Land use mapping of selected areas of county Durham, north-east England, by satellite remote sensing and field survey methods

Abstract

This thesis investigates the use of field and satellite data for agricultural land use mapping and land use change in six selected areas of County Durham, north-east England. These areas are selected to represent the major land use patterns of the County. The satellite data were obtained by the Thematic Mapper (TM) sensor on board Landsat-5 on May, 31, 1985, August, 1, 1990, and July, 10, 1992. TM data were geometrically corrected to the British National Grid, Land use maps for these study areas were produced by field survey to become the basis on which the research was based. These land use maps along with images were integrated in a Geographical Information System (GIS) called ARC / INFO. The total surface area of the study area is 5483.86 hectare, or 2.3% of the whole County Durham. The field area measurements were taken and the final hectare estimates were obtained for each land use / land cover type. The research demonstrated the ability of Landsat-TM to produce accurate land use maps of the study area. Results obtained emphasised that Satellite data when integrated in a Geographical Information System (GIS), can be used for mapping relatively small agricultural fields. A land cover classification scheme appropriate for the study area was applied. Number of land use land cover classes produced varied from one study area to another and from one image date to another also (18-25 classes). These detailed classes were generalised to their broader classes. Using the Landsat-TM data, classification accuracy varied from one study area to another and from one date to another also, but it was not less than 80%. The classification accuracy was assessed using the grid-by-grid overlay method. The study analysed the spectral properties of each land use land cover over the three image dates, and the importance of contextual and reference data to assess and improve the classification accuracy. Results obtained by this study showed that the selected study areas can represent the major agricultural crops in County Durham by a percentage of 81-96%. Results of the study also suggests that a data base in digital format of land use, topography, geology, soil and climate is essential to set-up a special model for the County to be used for faster, easier and more efficient updating of land use data of the County,