SNOWMELT RUNOFF MODELLING AND SEDIMENT STUDIES IN SATLUJ BASIN USING REMOTE SENSING AND GIS

Abstract

Most of the big rivers in the world originate from the mountainous areas and are considered important sources of water for the large population of the world. Himalayas provide one of the world's largest renewable supplies of fresh water. The chain of the Himalayan mountains act as an effective barrier to the summer monsoon and westerly winter disturbances. Due to this there is massive summer monsoon precipitation especially in the eastern and central parts of the Himalayas and snowfall at higher altitudes in the winters. The major river systems of India, the Indus, Ganges and Brahmputra have their origin in the Himalayan mountain region and snow accumulated during winter becomes a major source of runoff for these rivers during summer. The spring and summer runoff of the Himalayan rivers is mostly from snowmelt is considered a dependable source of water for irrigation, hydroelectric power and drinking water supply. The availability of high runoff coupled with wide variations in elevations provides a large potential for hydroelectric power. India is endowed with enormous economically exploitable and viable hydropower potential mostly in Himalayan region. This is assessed to be about 84,000 MW at 60% load factor. However, only about 20% of the hydroelectric potential have been harnessed so far and 10% is under various stages of development. Himalayan region is witnessing a large economic growth and demands for irrigation water, hydro-electricity are ever increasing. The provision of suitable flood control measure in the lowland areas is also an area of concern. These require proper understanding of the hydrological processes of the snowfed basins of the Himalayan region. Suitable hydrological models have to be developed to simulate these processes for generating the streamflow from such basins using observed flow records. Such studies are much needed for the planning, design, development and management of water resources of the Himalayan region. At the same time, transportation of the high sediments from the Himalayan basins is also one ofthe serious problems for any water resources development project. The capacity of the reservoirs located in this region reduces at an alarming rate due to deposition of sediment, thus adversely affecting the optimal operation of the water resources schemes. The present study is therefore focussed on snowmelt runoff modelling and sediment studies. An extensive literature survey has been made on snowmelt runoff processes and their modelling have been also reviewed. The relevant literature on sediment studies has also been reviewed. The remote sensing and GIS techniques have good potential for application in both these areas. The present study also includes a detailed review of such applications. The review of the studies in the Himalayan region shows that the modelling of streamflow for the snowfed basins and sediment studies are very limited. Therefore, it is planned to consider these important aspects for a Himalayan basin in the present study. Important snow covered basins in the western Himalayas are Satluj, Beas, Chenab, Ravi and Jhelum. Satluj basin has a significant contribution from snow/ice and has a reasonably good hydrometerological network. The hydropower potential of Satluj basin is 8634 MW, which constitutes almost 50% potential of the state of Himachal Pradesh. The Bhakra Nangal project on the Satluj River is a major project producing large amounts of hydroelectric power besides irrigation of large areas in Northern India. The Bhakra dam on Satluj River is considered a boon for India. Some of the other prestigious projects such as Nathpa-Jhakri, Baspa, Kol dam and Karchham Wangtoo are under construction. Therefore, for ensuring representativeness and possible operational use, the study area chosen for this study is Satluj basin (Indian part) up to Bhakra dam. ii The proper estimation of snow and glacier contribution in annual flow of a snowfed river is very important for planning and development of water resources. At present there is no systematic method available for this purpose for a snowfed basin in Himalayas having limited information on solid precipitation. In the present study, a methodology has been developed using a water balance approach for estimating snow and glacier contribution in the annual flows of a snowfed river. For this purpose, water budget analysis for the Satluj basin was made for a period of 10 years (Oct. 1986 - Sept. 1996). Precipitation data of 10 stations were used to compute total input to the basin over the water budget period. Total volume of flow for the same period was computed using discharge records at Bhakra dam. Evapotranspiration losses from the basin were estimated using a relationship developed between temperature and evapotranspiration. Snowcover and snow free areas needed in evapotranspiration computation was obtained using remote sensing data (Landsat-MSS, IRS-1A/1B-LISS-II) and image processing have been carried out on ERDAS IMAGINE software. It was found that generally a major part of the basin (65%) is covered with snow in the months ofMarch/April, which reduces to about 20% in September/October. The average contribution of snow and glacier melt in the annual flow of Satluj River at Bhakra was found to be about 59% and remaining 41% was from the rain. It shows that snow and glaciers provide a substantial contribution to the flows of Satluj River. For all the sites upstream of Bhakra dam, the contribution of snow melt runoff in total runoff would be higher than 59% due to higher percentage of snow covered area in upper basins. These findings, of the present study are thus very relevant for all the projects planned/under execution in the upstream of Bhakra dam. The Himalayan basins have runoff contribution for rainfall as well as snow and ice. Inspite of significant contribution of snow and ice in the Himalayan rivers, no attempts have been made to develop a simple scientifically based snowmelt model for the in Himalayan basins. Therefore, development of a snowmelt runoff model, which caters for the sparse network, limited availability of data and rugged topography of the Himalayan basins and also handles rainfall, is of utmost importance. Therefore, simple conceptual snowmelt model (SNOWMOD) based on temperature index approach has been developed for the Satluj basin having range of elevation from 600 m to 6000 m. The basin was divided into suitable number of elevation bands. Daily precipitation and temperature are the basic inputs to the model beside some useful information derived from topographical maps and satellite data. The temperature was extrapolated to different elevation bands using temperature lapse rate. Snowcovered area in the basin was determined using remote sensing data from Landsat-MSS, IRS-1A/1B-LISS II and IRS-WiFS. Contour and spot height data from Survey of India toposheets have been used to generate digital elevation model (DEM), which was used subsequently to prepare the area elevation curves. For this purpose, ILWIS GIS software has been used. Keeping in view, the physiographic features of the basin and elevation of meteorological stations, each elevation band was assigned to a representative meteorological station. The model structure is simple and involves use of relevant physical parameters. Only four parameters related with flow routing for snow covered and snow free areas need calibration. The model computes daily runoff from the snowcovered area and snow free area separately. The values of parameters representing degree day factor and runoff coefficients have been decided on the basis of literature review and have been varied from month to month to account for seasonal effect/behaviour of temperature and rainfall. The routing of surface and sub-surface flow is made using concept of linear cascade of reservoirs. The total outflow from the basin was computed by summing the different components of runoff. The model was calibrated using the dataset for a period of three years (1985-86 to 1987-88) and model parameters for structural routing were optimised. Because hydrological response of snow covered area IV and snow free area is very much different, therefore separate parameters were considered for these parts of the basin. Using the optimised parameters, streamflow simulations were made for two data sets, each containing 3 years period (1987-88 to 1990-91) and (1996-97 to 1998-99). The accuracy of the streamflow simulation were also determined using different criteria such as shape of the outflow hydrograph, efficiency, difference in volume. In all the three cases (one for calibration and two for simulation), model successfully simulated the observed flow and efficiency of the model varied between 81- 94%. Increasing population and rapid development activities in the geologically fragile Himalayan region are causing major land use changes and hence increasing problems of soil erosion and sedimentation. In recent years, the ecology and hydrology of the Himalayan region has been greatly affected from intensive deforestation, large-scale road construction, mining and cultivation on steep slopes etc. The study of erosionsedimentation processes and procedures for estimation of the sediment yield and sedimentation forms an important part of the present study. For the assessment of sediment yield, two approaches have been used: (i) relationship between suspended sediment load and discharge, and (ii) empirical relationship consisting of rainfall and physiographical data of the basin. The first approach was used for Satluj basin at different locations such as at Suni (52,983 km ), at Kasol (53,768 km2) and also for the intermediate basin between Kasol and Suni (785 km ). A relationship between sediment and discharge was developed using data of a period of three years (1991-1993). It was applied for the data sets of two independent years, 1994 and 1996, for estimation of sediment yield. The second approach, whichgives annual sediment yield, was used for a small intermediate basin because of the following two reasons: (a) this part is highly soil erosion prone area in the whole basin and; (b) the required physiographical data was available only for this part of the basin. For estimation of the sediment yield using the empirical relationship, various physiographical parameters representing land use, slope, and drainage density were generated using ILWIS GIS system. The annual sediment yield for the intermediate basin was estimated for 3 years and compared with observed sediment yield. The difference between computed and observed sediment yield and its trend was found to be related to orography. The available empirical relationship was therefore, revised by incorporating an orographic factor in the equation. This orographic factor represented the spatial variation of the rainfall in the basin. Using the revised empirical relationship, the sediment yield was estimated for two independent years and a good correlation was found between computed and observed sediment yield. The sediment generated from the Satluj basin is deposited in the Bhakra reservoir located in the foothills of the Himalayas. Reduction in the storage capacity of a reservoir beyond a permissible limit hampers the purpose of the reservoir for which it was designed. Thus, assessment of sediment deposition becomes very essential for the management and operation of such reservoirs. At present, some conventional methods such as hydrographic survey and inflow-outflow approaches are used for estimation of sediment deposition in a reservoir. But, these methods are cumbersome, data intensive, time consuming and expensive as well. There is a long felt need for developing a simple method, which requires less time and is also, cost effective. In the present study, a remote-sensing based approach has been developed for the assessment of sedimentation in a reservoir and applied for the Bhakra reservoir. Multi-date remote sensing data (IRS-1B-LISS II) was used to obtain the information on water spread area of the reservoir, which was used for computing the sedimentation rate. The revised capacity of the reservoir between maximum and minimum levels for the two periods, i.e., 1988-89 and 1996-97 was computed using the VI trapezoidal formula. The loss in reservoir capacity due to deposition ofsediments for a period of 24 years (1965-1989) was determined and it was found to be 491.315 Mm3. For aperiod of 32 years (1965-1997) it is worked out to be 807.35 Mm3. Assuming the uniform rate of sedimentation in the reservoir, the rate of sedimentation was found to be 20.47M m per year and 25.23 M m3 per year over a period of 24 years and 32 years respectively. The corresponding value of average rate of sedimentation for this reservoir using hydrographic survey were 21.93 Mm3 and 20.84 Mm3 per year. The results obtained using remote sensing based approach reasonably agree with the results obtained from hydrographic survey. The present study has resulted in (i) development and application of a simple and systematic approach for estimation ofsnow and ice melt contributions in the streamflow of Satluj basins (ii) development ofsimple conceptual model for daily streamflow simulation for Himalayan rivers having rain as well as snow/ice melt contributions, and its application in calibration as well as simulation model for Satluj basin (iii) simple approach for sediment yield estimation incorporating orographic factor, and (iv) remote sensing based approach for assessment of sedimentation in reservoirs. The application of these models and approaches for Satluj basin has been quite satisfactory. Based on the present research work, the following recommendations are made for future studies and applications, particularly in Himalayan region. The water balance approach developed in this study can be applied for estimation of snow and ice contributions for the other Himalayan basins too. The snowmelt model (SNOWMOD) was developed and applied for Satluj basin for streamflow simulation. There is scope for extending such studies for the other Himalayan basins. SNOWMOD may be applied for streamflow forecasting in the Himalayan rivers, which would require an additional information on the meteorological parameters in advance. Vll Sediment yield and sedimentation studies in the other parts of the Himalayan region can also be estimated with the approaches developed and presented in this study. There is also much scope for application of the remote sensing based approach for assessment of sedimentation in reservoirs located in Himalayan region and elsewhere. Considering the time and cost involved the hydrographic surveys may be conducted at longer interval and the remote sensing based sedimentation surveys may be carried out at shorter intervals. Such practice will be very economical and quicker without any loss in the accuracy of results.