DESC0023920

Meeting current and next-generation clean energy targets will require significant advancements in tools and models to accurately project long-term water supply in reservoirs and rivers with generation potential. Utilities and water managers across sectors currently rely on historical data or traditional process-based models to inform the next 50 years of water supply availability and the impact of climate change, despite strong evidence that the past no longer represents the future. Water managers planning on the long-term horizon need help processing and parsing through the vast amount of data available, and transforming it into actionable information for making decisions. Upstream Tech proposes a cutting edge solution, HydroForecast Long-term, that combines the most accurate streamflow prediction modeling system with a flexible and scalable data architecture to generate water supply projections out to the year 2100. Upstream Tech has a proven industry record of producing high quality, operational forecasts within the hydropower and water utilities sectors, with a high emphasis on integrating and contributing to academic research at the forefront of hydrology. Our approach for Phase I has four objectives: 1) create a prototype of HydroForecast Long-term, building the neural network prediction model, 2) build an automated data input pipeline that processes large amounts of data from the latest global temperature and precipitation climate models; 3) benchmark the accuracy of this model over the recent two decades over a large set of diverse basins, and 4) create a set of output visuals and summary metrics informed by customer feedback that connect the data to critical decision points. At the end of Phase I, a prototype of HydroForecast Long-term will be implemented, validated for accuracy across a diverse set of basins, and packaged into a set of visuals and key metrics based on feedback from target market customers. Two user groups will directly benefit from the Phase I achievements: hydropower operators will have better data to inform long-term portfolio management and expected supply for generation and investment, helping support a resilient, renewable-powered grid even as the climate changes; and, water supply utilities will have forward looking data to use in their Water Supply Plans for meeting municipal and regional demands. More broadly, any industry segment investing in long-term property and infrastructure will benefit from data that is more accurate and easier to access and understand. If this work is funded beyond a Phase I, additional input data sources and data analytic tools will support applications in agriculture, broader renewable energy development, city planning, water-intensive industries, and environmental groups, sectors who need similar data to inform long-term water availability, but with unique questions and data packaging needs.