To address significant challenges in analyzing the biological resource impacts of solar development in the Mojave and Colorado Desert Region, and the lack of detailed distribution information for sensitive plants, I am working with Patrick McIntyre on a project that integrates improved data on species’ occurrences, rigorous habitat distribution models, and field validation in order to generate novel habitat distribution maps for a suite of rare Mojave plant taxa. We are applying innovative techniques by which predictive habitat models can be generated for rare species, and are assessing the accuracy and usefulness of such models in identifying habitat areas beyond those places where modeled taxa are currently known to exist. Funded by the California Energy Commission, we coordinated with the Jepson Herbarium (at the University of California, Berkeley) to include recently georeferenced data on rare plant locations.
We applied species distribution models to 8 carefully selected rare plant species representative of the types impacted by utility scale solar energy development. Models were constructed using new methods to reduce the biases of spatial sorting and spatial autocorrelation that commonly plague rare species distributions (Hijmans 2012). In 2011 and 2012 we found novel occurrences for all studied species.
Currently we have a report for the California Energy Commission in review and we are drafting two manuscripts for peer review stemming from this work. The first manuscript will present practical considerations for the generation of distribution models for rare species and their assessment using a paired review and application of our methods for rare desert species. The second manuscript evaluates the effects of spatial clustering on model scores and the resultant interpretation of rare species models.
Map modified from DRECP