Forecasting reservoir inflows using remotely sensed precipitation estimates: a pilot study for the River Naryn, Kyrgyzstan

This study explores the feasibility of applying remotely sensed precipitation estimates (in this case from the Tropical Rainfall Measuring Mission [TRMM]) for forecasting inflows to the strategically important Toktogul reservoir in the Naryn basin, Kyrgyzstan. Correlations between observed precipitation at Naryn and 0.5° TRMM totals is weaker for daily (r = 0.25) than monthly (r = 0.93) totals, but the Naryn gauge is representative of monthly TRMM precipitation estimates across ~60% of the basin. We evaluate predictability of monthly inflows given TRMM estimates, air temperature and antecedent flows. Regression model skill was superior to the zero order forecast (mean flow) for lead-times up to three months, and had lower errors in estimated peaks. Over 80% of the variance in monthly inflows is explained with three-month lead, and up to 65% for summer half-year average. The analysis also reveals zones that are delivering highest predictability and hence candidate areas for surface network expansion.