Loading...
Thumbnail Image
Publication

DEVELOPMENT OF A DECISION SUPPORT SYSTEM WEBTOOL FOR HISTORIC AND FUTURE LOW FLOW ESTIMATION IN THE NORTHEAST UNITED STATES WITH APPLICATIONS OF MACHINE LEARNING FOR ADVANCING PHYSICAL AND STATISTICAL METHODOLOGIES

Abstract
Droughts are a global challenge and anthropogenic climate change is expected to increase the frequency and severity of extreme low flow events. A major challenge for resource managers is how best to incorporate future climate change projections into low flow event estimations, especially in ungaged basins. Using both physically based hydrology models and statistical models, this dissertation contributes novel methodologies to three key challenges associated with 7-day, 10-year low flow (7Q10) estimation in the northeast United States. Chapter 2 builds upon statistically based 7Q10 estimation in ungaged basins by comparing multiple machine learning algorithms to classical statistical methodologies. This chapter’s objective is to identify a robust statistical methodology applicable for the entire northeast U.S. that includes statistically significant climate variables that allow for the incorporation of climate change. Results suggest that the random forest method can provide regional 7Q10 estimates with similar errors to current, state-by-state 7Q10 estimates. Chapter 3 tests the applicability of a novel machine learning algorithm, Fuzzy C-Means clustering, to calibrate rainfall-runoff models in ungaged basins for both daily streamflow and 7Q10 estimation. Future updates to national rainfall-runoff models, which can directly incorporate climate change projections into calculations, will allow these models to be created in ungaged basins, but they will require extensive calibration and/or verification. Results suggest that this methodology significantly improves daily streamflow estimation but fails to improve 7Q10 estimation. Chapter 4 summarizes the development of a stakeholder-driven decision support system (DSS) web-application for calculating the 7Q10 at gages and estimating the 7Q10 in ungaged basins with projected climate changes. By incorporating the statistical model from Chapter 2 into the DSS and comparing the results to the physical modeling from Chapter 3, the DSS can estimate the impact of future temperature and precipitation changes on 7Q10s. This work highlights advancements in physical and statistical modeling techniques for 7Q10 estimation in ungaged basins and assists resource managers in addressing a growing need for incorporating anticipated climate change into drought calculations.
Type
dissertation
Date
2024
Publisher
Rights
License
Research Projects
Organizational Units
Journal Issue
Embargo
Publisher Version
Embedded videos
Collections