Hydrological Modelling and data assimilation of Satellite Snow Cover Area using a Land Surface Model, VIC

Abstract

NBSSLUP soil map and land use land cover map of ISRO-GBP project for year 2014 were used for generating the soil parameters and vegetation parameters respectively. The threshold temperature i.e. the minimum rain temperature is -0.5°C and maximum snow temperature is about +0.5°C at which VIC can generate snow fluxes. Hydrological simulations were done using both NCEP and IMD based meteorological Forcing datasets, but very few snow fluxes were obtained using IMD data met forcing, whereas NCEP based met forcing has given significantly better snow fluxes throughout the simulation years as the temperature resolution as given by IMD data is 0.5°C and rainfall resolution of 0.25°C. The simulated discharge has been validated using observed data from BBMB (Bhakra Beas Management Board) and coefficient of Correlation(R 2 ) measured for (2003)(2004)(2005)(2006) was 0.67 and 0.61 for the year 2006.But as VIC does not consider snowmelt runoff as a part of the total discharge, snowmelt runoff has been estimated for the simulation both with and without D.A. The snow fluxes as generated from VIC gives basin average estimates of Snow Cover, SWE, Snow Depth and Snow melt. It has been observed to be overestimated when model predicted snow cover is compared with MODIS SCA of 500 m resolution from MOD10A2 for each year. So MODIS 8-day snow cover area has been assimilated directly into the model state as well as by using EnKF after every 8 days for the year 2006.D.I Technique performed well as compared to EnKF. R 2 between Model SCA and MODIS SCA is estimated as 0.73 after D.I with Root Mean Square Error (RMSE) of +0.19. After direct Insertion of D.A, SCA has been reduced comparatively which resulted in 7% reduction of annual snowmelt contribution to total discharge.The assimilation of MODIS SCA data hence improved the snow cover area (SCA) fraction and finally updated other snow components.

An Integrated Shadow-Adjusted Snow-Aging Index for Alpine Regions

“…Many researches have explored using correlation analysis, sensitivity and response models of snow melting and data pertaining to other parameters mentioned above [38,39]. The classical degree-day model establishes a linear correlation between accumulated temperature and snow melting [40].Theoretically, from these parameters, the continuous monitoring of SAP can be achieved. However, extracting single-dimensional and single-index SAP parameters is not enough to understand the snow ablation process, which is complex and rapidly changing because of multiple factors.…”

“…Many researches have explored using correlation analysis, sensitivity and response models of snow melting and data pertaining to other parameters mentioned above [38,39]. The classical degree-day model establishes a linear correlation between accumulated temperature and snow melting [40].…”