GEOSPATIAL MODELING OF SNOW-MET PARAMETERS AND ESTIMATION OF ENERGY FLUXES

Abstract

Surface energy fluxes of the cryospheric regions have many applications in climatology, hydrology, glaciology, snow avalanche forecasting and other snow/ice related studies. The components of the surface energy fluxes are net shortwave radiation flux, net longwave radiation flux, sensible heat flux, latent heat flux and sub surface heat flux. A number of snow-meteorological parameters may be used in estimation of these components of surface energy fluxes. These include air temperature, surface temperature, relative humidity, wind speed, atmospheric pressure, snow depth, albedo, cloud amount etc.. Cryospheric regions have limitations of poor monitoring using sparse in situ snow-met observations, which may therefore not characterize spatial variation of energy fluxes over a large snow/ice covered regions. Attempts are being made to model snow-met parameters at spatial scale and to estimate surface energy fluxes of large snow/ice covered regions using satellite remote sensing in conjunction with in situ observations. The present research focuses on geo-spatial modeling of snow depth, albedo, surface temperature and estimate surface energy fluxes using in situ as well as remote sensing observations for the snow/ice covered cryospheric regions of Antarctica and Western Himalaya. The work carried out in this thesis consists of three parts. The first part covers the geospatial modeling of few snow-met parameters e.g. snow depth, albedo and surface temperature. The second part focuses on the estimation of surface energy fluxes using in situ recorded snow-met data and presents analysis of the temporal variability of energy fluxes for four years. The third part presents the spatial estimation of surface energy fluxes, their evaluation using in situ recorded data and analysis of spatial variation in north and south aspect slopes in the mountain topography. A novel algorithm for geospatial interpolation of snow depth in Western Himalaya has been developed using snow depth data of manual observatories and digital elevation model. The algorithm improves upon the limitations of earlier published snow depth interpolation algorithm. The algorithm has been used to produce snow depth maps at spatial resolution of 0.5 km. These maps have been validated at few remote locations of the study area and an overall correlation coefficient of 0.71 and RMSE of 42 cm between estimated and in situ collected snow depth have been obtained. The proposed algorithm has advantages over the previous models of snow depth in Western Geospatial modeling of snow-met parameters and estimation of energy fluxes [ii] Himalaya by having higher spatial resolution and applicable for all snow thicknesses. Algorithms for direct retrieval of snow broadband albedo using AWiFS and MODIS data have been developed and presented. In situ measurements of spectral reflectance and transmitted solar irradiance using data collected from spectroradiometer in field have been used for the development of snow broadband albedo from narrow band AWiFS and MODIS data. The retrieved albedo from AWiFS and MODIS data has been validated with in situ measurements. The overall R2 and RMSE values between estimated and in situ recorded albedo values for AWiFS sensor have been observed as 0.94 and 0.03. The corresponding values for MODIS sensor have been observed as 0.88 and 0.026 respectively. The algorithm developed for estimation of broad band albedo using narrow band reflectance of AWiFS images is probably one of the first attempts in this direction. Another algorithm has been developed for the estimation of surface temperature in the study area of Antarctica using split-window technique. The R2 and RMSE of 0.99 and 0.8°C respectively have been obtained between estimated and in situ recorded surface temperature. Surface energy fluxes have been estimated using in situ as well as remote sensing data. In situ data have been used to estimate surface energy fluxes at the edge of the Antarctic ice sheet. A four-year analysis of the meteorological parameters, radiative and turbulent energy fluxes have been presented. The energy fluxes have been analysed for summer season, winter season and transition periods. It has been observed that the meteorological conditions at the observation site have generally been characterised by mild air temperature (annual mean -10.2 °C), low relative humidity (annual mean 50%) and high katabatic winds (annual mean 8.3 m s-1). Net radiative flux has been the main heat source to the glacier during summer (summer mean 46.8 Wm-2) and heat sink during winter (winter mean -42.2 Wm-2). Sensible heat flux (annual mean 32 Wm-2) has been the heat source and latent heat flux (annual mean -61 Wm-2) has been the heat sink to the glacier surface throughout the year. The study highlights the high latent heat flux at the edge of the ice sheet compared to other coastal locations in Dronning Maudland of Antarctica. This may be due to mild temperature, low relative humidity and high katabatic wind compared to other locations. High latent heat flux causes high sublimation rate equivalent to 5.29 cm w.eq per month at the study site. This may contribute to higher rate of ablation of the ice sheet at the location. Abstract [iii] Surface energy fluxes estimated using in situ recorded data do not characterize spatial variation of energy fluxes of large snow/ice covered regions. So, energy fluxes have been estimated at spatial level using MODIS data and in situ recorded snowmeteorological data in Western Himalaya and Antarctica. Incoming shortwave radiation flux and net shortwave radiation flux have been estimated at spatial resolution of 0.5 km whereas net longwave radiation flux and net radiation flux have been estimated at spatial resolution of 1.0 km. These estimated energy fluxes have been evaluated at sampled locations using automatic weather stations data. RMSE in estimation of incoming shortwave radiation flux, net shortwave radiation flux and net radiation flux have been found to be 75 W m-2, 84.9 W m-2, 90 W m-2 respectively in Western Himalaya and 105 W m-2, 75 W m-2, 81 W m-2 respectively in Antarctica. In this research, the incoming shortwave radiation flux has been estimated at higher spatial resolution than those reported in earlier studies and at a comparable accuracy of 14-27% of the mean values. As there is no access to data from any other source on spatial estimation of net shortwave radiation and net radiation fluxes for snow/ice covered region, the results of this study therefore have not been evaluated any further. Spatial and temporal variations of energy fluxes on north and south aspect slopes of mountain topography in Western Himalaya have also been assessed. Incoming shortwave radiation flux on south aspect slopes has been observed to be higher than those observed on north aspect slopes for the study period. This may be due to low incidence angle (i) of solar radiation on south aspect slopes as compared to that on the north aspect slopes. Temporal variation in the incoming shortwave radiation flux has been found to be in accordance with the variation of solar zenith angle. However, the temporal variation of net shortwave radiation flux has been found to depend on incoming shortwave radiation flux and the albedo of the snow cover. The significant findings of this research include, 1. Development of algorithm for geospatial interpolation of snow depth, which has advantages over the previously reported algorithms. The algorithm shall have direct applications in spatial estimation of sub-surface heat flux, avalanche forecasting, hydrological and glaciological studies. 2. Development of an algorithm for broad band albedo of snow cover from narrow bands reflectance of MODIS and AWiFS images. The algorithm can be used in various snow studies for estimating radiative energy fluxes and snowmelt run-off modeling. The Geospatial modeling of snow-met parameters and estimation of energy fluxes [iv] study has focused on estimation of the sublimation and melt of the ice sheet for four years using in situ recorded snow-met parameters. The study highlights the high ablation rate of the ice sheet near Schirmacher Oasis, compared to other parts of East Antarctica. 3. Estimation of surface energy fluxes at spatial level using remote sensing technique and validation with in situ recorded data. The results may be directly applicable to various snow/ice studies in the cryospheric regions.