Initial Results from the Atmospheric River Retrospective Forecasting Experiment: Forecasting West-Coast Heavy Precipitation Events

The NOAA Hydrometeorology Testbed (HMT) recently partnered with the Hydrometeorological Prediction Center to conduct the Atmospheric River Retrospective Forecasting Experiment (ARRFEX). The experiment consisted of a retrospective analysis of 8 atmospheric river (AR) events that impacted the U.S. West coast during the 2010-2012 cool seasons. For each case, a collaborative team forecasters and researchers were provided a wide range of operational and experimental numerical model guidance data, including the HMT-West ensemble and ESRL 2^nd Generation Reforecast Data. Using this experimental guidance, participants were asked to: (1) proactively identify the timing (e.g., duration plus start and end times) of landfalling ARs, (2) create 72-hour cumulative QPFs for Days 1-3, and (3) create 24-hour probabilistic QPF (pQPF) of greater than 3” at both 3 and 5 day lead times.

The primary objective of the project was to examine techniques to improve the prediction of heavy precipitation events associated with west-coast atmospheric rivers. Specifically, the goals of the experiment were to: (1) explore the viability of pQPF guidance and forecasts at medium-range lead times, (2) examine the utility of high resolution ensembles (e.g. HMT-West ensemble) for QPF/pQPF at short and medium range lead times, and (3) examine the utility of the ESRL Reforecast Data and other experimental techniques (e.g. standardized anomalies) for QPF/pQPF at short and medium range lead times.

Results of both subjective and objective verification of the various guidance and forecast products will be presented. The viability of pQPF products at short and medium-range lead times, as well as the potential effectiveness of higher resolution ensembles, reforecasts, and standardized forecast anomaly tools to improve QPF of heavy-precipitation events will also be discussed.