Page 58 - GIS for Science Technology Showcases

MODELING GLOBAL
STREAMFLOW
A worldwide service for forecasting water volume
Steve Kopp, Esri
During the last 20 years, issues related to flood, drought, and water affected more than 3 billion people, caused nearly $700 billion in economic damage, and killed an estimated 166,000 people. In most places in the world, people don't know how much water will be in their rivers next week or have reliable historic information to understand past floods or droughts. To help address these problems, Esri, in collaboration with Brigham Young University (BYU), the European Center for Medium-Range Weather Forecasts (ECMWF), the GEO Global Water Sustainability (GEOGloWS) program, and the World Bank, embarked on a plan to provide global streamflow forecasts and historic streamflow information with easy-to-use web maps. These maps and a related web service now provide access to forecast and historic flow for more than 1 million rivers around the world.
The backbone of the project is the ECMWF global runoff forecast and historic modeled runoff. The global runoff forecast is an 18 km resolution, 51-member ensemble model. The historic modeled runoff is a 31 km resolution, 40-year estimate of historic streamflow. BYU and Esri together transformed the surface runoff measurements into the quantity of water flowing in a stream. This transformation is done through a series of GIS steps, computing watersheds from digital elevation models (DEMs), aggregating runoff into watersheds, transforming land runoff into streamflow, and using the Routing Application for Parallel computation of Discharge (RAPID) model to route the flow downstream. The daily processing workflow, which was piloted as a research project at BYU, was successfully transitioned to ECMWF. Now, each day, a new forecast is published and made available through a REST API hosted by ECMWF and a web service hosted by Esri.
Knowing about local water is important to public safety, health, food, energy, and more. By providing free and open web services, an API, and easy-to-use configurable web applications, streamflow forecasting will become as common as a weather forecast. As an example, when Latin America was struck by two Category 4 hurricanes, Eta and Iota, only two weeks apart, the Central American disaster prevention and preparedness organization (CEPREDENAC) used the streamflow service to improve situational awareness and emergency management efforts by understanding how much water would be in the rivers each day as Hurricane Iota approached and moved through the region.
Thanks to our project collaborators: Brigham Young University, European Center for Medium-Range Weather Forecasts, GEO Global Water Sustainability, National Aeronautics and Space Administration, National Oceanic and Atmospheric Administration, SERVIR, World Bank, U.S. Agency for International Development, European Commission, Copernicus, AquaVEO, and International Centre for Integrated Mountain Development.
192 GIS for Science