Abstract. A precise knowledge of the spatial distribution of taxa is essential for decision-making processes in land management and biodiversity conservation, both for present and under future global change scenarios. This is a key base for several scientific disciplines (e.g. macro-ecology, biogeography, evolutionary biology, spatial planning, or environmental impact assessment) that rely on species distribution maps. An atlas summarizing the distribution of European amphibians and reptiles with 50 × 50 km resolution maps based on ca. 85 000 grid records was published by the Societas Europaea Herpetologica (SEH) in 1997. Since then, more detailed species distribution maps covering large parts of Europe became available, while taxonomic progress has led to a plethora of taxonomic changes including new species descriptions. To account for these progresses, we compiled information from different data sources: published in books and websites, ongoing national atlases, personal data kindly provided to the SEH, the 1997 European Atlas, and the Global Biodiversity Information Facility (GBIF). Databases were homogenised, deleting all information except species names and coordinates, projected to the same coordinate system (WGS84) and transformed into a 50 × 50 km grid. The newly compiled database comprises more than 384 000 grid and locality records distributed across 40 countries. We calculated species richness maps as well as maps of Corrected Weighted Endemism and defined species distribution types (i.e. groups of species with similar distribution patterns) by hierarchical cluster analysis using Jaccard's index as association measure. Our analysis serves as a preliminary step towards an interactive, dynamic and online distributed database system (NA2RE system) of the current spatial distribution of European amphibians and reptiles. The NA2RE system will serve as well to monitor potential temporal changes in their distributions. Grid maps of all species are made available along with this paper as a tool for decision-making and conservation-related studies and actions. We also identify taxonomic and geographic gaps of knowledge that need to be filled, and we highlight the need to add temporal and altitudinal data for all records, to allow tracking potential species distribution changes as well as detailed modelling of the impacts of land use and climate change on European amphibians and reptiles.
Aim Maps of species ranges are among the most frequently used distribution data in biodiversity studies. As with any biological data, range maps have some level of measurement error, but this error is rarely quantified. We assessed the error associated with amphibian range maps by comparing them with point locality data.Location Global. MethodsThe maps published by the Global Amphibian Assessment were assessed against two data sets of species point localities: the Global Biodiversity Information Facility (GBIF), and a refined data set including recently published, high-quality presence data from both GBIF and other sources. Range fit was measured as the proportion of presence records falling within the range polygon(s) for each species.Results Using the high-quality point data provided better fit measures than using the raw GBIF data. Range fit was highly variable among continents, being highest for North American and European species (a fit of 84-94%), and lowest for Asian and South American species (a fit of 57-64%). At the global scale, 95% of amphibian point records were inside the ranges published in maps, or within 31 km of the range edge. However, differences among continents were striking, and more points were found far from range edges for South American and Asian species.Main conclusions The Global Amphibian Assessment range maps represent the known distribution of most amphibians well; this study provides measures of accuracy that can be useful for future research using amphibian maps as baseline data. Nevertheless, there is a need for greater investment in the continuous updating and improvement of maps, particularly in the megadiverse areas of tropical Asia and South America.
Aim Habitat loss and degradation are the factors threatening the largest number of amphibian species. However, quantitative measures of habitat availability only exist for a small subset of them. We evaluated the relationships between habitat availability, extinction risk and drivers of threat for the world's amphibians. We developed deductive habitat suitability models to estimate the extent of suitable habitat and the proportion of suitable habitat (PSH) inside the geographic range of each species, covering species and areas for which little or no high-resolution distribution data are available.Location Global.Methods We used information on habitat preferences to develop habitat suitability models at 300-m resolution, by integrating range maps with land cover and elevation. Model performance was assessed by comparing model output with point localities where species were recorded. We then used habitat availability as a surrogate of area of occupancy. Using the IUCN criteria, we identified species having narrow area of occupancy, for which extinction risk is likely underestimated. ResultsWe developed models for 5363 amphibians. Validation success of models was high (94%), being better for forest specialists and generalists than for open habitat specialists. Generalists had proportionally more habitat than forest or open habitat specialists. The PSH was lower for species having small geographical ranges, currently listed as threatened, and for which habitat loss is recognized as a threat. Differences in habitat availability among biogeographical realms were strong. We identified 61 forest species for which the extinction risk may be higher that currently assessed in the Red List, due to limited extent of suitable habitat.Main conclusions Habitat models can accurately predict amphibian distribution at fine scale and allow describing biogeographical patterns of habitat availability. The strong relationship between amount of suitable habitat and extinction threat may help the conservation assessment in species for which limited information is currently available.
Aim The incompleteness of information on biodiversity distribution is a major issue for ecology and conservation. Researchers have made many attempts to quantify the amount of biodiversity that still remains unknown. We evaluated whether models that integrate ecogeographical variables with measures of the effectiveness of sampling can be used to estimate biodiversity patterns (species richness) of reptiles in remote areas that have received limited surveys.Location The Western Palaearctic (Europe, Northern Africa, the Middle East and Central Asia). MethodsWe gathered data on the distribution of turtles, amphisbaenians and lizards. We used regression models integrating spatial autocorrelation (spatial eigenvector mapping and Bayesian autoregressive models) to analyse species richness, and identified relationships between species richness, ecogeographical features and large-scale measures of accessibility. ResultsThe two regression techniques were in agreement. Known species richness was dependent on ecogeographical factors, peaking in areas with high temperature and annual actual evapotraspiration, and intermediate cover of natural vegetation. However, richness declined sharply in the least accessible areas. Our models revealed regions where reptile richness is likely to be higher than currently known, particularly in the biodiversity hotspots in the south of the Arabian Peninsula, the Irano-Anatolian region, and the Central Asian mountains. An independent validation data set, with distribution data collected recently throughout the study region, confirmed that combining accessibility measures with ecogeographical variables allows a good estimate of reptile richness, even in remote areas that have received limited monitoring so far. Some remote regions that support very rich communities are covered very little by protected areas.Main conclusions Integrating accessibility measures into species distribution models allows biologists to identify areas where current knowledge underestimates the actual richness of reptiles. Our study identifies regions requiring future biodiversity research, proposes a novel approach to biodiversity prediction in poorly studied areas, and identifies potential regions for conservation.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.