Evaluation and economic value of winter weather forecasts

Derrick W Snyder, Purdue University

Abstract

State and local highway agencies spend millions of dollars each year to deploy winter operation teams to plow snow and de-ice roadways. Accurate and timely weather forecast information is critical for effective decision making. Students from Purdue University partnered with the Indiana Department of Transportation to create an experimental winter weather forecast service for the 2012-2013 winter season in Indiana to assist in achieving these goals. One forecast product, an hourly timeline of winter weather hazards produced daily, was evaluated for quality and economic value. Verification of the forecasts was performed with data from the Rapid Refresh numerical weather model. Two objective verification criteria were developed to evaluate the performance of the timeline forecasts. Using both criteria, the timeline forecasts had issues with reliability and discrimination, systematically over-forecasting the amount of winter weather that was observed while also missing significant winter weather events. Despite these quality issues, the forecasts still showed significant, but varied, economic value compared to climatology. Economic value of the forecasts was estimated to be $29.5 million or $4.1 million, depending on the verification criteria used. Limitations of this valuation system are discussed and a framework is developed for more thorough studies in the future.

Degree

M.S.

Advisors

Baldwin, Purdue University.

Subject Area

Atmospheric sciences

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS