GISdevelopment.net ---> AARS ---> ACRS 1989 ---> Agriculture & Soil Application of Remote Sensing and GIS in soil erosion hazard mapping and land suitability analysis in Phewa watershed - A case study Krishna P. Pradhan Scientist Agronomist Nepal Remote Sensing Centre Kathmandu Nepal Abstract The object of the study project is to demonstrate the application of land sat image processing and geographic information system in potential soil erosion hazard mapping and land suitability analysis even in the absence of very detailed information about the area . The land sat computer compatible tape of Phewa water sand from July 1979 was made available through UNITAR from Thailand Receiving station the digital image Processing was done in IBM PC using image Processing modules of EDAS Software Supervised classification technique was applied to extract land use cover information was the water sand .The image was dereferences taking ground control in the water sans units from 1:25000 scale topographical map of the water sand the cell size of 57 m was decided in order to obtained the same size with other GIS profiles. The existing maps Such as slope soil depth rock out crop were digitized in ARC vector mode systems the vector mode files from were converted in to raster using ARC /INFO command .Which creates in to a GRID with single variable file from polygons in ARC/INFO coverage and these data were transferred in the computer compatible tape all the files were stored in the IBM PC hard disk using the command DT Tape all the files rastarized using the same cell size of 57 meter as image in order to over lay exactly one another. The final map resulted from the image classification was combined with slope map to obtain of run off rate in the water shed . Due to the lack of very detailed data a very simple model was applied and prepare potential soil erosion hazard map. the following maps .i.e slope cover slope run off population and soil maps were combined five images of erosion rates were identified in the water shed . the following maps soil quality moisture availability rock outcrop and potential soil erosion hazard were overlaid each other to prepare land suitability for agriculture . The results obtained from this study can be served as a good example of application of remote sensing and Geographical information system for potential soil erosion hazard mapping and land suitability analysis this type of work can be done in any part of the country once the data are GIS format it can be retrived and manipulated at anytime. Introduction The basic needs of the people are food grains fuel wood and timber. The main emphasis should be given to the production of the food grains fulfill the basic need . There is no sufficient supply of food grains to fulfill the demand of the present population .to meet the demand of the population, it has been necessary to extend the agricultural land destroying the forest land .The high rate of deforestation may leads to the desertification of the area and erosion of soil by water which effect the environment and the productivity of land. The main reason of soil erosion are high precipitation high run off due to the steep slopes high population pressure on 1 land and clearing natural vegetation. Phewa lake which is located in the Phewa water shed is on the most prominent lakes in Nepal .it is remained un charged by the enthroned the land it receives about 10 ton of sediment the soil erosion which affect on the natural beauty as well as the life the lake it has been pointed pout that the soil erosion losss from grazing land is 174682 ton and from agricultural land. In the present study a Geographical information system technology was applied to amp the potential soil erosion hazard and the land suitable and remote sensing data. This type of study helps to locate the areas more vulnarable to soil erosion prone areas are located on the map then it is easy to make plane for better management of land such as afforestation plan terracing of agricultural land etc. Objectives The main Objectives of the study are as follows: To perform digital image processing of land sat CCI for land use cover types using ERDAS Software. To construct the data base for Geographic information system to digitized the existing maps such as slope soil depth rock out crops , maps with panchayat and sub watershed boundaries. To prepare potential soil erosion hazard and land suitability map. from the existing maps using GIS technology. Study area The Phewa water shed is located in the Kaski district of the western Development region. the water shed has an area of 12.5 km and drained in to on the Nepal most prominent lakes. the Phewa Lake. the water shed extend from 28 10 29 to 28 16 15 latitudes to 77 47 46 to 77 58 12 longitudes topography of the thr area is steep with the elevation ranging. from 850 to 2500 m from msl. The climate of the water shed is humid subtropics to humid temperature the mean temperature at the Phokara ranging from 12 to in December and January to 25C in July and August with minimum of about and maximum of about 34 C. Lumle Agricultural Center (1975 m) to the north-west of the watershed has a mean monthly temperature ranging between 3.5 C in January to 20C in August. The average annual train fall in Phewa water shed is about 4000 m in 1979. In the water shed of the land is intensively used for agricultural purpose .Two types of agricultural can be located land and bari land lost of the Khet land is located on the valley floor the agricultural crops of the Khet land is this area are rice in summer and wheat un winter seasons. Rice Crop in irrigated from streams. Bari land is located on the upper slopes of the water shed the main crops in this area are maize millet it summer and wheat in winter. The forest vegetation can be located on the higher elevation of the water shed area the main species area quarques semicarpifilia schins castanposis and Ainus sp. Fig.1. Material and methodology Data Processing The computer compatible tape was located into the CIPHER TAPE amd the study was displayed on the screen and the study area was located in ( fig.2 ). Image Enhancement Image enhancement is the modification of an image to alter its imapct on the view or to improve the interprêtability for a particular purpose .The object on a ground with similar spectral features may not distinguishable on a land sat image because the tone difference can observe by human eye. Therefore enhancement in necessary before classification. In this study, histogram equilization was performed. This enhances the image displayed on the screen by maximizing the color contrast. Mulispectral classification Multi spectral classification is a type of information extract process that analysis the spectral and assigns the pixels categories based on similar signature before geometric correction rectification of the image was performed a supervised classification was applied to produce a nine classification thematic map of principal of land use cover in the study area on the existing land use map of the study area used as a references .the training data were then used to assign a to each pixel in the study area. The purpose of classification of the image before geometric correction was to avoid the classification errors contributed by resembling process. Geometric correction Ground control points selection From the displayed image 35 ground control points were taken with in and a\outside the study area and recorded in Universal Transverse Marcato ( UTM ) coordinates, using 1:25000 scale topographical base map. The RMS ( root mean square ) error of 1.5 was given. Only five ground control point were rejected . Rectification The rectification of the image was done using the nearest neighbor re sampling scheme .the image was resample to a 57 by 57 meter grid. Finally the computer classified map of land use cover was also rectified using the same parameters which are used in image rectification .the land use cover map was over layer with slope map to go run off map. Entry of data in GIS All the existing resource maps were entered in to the ARDC INFO digital format via digitizing process .Digitizing is a process of tracing the boundaries of resource attributes on a map which is places on the digitizing table .Digitization produces a computerized files in vector form containing the location of each recource attributes in term of X and Y coordinates system. The following resource ,map were digitized in ARC /INFO at the scale of 1:250.000 with the UTM Projection. Population density map. Slope map Soil Map Soil depth map Rock out crop map The vector mode file from ARC INFO were converted in to raster from using programmers POLYGRID LINERED or POINTED depending on the features contained in ARC INFO coverage these progress result in so called single variable files the later were transferred in to ERDAS format using SVDERDAS progress this resulted 16 bit data files were copied to CCTS and loaded to IBM PC hard disk through CIPHER data file. Transfer utilities. All the files were raster using the same cell size 57 meter as rectified image in over in order to overlay exactly one other for the different map projection it is important to have the exists the files the same size same map projection and the same grid. Finally the different resource maps were overlaid each other to obtain a new map which used to analyze potential soil erosion hazard and land suitability fig 3 and 4 show the flow chart of analysis and over lay procedure. Fig.2. Fig.3. Flow chart of digital analysis of land sat data for land use / cover Fig.4. Flow chart for GIS analysis for potential soil erosion hazard and land suitability Results and discussions Soil Erosion Analysis Fig 5 shows a the computer classified land use cover map of water shed and the table 1 shows that 23.08 and the land is covered forest vegetation 64.51 agricultural land and 7.90 by grazing land shed . Table 1. Area of different land use cover types classification in Phewa water shed Land use / cover Area in ha Z Dense forest 2605.373 20.76 Terrace cultivation 3193.767 25.45 valley cultivation 1188.809 9.47 Lake 321.651 2.56 Forest 391.505 3.12 Crop land 3713.932 29.59 Shrub land 69.023 0.71 Builtup area 44.511 0.35 Grass land 1001.992 7.98 Total 12550.562 100.00 The land sat image was taken in July 1979 before rainy season classification result shows 2.56 of water . But the interpretation aerial photographs which are taken after the rainy season of the year shows 3.84 of water in the water shed after the rainy season water level has increased by 1.28 Z. The soil erodibility map derived from digitized soil map. The run off map was obtained from slope and land use/cover map. Fig 6 is the final map of potential soil erosion hazard obtained by combining run off soil erodibilty and population data.. The result shows that the water shed contains 24.28 % very low, 12.10 % low, 37.22 % medium, 17.65 % high and 4.92 % high very high potential erosion areas. Area having steep slope high run off high run off low vegetation cover, sandy soil erosion and high population pressure is considered as high potential soil erosion. During period of heavy rainfall a great amount of soil erosion take palace in the water shed and large amount of slit and clay particular place mostly from grazing land and agricultural land the soil particular removed mostly from amd streams and deposited in the lake. By studying the soil erosion map has a locate the area having grazing problem of soil erosion each cell on the map has a geographic location base on its coordinates in UTM system it will be very interesting to location the land owners with greatest problems of soil erosion if the data on land owner ship can be obtained . Fig.5. Fig.6. Land Suitability Analysis The moisture availabilty map was obtained by combining the soil erosion depth and the soil map. Moisture condition is one of the important factor for agricultural crops soil quality map was derived from soil map. Soil erosion data mositure availability soil quality data and the rock out crop data were over laid and the final map of land suitability prepared which is given in the fig 7. Table 2. Area of potential soil erosion hazard in Phewa water shed Erosion rate Area in ha. Z Very low 3046.912 24.28 Low 1518.258 12.10 Medium 4670.762 37.22 High 2215.493 17.65 Very high 616.985 4.92 Water 482.152 3.84 Total 12550.562 100.00 For a close different control the possible agggregetation of classes representing very different features and in some cases with no relevance at all MATRIX analysis was chosen as apposed to index programmer for instance in a sample addition one imagine different possibility of getting class number 9 going from extremes to another, e.g 9= 8 + 1, 9=1 + 8, 9= 4 + 5, etc. The number of combinations increases obviously a the number of files involved in the operation augments. The table 3 deposits the class values resulted from MATRIX and the corresponding recorded values .following a careful evaluation of the contribution of different factors to the problems a land having good quality in terms of texture depth water holding capacity organic matter on the soil low erosion is considered as suitable area for agriculture purpose. Table 4 shows that 2.10 and very highly suitable 29.08 highly suitable 9.42% medium suitable low suitable and 6.60 very low suitable for agriculture. The elimination factors like settlement areas industrial area transpiration to locate etc should be deducted from the suitability map to actual suitable. This land suitability map may be helpful to the management department to locate which needs better management to project from soil erosion the map that the agricultural land grass land need management practices. Fig.7. Table 3. & Table 4. MISSING Conclusions Digital analysis of land sat data for land use cover type classification in the Phewa water shed has given satisfactory result. This land use cover digital was used for erosion hazard mapping in the area using. This study project demonstrations how GIS can be locating potential soil erosion area and suitable land for agricultural even with get very detailed information . The result obtained from this study project may not be application directly in the field without thoroughly field checking. Never the less the model applied in this study the potential application of land sat digital data the GIS technology in soil erosion hazard mapping and land suitability analysis. Reference Balla, M.K. 1988, application of universal soil loss equation estimate soil loss in Phewa ral watershed forestry journal of institute of forestry Nepal Borrough, A.A 1986 Principles of Geographic information system for land resource assessment oxford university press. De Roo, A.P.J and Hazelhoff L.,1988 Assessing surface runoff and soil erosion in water sheds using GIS mapping technology in environment application of digital mapping. Jensen, Jhon R., 1986 introducing digital image prentice hall. Pradhan, K.P., 1984., Manual versus digital analysis of land sat data for land use cover mapping in the western part of Nepal case study Simonett. O., Turyatunga,F., Witt., 1987 environmental data base developments for assessment of deforestation soil erosion hazards and crop suitability a joint case study second annual international conference exhibits and workshop on GIS san Francisco 544- 553 .