While it is often the assumption in environmental modeling that finer-resolution modeling is preferred, especially if computation times are not prohibitive, few studies have assessed how climate grid resolution influences the Maxent-predicted habitat of desert vegetation species. Further, drought events can occur over longer or shorter terms with drought length potentially influencing species' habitat distributions. This study uses four higher-elevation Mojave Desert plant species experiencing known habitat contractions corresponding with climatic changes to assess how sensitive Maxent species distribution models are to using 5- and 10-year averaged climate data, as well as 800-m and 4-km resampled gridded climate data. Results show there are spatial differences in models despite relatively consistent clustering of three of the species' recorded field locations, whereas predicting habitat for the more broadly ranging species results in less certainty across all models. Overall, models were more sensitive to the spatial resolution of the climate data than to the climate time step used. When considering geographic areas with high relief, such as the Newberry Mountains in southern Nevada constituting the study area, the spatial resolution of climate data has a major influence on modeled habitat. As more fine-resolution climate data become available, researchers may need to establish more plots for field collection to assess specific microclimatic effects on vegetation.
Bidirectional recovery patterns of Mojave Desert vegetation in an aqueduct pipeline corridor after 36 years: I. Perennial shrubs and grasses
