Adv. Space Res. Vol. 13, No. 11, pp. (1 1)129—(1 1)134, 1993 0273—1177/93 $6.00 + 0.00 Printed in Great Britain. All rights reserved. Copyright © 1993 COSPAR GEOMORPHOLOGICAL FEATURES USING MSS AND TM DATA R. P. Singh,* K. K. Pahuja* and R. S. Chandel** * Department of Civil Engineering, Indian Institute of Technology, Kanpur- 208 016, India ** Department of Geology, Lucknow Universily, Lucknow, India ABSTRACT Landsat MSS and TM data have been analysed to classify various surface features. The classified image has been compared with the toposheet of the area prepared in the 1976. The digital remote sensing data of path 144 and row 41 of MSS quadrant ‘D’ has been analysed and compared with the results of visual interpretation. From the results it is clear that both the data have their own limitations in mapping various surfacial features. The detailed analysis and superiority of these data has been discussed in the present paper. INTRODUCTION The remote sensing data from the earth represents surface and subsurface information. Surface and subsurface information can only be extracted when the data is accurately analysed and interpreted. The analysis and interpretation are carried out visually and digitally using pattern recognition and classification techniques. Visual interpretation is mainly based on human experience and skill. Application of computer in remote sensing studies has gained momentum in developing countries mainly because of easy accessibility of personal computers. Image classification using computer analysis, auto- matically classifies all the pixels of an image into number of classes which can be used to represent various themes of land cover. Image classification can either be super- vised or unsupervised. Physical planners require up—to—date information for development planning /1/. In developing countries, physical landscapes are changing at a very rapid pace e.g. forestation, deforestation etc. With the help of computer classification, geomorphological changes can be mapped and up—to—date maps can be prepared of any area. In the present study, classification have been carried out by visual interpretation as well as by computer aided analysis. From the results it is clear that classification using computer can be used to classify the remote sensing data into maximum number of classes. TM data can be used for classification of more classes than those of MSS data, due to its better resolution and closely ranged spectral bands /2/. STUDY AREA The study area is bounded by latitude 26°30’ to 28°30’ and logitude 79°45’ to 81030~ which covers part of the area, but detailed CCT analysis has been carried out for Lucknow district. The regional slope of the area shows a south—east trend. The Ram— ganga, Gomti and Sarda river flows in the same direction. Gomti is a sluggish stream with intricate series of meanders. Tarai plains are well developed in the region. The north of this area is river Ghaghra. The mean elevation of Lucknow is ill meters above mean sea level and mean annual rainfall is 100 cm. Temperature ranges from 8—20°C in the winter and 15—45°C in the summer. METHODOLOGY Black and white paper prints on 1 : 250,000 scale of MSS band 2 and band 4 and TM band 5, 6 and 7 have been used for visual interpretation. Computer Compatible Tape (CCT) of MSS of 18th September 1986 has been used for computer analysis and for comp- arative study of the TM CCT of 10th November 1985 of path 144 and row 041 has been used. Survey of India topographical sheets 63B, 63B/13 and 63B/1 have been used for annotation and selection of training data sets. (11)129