You are here: » Information Zones » SIGs » MATSIG » Publications » Mat Sig Annual Report 2001 2002
Modelling & Advanced Techniques
MAT SIG Annual Report 2001-2002
The main achievement of MATSIG this year was the organisation of a two-day meeting held at the Department of Geography, University of Southampton, entitled ‘Uncertainty in Remote Sensing and GIS’. The meeting was held on 3rd and 4th July 2001 and was organised on behalf of both MATSIG and GISSIG by Professor Giles Foody and Dr Pete Atkinson. The conference convenor was Andrew Tatem and he was helped out by numerous other post-graduates from the Department of Geography. I would like to thank all the above for organising an excellent meeting that provided a much needed forum for the presentation of the state-of-the-art of research in the field of statistical uncertainty modelling using remote sensing, GIS and other spatial data.
The meeting was attended by an excellent group of national and international speakers including Professor Paul Curran, Professor Curtis Woodcock, from the University of Boston USA, Dr. Jennifer Dungan, NASA Ames Research Center, USA, and Dr. Gerard Heuvelink, University of Amsterdam, The Netherlands. For me the highlights were provided by Curtis Woodcock who gave a stimulating talk on the diverse sources of uncertainty in remotely sensed data, Gerard Heuvelink who outlined the problems of applying error propagation to environmental models due to our inability to obtain errors on the diverse array of input parameters, and Andrew Tatum who outlined a novel method of increasing the spatial resolution of classified imagery by applying spatial decision riles to land-cover proportions estimates obtained from fuzzy classification. The latter talk was both humorous and original.
A particular attraction of the meeting was the informal social event organised on the evening of 3rd July whereby we were all bussed to a pub, provided with food and drink, and then played each other at group skittles. The meeting was sponsored by the Ordnance Survey, the Remote Sensing and Photogrammetry Society, Taylor and Francis, Erdas UK, PCI Research Systems, RGS-IBG, and the University of Southampton and I would like to thank them for a very enjoyable two days.
MATSIG is currently discussing the details of the meetings that it will be organising over the next couple of years. The next meeting will probably be in collaboration with the British Hydrological Society and will consider the role of land surface processes in spatial modelling of floods in order to further our understanding of flood generation processes and to improve flood forecasting.
Nick Drake
