Format: ORAL

Alejandra Zarzo-Arias1,2,3, Cristina Ronquillo1, Mario Mingarro1,4, Victoria Ochoa5, Rubn G. Mateo3,6, Fernando T. Maestre7,8, Joaqun Hortal1

1 Departamento de Biogeografía y Cambio Global, Museo Nacional de Ciencias Naturales, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain 2 Departamento de Biología de Organismos y Sistemas, Facultad de Biología, Universidad de Oviedo 3 Departamento de Biología (Botánica), Universidad Autónoma de Madrid, Madrid, Spain 4 Estación Experimental de Zonas Áridas, Consejo Superior de Investigaciones Científicas (CSIC), Almería, Spain 5 Instituto de Ciencias Agrarias, Consejo Superior de Investigaciones Científicas (CSIC), Madrid, Spain 6 Centro de Investigación en Biodiversidad y Cambio Global (CIBC-UAM), Universidad Autónoma de Madrid, Madrid, Spain 7 Instituto Multidisciplinar del Estudio del Medio Ramón Margalef, Universidad de Alicante, Alicante, Spain 8 Departamento de Ecología, Universidad de Alicante, Alicante, Spain

Alterations in climate and land use are key drivers of global environmental change in drylands. Understanding the responses of dryland biodiversity to these novel stressors is imperative due to their extensive coverage and their importance in providing key ecosystems services and supporting many human populations. Such importance is further exacerbated due to the high number of endemic species that are unique to these environments. In this study, we explore the Grinnellian niches of the more than 1,500 plant species that inhabit the most important dryland ecosystems worldwide, extracted from BIODESERT dataset. We downloaded and filtered species occurrence locations from GBIF and BIEN data portals. Then, we used ensembles of Environmental Niche Models (ENMs) to represent the Grinnellian niches of each species, built on an individual species-specific selection of uncorrelated predictors from a set of 74 variables accounting for climate, soil, and topography, at a 10km resolution for the entire world. To ensure a comprehensive representation of the environmental space, we generated pseudo-absences distributed proportionally across 10 strata derived from a Principal Component Analysis (PCA) of the original pool of variables. Finally, for each species, we projected the models onto each of the seven biogeographical realms where the species is currently found, with a presence threshold of at least 1%. This allowed us to validate model projections against the empirical data on dryland communities from the BIODESERT dataset and assess the extent to which potentially suitable locations align with the actual habitats of species adapted to dry environments. This analysis contributes to and may enhance a deeper understanding of the relationship between species potential distributions and the establishment and sustainability of local populations in drylands, which can ultimately help informing ecosystem-level adaptation strategies to global change in extreme environments.