TR(BR)-13/97-98 : Development of relationship between glacier melt runoff and meteorological parameters

Abstract

In order to assess the predictive significance of the meteorological parameters to forecast the discharge for a high altitude glacierized basin, correlations have been determined between mean daily discharge and precipitation and temperature with a time lag of 0 to 3 days. Dokriani glacier basin located in the Garhwal Himalayas is considered for this purpose. Total drainage area of this glacier is about 16.13 km'', out of which about 9.66 km' (60%) is covered by snow and ice. Discharge-auto correlations are also attempted. Meteorological data used were collected by establishing a standard meteorological observatory at an altitude of about 4000m near the snout. In order to get discharge data from the study basin, an automatic water level recorder was installed at the gauging site established just very close to the snout of glacier. Continuous records of discharge were collected for the summer period when intensive melting takes place. The data collected for summer period during last three years (1995-1997) was used for the present analysis.

During the ablation period, the relations between the glacier melt water yield and meteorological parameters are not stationary. Therefore, correlation and auto-correlations have been developed both for monthly and seasonal basis for different years. Discharge autocorrelation was found to be very high for each month and for a season as a whole. It suggests a strong storage and drainage characteristic of the glacier located in the study basin which results in a delayed response of the input available at the glacier surface either in the form of rain or melt water to the runoff at the outlet of the glacier. For different months and seasons, changes in the correlations between discharge and temperature, discharge and rain with different time lags are noticed. Variation in physical features of glacier with time, weather conditions, rainfall and its distribution over the basin have attributed for the changes in correlations. Multiple linear regression equations are developed using discharge as dependent and temperature with lag 0 and 1 day and rainfall with lag 0 and 1 day as independent variables. To arrive at statistical significance of the independent variables in the simple regression equations. stepwise regression approach was used.