Introducing CAFE: The Calculation Assistant for Flow Extremes Decision Support Tool

Natural resource managers require estimates of hydrologic extremes on rivers (e.g., floods and droughts) for many reasons, including establishing water rights, setting minimum flow targets for protecting aquatic species, and developing flood/drought management plans for municipalities. Despite the USGS National Water Information System's extensive network, estimates of hydrologic extremes are often needed on rivers with either limited historical data or at locations where no gaging stations exist. In addition, state water resource managers often lack readily accessible and reliable tools to support estimation of possible changes in future extreme flows related to climate change.

This project extends previous research efforts by the Northeast Climate Adaptation Science Center to forecast climate change impacts on hydrologic extremes in the Northeast U.S. by developing a web-based decision support tool. Three avenues of research were pursued. First, we explored whether readily available “off the shelf” hydrology models could be calibrated with artificial intelligence to estimate low flows in locations without gaging stations. Next, we investigated the application of various artificial intelligence algorithms to better estimate low flows in comparison to current USGS estimates. Finally, we examined the application of more well-established, statistical approaches for estimating the impacts of climate change on future extreme flows. To accomplish these goals, we also conducted interviews with resource managers in the Northeast U.S. to identify current techniques being applied for calculating extreme flows and their effectiveness. These interviews established a stakeholder user-group that provided insight into the development of a new webtool called CAFÉ- Calculation Assistant for Flow Extremes. The webtool estimates hydrologic extremes on any free-flowing stream throughout the Northeast and acknowledges the impacts of climate change on extremes in two ways: 1) By incorporating a guided process that allows the user to utilize more recent historical flows to calculate extreme flows, and 2) By statistically incorporating forecasts of precipitation and temperature changes from global/regional climate change projections into extreme flow calculations.

Andrew DelSanto is a former NECASC fellow and graduate student in the department of Civil and Environmental Engineering at UMass Amherst. Since completing his Ph.D., he has continued in academia as a tenure-track faculty member in the Sleith College of Engineering at Western New England University in Springfield, MA. His research focuses on the application of physically based hydrology models (e.g., rainfall-runoff models) and machine learning algorithms for the analysis of extreme hydrologic flows (e.g., floods and droughts on rivers) with the incorporation of climate change for improved infrastructure design.

Richard Palmer served as the Department Head of the Department of Civil and Environmental Engineering at UMass Amherst from 2008 to 2019. He is also the founding Director of the Northeast Climate Adaptation Science Center and served in that role from 2012 to 2021. His research interests have focused on hydrologic extremes, such as droughts and floods, water resources planning and management, and climate change. He is now professor emeritus at UMass Amherst and lives in Seattle, Washington.