UC Berkeley

All forms of energy generation can have intensive or extensive land use requirements, causing habitat and biodiversity loss in sensitive and diverse ecosystems globally. With the rapid transformation and growth of the energy sector in countries worldwide, understanding the impacts of past practices and charting the trajectory of future development projects is imperative for preventing negative environmental consequences. This dissertation contributes modeling strategies for integrating environmental impacts in renewable energy planning processes and spatially-explicit empirical methods for identifying and quantifying land use and land cover impacts related to renewable energy development.

To explore land use and energy conflicts in a jurisdiction that is in the midst of a large-scale low-carbon energy transition, I ask the following: (1) is it possible to meet California's ambitious renewable energy targets without using high conservation-value land? (2) what are the system costs of low-impact renewable energy development? I find that while trade-offs between conservation value and renewable resource quality exist, restricting development to low-impact land is not only possible, but incurs only negligible economic cost increases. Given this possibility, I use California as a case study to identify decision-making opportunities in energy planning processes for integrating conservation and land use values and avoiding conservation-climate conflicts.

Extending the spatial methods developed for California to countries in Africa that are planning renewable energy expansion, I ask, what is the potential for low-environmental-impact, socially-responsible, and cost-effective development of wind and solar energy in emerging economies in Africa? Using a multi-criteria analysis approach, I find that "no-regrets'' options---specifically areas that are low-cost, low-environmental impact, and highly accessible--exist such that significant fractions of demand can be quickly served with low-impact resources without large additional cost.

Despite the magnitude and pace of hydropower expansion in highly biodiverse aquatic and terrestrial ecosystems in Southeast Asia, Africa, and Latin America, the potential indirect land use and land cover change resulting from hydropower development is poorly understood. To fill this gap, I ask, what are the indirect deforestation and land use impacts of utility-scale hydropower development in the Brazilian Amazon? Do siting choices and pre-existing land use and land cover affect the extent of the impact? Using scalable remote sensing and spatial econometric methods for causal inference, I find a 11-59% increase in indirect deforestation due to hydropower development. These findings can contribute to estimates of potential future terrestrial impacts from the hundreds of hydropower plants planned and proposed in relatively intact areas of the Brazilian Amazon.