| The dominant role of climate in species distribution modeling is supported by a large body of literature and is often considered the single most important factor in limiting species' ranges. Other factors, such as biotic interactions, are often assumed to be included via abiotic proxies. However, through climate change, these relationships could be decoupled or species could adapt to novel environments, leaving gaps in predicted future niches. To test the utility of including such factors, I modeled the distributions of two isolated populations of the endangered giant kangaroo rat (Dipodomys ingens) and local California ground squirrel (Otospermophilus beecheyi), a potential novel competitor. Modeling current and future distributions of both species identified potential niche overlap. I predicted that modeling the populations separately would improve the predictive accuracy of the models, as they have adapted over time to altered climatic regimes. Niche overlap between the populations was low, indicating that they experience slightly different climatic regimes. Thus, model accuracy would improve with population-level modeling. Future projections of range are dependent on model scope, but retractions in currently occupied areas will occur. We can use these future range estimates to protect critical habitat from further development, in hopes that we can protect the giant kangaroo rat within an endangered ecosystem. |