UC Berkeley

Anthropogenic climate and land-use change are widely considered to be the two greatest threats to global biodiversity and ecosystem functioning. Although species may respond to changing environments in a variety of ways, the majority of species are expected to shift their distributions to track ecologically favorable habitats. The extent of these shifts may be heterogenous across species or even within a species occupying different regions, but our knowledge of what drives this heterogeneity is still sorely lacking. In particular, the effects of climate and land-use change on shifting species distributions have largely been studied independently. In more realistic scenarios, most species will experience both drivers simultaneously, creating the potential for heterogeneous changes in species composition across regions with different histories of climate and land-use change, as well as species with different sensitivities to that change.

To investigate the combined effects of climate and land-use change on species occupancy and community composition, I surveyed avian diversity at 71 sites in the California Central Valley and Los Angeles. These sites were originally surveyed in the early 1900s by Joseph Grinnell and colleagues from the Museum of Vertebrate Zoology, providing a unique opportunity to directly compare bird occupancy and its relation to site-specific climate and land-use covariates across a century of change. Historic surveys paired with modern resurveys documented changing distributions by 148 bird species across two regions with similar initial species and habitat composition but differing patterns of climate and land-use change.

Our knowledge of species’ ecological and life history traits as predictors of sensitivity to land-use change is well developed, but the same cannot be said for traits as predictors of climate-induced range shifts. There is strong theoretical support that range shifts under climate change may be mediated by traits that facilitate dispersal and population establishment, but empirical evidence for this relationship has been mixed. In my first chapter, I conducted a meta-analysis of studies that analyzed climate-induced range shifts as a function of species’ traits, with the goal of identifying which traits have provided the strongest results. I show that the majority of traits failed to predict range shifts consistently across studies, while a much smaller number of traits, particularly habitat breadth, had strong empirical support. In subsequent chapters, I use the most informative species’ traits revealed by this meta-analysis to explore differences in species-specific occupancy change across my Central Valley and Los Angeles survey sites.

Both climate and land-use change are expected to favor exotic and generalist species over specialists, resulting in declining occupancy and diversity. Using occurrence data for 122 bird species from 41 resurvey sites in the California Central Valley, my second chapter examined how occupancy and diversity changed over the past 100 years in this predominantly agricultural region, as well as the relative influence of climate versus land-use covariates to occupancy within each survey period. In contrast to expectations that occupancy and diversity would decline, both remained stable: species that increased in occupancy (predominantly habitat generalists or human adapters) roughly balanced species that decreased in occupancy (predominantly open-habitat species), resulting in no significant change in average species richness or between-site diversity. Equally surprising was the far greater importance of water availability (precipitation and water cover) to occupancy within each survey period as compared to urbanization or agricultural cover.

The relative importance of climate versus land-use change to changes in species occupancy is poorly known and may result in heterogeneous patterns of diversity change across regions with different climate and land-use change histories. In my third chapter, I used occurrence data for 148 bird species across the full set of 42 Central Valley sites and 29 Los Angeles sites. I directly assessed the relative importance of climate and land-use covariates to probabilities of occupancy, persistence, and colonization, as well as compared patterns of occupancy and diversity change between the two regions. Climate and land-use covariates were both important, but to different aspects of occupancy and turnover, with climate driving initial occupancy, climate and land-use both having similar influence on colonization, and urban cover emerging as the biggest driver of local persistence. In contrast to stability in occupancy and diversity in the Central Valley, both declined in Los Angeles. This diversity loss resulted from a greater amount of urban development in Los Angeles and larger species-specific occupancy decreases in particularly forest and open-habitat species.

Overall, my dissertation provides a detailed picture of how bird species have responded to over a century of climate and land-use change. These changes were highly heterogeneous across species – driven by traits, namely habitat preference – and across very similar ecological regions – driven by divergent patterns in both climate change and urbanization. As we continue to create range projections and plans to conserve future diversity, it will be important to utilize knowledge generated by long-term historical datasets and to incorporate multiple drivers of heterogeneity in species’ responses to global change.