Probabilistic Streamflow Forecasts to Advance Flood Preparedness: Statistical Applications and User Perspectives
Year: 2021
Type: dissertation
Abstract: Disaster planning has historically allocated minimal effort and finances toward advanced \npreparedness; however, evidence supports reduced vulnerability to flood events through \nappropriate early actions, saving lives and money. Among other requirements, effective early \naction systems necessitate the availability of high-quality forecasts to inform decision making. \nChapter 2 of this work evaluates the ability of statistical and physically based season-ahead \nprediction models to appropriately trigger flood early preparedness actions for the flood-prone \nMarañón River and Piura River in Peru. A tailored statistical streamflow forecast developed in \nthis work demonstrates superior performance to an operational forecast at both study locations. \nContinued efforts should focus on applying this season-ahead prediction framework to additional \nflood-prone locations where early actions may be warranted and current forecast capacity is \nlimited. \n \nChapter 3 leverages an online questionnaire to explore perceptions of probabilistic river forecasts \namong county-level emergency managers in the Upper Midwest, specifically focusing on factors \naffecting use of probabilistic river forecasts for flood preparedness. Forecast use is found to be \nstatistically significantly related to emergency management budget, availability of forecast \nlocations, trainings, frequency of flood preparedness, perceived degree of flood preparedness, \nperceived value of forecasts, and forecast familiarity. Additionally, a multiple logistic regression \nmodel predicts 68% of the variance in forecast use, identifying significant positive associations \nwith forecast familiarity and full-time employment status and a negative association with \ncommunication with forecast providers. While results suggest forecasts provide value to \nemergency managers, many respondents indicate further improvements are possible. Further \ninvestigation into the forecast use-flood preparedness relationship is warranted.
Author: Colin P. Keating
Institution: University of Wisconsin–Madison
Language: English
Subfield: Global and Planetary Change
Field: Environmental Science
Domain: Physical Sciences
Open Access status: green
Awards: G1819919265, G245808252, G7847486505
OpenAlex ID: W3183654686
