Exploring Precipitation Biases for U.S. Landfalling Tropical Cyclones in ECMWF Forecasts

Precipitation estimates provided by the European Centre for Medium-Range Weather Forecasts (ECMWF) are an important tool to anticipate and provide early warnings of high impact rainfall produced by tropical cyclones (TCs) that make landfall in the US. It has been shown that these precipitation estimates contain systematic biases in the short-term (1-6 days) for both deterministic and ensemble forecast from ECMWF. Much of the effort of reducing biases in the tropical cyclone forecasts has been directed to reduce the intensity and track forecast errors, but little attention has been paid to reduce errors in precipitation estimates from TCs. In this study we use deterministic and reforecast from ECMWF to gain insight into the nature of the errors in the precipitation that is produced from those TCs that potentially make landfall over the US coastal regions. We compare those model estimates against the information provided by the Climatology-Calibrated Precipitation Analysis (CCPA) as ground daily rainfall observations. Additionally, we use simple bias correction and quantile mapping as a statistical approach to produce corrected estimates of the precipitation that is produced during days that are affected by named TCs over southern US.