Robin Wilson

A PhD student studying complexity in remote sensing

My PhD is entitled Enhancing scientific value from remotely-sensed data through improved pre-processing methods and models: towards an integrated UK Environmental Change Observatory. That may be a bit difficult to understand - and it's certainly a mouthful to say - so in plainer language, I'm working on trying to improve the pre-processing that must be done before satellite or airborne images can be used for scientific purposes. So, for example, if you want to estimate water stress in crops from a satellite image you will have to geometrically correct the image so that you can be sure you're looking at the right parts of the field, correct for the effects of the atmosphere and (particularly for certain sensors) correct for angular effects. Only after this pre-processing has been completed can you process the image to extract information on the subject of interest (in this case the crop water stress). The quality of these pre-processing stages directly impacts on the quality of the final output, so improving these pre-processing stages should allow us to produce better (that is higher quality: more accurate, less uncertain) scientific data from satellite and airborne imagery.

I'm using a wide range of approaches in my PhD including:

  • simulation modelling - particularly ray-tracing models of the atmosphere
  • novel uses of existing datasets - for example, the wide range of MODIS products
  • development of novel instruments - particularly those which are cheap and can therefore be widely distributed
  • crowd-sourcing of atmospheric and ground data - again, to get a wide distribution of (hopefully high-quality) data

The overall aim of my PhD is provide the improved pre-processing techniques needed to allow the development of a UK Environmental Change Observatory (UK-ECO), described below.

UK-ECO - Vision for 2012
The UK will have an Environmental Change Observatory, based on multiple remote-sensing systems, which will allow monitoring of environmental change in over short, medium and long time scales, at a medium-high resolution across the whole of the UK. Data produced will be quality-assured and will aim to provide information on the changing dynamics of the environment, including estimates of carbon flux, the impact of climate change and biodiversity. Products derived from UK-ECORS will underpin research to meet the NERC environmental change research objectives and will provide accurate time series of data for a number of the 50 GCOS Essential Climate Variables (GCOS 2003). The products produced by UK-ECORS (both the corrected remotely-sensed data and the ecological outputs) will be available within a very short time after data collection, allowing them to be used as key decision support tools by a range of public and private bodies. The UK-ECORS will form the UK’s contribution to GEOSS, the Global Earth Observing System of Systems.