Examination of missed summer severe rainfall events over Ontario, Canada

Summer severe rainfall has high impacts on Canadian society and economy especially in highly populated areas such as the Great Lakes regions, but it has low predictability. Since summer severe rainfall often involves convective and/or small-scale phenomena, it is difficult to detect and predict in terms of precipitation amount, location, and timing. Through diagnosing a series of missed summer severe rainfall events in this project, our objective is to identify the causes of missed forecasts, improve our understanding of physical processes leading to severe precipitation, and provide information for operational meteorologists to make accurate and timely forecasts/nowcasts and warnings and for modelers to improve the operational GEM regional models.

The analyses on instability, moisture availability, and trigger mechanism have shown that (1) the GEM regional model (15 km resolution) forecasted central mean-sea-level pressure of the low pressure system is over-deepened by 1-5 hPa, mainly associated with shifting the heating maximum to lower levels; (2) the model predicted CAPE is smaller than the NARR analyzed, indicating that deep convection and severe precipitation induced by the parameterized convection appears to be weak and small in magnitude; (3) the GEM regional model underforecasts the conversion of water vapor into clouds; (4) the predicted low-level jet exit region is misplaced by the GEM regional model, resulting in a weaker convergence and less precipitation in the area of interest.

In the situation involved with three-cyclone interactions, the GEM regional model predicted cyclones are over-deepened, leading to over rotation of the binary cyclones which further results in one of the binary cyclones moving away from the area of interest. In addition, the model forecasted merged cyclone has a slower translational speed toward the area of interest. Because of these, the model predicted precipitation is shifted in location and reduced in amount, and the model predicted vertical differences of moisture and temperature are smaller, reducing the atmospheric instability.

Suggestions are made for operational forecasters and modelers to improve the predictions of summer severe rainfall.