Sediment Study in Western Part of Himalayan Watershed Using Remote Sensing and Gis

Abstract

Thesis Title: Sediment Study in Western part of Himalayan Watershed Using Remote Sensing and GIS Name of Advisor : Dr. B. R. Singh India is known for its wealth of natural resources. Soil erosion affects about 53% of the total land area of the country. Sedimentation and Soil erosion by water is pronounced critical problem in Himalayan terrain due to anthropogenic pressure on its mountainous landscape. The present study was carried out to sediment study in western part of Himalayan watershed using remote sensing and GIS from Rampur to Kasal are considered in Himachal state with the objectives of (i). To estimate the runoff ~nd sediment yield using Arc SWAT (Soil and Water Assessment tool). (ii) To validate the estimated runoff and sediment yield for the watershed. Application of Arc SWAT2005 in integration with Remote Sensing and GIS to estimate the runoff, sediment yield and other hydrological parameters of the area Rampur and Kasal an intermediate basin of Satluj river located in western Himalayan regions in India. The total geographical area of Satluj catchment upto Bhakhra dam is about 56,980 km2 ofwhich about 37,153 km2 1ies in Tibet. The rest about 19,827 km2 1ies in the Indian Territory. The major portion of the Satluj basin lies in the greater Himalayan range. The bed slope of Satluj from its source to Bhakhra dam site is quite uniform. The elevation of the bed is 914 mat Rampur. The catchment receives heavy rainfall during the monsoons from July to mid September; sometimes rainy season extends up to late September and very rarely up to early October. The average rainfall in the catchment is 1145 mm: Kasal is almost at the tip of the reservoir. This area comes in the lower shiwaliks with mountain peaks up to 2134 m height and the rainfall in this region is heavy i.e. 913 mm to 1635 mm. This area being more populated than the higher regions has poor forests and more cultivation and this sediment yield is the heaviest. The total length of the satluj River is 1 ,448 km and 1 02 km falls in the current study area. The OEM was generated to prepare slope map. The Soil Map was prepared from NBSS&LUP soil map with a scale of 1:500,000. About 60.45 percent ct soil area feE under Fine-Loamy soil ta\.rg Hydrological group-O crd 17.94 percent of soil are a falls under Course Loamy soil with hydrological Group-D. In the study area 26 sub watersheds are formed. The area delineated by the Arc SWAT interface was found to be 81761ha against the manually delineated area of 81790 ha. The error of calculation was found to be 0.01 %. Rampur to Kasal an intermediate basin of Satluj river was used to estimate the runoff and sediment yields for the data of mainly eleven years from 1990 to 2000. The calibration period reported the value of R2 is 0.55 and 0.92 for daily and monthly results. The values of Nash-Sutcliffe R2NS for daily and monthly results were found to be 0.74 and 0.87 respectively. The model estimated results showed the correlation during sediment calibration having R2 value are 0.35 & 0.40 for daily and monthly basis respectively. The graphical validation analysis showed the R2 values is 0.65 for monthly runoff values for 1998 to 1999 with RNs values 0.46 and 0.66 for daily and monthly respectively. The validated R2 value was observed to be 0.50 for monthly sediment yield values for 1998 to 1999 respectively. There is reasonably good results between estimated and observed daily and monthly data for these years and for the current study basin from Rampur to Kasal. The hydrological SWAT model (Soil & Water Assessment Tool) was applied for the estimation of the runoff and resulted sediment yield. The weakness of the model is to simulate sediment was due to the improper peak runoff simulation and the nature and accuracy of the measured sediment data. The model was validated for the daily and monthly surface runoff for the year 1998 to 2000 with the corresponding measured rainfall data. During the year 1995, 1996 and 1997 the simulated runoff was 992.10, 823.90 & 990.20 (mm) respectively as against the observed runoff of 1039.90, 1336.651 & 1272.035 (mm) from a total rainfall of 1456, 1358 & 1650 (mm) correspondingly. The graphical analysis showed the R2 values is 0.62 for monthly runoff values for 1998 to 2000 having R2NS values 0.46 and 0.66 for daily and monthly respectively. This might be due to less numbers of rain gauge stations within the study area, as in the upper reach of the basin. The other reason may be the inaccuracy of the observed data as in few months surface runoff has been observed. Validation has also been done for daily and monthly sediment yield for the year 1998, 1999 and 2000. It was observed that the model has predicted more sediment load in most of the events. The R2 value was observed as 0.50 for monthly sediment yield values for 1998 to 2000 respectively. The total simulated sediment yield was observed to be 40.124, 25.35 & 48.664 (t/ha) as against the observed sediment yield of 133.245, 145.850 & 140.565 (t/ha) respectively for the validated year 1998 to 2000.