  • Desender, K., De Bakker, D., Versteirt, V. & De Vos, B. (2002) A baseline study on forest ground beetle diversity and assemblages in Flanders (Belgium). 10th European Carabidologists' Meeting. (ed. by J. Szysko, P.J. Den Boer and T. Bauer), pp. 237245. Warsaw Agricultural University Press, Warsaw, Poland.
  • Desender, K., Dekoninck, W., Maes, D., Crevecoeur, L., Dufrêne, M., Jacobs, M., Lambrechts, J., Pollet, M., Stassen, E. & Thys, N. (2008) Een nieuwe verspreidingsatlas van de loopkevers en zandloopkevers (Carabidae) in België. Rapporten van het Instituut voor Natuur- en Bosonderzoek 13. Instituut voor Natuur- en Bosonderzoek, Brussel, Belgium.
  • Didham, R., Ghazoul, J., Stork, N. & Davis, A. (1996) Insects in fragmented forests: a functional approach. Trends in Ecology and Evolution, 11, 255260.
  • Didham, R., Hammond, P., Lawton, J., Eggleton, P. & Stork, N. (1998b) Beetle species responses to tropical forest fragmentation. Ecological Monographs, 68, 295323.
  • Didham, R., Lawton, J., Hammond, P. & Eggleton, P. (1998a) Trophic structure stability and extinction dynamics of beetles (Coleoptera) in tropical forest fragments. Philosophical Transactions of the Royal Society of London B, 353, 437451.
  • Didham, R., Tylianakis, J., Gemmel, N., Rand, T. & Ewers, R. (2007) Interactive effects of habitat modification and species invasion on native species decline. Trends in Ecology and Evolution, 22, 489496.
  • Ewers, R. & Didham, R. (2006) Confounding factors in the detection of species responses to habitat fragmentation. Biological Reviews, 81, 117142.
  • Ewers, R., Thorpe, S. & Didham, R. (2007) Synergistic interactions between edge and area effects in a heavily fragmented landscape. Ecology, 88, 96106.
  • Fagan, W., Cantrell, R. & Cosner, C. (1999) How habitat edges change species interactions. The American Naturalist, 153, 165182.
  • Fletcher, R., Ries, L., Battin, J. & Chalfoun, A. (2007) The role of habitat area and edge in fragmented landscapes: definitely distinct or inevitably intertwined? Canadian Journal of Zoology, 85, 10171030.
  • Galetti, M., Alves-Costa, C. & Cazetta, E. (2003) Effects of forest fragmentation anthropogenic edges and fruit colour on the consumption of ornitohocoric fruits. Biological Conservation, 111, 269273.
  • Gaublomme, E., Hendrickx, F., Dhuyvetter, H. & Desender, K. (2008) The effects of forest patch size and matrix type on changes in carabid beetle assemblages in an urbanized landscape. Biological Conservation, 141, 25852596.
  • Gaublomme, E., Maebe, K., Van Doninck, K., Dhuyvetter, H., Li, X., Desender, K. & Hendrickx, F. (2012) Loss of genetic diversity and increased genetic structuring in response to forest area reduction in a ground dwelling insect: a case study of the flightless carabid beetle Carabus problematicus (Coleoptera: Carabidae). Insect Conservation and Diversity. doi: 10.1111/icad.12002.
  • Gryseels, M. (1998) Nature and the green areas in the Brussels Capital Region (Belgium). Studiedocumenten van het KBIN, 93, 1533.
  • Halme, E. & Niëmela, J. (1993) Carabid beetles in fragments of coniferous forest. Annales Zoologici Fennici, 30, 1730.
  • Hanski, I. (1998) Metapopulation dynamics. Nature, 396, 4149.
  • Heliola, J., Koivula, M. & Niemelä, J. (2001) Distribution of carabid beetles (Coleoptera Carabidae) across a boreal forest-clearcut ecotone. Conservation Biology, 15, 370377.
  • Hendrickx, F., Maelfait, J.-P., Desender, K., Aviron, S., Bailey, D., Diekotter, T., Lens, L., Liira, J., Schweiger, O., Speelmans, M., Vandomme, V. & Bugter, R. (2009) Pervasive effects of dispersal limitation on within and among community species richness in agricultural landscapes. Global Ecology and Biogeography, 18, 607616.
  • Hendrickx, F., Maelfait, J.-P., van Wingerden, W., Schweiger, O., Speelmans, M., Aviron, S., Augenstein, I., Billeter, R., Bailey, D., Bukacek, R., Burel, F., Diekötter, T., Dirksen, J., Herzog, F., Liira, J., Roubalova, M., Vandomme, V. & Bugter, R. (2007) How landscape structure land-use intensity and habitat diversity affect components of total arthropod diversity in agricultural landscapes. Journal of Applied Ecology, 44, 340351.
  • Hermy, M., Honnay, O., Firbank, L., Grashof-Bokdam, C. & Lawesson, J. (1999) An ecological comparison between ancient and other forest plant species of Europe and the implications for forest conservation. Biological Conservation, 91, 922.
  • Honnay, O., Endels, P., Vereecken, H. & Hermy, M. (1999) The role of patch area and habitat diversity in explaining native plant species richness in disturbed suburban forest patches in northern Belgium. Diversity and Distributions, 5, 129141.
  • Honnay, O., Verheyen, K. & Hermy, M. (2002) Permeability of ancient forest edges for weedy plant species invasion. Forest Ecology and Management, 161, 109122.
  • Keller, I. & Largiader, C. (2003) Recent habitat fragmentation caused by major roads leads to reduction of gene flow and loss of genetic variability in ground beetles. Proceedings of the Royal Society of London B, 270, 417423.
  • Kotze, J.D., Lehvävirta, S., Koivula, M., O'Hara, R.B. & Spence, J.R. (2012) Effects of habitat edges and trampling on the distribution of ground beetles (Coleoptera, Carabidae) in urban forests. Journal of Insect Conservation, 16, 883897.
  • Laurance, W. & Yensen, E. (1991) Predicting the impacts of edge effects in fragmented habitats. Biological Conservation, 55, 7792.
  • Lövei, G., Magura, T., Tothmeresz, B. & Ködöböcz, V. (2006) The influence of matrix and edges on species richness patterns of ground beetles (Coleoptera: arabidae) in habitat islands. Global Ecology and Biogeography, 15, 283289.
  • Lövei, G. & Sunderland, K. (1996) Ecology and behavior of ground beetles (Coleoptera: Carabidae). Annual Review of Entomology, 41, 231256.
  • Magura, T., Lövei, G. & Tothmeresz, B. (2010) Does urbanization decrease diversity in ground beetle (Carabidae) assemblages? Global Ecology and Biogeography, 19, 1626.
  • Magura, T., Tothmeresz, B. & Molnar, T. (2001) Forest edge and diversity: carabids along forest-grassland transects. Biodiversity and Conservation, 10, 287300.
  • Margules, C., Milkovits, G. & Smith, G. (1994) Contrasting effects of habitat fragmentation on the scorpion Cercophonius squama and an amphipod. Ecology, 75, 20332042.
  • Martin, T. & Major, R. (2001) Changes in wolf spider (Araneae) assemblages across woodland-pasture boundaries in the central wheat-belt of New South Wales – Australia. Austral Ecology, 26, 264274.
  • Matthews, A., Dickman, C. & Major, R. (1999) The influence of fragments size and edge on nest predation in urban bushland. Ecography, 22, 349356.
  • Molnar, T., Magura, T., Tothmeresz, B. & Elek, Z. (2001) Ground beetles (Carabidae) and edge effect in oak-hornbeam forest and grassland transects. European Journal of Soil Biology, 37, 297300.
  • Niemelä, J., Kotze, J., Venn, S., Penev, L., Stoyanov, I., Spence, J., Hartley, D. & Montes de Oca, H. (2002) Carabid beetle assemblages (Coleoptera Carabidae) across urban-rural gradients: an international comparison. Landscape Ecology, 17, 387401.
  • Oksanen, J., Blanchet, F., Kindt, R., Legendre, P., Minchin, P., O'Hara, B., Simpson, G., Solymos, P., Stevens, M. & Wagner, H. (2012) vegan: Community Ecology Package R Package Version 2 0-3. <http://CRAN.R-project.org/package=vegan> 15th January 2013.
  • R Development Core Team (2012) R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. <http://www.R-project.org/> 15th January 2013.
  • Ries, L., Fletcher, R., Battin, J. & Sisk, T. (2004) Ecological responses to habitat edges: mechanisms models and variability explained. Annual Review of Ecology, Evolution and Systematics, 35, 491522.
  • Roland, J., Keyghobadi, N. & Fownes, S. (2000) Alpine Parnassus butterfly dispersal: effects of landscape and population size. Ecology, 81, 16421653.
  • Summerville, K. & Crist, T. (2004) Contrasting effect of habitat quantity and quality on moth communities in fragmented landscapes. Ecography, 27, 312.
  • Taboada, A., Kotze, J. & Salgado, J. (2004) Carabid beetle occurrence at the edges of oak and beech forests in NW Spain European. Journal of Entomology, 101, 555563.
  • Tack, G., Van den Bremt, P. & Hermy, M. (1993) Bossen van Vlaanderen: Een historische ecologische benadering. Davidsfonds, Leuven, Belgium.
  • Tscharntke, T., Steffan-Dewenter, I., Kruess, A. & Thies, C. (2002) Characteristics of insect populations on habitat fragmetns: a mini review. Ecological Research, 17, 229239.
  • Turin, H. (2000) De Nederlandse Loopkevers: verspreiding en oecologie. Nederlandse Fauna 3. KNNV Uitgeverij, Leiden, the Netherland.
  • Vitousek, P., Mooney, H., Luchenco, J. & Melilo, J. (1997) Human domination of earth's ecosystems. Science, 277, 494499.

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
icad12036-sup-0001-TableS1.docxWord document30KTable S1. Full species names of all species captured in the study, together with their acronym (Acronym), their habitat affinity (EU = generalist species and non-forest specialist species; FG = forest generalist species; FS = forest specialist species; OP = open landscape species), species scores on the unconstrained CA analysis depicted in Fig. 2 and their total number of individuals captured.

Please note: Neither the Editors nor Wiley Blackwell are responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.