Build imprecise supercomputers
The Oxford Martin Programme on
Modelling and Predicting Climate
The Programme on Modelling and Predicting Climate was established in 2010. Funding from the Oxford Martin School ceased in 2015. Although this programme no longer receives funding from the Oxford Martin School, members of the team are still actively involved in the School's community.
the project
We aim to increase the reliability of climate prediction through the development of novel mathematical techniques for use in next-generation earth-system models.
Climate predictions are used for making major decisions in climate mitigation, adaptation and geoengineering. The quality of a decision depends on how accurate the forecasts are and how well the uncertainty in the forecast can be estimated.
Whilst on the global scale climate prediction is broadly reliable, attention has turned to more regional climate prediction, where errors inherent in conventional modelling methods are more apparent.
We bring complex mathematical modelling techniques to understandings of climate dynamics and atmospheric physics. We play an active role in informing global debate and advancing the field of climate science through building more reliable models for weather and climate prediction.
publications
Opportunities for Energy Efficient Computing: A Study of Inexact General Purpose Processors for High-Performance and Big-data Applications
The implications of carbon dioxide and methane exchange for the heavy mitigation RCP2.6 scenario under two metrics
Stochastic modelling and energy-efficient computing for weather and climate prediction
Climate science: Breaks in trends
Reducing Risks of Future Disasters