Statistical downscaling of climate projections for hydrologic analysis and impact assessments

Recent advances in science and technology now provide useful tools as well as accurate and real-time datasets which can be used for more informed decision-making at different levels in various sectors. Advanced information on the weather systems and occurrence of extreme hydrologic events based on ENSO/SSTA measurements at Niño 3.4 region can guide stakeholders in the agricultural sectors regarding agricultural production activities and in planning climate change adaptation strategies. An efficient early warning system (EWS) can be established to forewarn on possible occurrence of extreme events. Timely implementation of response strategies and safety nets based on reliable climate projections is expected to significantly reduce the losses on crops, properties, and lives. Moreover, climate change including extreme climatic variability directly affects and impacts the environment, agriculture, livelihoods, assets and infrastructure. It threatens food security and ecological stability in a region. Impacts and vulnerability assessments, hydrologic frequency analysis, and climatic risk analysis for future conditions for a particular area require liner resolution data. Thus, the need to have more reliable and adequate climate projections based on global climate scenarios for the anticipated planning periods in the future involves the use of efficient downscaling techniques. Thus, general circulation model (GCM) projections have to be downscaled using a dynamical procedure like a regional climate model (RCM), or a statistical downscaling technique using regression modeling, stochastic weather generators, weather typing procedure, and transfer functions. Ready access to available historical weather data, and the availability of software for statistical downscaling will further advance quantitative assessments of climate change impacts in hydrology and water resources, agricultural production system, and also in climate adaptation planning and risk reduction management. In this study, statistical downscaling or climate projections is presented. Downscaled datasets on extreme events are presented by comparing the distributions, frequencies, magnitudes, and other statistics of annual maximum values in the series under different time periods. The effect on yield probabilities and on cropping calendar are also demonstrated for selected locations in the Philippines. Some priority issues and challenges as well as opportunities are also discussed.