Dr Ian Culverwell
Ian works on simulated satellite imagery - in particular, on developing "metrics" to quantify the difference between imagery from two different sources.
Areas of expertise
- Mathematical methods and modelling
- Geophysical fluid dynamics
- Data visualisation and analysis
- Software maintenance and development
- Release management
Publications by Dr Ian Culverwell
Current activities
Ian works on the development of the "metrics toolkit", which is a collection of Python software routines that are designed to quantify the difference between observed and simulated imagery, or in general between simulated imagery from any two different sources. At present, the contents of the toolkit include: time-series of (areally averaged) standard difference statistics, such as means and standard deviations; histograms of (for example) brightness temperatures; visual maps of such fields; and the Fractions Skill Score, which is one way of estimating the length scales over which one brightness temperature image, for example, has useful skill in predicting another. Further metrics are planned for inclusion. The hope is that a quantitative difference between two images could provide a more objective assessment of model skill than that produced by eye. This could be of use to Operational Meteorologists, especially in the context of ensemble forecasts, where ranking such metrically-quantified differences could allow the forecasters to identify and then concentrate on the ensemble members "closest" to the observations, since these are probably the ones that are more likely to be useful. It could also be of use to model developers, as it might allow model differences (which are sometimes small) to be assessed quantitatively rather than estimated qualitatively.
Career background
Ian joined the Met Office in 1995, after gaining a BSc in Physics and a PhD in Plasma Physics from Imperial College, London, and a Master's degree in Mathematics from Cambridge University. At the Met Office he was involved in the development of regional and global Numerical Weather Prediction models, the construction of new coupled atmosphere/ocean/sea-ice climate models, and the maintenance, development and release of radio occultation processing software, before starting his current job in 2022.