The 21 June 1997 flood: storm scale simulations and implications for operational forecasting

On 20-21 June 1997, a convective outbreak in Nebraska, Iowa, Illinois and Wisconsin resulted in 2 fatalities, 8 injuries and approximately $104 million in property and crop damage. The majority of the damage ($92 million) was the result of flooding in southeastern Wisconsin owing to nearly 250 mm of rain produced by training convection and to a lesser extent the passage of a persistent, elongated convective system or PECS. The flood event was analyzed and storm scale (5 and 1.67 km grid spacing) resolution model simulations at 00-24h, 12-36h and 24-48h ranges were produced to study the evolution and predictability of the rainfall.

Synoptic conditions corresponded closely to the mesohigh pattern frequently associated with heavy rainfall events. Despite the recognition by NWS forecasters of the potential for heavy rainfall, uncertainty concerning event magnitude and affected areas, exacerbated by poor operational model guidance, resulted in a failure to issue flash flood watches prior to the onset of flooding. Simulations of the event using 5 km grid spacing show that short-range forecasts (00-24h) were able to suggest focused precipitation in southeastern Wisconsin. Increased resolution (1.67 km grid spacing) was necessary to capture the magnitude of the event in this area. Precipitation forecast skill as measured by the Kuiper Skill Score exceeded 0.3 for precipitation thresholds of 25-125 mm. Initiation of the convection was tied to lift at the leading edge of a developing low level jet. A rapid loss of predictability resulted in part from poorer representations of the LLJ at 12 36 and 24-48h ranges.

The case illustrates the need for forecasters to form and critically assess “mental models” of a forecast problem, using a combination of numerical model and observational datasets. Consistent implementation of this procedure will reduce forecast risk associated with utilizing high resolution models, but significant constraints related to real-world application of such models and procedures remain.