Aim: Correlative species distribution models (SDMs) are among the most frequently used tools for conservation planning under climate and land-use changes. Conservation-focused climate change studies are often conducted on a national or local level and can use different sources of occurrence records (e.g., local databases, national biodiversity monitoring) collated at different geographic extents. However, little is known about how these restrictions in geographic space (i.e., Wallacean shortfall) can lead to restrictions in environmental space (i.e. Hutchinsonian shortfall) and accordingly affect conclusions about a species’ vulnerability to climate change.

Location: Americas with a focus on Mexico.

Methods: We present an example study constructing SDMs for three Mexican tree species (Alnus acuminata, Liquidambar styraciflua and Quercus xalapensis) using datasets collated at a global (Americas), national (Mexico) and local (cloud forests of eastern Mexico) level to demonstrate the potential effects of a Wallacean shortfall on the estimation of the environmental niche - and thus on a Hutchinsonian shortfall -, its projection in space and time and, consequently, on species’ potential vulnerability to climate change.

Results: The consequence of using the three datasets was species-specific and strongly depended on the extent to which the Wallacean shortfall affected estimations of environmental niches (i.e., Hutchinsonian shortfall). Where restrictions in geographic space lead to an underestimation of the environmental niche, vulnerability to climate change was estimated to be substantially higher. Additionally, the restrictions in geographic space may increase the likelihood of issues with non-analog climates, increasing model uncertainty.

Main Conclusion: We recommend to assess the extent to which a species’ entire realized environmental niche is captured within the target conservation area, and increasing the geographic extent, if needed, to account for environments and occurrences reflecting potential future conditions. This way, the risk of underestimating the climatic potential of the species (i.e., Hutchinsonian shortfall), as well as the errors induced by extrapolation into “locally novel” climates, can be minimised.

The global dataset was created by downloading all occurrence records of the target species from GBIF (GBIF.org 14 January 2020, GBIF Occurrence Download https://doi.org/10.15468/dl.g2yss3) and then cleaning the data by removing those with no geographical coordinates, coordinate uncertainty greater than 1000 m, or incorrect or duplicate coordinates; or with the observation dated prior to 1950. We therefore only kept records with no known coordinate issues, and for which the basis of observation in GBIF was reported as “human observation”, “observation”, “specimen”, “living specimen”, “literature occurrence”, and “material sample”. All records outside of the Americas were removed as these represent non-native locations often associated with (botanical) gardens and parks (e.g., Vetaas, 2002) or other urban areas (Booth, 2017). These highly human-modified environments (e.g. watering or shelter from extreme climate) rather represent the fundamental niche and are often not well reflected by global climate data, and would therefore introduce other niche dimensions or bias to the model.

The national dataset was identical to the global but with occurrence records restricted to Mexico.

The local dataset was compiled by more intensive resource collection, based on previous studies developed in the Laboratory of Plant-Atmosphere Interaction from the Institute of Ecology, National Autonomous University of Mexico (UNAM), the National Commission for the Knowledge and Use of Biodiversity (CONABIO), the National Forestry Commission (CONAFOR), and the United States Forest Service (USDA Forest Service). This local dataset is focused on the cloud forests of eastern Mexico (Figure S1). 

To limit the effects of spatial autocorrelation and sampling bias, we disaggregated the occurrence records in all datasets by removing occurrences closer than 5 km from each other using the R-package spThin (Aiello-Lammens et al., 2015). The final numbers of occurrences per species and dataset are given in Table 1.

See readme.txt for details on data.