In recent years there has been a marked increase in public awareness of environmental issues particularly the deforestation of the world's rain forest. To this end there is a need for accurate detection, inventory, monitoring and management for forest resources. This study sets to examine the extent to which synthetic aperture radar (SAR) imagery augment the information content of optical imagery in forest classification. The present mapping of forest areas by remote sensing involves the integration of remotely sensed images with other ancillary data. This integration of data sources would be input into a Geographical Information System (GIS). With the proper use of image analysis system it would provide user queries about the spatial data of the environment. Initial study area was intended to be the forest of Malaysia, but in anticipating the insurmountable difficulties that would be imposed by the Government of Malaysia with regards to topographic maps, aerial photographs, and other ancillary data which are restricted materials, an area in Britian was chosen. The chosen area was Forest of dean, near Bristol (west of England), which has close resemblance in terms of topography. Ordnance Survey maps, Forestry Commission database and its stock maps with Landsat Thematic Mapper and-Seasat Synthetic Aperture Radar acts as the data source. Processes of digitising, digital image processing, image transformation, data integration and finally classification were highlighted. This study examines and discusses the results of feature selection and maximum likelihood clasification and to what extent does the transformation of spectral data and the incorporation of ancillary data improves the classification of forest trees. On the basis of source data and the classifier used it is concluded that Seasat SAR do not make any significant contribution to the separability of forest trees, but terrain data improves the classification asccuracy. The refinements in training and class selection would certainly have further boasted the accuracy of forest tree classification. This study has establish a routine operational used of remote sensing, imaging radar and other ancillary data integration for the classification of forest trees and the results could be incorporated in a GIS. It recommends the used of new improved classifier and new generation of improved radar satellite data sources.