This paper aims to examine the possibility of applying a simple vector autoregressive model to forecast future hydrological series and evaluate the resulting impact on hydropower projects. Three projects are considered - in India, Sri Lanka, and Vietnam. The results are still tentative in terms of both methodology and implications; but the analysis shows that the calibrated dynamic forecasts of hydrological series are much different from the conventional reference points in the 90 percent dependable year. The paper also finds that hydrological discharges tend to increase with rainfall and decrease with temperature. The rainy season would likely have higher water levels, but in the lean season water resources would become even more limited. The amount of energy generated would be affected to a certain extent, but the project viability may not change so much.
Download