Key synoptic-scale features influencing the high-impact heavy rainfall in Beijing, China, on 21 July 2012

Abstract

This work examined quantitatively the key synoptic features influencing the high-impact heavy rainfall event in Beijing, China, on 21 July 2012 using both correlation analysis based on global ensemble forecasts (from TIGGE) and a method previously used for observation targeting. The global models were able to capture the domain-averaged rainfall of >100 mm but underestimated rainfall beyond 200 mm with an apparent time lag. In this particular case, the ensemble forecasts of the National Centres for Environmental Prediction (NCEP) had apparently better performance than those of the European Centre for Medium-Range Weather Forecasts (ECMWF) and the China Meteorological Administration (CMA), likely because of their high accuracies in capturing the key synoptic features influencing the rainfall event. Linear correlation coefficients between the 24-h domain-averaged precipitation in Beijing and various variables during the rainfall were calculated based on the grand ensemble forecasts from ECMWF, NCEP and CMA. The results showed that the distribution of the precipitation was associated with the strength and the location of a mid-level trough in the westerly flow and the associated low-level low. The dominant system was the low-level low, and a stronger low with a location closer to the Beijing area was associated with heavier rainfall, likely caused by stronger low-level lifting. These relationships can be clearly seen by comparing a good member with a bad member of the grand ensemble. The importance of the trough in the westerly flow and the low-level low was further confirmed by the sensitive area identified through sensitivity analyses with conditional nonlinear optimal perturbation method.