Ed's broad research interest is how applied mathematics and statistics can be used together to solve critical environmental problems.  His current work at the Lancaster Environment Centre is to understand better, and quantify the uncertainties of, the key parameters and processes in current numerical models of atmospheric chemistry and transport.  This will lead to a better grasp of the uncertainty in important model outputs such as tropospheric ozone levels and methane lifetime, and will help to explain differences in these outputs between different models.  To achieve these goals, he is using cutting edge methods of: (i) statistical emulation, (ii) sensitivity analysis, (iii) structural uncertainty estimation and, (iv) parameter estimation, such as Generalised Linear Uncertainty Estimation and calibration-emulation techniques.

Ed's previous post-doc at Arizona State University in the US used field data, models, and hierarchical Bayesian statistics to improve our understanding of how and why multi-annual rates of photosynethesis and soil respiration will change under a future climate of elevated CO₂ and warming. He also built a novel model of soil CO₂ production and transport at multiple depths and times to test commonly adopted steady-state assumptions, under different environmental and meteorological conditions.