James works on cloud detection algorithms applied to satellite imagery.
Areas of expertise:
Cloud detection algorithms
James works primarily with imagery from the Meteosat Second Generation satellites. These geostationary satellites operated by EUMETSAT have been providing high-resolution images over Europe, Africa and the Atlantic every 15 minutes at solar and infrared wavelengths since 2004. The image data contain a wealth of information relating to cloud parameters, surface properties, and atmospheric variables such as temperature and humidity. Various quantitative imagery products derived from the data provide guidance for forecasters and also form inputs to the Met Office's automated nowcasting systems. In addition to imagery products, the satellite observations in areas of clear-sky are assimilated into the Met Office UM, and work is underway to enable assimilation of radiances in cloudy regions. A fundamental pre-processing step for these activities is the accurate identification of cloud-contaminated pixels in the Meteosat imagery. Developing and improving algorithms to carry out this task forms the focus of James's work.
James is also working on improvements to the RTTOV fast radiative transfer model. The capability to simulate short (visible and near-infrared) wavelengths is necessary in order to assimilate satellite radiances at these wavelengths into NWP models. Assimilation of such satellite observations will improve the analysis of aerosol distributions. At present RTTOV can simulate infrared and microwave radiances: James is involved in extending the capabilities of RTTOV to the simulation of visible and near-infrared wavelengths.
James joined the Satellite Applications team in 2007 after obtaining an MSc in Remote Sensing and Image Processing from Edinburgh University. As an undergraduate James studied Maths at Cambridge University.