Volume 17, Issue 2 p. 259-272
ORIGINAL ARTICLE
Open Access

Urbanization reduces Orthoptera diversity and changes community structure towards mobile species

Nadja Pernat

Corresponding Author

Nadja Pernat

Institute of Landscape Ecology, University of Münster, Münster, Germany

Centre for Integrative Biodiversity Research and Applied Ecology (CIBRA), University of Münster, Münster, Germany

Correspondence

Nadja Pernat, Institute of Landscape Ecology, University of Münster, Heisenbergstrasse 2, Münster 48149 Germany.

Email: [email protected]

Contribution: Data curation, Writing - original draft, Writing - review & editing

Search for more papers by this author
Sascha Buchholz

Sascha Buchholz

Institute of Landscape Ecology, University of Münster, Münster, Germany

Centre for Integrative Biodiversity Research and Applied Ecology (CIBRA), University of Münster, Münster, Germany

Contribution: Conceptualization, ​Investigation, Funding acquisition, Writing - review & editing, Project administration, Data curation, Resources

Search for more papers by this author
Jens Schirmel

Jens Schirmel

RPTU Kaiserslautern-Landau, iES Landau, Institute for Environmental Sciences, Landau, Germany

Contribution: Conceptualization, Formal analysis, Validation, Data curation, Writing - review & editing, Supervision

Search for more papers by this author
First published: 13 March 2024
Editor/Associate Editor: Catherine Matilda (Tilly) Collins

Abstract

  1. Urban land use is regarded as one major driver for insect declines worldwide. We investigated how Orthoptera diversity and traits respond to the urban matrix and local vegetation parameters.
  2. Orthoptera were collected using a combined method of box-quadrat sampling and pitfall trapping. We sampled 42 dry grasslands in Berlin, Germany, along gradients of urbanization (measured as proportion of sealed surface in the surrounding of the sites) and connectivity. We further included patch size and local vegetation parameters (plant richness, total vegetation cover, herb cover, neophyte cover) in our models.
  3. Urbanization was the main driver explaining Orthoptera species richness, Simpson diversity and functional diversity (functional dispersion), which all decreased with increasing proportion of sealed surface in the surrounding. Urbanization also influenced the Orthoptera species composition. Thereby, functional trait analyses revealed that the urban matrix acted as a habitat filter sorting towards mobile species. However, responses of the most common species showed that the effects of the urban matrix and vegetation parameters were species specific.
  4. Our results demonstrate predominantly negative effects of urbanization on Orthoptera diversity. Nevertheless, urban environments can provide important habitats, in particular for mobile species. Even small and isolated patches of dry grasslands can contribute to Orthoptera conservation in cities.

INTRODUCTION

The pressing problem of global biodiversity loss is manifested in a measurable decline in insect diversity and biomass (Cardoso et al., 2020; Díaz et al., 2019; Wagner et al., 2021). Human-induced land transformation is a major cause of this trend, including the loss and fragmentation of natural habitats associated with urban development (Wagner et al., 2021). Indeed, evidence suggests foremost negative effects of urbanization on arthropod species diversity and abundances as well as communities (Fenoglio et al., 2021; Piano et al., 2020). However, urban areas can also present refuges for endangered or threatened species and for those that migrate to urban areas from highly intensified agricultural areas (Buchholz et al., 2018; Grossmann et al., 2022; Soanes & Lentini, 2019). These heterogeneous findings may result from highly intertwined variables of the urban matrix and from different spatial scales, habitat types and metacommunities (Piano et al., 2020).

Examples for negative effects of urban development on the diversity and abundance include various arthropod groups such as wild bees (Cardoso & Gonçalves, 2018; Herrmann et al., 2023), beetles (Martinson & Raupp, 2013) or butterflies (Herrmann et al., 2023; Ramírez-Restrepo & MacGregor-Fors, 2016,). Similarly, alterations in the species composition in relation to urbanization have been found, for example, for different taxonomic groups in different strata such as spiders in shrub, canopy and bark microhabitats (Chatelain et al., 2023), in overall arthropod litter communities (Van Nuland & Whitlow, 2014) or grassland grasshopper assemblages (Eckert et al., 2017). Cross-taxon approaches also confirmed general negative trends for species diversity (Fenoglio et al., 2020), with the exception of hematophagous and synanthropic arthropods or those that thrive uncontrolled without appropriate insect predators, such as aphids (Chatelain et al., 2023; Fenoglio et al., 2020; Korányi et al., 2020). Urban habitat characteristics associated with the filtering out of species from the regional pool include fragmentation, isolation, connectivity and size of patches, the disturbance by humans and pets as well as climatic conditions and soil characteristics (Fenoglio et al., 2021). Buchholz et al. (2020) reported a negative correlation of taxonomic and functional diversity in bee communities with isolation of grasslands, specifically for endangered bee species. Urban warming presents a potent morphological and physiological filter, for example, body size or heat tolerance (Diamond et al., 2017; Merckx et al., 2018). Fournier et al. (2020) found that in Zurich, Switzerland, mainly mobile bee and carabid beetle species with a tolerance against drought were selected from the regional pool. A trend towards generalist, synanthropic species with high mobility is vaguely emerging (Chatelain et al., 2023; Hahs et al., 2023), with the change in urban species communities being the result of true species change rather than a filtered sample of the rural species pool (Knop, 2016). Similarly, just as declines cannot be generalized to all insect groups, the effects of urbanization on them are not uniform.

With no halt to land grabbing by settlement and urban land in the near future, it is not only important to understand what impacts species diversity and abundance in the city but also reveal the adaptive mechanisms of species to the urban environment in order to develop strategic and workable management and conservation plans. The urban matrix not only filters out species by lacking key features of natural areas to which their evolutionary blueprint is adapted but also by exposing them to novel stressors that are highly concentrated in cities, such as light, noise and pollutants like microplastics (Penone et al., 2013). In addition, novel species communities, for example, caused by biological invasions, can alter biotic species interactions, resulting in even more environmental stress due to increased predation risk or altered food sources (Valentine et al., 2020). These cumulative factors exert strong selection pressures on species' functional traits (Hahs et al., 2023), shaping urban eco-evolutionary processes that may be expressed in plastic or genetic-driven phenotypic changes (Alberti, 2015).

Grasshoppers are sensitive to land use changes, but also feature representatives that can thrive in highly modified habitats, making them suitable indicator species and study subjects for urbanization effects (Rech et al., 2022; Schirmel et al., 2011). Because Orthoptera species are closely associated with grassland plant communities and plant food quality, this group is also suitable for grassland quality assessment (Nakajima & Miyashita, 2021; Schirmel et al., 2019). There are few studies examining the biological and functional diversity of grasshopper species in relation to urban habitat parameters and urban matrix variables. Most studies reported decreasing overall diversity of grasshopper species with increasing urbanization levels (Cherrill, 2015; Melliger et al., 2017; Penone et al., 2013; Piano et al., 2020), but some failed to establish these relationships (Huchler et al., 2022). Vegetation structure and its mowing regime have been shown to be important for urban grasshopper diversity (Eckert et al., 2017; Huchler et al., 2022), but this also applies to rural habitats (Ogan et al., 2022; Schirmel et al., 2019). Patch size is relevant for species richness (Huchler et al., 2022), suggesting that species–area relationships also exist in urban areas similar to rural landscape features (Melliger et al., 2017). However, urbanization affects single grasshopper species differently, which could result in generalists and mobile species being present more and in higher numbers in urban ecosystems than sedentary species and specialists (Penone et al., 2013). For example, Ancillotto and Labadessa (2024) discovered that the local extinction of Orthoptera species in Rome over time is correlated with low mobility and more specialized climatic niches. Conflicting findings often result from the initial rarity of less mobile species in urban areas (Huchler et al., 2022; Melliger et al., 2017), and the scarcity of studies on grasshopper communities in urban areas prevent reliable conclusions.

In this study, we investigate the (functional) diversity of Orthoptera in the German capital Berlin and how environmental factors associated with urbanization influence grasshopper diversity, community composition and trait selection. Due to the potentially positive association between urbanized habitats and the prevalence of generalist and mobile grasshopper species, dispersal and flying ability, habitat size, habitat specialization as well as body size were considered as meaningful functional traits. By collecting Orthoptera species on dry grasslands with varying degrees of urbanization throughout the city of Berlin, we relate urban matrix parameters to different indices of biological and functional diversity. We hypothesize that urbanization and isolation: (i) reduce Orthoptera diversity and density, (ii) elicit species-specific responses in numbers of individuals due to differences in habitat requirements and (iii) affect traits as indicated by changes in community-weighted means.

MATERIALS AND METHODS

Study area and site selection

The study area was the capital Berlin, with an area of 891 km2 and a population of 3.8 million in 2021 the largest city in Germany. Approximately 59% of Berlin's area is built-up, and 41% is green and open space, including forests (18%) and grassland (5%; SenStadtUm, 2016). From the CityScapeLab Berlin, an experimental research platform designed to investigate the effects of urbanization on biodiversity and biotic interactions (von der Lippe et al., 2020), 42 study sites of dry grassland type were selected. These sites extend across the outskirts of Berlin and have emerged and grown on sandy soils on ruderal sites, along roadsides, in forest clearings, or near forests. They are extensively managed by mowing no more than once a year, and no fertilization or irrigation is applied. All sites present the phytosociological vegetation type of Sedo-Scleranthetea (Leuschner & Ellenberg, 2017).

Environmental variables

We applied seven environmental variables at two spatial scales referring to the framework of the CityScapeLab Berlin (von der Lippe et al., 2020). To analyse effects of the urban matrix, we used the proportion of sealed surface within a 500-m buffer around each dry grassland patch as urbanization variable. Share of dry grassland in the surrounding—also within a 500-m buffer—was calculated as connectivity variable. Further, we measured the patch size. For these three measurements we used GIS analyses using QGIS Version 2.18.0 (QGIS Development Team, 2016). At the local scale, we characterized habitat structure by measuring the richness of all vascular plants per site (plant richness), cover of the total vegetation (total vegetation cover), herbal plant species (herb cover) and neophytes (neophyte cover). Classification of plant species into native and nonnative species (neophytes) was based on Seitz et al. (2012).

Sampling of Orthoptera

Orthoptera were sampled in summer 2017 using pitfall traps and a box-quadrat. The box-quadrat is a very effective sampling method for Orthoptera especially in grasslands (Gardiner & Hill, 2006). We used a box-quadrat with a ground area of 1 m2 with gauze-covered sides of 0.8 m height. Box-quadrat sampling was done once per site in September/October, and the box-quadrat was randomly set up 10 times per site (total area sampled site: 15 m2). All individuals within the box-quadrat were captured and directly identified in the field. Afterwards, individuals were released. We additionally used pitfall trapping as it is a useful sampling method in particular for ground-dwelling Orthoptera in open habitats (Schirmel et al., 2010). Pitfall trapping was done during two sampling periods in May/June and August/September for 108 days. Per site, four pitfall traps (opening diameter = 9 cm) were installed and filled with a 4% formalin solution as trapping liquid. Individuals were identified in the lab. As a determination key, we used Bellmann (2006). For statistical analyses, we summed the number of individuals from both the box-quadrat sampling and the pitfall trapping (summing up both sampling periods) to obtain one data point per site.

Data analysis

Functional dispersion (FDis) was used as an index for functional diversity. FDis is unaffected by species richness and considers species relative abundances by estimating their dispersion in a multidimensional trait space (Laliberté & Legendre, 2010). Trait data of Orthoptera included in the calculation of FDis were dispersal ability—the potential movement of one individual from one habitat to another—(1 = low, 2 = medium, 3 = high), flying ability (1 = no, 2 = partly, 3 = yes), required habitat size (1 = small, 2 = medium, 3 = large), habitat specialization (1 = generalist, 2 = specialist) and body size (continuous; according to Maas et al., 2002; Reinhardt et al., 2005). We further calculated the community-weighted means (CWM) of these traits using the FD package (Laliberté et al., 2014).

We used generalized linear models (GLM) to relate Orthoptera diversity, total number of all individuals and of the most common species (Chorthippus dorsatus, Oedipoda caerulescens, Myrmeleotettix maculatus, Chorthippus brunneus and Chorthippus biguttulus) and CWM of traits to urbanization parameters. Explanatory variables in the full models were the proportion of sealed surface in a buffer with 500-m radius around the sites (‘Urban’), the connectivity of the sites within these buffer (‘Connectivity’), the size of the sites (‘Patch size’), the richness of all vascular plants per site (‘Plant richness’) and the cover of the total vegetation (‘Total vegetation cover’), herbal plant species (‘Herb cover’) and neophytes (‘Neophyte cover’) in each site. For count data (species richness, number of individuals), we used GLM with negative binomial error distribution because overdispersion was detected (Zuur, 2009) and for indices (Simpson index, FDis, CWM) Gaussian error distribution. Model selection was done using an information-theoretic approach to multimodel inference (Burnham & Anderson, 2002). For automated model selection, we used the ‘dredge’ function (R package MuMln, Bartoń, 2023). We used the AICc for small sample sizes and selected top-ranked models within Δ AICc <2. Averaged parameter estimates from this top set of models were then produced using the ‘model.avg’ function. We used the full model average, where parameter estimates are averaged over all top-ranked models (ΔAICc <2). This method reduces model selection bias and is appropriate to determine which factors have the strongest effect on the response variable (Burnham & Anderson, 2002; Nakagawa & Freckleton, 2011). Collinearity in the explanatory variables was assessed calculating variation inflation factors (VIF). In all models, predictor variables had VIF values <2.0 indicating low collinearity. Potential outliers and highly influential data points were inspected using model diagnostic plots and the Cook's distance. We have not observed extreme values (Cook's distance >0.5) in any model.

The species composition of Orthoptera was analysed using redundancy analysis (RDA; command ‘rda’ in R package vegan: Oksanen et al., 2022). For the multivariate analyses, we used a log-transformation for the number of individuals of the species. We included the same explanatory variables in the multivariate analysis as in the univariate GLM. Their significance was tested using a permutation test (library ‘vegan’).

RESULTS

General results

In total, we sampled 3750 Orthoptera individuals and 25 species. About 97% of all sampled individuals belonged to the suborder Caelifera (3645 individuals with 15 species), while only about 3% (105 individuals with 10 species) were Ensifera. By far the most common species was Chorthippus mollis (N = 2208) followed by C. dorsatus (N = 415), O. caerulescens (N = 207), M. maculatus (N = 198), C. brunneus (N = 176) and C. biguttulus (N = 144). Most common Ensifera species were Platycleis albopunctata (N = 20), Conocephalus fuscus (N = 19), Gryllus campestris (N = 18) and Bicolorana bicolor (N = 17; Appendix 1).

Orthoptera diversity and individual numbers

Species richness, Simpson diversity and functional diversity (FDis) of Orthoptera were all affected by urbanization and decreased with increasing amount of sealed surface in the surrounding landscape (Table 1, Figure 1a–c). Functional diversity was additionally related to the total vegetation cover and increased towards more densely vegetated dry meadows (Table 1, Figure 1d). In contrast, neither the total number of Orthoptera individuals nor the number of individuals without the dominant species C. mollis were statistically significantly related to any parameter (Table 1).

TABLE 1. Model-averaging results of the top-ranked models for diversity and densities of Orthoptera in dry meadows around Berlin.
Predictor Standardized coefficient Adjusted SE z p
Species richness
Intercept 1.846 0.321 5.745 <0.001
Urban −0.011 0.004 2.970 0.003
Herb cover 0.001 0.003 0.331 0.7441
Total vegetation cover 0.001 0.003 0.317 0.752
Total individuals
Intercept 3.892 0.960 4.055 <0.001
Total vegetation cover 0.009 0.010 0.811 0.417
Plant richness −0.004 0.010 0.397 0.691
Neophyte cover −0.003 0.009 0.308 0.758
Herb cover −0.001 0.004 0.201 0.841
Connectivity −0.443 1.802 0.246 0.806
Total individuals without C. mollis
Intercept 4.052 0.742 5.461 <0.001
Plant richness −0.018 0.021 0.873 0.382
Neophyte cover −0.011 0.019 0.553 0.581
Total vegetation cover 0.001 0.0106 0.240 0.810
Simpson diversity
Intercept 0.295 0.210 1.404 0.160
Total vegetation cover 0.004 0.002 1.482 0.138
Urban −0.005 0.001 3.746 <0.001
Connectivity 0.228 0.563 0.405 0.685
Patch size 0.000 0.000 0.349 0.727
Plant richness −0.003 0.001 0.222 0.824
Functional diversity (FDis)
Intercept 0.069 0.302 0.229 0.819
Total vegetation cover 0.011 0.004 2.802 0.005
Patch size −0.000 0.000 0.755 0.450
Urban −0.011 0.002 5.024 <0.001
Connectivity 0.649 1.159 0.560 0.576
Herb cover 0.002 0.000 0.531 0.595
  • Note: Statistically significant results are shown in bold.
Details are in the caption following the image
Statistically significant negative relationships of Orthoptera: (a) species richness; (b) Simpson diversity; (c) functional diversity (FDis) with the amount of sealed surface in 500 m surrounding of the dry meadows; (d) significant positive relationship between functional diversity (FDis) and the total vegetation cover.

The number of individuals of the most common species showed differential responses to the environmental parameters with the following statistically significant associations. While the dominant species C. mollis was unrelated to any of the parameters, the number of individuals of C. dorsatus grew with increasing total vegetation cover (Table 2, Figure 2a). Also O. caerulescens was related to the vegetation and the number of individuals decreased with increasing herb cover and plant richness (Table 2, Figure 2b,c). Myrmeleotettix maculatus, C. brunneus and C. biguttulus were all negatively affected by urbanization with a decreasing number of individuals with an increasing amount of sealed surface in the surrounding (Table 2, Figure 2e–g). Number of individuals of M. maculatus further increased with higher connectivity of the sites (Table 2, Figure 2d). The number of individuals of C. biguttulus increased with increasing plant richness (Table 2, Figure 2h).

TABLE 2. Model-averaging results of the top-ranked models for densities of the most common Orthoptera species in dry meadows around Berlin.
Predictor Standardized coefficient Adjusted SE z p
Chorthippus mollis (N = 2208)
Intercept 2.570 1.074 2.392 0.017
Connectivity −5.037 4.574 1.101 0.271
Total vegetation cover 0.017 0.012 1.394 0.164
Herb cover −0.001 0.005 0.250 0.802
Chorthippus dorsatus (N = 415)
Intercept −6.963 2.062 3.377 <0.001
Total vegetation cover 0.098 0.023 4.269 <0.001
Connectivity −3.374 6.309 0.535 0.593
Neophyte cover −0.006 0.019 0.324 0.746
Herb cover 0.002 0.009 0.233 0.816
Oedipoda caeruluscens (N = 207)
Intercept 8.292 1.749 4.741 <0.001
Herb cover −0.091 0.028 3.262 0.001
Neophyte cover 0.053 0.056 0.949 0.343
Plant richness −0.095 0.041 2.300 0.021
Myrmeleotettix maculats (N = 208)
Intercept 5.253 2.884 1.822 0.069
Connectivity 41.538 12.774 3.252 0.001
Plant richness −0.067 0.076 0.895 0.371
Urban −0.083 0.034 2.438 0.015
Herb cover −0.024 0.038 0.632 0.527
Total vegetation cover −0.013 0.029 0.437 0.662
Chorthippus brunneus (N = 176)
Intercept 1.935 0.354 5.308 <0.001
Urban −0.030 0.013 2.288 0.022
Patch size −0.000 0.000 0.281 0.779
Chorthippus biguttulus (N = 144)
Intercept −0.919 1.255 0.733 0.464
Plant richness 0.078 0.029 2.667 0.008
Urban −0.037 0.014 2.637 0.008
Neophyte cover −0.024 0.046 0.532 0.595
Total vegetation cover 0.003 0.011 0.273 0.785
  • Note: Statistically significant results are shown in bold.
Details are in the caption following the image
Statistically significant relationships between the densities of the most common Orthoptera species with measured environmental parameters.

CWM of traits

The following statistically significant correlations were found when analysing the CWM. The CWM of the flying ability increased with increasing urbanization (Table 3, Figure 3b). The CWM of dispersal ability and the CWM of the required habitat size were affected by the connectivity and both decreased towards highly connected sites (Table 3, Figure 3a,c). The CWM of body size decreased with increasing plant richness (Table 3, Figure 3d).

TABLE 3. Model-averaging results of the top-ranked models for CWM of traits of Orthoptera in dry meadows around Berlin.
Predictor Standardized coefficient Adjusted SE z p
Dispersal ability
Intercept 2.837 0.088 32.092 <0.001
Connectivity −1.766 0.662 2.577 0.001
Urban 0.001 0.001 1.369 0.171
Plant richness 0.001 0.002 0.617 0.537
Flying ability
Intercept 3.089 0.129 23.912 <0.001
Herb cover −0.002 0.002 1.078 0.281
Urban 0.003 0.001 2.472 0.013
Total vegetation cover −0.001 0.002 0.706 0.480
Required habitat size
Intercept 1.813 0.124 14.599 <0.001
Connectivity −1.750 0.615 2.847 0.004
Plant richness 0.001 0.002 0.591 0.554
Urban 0.000 0.000 0.524 0.600
Total vegetation cover 0.000 0.001 0.459 0.646
Herb cover 0.000 0.000 0.177 0.860
Neophyte cover 0.000 0.000 0.157 0.875
Body size
Intercept 19.799 1.083 18.276 <0.001
Connectivity −3.88 4.985 0.779 0.436
Plant richness −0.042 0.019 2.190 0.029
Total vegetation cover 0.007 0.011 0.577 0.564
Herb cover 0.002 0.007 0.317 0.751
  • Note: Statistically significant results are shown in bold.
Details are in the caption following the image
Statistically significant relationships between the community-weighted means (CWM) of traits of Orthoptera and measured environmental parameters.

Orthoptera species composition

The RDA explained 23.6% of the total variation. Orthoptera species composition was significantly affected by urbanization, plant richness and the total vegetation cover (Table 4). In contrast, patch size, connectivity as well as neophyte and herb cover had no significant influence (Table 4). Almost all Orthoptera species were related to less urbanized sites, which was especially pronounced for M. maculatus, C. brunneus, C. biguttulus and P. parallelus (Figure 4). Chorthippus dorsatus was related to dry meadows with dense and diverse vegetation while the opposite was found for O. caerulescens, M. maculatus and P. albopunctata.

TABLE 4. Species composition of Orthoptera in relation to urbanization, connectivity and local environmental parameters.
Predictor F p
Neophyte cover 0.807 0.593
Connectivity 1.681 0.100
Patch size 0.384 0.930
Urban 1.811 0.046
Plant richness 1.856 0.039
Total vegetation cover 2.443 0.013
Herb cover 1.490 0.155
  • Note: Statistically significant results are shown in bold.
Details are in the caption following the image
Redundancy analysis (RDA) for the Orthoptera species composition in relation to measured environmental parameters in urban meadows in Berlin. (a) Sites and environment. Only environmental variables with statistically significant effects are shown. (b) Orthoptera species. Bb, Bicolorana bicolor; Ca, Chorthippus apricarius; Cal, Chorthippus albomarginatus; Cf, Conocephalus fuscus; Ci, Calliptamus italicus; Dv, Decticus verrucivorus; Eb, Euthystera brachyptera; Gc, Gryllus campestris; Lp, Leptophyes punctatissima; Mm, Meconema meridionale; Ns, Nemobius sylvestris; Pf, Phaneroptera falcata; Ts, Tetrix subulata; Tt, Tetrix tenuicornis; Tv, Tettigonia viridissima. For test statistics, see Table 4.

DISCUSSION

We found that urbanization had a negative effect on alpha and functional diversity of dry grassland Orthoptera. However, responses to urbanization were species specific as well as responses to connectivity and local vegetation parameters. Functional trait analysis showed that the CWM of flight ability increased with increasing urbanization and that dispersal ability decreased with increasing connectivity indicating species sorting towards more mobile Orthoptera in isolated urban grassland sites.

Increasing urbanization has been linked to declines in species diversity in many studies (Chatelain et al., 2023; Fenoglio et al., 2021), including investigations of grasshoppers (Piano et al., 2020). Interestingly, Piano et al. (2020) provided the explanation that lower diversity cannot be explained by a decrease in species numbers per se but also by a decrease in the number of individuals. While we did not find any differences in total abundances along the urbanization gradient, and the functional analyses in particular indicated that a habitat filter select grasshopper species in cities. This does not generally suggest a decline in species but we found a decline of less mobile species on habitat islands in the sealed environment. In this case, a decrease in species diversity resulting from the selection of biological traits such as mobility can be assumed (Fournier et al., 2020). Responses from less mobile species to urbanization are known from other invertebrate groups such as wild bees (Hahs et al., 2023) but also from grasshoppers (Merckx et al., 2018; Penone et al., 2013). In this context, it can be assumed that small species in particular are at a distinct disadvantage because they often have smaller mobility ranges (Poniatowski et al., 2020). In line with that, Merckx et al. (2018) found an increase in the CWM of body size of Orthoptera with increasing urbanization, which was explained by the positive correlation of dispersal ability and size for this insect group. However, our results also showed that increased connectivity may mitigate the decline of less mobile species. For example, M. maculatus—one of the smallest grasshopper species in Germany—was shown to be negatively affected by urbanization but strongly benefitted from a high connectivity of the studied dry grassland sites (Figure 2d,e). Species conservation often argues the need to create corridors or stepping stone habitats, especially in highly fragmented landscapes (Baum et al., 2004; Simberloff et al., 1992). In cities, this is often in contrast to the intended urban densification but on the other hand, it is often the unintended structures that provide opportunities, such as railroad tracks and streets that are important migration corridors for grasshoppers (‘street corridor effect’, see Hahs et al., 2023; Penone et al., 2013). This importance is further enhanced when railroad and street verges are less intensively managed or even include industry and transport brownfields (such as abandoned track fields), as is the case in the Berlin metropolitan region (Eckert et al., 2017).

While some studies found a shift towards smaller body sizes in cities for invertebrates (Merckx et al., 2018; Weller & Ganzhorn, 2004), we only found a relationship between smaller body sizes and higher plant richness. This is mainly related to the abundant occurrence of the relatively large species O. caerulescens in open, sparsely vegetated habitats with low plant richness (see above). Regarding traits, we finally observed that low connectivity is associated with a greater required habitat size, meaning that species depending on larger sizes thrive well also in isolated patches. Interestingly, we did not find any relations between patch size and required habitat size in our analyses. One possible explanation would be the implications of the habitat amount hypothesis (Fahrig & Triantis, 2013), in which species diversity is less related to the degree of isolation and patch size than to the number of potential habitats within a radius of sampling areas. Transferring this hypothesis to functional diversity, it would make sense that there is no relationship between actual and required habitat size: some of the sampling areas are located in the outskirts of the city and are surrounded by nature and landscape reserves, which may have a stronger effect on the trait than the degree of fragmentation approximated by connectivity and patch area. Accordingly, smaller and isolated patches can also contribute to urban Orthoptera conservation (Watling et al., 2020). However, the preservation of large and connected patches of natural habitats should be preferred whenever possible (Piano et al., 2020). Furthermore, it is conceivable that phenotypic adaptations to the urban environment are already evident and that the required habitat size is already undergoing a shift.

From an ecological perspective, the physiologically predetermined niche breadth width often plays a role in the selection of species in the city, and thus a decrease in habitat specialists—species with a narrow niche breadth—is predominantly assumed (Ancillotto & Labadessa, 2024; Bonier et al., 2007; Penone et al., 2013). We can only partially confirm this concept. While M. maculatus, a typical dry grassland specialist (Detzel, 1998), was shown to be an urbanization loser, O. caerulescens—another xerophilous habitat specialist (Detzel, 1998)—was not affected. However, both species strongly differ in mobility. High mobility can compensate possible disadvantages of a strong ecological specialization (Poniatowski et al., 2020), and given that Oedipoda is a relatively large species with high dispersion capacity (Detzel, 1998) it may find more suitable habitats, especially in times of global warming (Ogan et al., 2022). As a typical pioneer species, O. caerulescens prefers open habitats with sparse vegetation (which explains the negative relationship with plant richness in our study) and railroad verges, even in highly urbanized areas, represent important habitat analogous (Schlumprecht & Waeber, 2003). Most of the other abundant species in this study actually yield no surprising results from an ecological point of view. The increase of C. dorsatus with increasing herb layer cover is not surprising, since it is a typical species of mesophilic grasslands (Detzel, 1998)—which is generally the same for C. biguttulus (Detzel, 1998). Finally, C. mollis—a further typical grassland species (Detzel, 1998)—is distributed over the entire spectrum of vegetation structure and is also completely independent of urbanization effects, which makes it a typical urban species for this geographic area. The low proportion of detected Ensifera species in our study can, on the one hand, be explained due to their preference for usually tall vegetation (Ingrisch & Köhler, 1998). Another explanation is our sampling methodology with the use of pitfall traps and the box-quadrat. Both sampling techniques are biased towards ground and herb-dwelling species while species occurring on higher vegetation (i.e., many Ensifera) are underrepresented (Schirmel et al., 2010).

Leaving the functional level, the urban matrix effects or the habitat filter become much less clear and Huchler et al. (2022) as well as Ogan et al. (2022) already postulated that urbanization cannot be the only driver of the urban habitat filter. Indeed, both grasshopper species diversity and abundance of most species as well as the species composition responded strongly to habitat characteristics at the local level, too. Eckert et al. (2017) have already demonstrated the latter for urban brownfields, where grasshopper diversity was even best explained by local characteristics. Habitat heterogeneity is a major driver for species diversity and especially functional diversity can be much higher in structurally rich habitats because many more resources and structures allow for coexistence of many species with different biological traits, which in turn leads to more complex species communities (Schirmel et al., 2012). Increasing habitat heterogeneity in urban habitats may therefore help to counteract to a certain extent the often-described homogenization of urban species communities (Piano et al., 2020). Huchler et al. (2022) have correctly stated that urban species communities are significantly shaped by management and its consequences for habitat structure. This becomes even more evident when not only diverse and multifaceted plant communities with high and dense cover are considered but also highly disturbed areas with open bare ground (Schwarz & Fartmann, 2022).

CONCLUSIONS

We found that urbanization strongly reduced Orthoptera diversity and altered the species composition towards mobile species. This pattern is consistent with several studies of arthropods in urban environments and shows a clear trend suggesting targeted management to protect habitat especially of less mobile Orthoptera species. Our results also showed species-specific relationships with the urban matrix and local vegetation parameters. Besides generalist species, even specialists such as O. caerulescens can use urban habitats. This highlights the importance of studying effects of urbanization not only at the community level but also for individual species. Even though species have a high degree of similarity in their functional characteristics, individual traits determine survival in the urban landscape. Thus, locally adapted management of dry grassland sites, even small sites and isolated ones, can be important for Orthoptera conservation. In addition to further necessary species-specific studies, we also found some unexpected relationships worthy of investigation—for example, it is still unclear whether not only local (e.g., patch size and degree of isolation) but also landscape-level factors (habitat density in the surrounding area) have an impact on taxonomic and functional diversity.

AUTHOR CONTRIBUTIONS

Nadja Pernat: Data curation; writing – original draft; writing – review and editing. Sascha Buchholz: Conceptualization; investigation; funding acquisition; writing – review and editing; project administration; data curation; resources. Jens Schirmel: Conceptualization; formal analysis; validation; data curation; writing – review and editing; supervision.

ACKNOWLEDGEMENTS

We thank Felix Wedel for help with the field work. Open Access funding enabled and organized by Projekt DEAL.

    FUNDING INFORMATION

    The study was funded by the German Federal Ministry of Education and Research (BMBF) within the Collaborative Project ‘Bridging in Biodiversity Science—BIBS’ (funding number 01LC1501).

    CONFLICT OF INTEREST STATEMENT

    The authors declare that they have no conflicts of interest.

    APPENDIX 1: Species list, respective number of individuals and corresponding trait data

    Individuals % sites Dispersal ability Flying ability Required habitat size Body size
    Caelifera
    Caliptamus italicus 20 17 2 3 3 23
    Chorthippus albomarginatus 16 5 3 3 2 20
    Chorthippus apricarius 60 24 2 1 2 19
    Chorthippus biguttulus 144 60 3 3 2 20
    Chorthippus brunneus 176 62 3 3 1 22
    Chorthippus dorsatus 415 74 3 3 2 23
    Chorthippus mollis 2208 100 3 3 2 18
    Euthystera brachyptera 2 2 2 2 2 22
    Myrmeleotettix maculatus 198 31 2 3 1 16
    Oedipoda caerulescens 207 43 3 3 2 25
    Omocestus haemorrhoidalis 46 19 1 3 1 18
    Pseuchorthippus paralellus 105 43 3 2 2 20
    Stenobothrus lineatus 30 29 2 3 2 24
    Tetrix subulata 1 2 3 3 1 12
    Tetrix tenuicornis 17 2 2 3 1 10
    Ensifera
    Bicolaran bicolor 17 10 3 1 2 18
    Conocephalus fuscus 19 14 3 3 1 20
    Decticus verrucivorus 6 5 1 3 3 44
    Gryllus campestris 18 12 1 1 2 26
    Leptophyes punctatissima 3 7 2 1 1 17
    Meconema meridionale 1 2 3 1 1 15
    Nemobius sylvestris 8 7 1 1 2 10
    Phaneroptera falcata 11 12 3 3 2 20
    Platycleis albopunctata 20 17 2 3 3 24
    Tettigonia viridissima 2 2 3 3 3 42

    DATA AVAILABILITY STATEMENT

    The data that support the findings of this study are available from the corresponding author upon reasonable request.