ISPRS Archives XXXVIII-8/W3 Workshop Proceedings: Impact of Climate Change on Agriculture 344 ALBEDO-RAINFALL FEEDBACK OVER INDIAN MONSOON REGION USING LONG TERM OBSERVATIONS

Albedo determines radiation balance of land (soil-canopy complex) surface and influence boundary layer structure of the atmosphere. Accurate surface albedo determination is important for weather forecasting, climate projection and ecosystem modelling. Albedo-rainfall feedback relationship has not been studied so far using observations on spatial scale over Indian monsoon region due to lack of consistent, systematic and simultaneous long-term measurements of both. The present study used tenday satellite (e.g. NOAA) based Pathfinder AVHRR Land (PAL) datasets between 1981 and 2000 over India (68-100°E, 5-40°N) at 8 km. Land surface albedo was computed using linear transformation of red and near infrared (NIR) surface reflectances. The cloud effects were removed using a smoothening filter with harmonic analysis applied to time series data in each year. The monthly, annual and long term means were computed from ten-day reconstructed albedo. The mean and coefficient of variation (CV) of surface albedo over seventeen years, averaged over Indian land region, were found to show a decreasing (0.15 to 0.14 and 39 to 34%, respectively) trend between 1981 and 2000. Further analysis showed that, among all the land use patterns, the inter-annual variation of albedo of Himalayan snow cover showed a prominent reducing trend (0.42 – 0.35). This could be due to increase in snow melting period and snow melt area. The annual rainfall measured at IMD (India Meteorological Department) gauging stations, interpolated and averaged over 35×33 grids (1° x 1°) also show declining trend (1300 to 1150 mm) during this period, suggesting strong coupling between snow albedo and rainfall. A strong inverse exponential relation (correlation coefficient r=0.95, n=100) was found between annual rainfall and annual albedo over seven rainfall zones.