UC Berkeley

Land system sciences offer techniques to study tradeoffs and solutions for ecosystems that are vulnerable to–and solutions for–climate change. This dissertation investigates tradeoffs for fire, forest, and carbon management in California. Throughout this work, I focus on methods to improve inference using remote sensing, spatial data analysis, and quasi-experimental methods. In the first chapter, I test assumptions about green landscapes as solutions to threats of fire. I use remote sensing and causal inference techniques to test whether green landscapes can alter fire severity and limit fire spread around communities, using golf courses as an example. I find that golf courses act as buffers, reducing fire severity by 49% relative to similar burned vegetation and that golf courses limit fire spread better than other vegetated landscape like regional parks. In the second chapter, I evaluate models used to make management decisions and discuss how errors from these models have landscape and policy implications. I conduct a model validation study for the widely used Forest Vegetation Simulator and find a significant overestimate in predicted growth rates relative to observed growth. I find that these errors in carbon growth rate are most pronounced on national forest lands, for stands aged 50-100 years old, forests managed as reserves, and for the least productive forest sites. Model assumptions of higher productivity than the observed growth will bias management away from more proactive silviculture techniques and can overestimate their impacts as a natural climate solution. In the third chapter, I document how land use policies for forests create tradeoffs that may undermine policymakers’ objectives. I explore how operationalizing carbon management in forests creates tradeoffs between sustaining high levels of biomass, and carbon, and removing biomass to promote forest structures that are more resistant to disturbance. I find that forests managed for carbon credits are not changing fuels like forests managing for fire risk but are having changes in vegetation greenness as if they were. Throughout this work, I demonstrate how the application of rigorous methods of measurement and estimation can support improved decisions that deliver climate and ecosystem benefits for California.