Using MODE to Calculate the WPC, NBM, and HAFS QPF Displacement Error for Landfalling Tropical Cyclones

Many different hazards are associated with landfalling tropical cyclones. Extreme wind and storm surge are two of the most common hazards linked to tropical cyclones that impact land. However, tropical cyclones often produce torrential rainfall that can lead to flash flooding and be just as or even more dangerous than wind or storm surge. Hazardous rainfall can affect areas not just along the coast, but further inland as well. A complicating factor in the prediction of rainfall produced by tropical cyclones is that amounts and impacts are not directly related to the strength of the storm and can depend on storm speed, storm size, and regional geography, to name a few.

The Weather Prediction Center (WPC) is America’s “go-to” center for heavy to excessive precipitation events. WPC collaborates with both the National Hurricane Center (NHC) and local Weather Forecast Offices (WFOs) regarding the precipitation forecast of landfalling cyclones. Given the dangerous impacts of heavy rainfall and the potential for flash flooding, an accurate precipitation forecast is critical for each landfalling tropical cyclone.

This study uses the Model Evaluation Tools (MET) Method for Object Based Diagnostic Evaluation (MODE) tool from the Developmental Testbed Center (DTC) to evaluate quantitative precipitation forecasts (QPF) from WPC, the National Blend of Models (NBM), and the Hurricane Analysis and Forecast System (HAFS) over the past several hurricane seasons. Select landfalling storms are examined. The MODE tool identifies and compares both the forecast QPF objects and observed rainfall objects at different thresholds and produces a number of statistics based on the comparison of each object. The centroid distance, a statistic produced by MODE defined as the distance between the centroids of a matched forecast and observed object, are used to determine the QPF displacement error of high impact rainfall. Results from this study will help inform forecasters of potential biases for future landfalling tropical cyclone precipitation forecasts.