Global change is driving biodiversity loss globally and Mediterranean-type ecosystems, like those found in the California Floristic Province (CFP), are among the most diverse and threatened natural systems. Habitat loss due to climate and land use change amplifies species’ extinction risks, with some species being more exposed than others due to their rarity as well as their geographic distributions. As we attempt to assess and model species’ extinction risk, it is imperative to address and communicate the uncertainty introduced by each modeling component, especially as new approaches emerge, and assess this uncertainty spatially. While estimating species-level vulnerability is essential for prioritization, it is equally important to understand how habitat dynamics may vary across species ranges in response to global change. In this dissertation, I first assessed the relationships between species’ rarity, geography, and exposure to climate and land use change for 106 plant species found in the CFP. I found that although small-ranged species in flat areas with high climate velocity tend to be the most exposed to climate change, those found at low elevations are at the highest risk of habitat loss via land use change. Next, I evaluated predictions of projected habitat suitability under different climate models, emissions scenarios, and species distribution modeling consensus (model averaging) approaches. Surprisingly, these results showed that vulnerability predictions made by different model averaging techniques can differ substantially, if the models are projected into geographic areas outside of species’ ranges. Finally, I assessed how habitat suitability change varied across species ranges as a function of geographic features, including latitude, distance to coast, elevation, and topographic heterogeneity. I found substantial variation in the geography of species habitat change, with some, mostly montane species, showing potential refugia in high elevation areas regardless of climate model, while other species, such as those found in coastal sage scrub and chaparral, exhibited variability in the location of refugia depending on the magnitude and direction of climatic change. This dissertation emphasizes the spatial heterogeneity of global change in the CFP and the importance of conducting geographic assessments of species’ vulnerability.

While most restoration projects to date have focused on a specific system in a specific location and are rarely greater than 5,000 hectares in extent, calls for restoration at broader scales with a focus on restoring complexity, and in diverse ecosystems are growing. However, ecological restoration at broader scales, encompassing multiple habitat types, management goals, and complex cross-scale interactions presents its own suite of challenges. To address these challenges I first developed a long-term adaptive ecological monitoring framework for landscape restoration projects. Then I assessed changing fire regimes in California woodlands at regional scales and how this may influence current vegetation structure. I also evaluated the current conditions and restoration needs of California’s iconic and extensive Blue oak woodlands and savannas.