Thumbnail Image

Theses

Browse

20201
Reset filters
Subject: Soil and Water Conservation × Author: Bajirao, Tarate Suryakant ×
Reset filters

Search Results

Now showing 1 - 1 of 1
  • Thumbnail Image
    ThesisItemOpen Access
    Geospatial technology and soft computing for hydrological modelling of Koyna river basin
    (G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2020-02) Bajirao, Tarate Suryakant; Pravendra Kumar
    Conservation of natural resources plays an important role in sustainable development of agriculture. In this study, geospatial technology like remote sensing and geographical information system (GIS) and soft computing techniques like artificial neural network (ANN), adaptive neuro-fuzzy inference system (ANFIS), wavelet coupled artificial neural network (WANN) and wavelet coupled adaptive neuro-fuzzy inference system (WANFIS) were employed for hydrological modelling of Koyna river basin, Maharashtra. Different basin characteristics like topography, slope, runoff characteristics etc. were analyzed and different subwatersheds were prioritized to adopt conservation practices to prevent critically degraded area from further erosion. Original rainfall, runoff and suspended sediment concentration (SSC) time series data were decomposed with discrete wavelet transform using different mother wavelets into different multi-frequency sub-signals. Hybrid, WANN and WANFIS models were developed by coupling wavelet transformed data to single ANN and ANFIS, respectively. The performance of the developed models was assessed using qualitative and quantitative performance evaluation criteria. The impact of land use/land cover change on rainfall-runoff transformation was analyzed using natural resource conservation service curve number (NRCSCN) method. Different morphometric parameters indicated young stage of basin topography. Still the soil erosion process is active in nature. It was observed that Coiflet wavelet coupled ANFIS (WANFIS) models performed the best for daily runoff and SSC predictions. Due to spatiotemporal variability of land use/land cover over 13 years, average runoff coefficient increased from 0.41 to 0.43. The performance of data driven models was observed to be satisfactory for daily runoff and SSC predictions. This analysis is useful for decision makers to start natural resources conservation practices on priority basis and it can also be useful to predict daily runoff and SSC of the Koyna river basin.