Understanding species status and forecasting range shifts for plants and animals requires accurate species distribution information, particularly at the margins of species ranges. However, most distribution studies rely on depauperate species occurrence datasets from herbarium records and public databases, along with statistically derived “pseudoabsences”. True absence data are rarely available. Most distribution analyses rely on statistical species distribution models (SDM). While environmental covariates used to fit SDMs are increasingly precise due to satellite data, the availability of species occurrence records is still a large source of bias in model predictions.
Joshua trees are iconic and keystone species of the desert Southwest and were recently petitioned for listing under the Endangered Species Act by the U.S. Fish and Wildlife Service (FWS) due to multiple threats across the range of both species. Scientists produced the first hybrid distribution models for eastern and western Joshua trees (Yucca brevifolia and Y. jaegeriana, respectively) using empirical data supplemented by statistical species distribution modeling. This method combines a high-resolution grid for intensive visual searches of publicly available satellite imagery evaluated with ground truthing and other sources of distribution data to derive the current distribution of adult Joshua trees across most of their range with high accuracy. While most of the distributions were mapped empirically at a square quarter kilometer resolution (500 x 500 meter) across the more than 600,000 cells of the study area, one large and important area has unacceptable remote sensing imagery quality due to national security. Empirically derived data were used to model distributions within the region of poor image quality to create a complete model across the ranges of Y. jaegeriana and Y. brevifolia. The statistically modeled area is of particular ecological and management significance because the two species distributions overlap there and it guides where hybrid plants with characteristics shared by both species may occur. The models resulting from this work identify key environmental gradients shaping the distributions of each species. Annual temperature, precipitation, and aridity best predicted occupied habitat across the species’ ranges. While the environmental niches for Y. brevifolia and Y. jaegeriana were similar in terms of total aridity, they differed with respect to seasonal precipitation and temperature tolerances, suggesting a potential for disparate responses to climate change.
The maps are the primary distribution information available for the FWS Species Status Assessment. The high-resolution distributions can be used to empirically track changes in Joshua tree populations at the leading and trailing edges of the species distributions in relation to climate change and habitat disturbances. The associated manuscript is available here.
%20We.jpg)
Joshua Trees in Inyo Mountains above Eureka Valley, California. Photo credit: Ken Kole (U.S. Geological Survey).
Disclaimer: Any use of trade, firm, or product names is for descriptive purposes only and does not imply endorsement by the U.S. Government.