Research Areas
Designing sustainable and renewable energy systems. Designing renewable energy systems means designing systems that can deal with and even thrive on the variability of solar and wind power. In a fully sector-coupled energy system where heat and transport are partially or fully electrified and synthetic fuels replace fossil ones, different strategies to balance renewable generation are available and come with their advantages and disadvantages: from continent-spanning electricity grids, to large-scale hydrogen storage, to demand response through heat electrification. All of this requires models that depict the necessary spatial and temporal resolution. Climate change is not the only urgent global problem. The design of truly sustainable energy systems also requires examining trade-offs between the energy transition and other issues, such as land use, material requirements, ecosystem impacts, and ramifications for society. As part of efforts in this area, we develop deeper integration of our models with those from other fields, such as industrial ecology.
Quantifying the spatio-temporal variability of energy supply and demand. Understanding the weather-dependent variability of renewable electricity generation and energy demand. This involves working at spatial scales ranging from individual buildings to entire continents, and at time scales ranging from minutes to decades. These efforts rely on links to meteorology and climate science, and on gathering crowdsourced measured data on solar and wind system performance from around the world. With Iain Staffell at Imperial College London, I develop one of the world’s leading suites of validated, decade-spanning, hourly power generation simulations, Renewables.ninja.
Enhancing the use of computer models in decision support and planning. Computer models used in policy and planning are powerful tools, and with this power comes responsibility. We examine empirically and theoretically how models are developed and used in different contexts. Our aim is to improve the models themselves, but also their translation into impact, for example at the science-policy interface. Despite increasing adoption of open models and data, transparency and understandability of models needs further improvement, and issues such as power dynamics and ethics in model use remain contested.
Developing open scientific software and data. My team and I have long-standing experience developing scientific software. Our aim is to develop scientific models with direct relevance for decision support, for example, by combining carefully validated scientific models with easy-to-use web interfaces. We have considerable experience managing and leading the development of open-source scientific software across multiple institutions. → More on our software projects.