Operational Water Prediction in Highly Regulated, Transboundary Watersheds using Multi-Basin Modelling, Earth Observations and Co-production
Useful, actionable and timely forecasts of surface water resources across time and space can provide informed science-based information for policymakers. In developing countries, sparse in-situ monitoring networks, limited data access and rapid growth of reservoirs and dams inhibit the accurate estimation and verification of forecast information, particularly at transboundary watersheds where rivers cross administrative borders. Open-source watershed modelling, publicly available satellite-based observations and climate data provide promising tools and data for estimating and forecasting surface water resources in data poor region. The vast amount of available information and technology, nevertheless, does not imply absolute translations into actionable applications in local areas. Estimation uncertainties, data latency, accuracy levels, stakeholders’ capacity, governance culture, and needs can hinder the pathway to bring more sustainable societal benefits. The objectives of this dissertation were to quantify and forecast availability of surface water resources in highly regulated, transboundary watersheds across time and space while improving the potentials of multi-basin watershed modelling, earth observations and co-production to inform decisions on operational water prediction. Three studies were performed. Results demonstrate that non-regulated surface water resources can be reasonably estimated through innovative approaches using regionalization derived from similarity of physiography and climate, and flow correlation of water levels derived from satellite altimetry. Results also indicate that highly-regulated surface water resources can be competently estimated when including storage and operation of all existing upstream reservoirs using an Integrated Reservoir Operation Scheme with support of remote sensing imagery. Quality of hydrological forecasts is found to vary geographically and deteriorate with increasing lead time. Inclusion of reservoir operation, nevertheless, can improve the quality and usefulness of forecast information at locations affected by upstream dam activities. Finally, participatory co-production approach in highly challenging transboundary watersheds of the Greater Mekong region provides important unprecedented implications in bridging the gap between science and practice. The findings from the dissertation can be beneficial for future researches in transboundary water resources management, particularly in developing nations, towards enhancing resilience and catalyzing adaptation to social and environmental change of climate-vulnerable marginalized populations.