Plant species extinction debt in a temperate biodiversity hotspot: Community, species and functional traits approaches

Destruction and fragmentation of (semi-) natural habitats are considered the main causes of biodiversity loss worldwide. Plant species may exhibit a slow response to fragmentation, resulting in the development of an extinction debt in fragmented plant communities. The detection of extinction debt is of primary importance in habitat conservation strategies. We applied two different approaches proposed in the literature to identify extinction debt in South-East Belgium calcareous grasslands. The first method compared species richness between stable and fragmented habitat patches. The second explored correlations between current species richness and current and past landscape configurations using multiple regression analyses. We subsequently examined results generated by both methods. In addition, we proposed techniques to identify species that are more likely to support extinction debt and associated functional traits. We estimated a respective extinction debt of approximately 28% and 35% of the total and specialist species richness. Similar results were obtained from both methods. We identified 15 threatened specialist species under the current landscape configuration. It is likely the landscape configuration no longer supports the species habitat requirements. We demonstrated that non-clonal species are most threatened, as well as taxa that cannot persist in degraded habitats and form only sparsely distributed populations. We discussed our results in light of other studies in similar habitats, and the overall implications for habitat conservation.     

Indirect evidence for an extinction debt of grassland butterflies half century after habitat loss

Habitat loss often leads to a substantial decline in species richness. However, the extinction of species is typically not instant, but rather involves a time lag. Species richness in recently disturbed habitats is therefore expected to reflect past rather than current habitat availability, with the set of species eventually going extinct representing extinction debt. We explored current species richness of butterflies (Lepidoptera: Rhopalocera) and burnet moths (Lepidoptera: Zygaenidae) with respect to past and current habitat characteristics. The study was conducted in calcareous grasslands, a habitat that has suffered a rapid decline in Europe during the last century. We showed that species richness of habitat specialists correlated positively with both past (75 years ago) and current habitat area, and was best explained by a model incorporating both of these variables. The independent effect of past habitat area presumably indicates the presence of extinction debt in the study region. Species requiring large habitat areas were more likely to show signs of extinction debt. Species richness of other grassland butterflies (those not confined to the focal habitat type) was not associated with past or current habitat area, the pattern indicating their lower sensitivity to changes in focal habitats. The response to habitat loss in specialist butterflies appears faster than in specialist plants, studied in the same landscape earlier. We conclude that linking patterns of species diversity with the temporal dynamics of habitats substantially improves the understanding of the variation in species diversity, and thus helps to identify species of conservation concern.     

Detection of extinction debt depends on scale and specialisation

Many plants can persist in landscapes for a long time after focal habitats have disappeared or become fragmented, which might contribute to an extinction debt. Delayed responses of plant occurrence have recently received great attention, particularly in conservation, although evidence for extinction debts is incongruent. Here we asked if we could detect an extinction debt for plant species after 100 years of fragmentation, depending on regional or local (gamma or alpha respectively) diversity measure used, and if all plant species or only habitat specialists were investigated. Historical and contemporary grassland patterns were analysed in 33 rural landscapes (each 1 km² in diameter) in south-eastern Sweden. Results show that managed semi-natural grassland had declined from 39% to 3% in 100 years. Diversity measured at regional scale was best explained by grassland extent 100 years ago, for both all species and grassland specialists. Present-day management, but neither present nor past grassland extent, was important for grassland specialists’ occurrence at the local scale, although present-day grassland proportion had a positive influence on species richness at the local scale. We found evidence of an extinction debt at both local and regional scale when all species were included in the analysis, but not for grassland specialist species at the local scale. However, the extinction debt is still to be settled for grassland specialists at the regional scale, and therefore the estimation of extinction debts in fragmented habitats presents one of the greatest challenges for conservation today and in the future.     

Landscape patterns influencing bird assemblages in a fragmented neotropical cloud forest

The cloud forest is one of the rarest and most threatened ecosystems in Mexico, although it contributes highly to the country's biological diversity and provides important ecological services. It is a naturally fragmented ecosystem, but anthropogenic deforestation and fragmentation has been severe. Consequently, it is essential to identify landscape patterns critical for the conservation of cloud forest. In order to understand how landscape patterns affect diversity in this ecosystem, this study explores the consequences of cloud forest fragmentation on bird diversity in eastern Mexico. I analysed the response of bird species richness and abundance as a function of forest fragment size, shape, topographical complexity, altitudinal range, connectivity, and proportion of landscape forested area in a system of 13 cloud forest fragments. Fragment shape was the main characteristic positively related to species richness in the bird community, but a differential response to landscape patterns was also detected. Fragment size was the main characteristic influencing the segment of the bird community depending mostly on forest, that is to say, forest interior and generalist species. In contrast, the extent of forest edge, expressed as fragment shape, produced a positive response of forest border species. Both, forest dependent and border dependent species positively responded to the extent of their suitable habitat. The immediate and most effective ecologically oriented conservation strategy for the region is the conservation of the largest cloud forest fragments.     

The influence of patch area and connectivity on avian communities in urban revegetation

Landscape restoration through revegetation is being increasingly used in the conservation management of degraded landscapes. To effectively plan restoration programs information is required on how the landscape context of revegetation influences biodiversity gains. Here, we investigate the relative influence of patch area and connectivity on bird species richness and abundance within urban revegetation patches in Brisbane, Australia. We carried out bird surveys at 20 revegetation sites, and used hierarchical partitioning and model selection to test the relative importance of patch area (the area of revegetation including all directly connected remnant vegetation) and landscape connectivity (the vegetated area connected by less than 10 m, 20 m, 30 m, 40 m and 50 m cleared gaps). We controlled for a number of possible confounding variables within the hierarchical partitioning procedure. Both the hierarchical partitioning and model selection procedures indicated that connectivity had an important influence on bird species richness. Patch area in combination with connectivity were important influencing factors on overall bird abundance. We also carried out the hierarchical partitioning procedure for bird abundance data within a range of feeding guilds, yielding results specific to species groups. Overall our data suggest that greater connectivity enhances the habitat area that colonists can arrive from (resulting in greater species richness), whereas increased patch area allows for increased abundance by expanding the habitat available to species already present in a patch. A combined approach where connectivity and overall habitat area is enhanced across the landscape is likely to be necessary to meet long-term conservation objectives.     

Does the heathland flora in north-western Belgium show an extinction debt?

The large losses of heathland area since the end of the 18th century can be expected to have resulted in the decline or even extinction of many characteristic heathland species. Historical data on plant species distribution patterns can provide valuable information in this context. Therefore, the aims of this research are to study how the loss of heathland area has changed the presence of heathland and forest plant species in north-western Belgium using historical plant distribution data, and to test whether the heathland flora shows an extinction debt. Furthermore, plant traits determining extinction sensitivity are investigated.Our results revealed that, despite the dramatic reduction of heathland area (more than 99% of heathland was destroyed over a 230-year period), the loss of heathland species is relatively limited (11%) and is comparable with that of forest species (11%). Heathland species that have a long-term persistent seed bank or can propagate vegetatively are least sensitive to extinction. For forest species, on the other hand, growth form is the key determinant for extinction sensitivity. Long-lived woody species have a greater chance of persisting.The relatively low extinction numbers probably represent an extinction debt and the full effects of habitat loss may not have been fully manifested yet. Consequently, future extinctions are expected to occur unless environmental conditions are improved. Therefore, heathland restoration and prevention of further heathland area losses is required.     

Plant species richness in midfield islets and road verges - The effect of landscape fragmentation

Small marginal habitats in the rural landscape may play an important role for plant species richness as refugias. Little is known how the surrounding landscape and landscape history influence these patterns. I analysed how plant species richness was affected by isolation, habitat area, past and present land use, and if landscape context matters. Plant species occurrence in two different types of small marginal habitats were analysed, road verges and midfield islets. The study was conducted in two different agricultural landscapes in Sweden; one open modern agricultural landscape and one traditional rural landscape, and the results compared. Present and past land use, and landscape change was analysed using aerial photographs and old maps. There was a large grassland reduction more than 50 years ago in the modern landscape, when there still were quite a lot of grasslands left in the traditional landscape. Area and connectivity were more important for plant incidence in small remnant habitats in the modern landscape, compared to the less fragmented, traditional rural landscape. On the other hand there were more grassland specialists, 23% in the traditional landscape compared to 16%. Species richness became higher on midfield islet if grazing was re-introduced. The legacy of surrounding landscape remains in the species pool for a long time, atleast 50 years, even in small grassland fragments. Although small grassland remnants are more sensitive to fragmentation effects compared to larger grasslands, they still encompass a substantial part of the grassland species pool and may be valuable for reconstructing grassland management at a landscape scale.     

The search for fragmentation thresholds in a Southern Sydney Suburb

Fragmentation of habitat is recognised as the number one threat to biodiversity and as such has attracted considerable research. However, much of this research has been conducted in forestry and agricultural environments, with little research in urban areas. In this study, field surveys were conducted measuring the impact of fragmentation on bird, frog, plant and fungi species richness, within the fragmented urban landscape of southern Sydney. Of all fragmentation parameters examined, remnant area was the best and most significant predictor of species richness for all taxa studied. Remnant size thresholds, below which biodiversity declined rapidly, were observed at approximately 4 ha for bird and frog species richness and approximately 2 ha for plant and fungal species richness. A further threshold appears to exist at approximately 50 ha for the dominance of forest interior species. Further relationships were also observed for perimeter:area ratio, indicating the influence of various edge effects on all taxa. Isolation effects were observed in the form of an inverse linear relationship between distance to other large reserves and species richness for fungi, birds and frogs. Corridor connectivity also produced an overall positive relationship for birds, frogs and plants. It is concluded that the identification of fragmentation thresholds and relationships provides an important management tool for the design of networks aimed at conserving biodiversity in fragmented urban environments.     

Habitat fragmentation alters the structure of dung beetle communities in the Atlantic Forest

Habitat loss and fragmentation have turned into the most important threats to biodiversity and ecosystem function worldwide. Here we investigate the effects of habitat fragmentation and drastic changes in tree communities on dung beetle richness and community structure. This study was carried out in a severely fragmented 670-km² forest landscape of the Atlantic Forest of north-eastern Brazil. Sampling was carried out in 19 forest fragments between September 2007 and March 2008 with the use of pitfall traps and flight interception traps. A total of 5893 individuals and 30 species of dung beetle were collected. Fragment area and isolation were the most significant explanatory variables for predictable and conspicuous changes in dung beetle species richness. Smaller and isolated fragments presented lower number of species, but fragments with lower tree species richness and lower proportion of shade-tolerant species were also considerably impoverished in terms of dung beetle species richness. The body mass of dung beetles were explained by fragment area and the percentage of emergent trees with smaller and less stratified fragments being dominated by small-bodied dung beetles. An ordination analysis segregated dung beetle communities between small fragments (<100 ha) and the control area. Seventy-seven percent of the species were recorded in the control area and 22% of all species were unique to this habitat. Our findings indicate that large fragments in the Atlantic Forest appear to consist in a sort of irreplaceable habitats for particular groups of dung beetle species, as well as for the integrity of their communities.     

Low impact of present and historical landscape configuration on the genetics of fragmented Anthyllis vulneraria populations

Decreasing habitat fragment area and increasing isolation may cause loss of plant population genetic diversity and increased genetic differentiation between populations. We studied the relation between the historical and the present landscape configuration (i.e., patch area and patch connectivity), and the present management of calcareous grassland fragments on the one hand, and the within and between population genetic structure of 18 Anthyllis vulneraria populations on the other hand. Despite the long-time fragmentation history and the mainly selfing breeding system of the species, we detected very low genetic differentiation (Φ_st = 0.056) among habitat fragments and no significant isolation-by-distance relation. Average within fragment genetic diversity measured as molecular variance and expected heterozygosity, were relatively high (16.46 and 0.28, respectively), and weakly positively correlated with the current fragment area, most likely because larger fragments contained larger populations. We found no effects of the historical landscape configuration on the genetic diversity of the populations. Our data suggest that the consequences of habitat fragmentation for genetic differentiation and genetic diversity of A. vulneraria are relatively minor which is very likely due to the historical high levels of seed exchange among fragments through grazing and roaming livestock. This study provides indirect evidence that nature management by grazing not only positively affects habitat quality but that it might also mitigate the genetic consequences of habitat fragmentation. From the conservation point of view, this study illustrates the importance of grazing and of the regular transport of livestock between fragments to prevent the long-term effects of fragmentation on the genetic diversity of the populations studied.     

Time-lag in biological responses to landscape changes in a highly dynamic Atlantic forest region

Time-lagged responses of biological variables to landscape modifications are widely recognized, but rarely considered in ecological studies. In order to test for the existence of time-lags in the response of trees, small mammals, birds and frogs to changes in fragment area and connectivity, we studied a fragmented and highly dynamic landscape in the Atlantic forest region. We also investigated the biological correlates associated with differential responses among taxonomic groups. Species richness and abundance for four taxonomic groups were measured in 21 secondary forest fragments during the same period (2000-2002), following a standardized protocol. Data analyses were based on power regressions and model selection procedures. The model inputs included present (2000) and past (1962, 1981) fragment areas and connectivity, as well as observed changes in these parameters. Although past landscape structure was particularly relevant for trees, all taxonomic groups (except small mammals) were affected by landscape dynamics, exhibiting a time-lagged response. Furthermore, fragment area was more important for species groups with lower dispersal capacity, while species with higher dispersal ability had stronger responses to connectivity measures. Although these secondary forest fragments still maintain a large fraction of their original biodiversity, the delay in biological response combined with high rates of deforestation and fast forest regeneration imply in a reduction in the average age of the forest. This also indicates that future species losses are likely, especially those that are more strictly-forest dwellers. Conservation actions should be implemented to reduce species extinction, to maintain old-growth forests and to favour the regeneration process. Our results demonstrate that landscape history can strongly affect the present distribution pattern of species in fragmented landscapes, and should be considered in conservation planning.     

Quantifying the role of multiple landscape-scale drivers controlling epiphyte composition and richness in a conservation priority habitat (juniper scrub)

Recent concern over human-induced climate warming has activated bioclimatic research projecting the species-response to climate change scenarios. However, climate change is one of a range of human-induced environmental drivers controlling biodiversity, and for many species should be considered together within a framework of relevant stresses and threats. This paper critically assesses the sensitivity of epiphyte assemblages to regional gradients in climate, pollution regime and landscape-scale habitat structure (woodland extent and fragmentation). We examine lichen epiphytes associated with juniper scrub (a conservation priority habitat in Europe), sampled across a network of protected sites in Britain (Special Areas of Conservation). Results point to significant differences in associated epiphyte diversity between conservation priority sites. Historic woodland structure was identified as of greater importance than present-day woodland structure in controlling species composition and richness, pointing to an extinction debt among lichen epiphytes. Climatic setting was important in controlling species composition, but not species richness. However, we demonstrate that pollution regime exerts the dominant controlling force for epiphyte assemblages across regional gradients. As a corollary, we caution that for many species groups - for example those sensitive to pollutants, or landscape structure - an exclusive focus on climate is restricting, and that climate change models should expand to include a range of multiple interacting factors.     

Anthropogenic determinants of primate and carnivore local extinctions in a fragmented forest landscape of southern Amazonia

Habitat fragmentation has been shown to influence the abundance, movements and persistence of many species. Here, we examine the effects of forest patch and landscape metrics, and levels of forest disturbance on the patterns of local extinction of five primate and 14 carnivore species within 129 forest patches in a highly fragmented forest landscape of southern Brazilian Amazonia. Classic habitat area effects were the strongest predictors of species persistence, explaining between 42% and 55% of the overall variation in primate and carnivore species richness. Logistic regression models showed that anthropogenic disturbance, including surface wildfires, timber extraction and hunting pressure, had detrimental effects on the persistence of some species over and above those of fragment size. Different species ranged in their responses from highly sensitive to highly tolerant to forest fragmentation. Patterns of local extinction documented here were by no means chance events, and the nestedness of the overall species-by-site matrix was highly nonrandom in terms of the sets of species extirpated from the most to the least species-rich forest patches.     

Insect communities and biotic interactions on fragmented calcareous grasslands—a mini review

We evaluate existing evidence for alternative hypotheses on the effects of fragmentation with special emphasis on insects of calcareous grasslands. Species richness of butterflies in general and the proportion of monophagous species in particular increase with fragment size. Habitat fragmentation disrupts plant-pollinator and predator—prey interactions in some cases. No evidence for changes in the outcome of competitive interactions exists for insects. Habitat connectivity increases inter-patch movement and population density and decreases extinction risk. Habitat quality changes with succession and management. Depending on life-history traits, insect species may profit from early, mid or late successional stages of calcareous grasslands. We conclude from the few well-designed and replicated studies that there is a bias towards modelling approaches and small-scale fragmentation experiments, so more large-scale studies on a community level are needed to quantify the functional roles of insects and their dependence on fragmented calcareous grasslands in the landscape mosaic.     

Small-scale plant species richness and evenness in semi-natural grasslands respond differently to habitat fragmentation

The study explores whether small-scale species diversity, species evenness and species richness in semi-natural grassland communities are similarly associated with present management regime and/or present and historical landscape context (percentage of different land-cover types in the surroundings). Species diversity, evenness and richness were recorded within 441 50 × 50 cm grassland plots in 4.5 × 4.5 km agricultural landscape on Öland, Sweden. Recent and historical land-cover maps (years 2004, 1959, 1938, 1835, and 1800) were used to characterize the present and past landscape context of the sampled vegetation plots. Partial regression and simultaneous autoregressive models were used to explore the relationships between species diversity measures (Shannon diversity, richness and evenness) and different explanatory variables while accounting for spatial autocorrelation in the data. The results indicated that species richness was relatively sensitive to grassland isolation, while the response of species evenness to isolation was characterized by a degree of inertia. Because the richness and evenness components of species diversity may respond differently to habitat fragmentation, we suggest that monitoring projects and empirical studies that focus on changes in biodiversity in semi-natural grasslands should include the assessment of species evenness - as a complement to the assessment of species richness. In addition, our results indicated that the development and persistence of a species-rich and even grassland vegetation was favoured in areas that have historically (in the 19th century) been surrounded by grasslands. Information on landscape history should, whenever possible, be incorporated into the planning of strategies for grassland conservation.     

Plant species richness in the Chaco Serrano Woodland from central Argentina: Ecological traits and habitat fragmentation effects

The Chaco Serrano Woodland from central Argentina has been dramatically reduced during the past 30 years, and is currently confined to several isolates of different size. In this study, we evaluated the effects of forest size, isolation and edge formation on plant species richness. Furthermore, we tested whether plants species with particular ecological traits were differentially affected by habitat fragmentation. Habitat area showed the highest explanatory value for plant species richness in stepwise multiple regressions. The effect of area was most pronounced for rare species, suggesting that large forests are necessary to preserve species with low local or regional abundance. Differences between edge and interior of Chaco Serrano were more pronounced for native and shrub species richness. The analysis of individual species cover revealed that native and biotically pollinated plants were less abundant in woodland edges. Our results showed that forest transformation into smaller remnants has lead to an impoverishment of plant communities, with particular subsets of species defined by ecological traits (rarity, origin and pollination mode) being more susceptible.     

The severe decline of butterflies on western German calcareous grasslands during the last 30 years: A conservation problem

Butterflies and burnet moths are a suitable model species group with which to analyse the general decline of invertebrate biodiversity over the last few decades. In this study, we analysed which ecological groups of butterflies and burnet moths are most affected and how the recent modifications of the landscape have influenced them. Therefore, we studied the species composition of seven calcareous grassland remnants in south-western Germany in 1972 and 2001. We observed a strong change in the community composition and a severe decline in species richness. In general, the incidence of the autochthonous non-ubiquitous species declined by more than 50%, whereas ubiquitous species showed no significant difference in numbers during this period. Especially affected by the decline were those species which need structured habitats, those which are poor dispersers, species which need habitat sizes of 16 ha and more, monophagous species, K strategists and Red Data Book species. Most probably, either habitat outside the reserves is affecting dynamics within the reserves or loss of habitat outside the protected areas has reduced the overall area and connectivity of habitat for some species, increasing extinction rates and reducing colonisation rates in metapopulations. We conclude that these negative trends can only be stopped or even reversed if the landscape structure is made less hostile for species with conservation interest.     

Landscape-level thresholds of habitat cover for woodland-dependent birds

Theory suggests that a disproportionate loss of species occurs when total habitat cover decreases to 10-30% of the landscape. To date, little empirical evidence has been collected to test for such thresholds in habitat cover, especially at the landscape scale. Here, we present empirical data on the species richness of woodland-dependent birds collected systematically from 24 landscapes (each 100 km²) that sample a gradient in habitat cover from <2% to 60%. To compare the relative effects of habitat cover and habitat configuration, landscapes with similar amounts of habitat but contrasting configuration (i.e., aggregated versus dispersed) were surveyed and the richness of woodland-dependent birds collated for each landscape. The relationship between species richness, habitat cover and habitat configuration was examined using analysis of co-variance (ANCOVA), multiple linear regression and univariate non-linear modelling. There was a significant effect of habitat cover (co-variate) in the ANCOVA, but the main treatment effect of configuration was not significant. However, comparison of non-linear models indicated that the shape of the response curve of species loss with decreasing habitat cover differed between aggregated and dispersed landscapes. Species richness was significantly related to habitat cover in all analyses, explaining between 55% and 60% of the variance in regression models. Mean patch shape complexity and the extent of habitat aggregation were also significant explanatory variables, but explained less than 10% of the variance in richness of woodland birds. Biogeographic variables (range in elevation and geographic location) explained up to 14% of the variance in species richness. There was strong evidence for a threshold response in species richness: non-linear models (broken-stick, exponential, inverse) exhibiting a sharp decline in species richness in landscapes with less than 10% habitat cover provided a better fit to the observed data than linear models. To our knowledge, this is the first empirical demonstration of landscape-level thresholds in species richness. We emphasise that thresholds in species richness denote multiple species’ extinction events, the end point of the process of species decline. For viable populations, habitat cover must be maintained well above the threshold level. Finally, thresholds of assemblage measures, such as species richness, potentially mask compositional changes in the avifauna community and may also conceal the loss of species with greater sensitivity to landscape change.     

The effects of forest patch size and matrix type on changes in carabid beetle assemblages in an urbanized landscape

In this study we compared ground beetles (Carabidae) from a range of different forest fragments along an urbanization gradient in Brussels, Belgium. We address the following questions: (i) How does the degree of urbanization in the surrounding habitat affect forest beetles, and does it interact with the effects of patch size and distance to forest edge? (ii) Do these factors have a different effect at the level of individual species, habitat affinity groups or total community? During 2002 we sampled 13 forest plots in 10 forest patches, ranging in size from 5.27 to 4383 ha. The beetles were captured using transects of pitfall traps from the edge to a distance of 100 m into each woodland and identified to species level. Effects of urbanization, forest size and forest edge were evaluated on total species number, abundance and habitat affinity groups and ten abundant, widespread model carabid species. Overall, the effects of urbanization, forest size and edge effects slightly influenced total species richness and abundance but appeared to have a major effect on ground beetle assemblages through species specific responses. More urbanized sites had significantly fewer forest specialists and more generalist species. Large forest fragments were favoured by forest specialist species while generalist species and species frequently associated with forest (forest generalists) dominated the smaller forests. Forest edges mainly harboured generalist species while forest specialist species were more frequent into the forests if the forest patches were large enough, otherwise they disappeared due to the destruction or impoverishment of their habitat. Our results show the importance of differentiating between habitat affinity, especially habitat generalists versus specialists, the latter having a higher value in nature conservation, and merely the quantity of species represented in human-dominated areas.     

Response of a desert lizard community to habitat degradation: Do ideas about habitat specialists/generalists hold?

We report the impact of human-induced desertification on the species richness, abundance, and composition of sand dune flora and herpetofauna of North Sinai, Egypt. Our hypothesis was that degraded habitats would have reduced vegetation complexity, richness, and abundance, and consequently lower reptile species richness and abundance. We also hypothesized that desert lizards would not follow the typical generalist/specialist responses to habitat degradation found in other biomes. Instead, we predicted that because vegetation loss intensifies the environmental extremity of deserts, those species specialized for open and sandy environments would be more likely to persist in desertified habitats than would desert generalists. Our results showed that areas protected from vegetation loss did not have significantly higher vegetation richness or abundance except for only a few species. However, protected sites did have significantly higher percent vegetation cover and height. Habitat protection clearly had strong effects on the reptile community as species richness and abundances were significantly higher in protected sites. The composition of the reptile community between protected and unprotected sites differed significantly. Contrary to past studies in other environments, desert generalist species were not able to persist in degraded sites and were only found in protected sites. Specialist species were ubiquitous in that they occurred in both areas protected and unprotected from vegetation loss. We propose that the effects of disturbance on species composition (specialists or generalists) depends on whether the disturbance exacerbates or reduces environmental harshness and the conditions that favor specialization. In extreme environments, specialist and generalist responses to habitat degradation are opposite to that of more productive environments.