Extinction Debt Research Articles

Connectivity affects species turnover in soil microarthropod communities during Mediterranean forest establishment

AbstractEnvironmental conditions and the spatial arrangement of habitats are thought to play a major role in shaping community assembly during habitat construction and may be key in establishing time lags between habitat establishment and species colonization (colonization credits), and between local species extinction and habitat loss (extinction debts). Despite the importance of soil organisms in ecosystem functioning and their limited dispersal ability, these kinds of shifts in microarthropod communities have rarely been explored. We assessed the roles of the environment and connectivity in soil microarthropod community composition during habitat regeneration in Mediterranean forests. We selected three different forest patches in six study areas: one long‐established (pre‐1956; LONG) control forest patch, and two recent forest patches (post‐1956), one connected (CONN) to and one isolated (ISOL) from the LONG. In each patch, we determined soil temperature, moisture, pH, and fertility and collected soil samples for microarthropod extraction. Our results indicate that the densities of fauna such as oribatid mites and proturans with limited mobility were lower in ISOL than in other forest types; as well, oribatid populations in ISOL were less diverse. Arthropods with greater mobility than the precedent groups, such as collembolans and non‐oribatid mites, were present in lower densities in LONG than in other forest types. Oribatid assemblages in the two recent forest patches were more similar to one another than to those in the LONG. Recent forests were mostly dominated by known pioneer species, while LONG was dominated by the typical species found in well‐developed organic soils. In conclusion, oribatid density and richness had recovered surprisingly well in CONN, more so than in ISOL and to a similar extent as LONG. However, oribatid assemblages in the recent forests are different from those in long‐established forests due to the potential colonization credits and extinction debts triggered by soil development.

Different extinction debts among plants and arthropods after loss of grassland amount and connectivity

A decrease in habitat amount and connectivity causes immediate or delayed species extinctions in transformed landscapes due to reduced functional connections among populations and altered environmental conditions. We assessed the effects of present and historical grassland amount and connectivity as well as local habitat factors typical of the present landscape on the current species richness of grassland specialist and generalist plants and arthropods in grassland fragments. We surveyed herbaceous plants, ants, orthopterans, true bugs and rove beetles in 60 dry grassland fragments in Hungary. We recorded the area of the focal grassland, the slope and the cover of woody vegetation. By using habitat maps of the present and historical landscape, we calculated grassland amount and connectivity for four time periods covering 158 years (1858–2016). We found evidence for an unpaid extinction debt in specialist plants, suggesting that they have not come to equilibrium with the grassland amount and connectivity of the present landscape yet. This localised and typically long-lived group responded slowly to the landscape changes. Specialist arthropod taxa with short generation times responded much faster to habitat loss than plants and did not show an extinction debt. Generalist plants and animals adapted to a wide range of habitats were affected by the landscape-scale decline of grassland habitats. Despite decreased habitat connectivity, grassland fragments with dry environmental conditions and high environmental heterogeneity can sustain specialist plants in transformed landscapes. Unpaid extinction debt should be considered an early warning signal: Restoration of grassland connectivity is necessary to halt ongoing extinction processes.

Slow loss of a foundation species in agricultural landscapes: Effects of nutrients, land clearing, and other factors

Foundation species strongly influence other organisms and declines in foundation species due to disturbance can have impacts that extend throughout communities. Spinifex grasses (Triodia spp.) are foundation species of arid Australia that provide resources to many birds, mammals, reptiles and invertebrates. We tested drivers of Triodia scariosa presence and abundance at local and landscape scales using a Bayesian structural equation model and species distribution modelling, based on the mapped distribution of 11,887 spinifex plants in remnant vegetation across a ~12,000 km2 agricultural region. At the local scale (<1 km), there was a negative effect of soil phosphorus on spinifex abundance, particularly when the prevalence of other grasses was high. These results suggest that the combination of other grasses and nutrient enrichment promote hostile conditions for spinifex in small agricultural remnants, possibly through altered competitive interactions. However, at the landscape scale (100 + km), there was a positive effect of phosphorus on spinifex abundance, reinforcing the importance of multi-scale studies. Spinifex presence at the landscape scale was best determined by proximity to nature reserves. Sites closest to reserves were cleared later than more distant sites, which suggests a trajectory of decline consistent with an extinction debt. Our results correspond with a global phenomenon of delayed biodiversity loss and identify key threats to foundation species in human-dominated landscapes. Given that our proposed mechanisms of loss, nutrient enrichment and possible competition are widespread in agricultural regions throughout the world, understanding how to mitigate threats to foundation species should be a priority.

Looming extinctions due to invasive species: Irreversible loss of ecological strategy and evolutionary history.

Biological invasions are one of the main drivers of biodiversity decline worldwide. However, many associated extinctions are yet to occur, meaning that the ecological debt caused by invasive species could be considerable for biodiversity. We explore extinction scenarios due to invasive species and investigate whether paying off the current extinction debt will shift the global composition of mammals and birds in terms of ecological strategy and evolutionary history. Current studies mostly focus on the number of species potentially at risk due to invasions without taking into account species characteristics in terms of ecological or phylogenetic properties. We found that 11% of phylogenetic diversity worldwide is represented by invasive-threatened species. Furthermore, 14% of worldwide trait diversity is hosted by invasive-threatened mammals and 40% by invasive-threatened birds, with Neotropical and Oceanian realms being primary risk hotspots. Projected extinctions of invasive-threatened species result in a smaller reduction in ecological strategy space and evolutionary history than expected under randomized extinction scenarios. This can be explained by the strong pattern in the clustering of ecological profiles and families impacted by invasive alien species (IAS). However, our results confirm that IAS are likely to cause the selective loss of species with unique evolutionary and ecological profiles. Our results also suggest a global shift in species composition away from those with large body mass, which mostly feed in the lower foraging strata and have an herbivorous diet (mammals). Our findings demonstrate the potential impact of biological invasions on phylogenetic and trait dimensions of diversity, especially in the Oceanian realm. We therefore call for a more systematic integration of all facets of diversity when investigating the consequences of biological invasions in future studies. This would help to establish spatial prioritizations regarding IAS threats worldwide and anticipate the consequences of losing specific ecological profiles in the invaded community.

Changes of rock glacier vegetation in 25 years of climate warming in the Italian Alps

Rock glaciers are periglacial landforms sensitive to climate change, and a harsh environment for vegetation colonization due to the potential occurrence of surface instability. Changes of rock glacier vegetation would provide evidence of the importance of climate change impacts, as the limitation provided by physical disturbance could be overcome by plant responses to warming.We assess, through the resurvey of ten rock glaciers (six active, four inactive) in the Italian Central Alps, the vegetation changes in response to 25 years of climate warming. Vegetation was analyzed through the phytosociological method in 1995 and 2020 adopting the same field protocol for both surveys with 1208 relevés. The relation with climate was assessed by multivariate analysis.Air warming occurred in the study area and vegetation changed both in active and inactive rock glaciers, with increases of species richness, cover, changes in floristic composition. After 25 years the differences between active and inactive rock glaciers persisted but with an appreciable decrease mainly due to the larger vegetation changes showed by active rock glaciers. The multivariate analysis confirmed the relation between vegetation changes and air warming. The observed vegetation changes were compatible also with the impact of the decrease of the physical disturbance limitation likely due rock glacier stabilization, in particular for the active rock glaciers located at lower elevations, while the opposite trend could be hypothesized for those at higher elevations. The changes of floristic composition allowed assess that a gap-filling process, instead of species upward migration, produced the vegetation changes, with species persistence coupled to the recruitment of new species from the neighbor communities. Persistence of climate warming in future could lead to vegetation homogenization and biodiversity loss due to the extinction debt of several alpine species.

SNAPSHOT USA 2019: a coordinated national camera trap survey of the United States.

With the accelerating pace of global change, it is imperative that we obtain rapid inventories of the status and distribution of wildlife for ecological inferences and conservation planning. To address this challenge, we launched the SNAPSHOT USA project, a collaborative survey of terrestrial wildlife populations using camera traps across the United States. For our first annual survey, we compiled data across all 50 states during a 14-week period (17 August-24 November of 2019). We sampled wildlife at 1,509 camera trap sites from 110 camera trap arrays covering 12 different ecoregions across four development zones. This effort resulted in 166,036 unique detections of 83 species of mammals and 17 species of birds. All images were processed through the Smithsonian's eMammal camera trap data repository and included an expert review phase to ensure taxonomic accuracy of data, resulting in each picture being reviewed at least twice. The results represent a timely and standardized camera trap survey of the United States. All of the 2019 survey data are made available herein. We are currently repeating surveys in fall 2020, opening up the opportunity to other institutions and cooperators to expand coverage of all the urban-wild gradients and ecophysiographic regions of the country. Future data will be available as the database is updated at eMammal.si.edu/snapshot-usa, as will future data paper submissions. These data will be useful for local and macroecological research including the examination of community assembly, effects of environmental and anthropogenic landscape variables, effects of fragmentation and extinction debt dynamics, as well as species-specific population dynamics and conservation action plans. There are no copyright restrictions; please cite this paper when using the data for publication.

Towards quantifying the mass extinction debt of the Anthropocene.

To make sense of our present biodiversity crises, the modern rate of species extinctions is commonly compared to a benchmark, or 'background,' rate derived from the fossil record. These estimates are critical for bounding the scale of modern diversity loss, but are yet to fully account for the fundamental structure of extinction rates through time. Namely, a substantial fraction of extinctions within the fossil record occurs within relatively short-lived extinction pulses, and not during intervals characterized by background rates of extinction. Accordingly, it is more appropriate to compare the modern event to these pulses than to the long-term average rate. Unfortunately, neither the duration of extinction pulses in the geological record nor the ultimate magnitude of the extinction pulse today is resolved, making assessments of their relative sizes difficult. In addition, the common metric used to compare current and past extinction rates does not correct for large differences in observation duration. Here, we propose a new predictive metric that may be used to ascertain the ultimate extent of the ongoing extinction threat, building on the observation that extinction magnitude in the marine fossil record is correlated to the magnitude of sedimentary turnover. Thus, we propose that the ultimate number of species destined for extinction today can be predicted by way of a quantitative appraisal of humanity's modification of ecosystems as recorded in sediments-that is, by comparing our future rock record with that of the past. The ubiquity of habitat disruption worldwide suggests that a profound mass extinction debt exists today, but one that might yet be averted by preserving and restoring ecosystems and their geological traces.

Imperfect detection biases extinction‐debt assessments

AbstractFreshwater ecosystems have been substantially altered, threatening the survival and recovery of aquatic species at risk. Estimating the likelihood and magnitude of future extinctions (extinction debt; ED) is integral for conserving biodiversity and requires accurate species composition lists. Using species‐area relationships, we estimated ED for fishes in historically disturbed wetlands in the Lake Erie basin. Then, we used simulated data sets to assess how ED varied when species lists used to derive species‐area relationships had an increasing proportion of undetected species. When species lists were incomplete, ranging from 0.99 to 0.75, 15% fewer wetlands were estimated to have species in ED and, on average, 50% fewer species were expected to go extinct per wetland. Imperfect detection ultimately biased conservation prioritization among wetlands. Our findings suggest that if imperfect detection is not accounted for when projecting future extinctions, the severity of future species loss across a landscape, and the subsequent need for immediate restorative action, can be greatly underestimated.

Relict of primeval forests in an intensively farmed landscape: what affects the survival of the hermit beetle (Osmoderma barnabita) (Coleoptera: Scarabaeidae) in pollard willows?

The hermit beetle (Osmoderma barnabita) is an internationally protected specialist of tree hollows considered a relic of primeval forests. The backbone of its distribution in eastern Czechia, however, is a system of pollard willows in intensively farmed lowlands. Pollarding, a traditional agroforestry practice inducing formation of tree hollows, was partly abandoned in the area during the twentieth century. To assess the state of the system of pollard willows and to investigate parameters of pollards and their stands which affect the beetle’s presence. We inventoried pollard trees and beetle distribution across ca 2000 km2 of lowlands along the Dyje, Morava and Odra rivers and their tributaries in eastern Czechia and westernmost Slovakia. We found 10 441 pollard willows in 324 stands: most stands contained trees of large diameters and poor health; young trees were rare. Probability of O. barnabita presence increased with number of trees in stand, decreased with distance to the nearest occupied stand, and was further affected by the health state of trees. Our results show that high hollow density in pollards allows for the existence of specialized, dispersal limited forest organisms in virtually deforested landscapes. They suggest stands should contain at least 200–300 pollard willows to sustain the beetle population. Although the beetle is still widespread over the study area, the existing populations are isolated and subjected to extinction debt. Large-scale restoration of pollarding practice, planting and pollarding of numerous young trees, and increasing connectivity is vital to facilitate the survival of this hermit beetle population but also support other hollow associated organisms.

Estimating the impacts of habitat loss induced by urbanization on bird local extinctions

Habitat loss and fragmentation can be long-term processes driving extinction debts, especially in cities. Because of a lack of temporal data, long-term impacts are generally investigated less frequently than short-term impacts. However, these effects are equally important for understanding biodiversity loss. We developed a Bayesian Island Biogeography model combining historical imagery and current species distribution to investigate the long-term impacts of urbanization-induced habitat loss and isolation on birds. We surveyed 101 bird species in 43 urban forest patches in a South American city with different combinations of size, isolation, and time since the last abrupt change in area. Based on the current and historical landscape configuration, we estimated the temporal dynamics between species extinction rate and area, and between immigration rate and isolation. Additionally, we used model parameters to forecast bird extinction debts. Large patches had lower extinction rates than small patches, and immigration rates were lower in more isolated patches. Furthermore, current bird species richness was better explained by current landscape configuration than by past configuration, suggesting that birds rapidly respond to changes in urban landscapes and there are no apparent extinction debts. Indeed, we found that habitat loss reduces species diversity to a new equilibrium in relatively short period (less than 10 years), and model projections indicate that bird richness will remain similar to the present if patch area is maintained. Habitat loss is the most important cause of species local extinctions in urban areas, and tropical cities could rapidly promote biodiversity by restoring forest cover.

Gharial nesting in a reservoir is limited by reduced river flow and by increased bank vegetation

The gharial (Gavialis gangeticus Gmelin) is a fish-eating specialist crocodylian, endemic to south Asia, and critically endangered in its few remaining wild localities. A secondary gharial population resides in riverine-reservoir habitat adjacent to the Nepal border, within the Katerniaghat Wildlife Sanctuary (KWS), and nests along a 10 km riverbank of the Girwa River. A natural channel shift in the mainstream Karnali River (upstream in Nepal) has reduced seasonal flow in the Girwa stretch where gharials nest, coincident with a gradual loss of nest sites, which in turn was related to an overall shift to woody vegetation at these sites. To understand how these changes in riparian vegetation on riverbanks were related to gharial nesting, we sampled vegetation at these sites from 2017 to 2019, and derived an Enhanced Vegetation Index (EVI) from LANDSAT 8 satellite data to quantify riverside vegetation from 1988 through 2019. We found that sampled sites transitioned to woody cover, the number of nesting sites declined, and the number of nests were reduced by > 40%. At these sites, after the channel shift, woody vegetation replaced open sites that predominated prior to the channel shift. Our findings indicate that the lack of open riverbanks and the increase in woody vegetation at potential nesting sites threatens the reproductive success of the KWS gharial population. This population persists today in a regulated river ecosystem, and nests in an altered riparian habitat which appears to be increasingly unsuitable for the continued successful recruitment of breeding adults. This second-ranking, critically endangered remnant population may have incurred an "extinction debt" by living in a reservoir that will lead to its eventual extirpation.

Potential extinction debt due to habitat loss and fragmentation in subalpine moorland ecosystems

Habitat loss and fragmentation would often induce delayed extinction, referred to as extinction debt. Understanding potential extinction debts would allow us to reduce future extinction risk by restoring habitats or implementing conservation actions. Although growing empirical evidence has predicted extinction debts in various ecosystems exposed to direct human disturbances, potential extinction debts in natural ecosystems with minimal direct human disturbance are little studied. Ongoing climate change may cause habitat loss and fragmentation, particularly in natural ecosystems vulnerable to environmental change, potentially leading to future local extinctions. Recent climate change would lead to extended growing season caused by earlier snowmelt in spring, resulting in expansion of shrubby species and thereby habitat loss and fragmentation of mountainous moorlands. We examined the potential extinction debts of species diversity and functional diversity (FD; trait variation or multivariate trait differences within a community) in subalpine moorland ecosystems subjected to few direct human disturbances. Plant species richness for all species and for moorland specialists were primarily explained by the past kernel density of focal moorlands (a proxy for spatial clustering of moorlands around them) but not the past area of the focal moorlands, suggesting potential extinction debt in subalpine moorland ecosystems. The higher kernel density of the focal moorland in the past indicates that it was originally surrounded by more neighborhood moorlands and/or had been locally highly fragmented. Patterns in current plant species richness have been shaped by the historical spatial configuration of moorlands, which have disappeared over time. In contrast, we found no significant relationships between the FD and historical and current landscape variables depicting each moorland. The prevalence of trait convergence might result in a less sensitive response of FD to habitat loss and fragmentation compared to that of species richness. Our finding has an important implication that climate change induced by human activities may threaten biodiversity in natural ecosystems through habitat loss and fragmentation.

Past landscape structure drives the functional assemblages of plants and birds

Landscape structure is a major driver of biodiversity in agricultural landscapes. However, the response of biodiversity can be delayed after landscape changes. This study aimed to determine the effect of current and past landscape structure on plant and bird assemblages. We used a trait-based approach to understand their responses to landscape simplification and habitat fragmentation. We quantified landscape structure at three different years (1963, 1985, 2000) and sampled current plant and bird assemblages in twenty 1 km2 landscape windows located along the Seine Valley (France). For each window, we calculated plant and bird species richness, Community Weighted Variance (CWV), and Community Weighted Mean (CWM) of five functional traits related to dispersal capacity, reproduction, and life-cycle. We detected non-random patterns of traits for both taxa. Plant and bird species richness was lower in simple landscapes. The functional variance of plant traits was higher in landscapes simple in configuration. Both plant and bird assemblages strongly responded to past landscapes, especially their traits related to reproduction and life-cycle. It suggests that landscapes of the Seine valley will face a functional extinction debt. Further research is needed to better predict the delayed response of biodiversity expected to occur after landscape structure changes.

Doubling the longevity record of the American Three-toed Woodpecker (Picoides dorsalis)

The American Three-toed Woodpecker (Picoides dorsalis) is a relatively poorly known species due to its remote northern distribution, the low density of its populations, and its cryptic behavior. In this article, we report the observation of an 11-year-old individual, nearly 2 times older than the previous record for the species. This longevity is similar to values reported for the closely related Eurasian Three-toed Woodpecker (P. tridactylus). We report on the observation of a male that was captured, aged, color-banded, and photographed in 2010, and photographed again when it was resighted in 2019. This finding has potential implications for the modeling of population viability and the possibility of extinction debt in marginal habitat.

Plant species surplus in recent peri-urban forests: the role of forest connectivity, species’ habitat requirements and dispersal types

In peri-urban regions, the establishment of new forest patches can be affected by specific community assembly rules driving species extinction debts and colonization credits. This might result in species surpluses or deficits and shifts in species composition along forest establishment, whose association with human-driven landscape transformation remains largely unknown. We have compared plant assemblages in peri-urban Mediterranean forests of contrasting history (recent vs. preexisting; post- vs. pre-1956) according to species’ habitat requirements (forest, scrub-grassland, synanthropic or other) and dispersal mode (vertebrate- vs. non-vertebrate), while considering site and landscape characteristics. Species richness and cover were mostly associated to site characteristics, but also to forest connectivity and history. Connectivity increased total species cover, forest species richness and both richness and cover of vertebrate-dispersed species, while reducing non-forest species richness and synanthropic species cover. Regarding forest history, recent stands showed higher total and non-forest species richness, higher scrub-grassland species cover and lower cover of vertebrate-dispersed species than preexisting forests. Thus, recent stands exhibit a species surplus due to the extinction debt of scrub-grassland species and the transient abundance of synanthropic species, despite the observed colonization credit of vertebrate-dispersed species. These results emphasize the importance of forest connectivity when managing plant species assemblage in peri-urban forest patches but also highlight the need to consider recent history of habitat patches to manage time lags in biodiversity change in human-transformed landscapes.

Changes in a peri‐urban butterfly assemblage over 80 years near Melbourne, Australia

AbstractAlthough changes, particularly declines, in Australian terrestrial insects and other invertebrates have long been suspected and well‐documented for some species, the magnitude, rate and spatial extent of decline remain unclear. Here we use a combination of alternative, qualitative approaches (expert opinion, historical records and temporal replication of surveys) to standardised monitoring and mapping programs to investigate the extent of change of a peri‐urban butterfly assemblage. This assemblage, comprising 52 species, of which 46 are residents or seasonal immigrants, was studied at three spatial scales (local ~0.01 km2, intermediate ~9 km2 and regional ~100 km2) in the Eltham district near Melbourne based on presence/absence data over the past 40 (1981–2020) and 80 years (1941–2020). We then consider the causal factors or drivers that have led to changes, and we explore the timing and ecological patterns underpinning extirpations. Long‐term records reveal substantial changes (mostly decline) in composition and species richness of the 46 breeding species at all spatial scales and time frames analysed. Although the magnitude and rate of decline were higher at the smaller, local to intermediate scales (29–43% decline over 40 years, loss rate of 0.20–0.25 species/year) compared with the larger, regional scale (26% decline over 80 years, loss rate of 0.15 species/year), extirpations at the larger scale were more alarming because they are indicative of widespread population collapse. Declines in relative abundance and occupancy were also recorded at the intermediate and regional spatial scales. Further decline (extinction debt) is anticipated for several ecological specialists currently known from very few sites. Historical extirpations mostly involved obligate myrmecophilous lycaenids and appear to have been largely driven by a combination of urbanisation (habitat loss, degradation and fragmentation) and vegetation encroachment. More recent extirpations consist mainly of grass‐feeding and mistletoe‐feeding specialists and are more worrying because they have predominantly occurred within the past 20–30 years. An interaction of urbanisation (tree canopy death and decline of mistletoes and their host trees from ecosystem dysfunction) and climate change (water stress and heat waves) are considered to be the most likely drivers for these ecological guilds.

How do habitat amount and habitat fragmentation drive time-delayed responses of biodiversity to land-use change?

Land-use change is a root cause of the extinction crisis, but links between habitat change and biodiversity loss are not fully understood. While there is evidence that habitat loss is an important extinction driver, the relevance of habitat fragmentation remains debated. Moreover, while time delays of biodiversity responses to habitat transformation are well-documented, time-delayed effects have been ignored in the habitat loss versus fragmentation debate. Here, using a hierarchical Bayesian multi-species occupancy framework, we systematically tested for time-delayed responses of bird and mammal communities to habitat loss and to habitat fragmentation. We focused on the Argentine Chaco, where deforestation has been widespread recently. We used an extensive field dataset on birds and mammals, along with a time series of annual woodland maps from 1985 to 2016 covering recent and historical habitat transformations. Contemporary habitat amount explained bird and mammal occupancy better than past habitat amount. However, occupancy was affected more by the past rather than recent fragmentation, indicating a time-delayed response to fragmentation. Considering past landscape patterns is therefore crucial for understanding current biodiversity patterns. Not accounting for land-use history ignores the possibility of extinction debt and can thus obscure impacts of fragmentation, potentially explaining contrasting findings of habitat loss versus fragmentation studies.

Linear infrastructure habitats increase landscape-scale diversity of plants but not of flower-visiting insects

Habitats along linear infrastructure, such as roads and electrical transmission lines, can have high local biodiversity. To determine whether these habitats also contribute to landscape-scale biodiversity, we estimated species richness, evenness and phylogenetic diversity of plant, butterfly and bumblebee communities in 32 4 km2 landscapes with or without power line corridors, and with contrasting areas of road verges. Landscapes with power line corridors had on average six more plant species than landscapes without power lines, but there was no such effect for butterflies and bumblebees. Plant communities displayed considerable evenness in species abundances both in landscapes with and without power lines and high and low road verge densities. We hypothesize that the higher number of plant species in landscapes with power line corridors is due to these landscapes having a higher extinction debt than the landscapes without power line corridors, such that plant diversity is declining slower in landscapes with power lines. This calls for targeted conservation actions in semi-natural grasslands within landscapes with power line corridors to maintain biodiversity and prevent imminent population extinctions.

Biodiversity conservation in the sacred groves of north-west Ethiopia: diversity and community structure of woody species

Forest fragmentation is globally pervasive but especially severe in tropical forests, as exemplified by the state of Afrotropical montane forests in Ethiopia, where remnant forests almost exclusively exists as small, isolated fragments centered around churches, forming networks of partially intact ‘habitat islands.’ Church forests deliver ecosystem services such as forest products and erosion control, and are considered a benchmark for assessing forest biodiversity quality. However, safeguarding church forest integrity necessitates a better understanding of the factors determining their biodiversity and community structure. Therefore, we surveyed a network of church forests southeast of Lake Tana to assess factors explaining their tree species richness and composition. We found that the 24 church forests studied here represent about 20% of the ca. 270 tree species that occur in tropical northeast Africa. All surveyed forests however face strong extinction debts, with five tree species at risk of completely disappearing from the region and many more species only regenerating in a limited number of church forests. Church forest integrity partially reflected the anthropogenic pressure on the area, as forest regeneration tended to be stronger in fenced church forests. Seedling abundance depended on the local presence of large, mature conspecific trees and on the geographic distance to potential source populations of seeds, strongly suggest that metapopulation dynamics likely are important. We conclude that church forest conservation and minimizing further degradation of the landscape matrix are needed to help sustaining the ecological and socioeconomic potential of this unique network of remnant forests.

Robustness of a meta‐network to alternative habitat loss scenarios

Studying how habitat loss affects the tolerance of ecological networks to species extinction (i.e. their robustness) is key for our understanding of the influence of human activities on natural ecosystems. With networks typically occurring as local interaction networks interconnected in space (a meta‐network), we may ask how the loss of specific habitat fragments affects the overall robustness of the meta‐network. To address this question, for an empirical meta‐network of plants, herbivores and natural enemies we simulated the removal of habitat fragments in increasing and decreasing order of area, age and connectivity for plant extinction and the secondary extinction of herbivores, natural enemies and their interactions. Meta‐network robustness was characterized as the area under the curve of remnant species or interactions at the end of a fragment removal sequence. To pinpoint the effects of fragment area, age and connectivity, respectively, we compared the observed robustness for each removal scenario against that of a random sequence. The meta‐network was more robust to the loss of old (i.e. long‐fragmented), large, connected fragments than of young (i.e. recently fragmented), small, isolated fragments. Thus, young, small, isolated fragments may be particularly important to the conservation of species and interactions, while contrary to our expectations larger, more connected fragments contribute little to meta‐network robustness. Our findings highlight the importance of young, small, isolated fragments as sources of species and interactions unique to the regional level. These effects may largely result from an unpaid extinction debt, whereby younger fragments are likely to lose species over time. Yet, there may also be more long‐lasting effects from cultivated lands (e.g. water, fertilizers and restricted cattle grazing) and network complexity in small, isolated fragments. Such fragments may sustain important biological diversity in fragmented landscapes, but maintaining their conservation value may depend on adequate restoration strategies.