Citation

Yussouf N, Knopfmeier KH. Q. J. Roy. Meteorol. Soc. 2019; 145(723): 2385-2403.

Copyright

(Copyright © 2019, John Wiley and Sons)

DOI

10.1002/qj.3568

PMID

unavailable

Abstract

The vision of the National Oceanic and Atmospheric Administration's (NOAA) Warn-on-Forecast (WoF) research and development project is to provide very short-term probabilistic model guidance products that will aid the ability of National Weather Service forecasters to issue probabilistic warnings of severe convective hazards with higher accuracy and longer lead times. The experimental Warn-on-Forecast System (WoFS), which is under development, is a frequently cycled, regional, convective-scale, on-demand, ensemble data assimilation and prediction system. The system assimilates thunderstorm observations to forecast the life cycle and associated hazards of individual convective storms. Most of the initial research effort, since the beginning of the WoF project in 2009, has been focused on tornadic events, which constitute one of the most violent and difficult to predict weather threats. However, the system emerging from this project has significant potential to help with prediction of other hazards as well: in particular, flash-flood-producing, convectively driven intense rainfall events. This study focuses on the application of WoFS in 0-6-hr probabilistic rainfall forecasts of several flash-flood-producing heavy rainfall events during the spring and summer of 2015 and 2016.

RESULTS indicate that WoFS successfully initializes the convective storms in the ensemble. Visual inspections and probabilistic verification metrics show that WoFS predicts the intense rainfall that often leads to flash flooding at the correct location with higher accuracy in areal coverage and amount during the 0-3 hr forecast period than during the later 3-6 hr period. Overall results indicate that the frequently updated 0-6-hr ensemble forecast from the WoFS has the potential to highlight areas where intense rainfall can result in flash flooding and increase near-term situational awareness of flash flood threats.

Language: en

Keywords

convective-scale numerical modeling; ensemble data assimilation; ensemble forecasting; flash flood; heavy rainfall; Warn-on-Forecast