Weather Radar Analysis of Severe Storms Depth in Southern Brazil and Paraguay

Rasmussen and Houze (2016) summed up that South American cloud shields associated with mesoscale convective systems (MCSs) are 60% larger than those over the United States (Velasco and Fritsch 1987), the convection is deeper (Zipser et al. 2006), and they have more extensive and longer-lived precipitation areas than those over the United States and Africa (Durkee et al. 2009). Therefore, it is important to discover the morphologic and dynamic characteristics of these high-impact severe storms using radar data.

Nowadays, more than sixty operational radars are working in South America. In this work, we analyzed the depth of the storms that yield high-impact weather such as flash floods, gust winds, and precipitation estimation. The analysis was divided into storms that develop over the west of Brazil and Paraguay, and the ones initiated in the coast of the Atlantic Ocean under maritime circulation. At the west and central areas of Rio Grande do Sul, Santa Catarina, and Parana states it was observed supercells, squall lines and Mesoscale Convective Complex (MCC) with, on the average, echo tops of 12 km when considering a reflectivity factor greater than 30 dBz. In Paraguay, all of the eight flash floods events analyzed between 2015 and 2018 were caused by squall lines which had echo tops with, at least, 15 km with echoes greater 30 dBZ. Some squall lines were, in fact, an MCC when observed by satellite images leading up the Maddox (1980) concept. While the deep convection inside the MCS was supplied by enhanced moisture flux of warm and moist low-level air from the Amazon basin via the South American Low-Level Jet (SALLJ), in the east coast of Brazil intense flash floods occurred with shallow convection supplied by low-level maritime circulation, usually by the subtropical high pressure southwest flank. For example, on 09-11 January 2018 shallow convection yielded 400 mm during only three days, causing historic flooding in Florianopolis City. In this event, shallow convection developed in the form of several cells, parallel to the coast, for more than two hundred kilometers during about 6 hours, and with radar echoes not greater than 42 dBZ, and the clouds weren't higher than 6 km using a 30 dBZ threshold. This is a challenge to Quantitative Precipitation Estimation (QPE), because, in the majority of cases, there are much more flood events associated with deep convection than shallow clouds. So, all methods, even when using machine learning, tend to relate high values of rainfall with high values of radar reflectivity and very low satellite bright temperature. So, the shallow convection events are not accurately estimated carry on the unsuccessfully nowcasting and flash flooding forecasts.

For the severe storms that cause high wind gusts (with values higher than 20 ms-1), it has been found 12 to 15 km high clouds with intense low-level shear and 1000 to 2000 convective available potential energy (CAPE) for both squall lines and supercell. Also, it has been found higher is the SALLJ contribution deeper will be the storm. Severe storms associated with cold fronts are shallower than the ones developed with flux from SALLJ, but are not clear the differences in the severity of the downward wind gusts that causing an electric breakdown in several cities every year. The understanding of the storm depth could be a good predictor for QPE and nowcasting methods and useful to improve the parameterization of mesoscale numerical models.