Improving storm-scale 1-h probabilistic low-level rotation forecasts through machine learning
One focus of the NOAA Warn-on-Forecast (WoF) project is to provide probabilistic guidance for short-term (e.g., 0-1 h) tornado forecasts. Tornadoes are unresolvable with current operational models, but convection-allowing ensembles (CAEs) such as the 3-km NSSL Experimental WoF System for Ensembles (NEWS-e) provide forecasts of low-level rotation that can potentially discriminate between tornadic and non-tornadic storms. However, CAE forecasts often contain large errors in storm intensity, timing, and location. Machine learning methods can be applied to raw ensemble model output to mitigate systematic forecast biases, incorporate ensemble uncertainty, and include additional model variables to produce calibrated probability forecasts.
This study utilizes real-time NEWS-e forecasts generated during the 2016 and 2017 NOAA Hazardous Weather Testbed Spring Forecasting Experiments and low-level azimuthal shear analyses from the NSSL Multi-Radar/Multi-Sensor (MRMS) product suite as input data for training various machine learning algorithms to produce calibrated 1-h probabilistic forecasts of low-level rotation. Given that strong low-level rotation occurs infrequently, we are oversampling the grid points within the rotational tracks to produce a more balanced dataset, which has been shown to improve machine-learning performance. The reliability and discrimination ability of the forecasts is evaluated using traditional verification statistics [e.g., receiver operating characteristic (ROC), area under the ROC curve, brier skill score (BSS), and attribute and performance diagrams].