Modeling the Current and Potential Future Spatial Distributions of Honey Mesquite (Prosopis glandulosa torr.) in Mexico and the Southwestern United States

Keywords: honey mesquite, woody plant encroachment, species distribution modeling, CHELSA, , GBIF
Abstract Type: Paper Abstract

Abstract

Woody plant encroachment—the replacement of grasses by woody plants in drylands— is a globally observed phenomenon with diverse and often negative consequences for social-ecological systems. One encroaching woody plant species, honey mesquite (Prosopis glandulosa), is of particular concern in the arid and semiarid Southwestern U.S.A. and Mexico. Its remarkable ability to thrive in these regions has raised concerns about reduced perennial grass production and associated decreases in livestock carrying capacity as well as changes in biodiversity, soils, biogeochemical cycles, biogeophysical feedbacks, and other environmental characteristics. Sensible management of areas affected by honey mesquite encroachment requires data about its current distribution and anticipated expansion. However, this information is not adequately available. In this study, we address this issue by modeling a) the current and b) the potential future distribution of honey mesquite. To meet these objectives, we integrate occurrence points of honey mesquite obtained from the GBIF database with climate and topographic data obtained from the CHELSA dataset in Maxent, which we implemented in R. We thinned the occurrence points to minimize bias spatial autocorrelation, removed redundant variables using Pearson’s correlation coefficients and variance inflation factors, and then carefully calibrated and evaluated the model. Our findings indicate that honey mesquite currently occupies extensive areas In the Southwestern U.S.A. and Mexico, with a predicted expansion into northern areas and higher elevations. In the presentation, we will visualize this more specifically using maps and also present and discuss the relative importance of environmental factors in shaping current and potential future distributions.