Abstract:
Because of changing climate, inadequate data and anthropogenic effects on water resources, spatial and temporal information about water resources is becoming necessary worldwide and this limitation can be overcome with the availability of open access earth observation datasets. The purpose of the study was to estimate the stream flow of Hunza and Astore sub-basin using remote sensing datasets and validate results with the field observation data using R2. Pixel-based water balance was quantified by segregating precipitation into evapotranspiration, runoff, and potential ground-water infiltration. Open access remotely sensed Rainfall products (TRMM and CHIRPS), evapotranspiration dataset (SSEBop), leaf area index dataset (MODIS), and soil Water Index (SWI) dataset of The Advanced Scatterometer ( ASCAT) were used in the water balance equation for Rainfall-runoff estimation. In this study, precipitation dataset TRMM is used in water balance calculation because it represented a better relationship with gauge data R2= 0.74 for Astore basin and R2= 0.71 for Hunza basin as compared to CHIRPS precipitation data R2=0.68 for Astore basin and 0.66 for Hunza basin. Resulted values from the remote sensing model was compared and validated with field observation data using coefficient of determination R2. Results represented a good relationship among estimated runoff values from the water balance model and field data (WAPDA) with R2 value of 0.75 in Astore basin and R2 value of 0.70 in Hunza basin respectively. The presented approach can be used in ungauged basins without any need of field data. The study helped in estimating stream flow by fast and cost effective method without complex hydrological models requiring intensified data and tuning.
