Macro- and Microphysical Characteristics of Rain Cells Observed during SOS-CHUVA in Campinas, Brazil

Convective clouds are one of the primary ways in which the local meteorology impacts the functioning of metropolitan areas in the tropics. They can cause human and patrimonial damage through flash floodings, wind bursts, hailfall and lightning. The SOS-CHUVA project aims to study convective clouds with damage potential around the city of Campinas (São Paulo state), Brazil, focusing on their nowcasting. In this study we present a methodology to study their properties in a holistic way, by tracking rain cells with the ForTraCC algorithm applied to a X-band radar dataset. In this way, rain cells properties are stored in a Lagrangian framework. The Center of Activity (COA), the altitude with the highest average reflectivity in the cell's CAPPIs (Constant Altitude Plane Position Indicator), is presented as a new perspective to study convective clouds. This parameter most of the time varies between 2.0 km and 4.5 km, which is mostly below the local 0 °C level. It is shown that the combination of COA and the Vertically Integrated Liquid (VIL) provides a useful phase space to compare different rain cells. In our methodology, rain cells are detected in a relatively mature stage of their life cycle where they present high COA (around 4.0 km to 4.5 km) and high VIL (~3.0 kg m^-2). From this point on, the cells usually evolve towards lower COA and VIL values, indicating rain cell collapse and the beggining of the dissipation stage. The COA-VIL space is also used to constrain the microphysical study of the cells. By obtaining the polarimetric measurements at the most intense region within the COA level, it is possible to characterize the cells cores throughout their lifecycle. It is shown that reflectivy (Z), differential reflectivity (Z_dr) and differential phase shift (K_dp) tend to grow with increasing VIL and decreasing COA. The increase in the polarimetric variables with decaying COA indicates continued collection growth of the hydrometeors at the COA level, while evaporation likely dominates in the layers below. However, smaller VIL at later stages of the cells life cycles can offset the collection process and reduce the polarimetric variables values. Interestingly, the average Z at COA remains relatively constant at slightly above 40 dBZ regardless of COA or VIL values, indicating a balance between the collection and evaporation processes. From the polarimetric measurements, the droplet size distribution(DSD) was estimated using a normalized gamma function. For constant VIL values, it was shown that the shape parameter μ tends to decrease as COA decreases, indicating an evolution towards an exponential DSD. This indicates that the collection and evaporation processes are reaching balance and the size distribution is approaching the equilibium DSD.

Aside from the overall rain cells statistics, we also present a few case studies where the relation between COA and the rain cells characteristics is further explored. It was shown that high COA (> 3.5 km) is likely associated to Z_dr columns and lightning initiation, which is accompanied be lower values of the DSD shape parameter (even reaching values < 0 in the methodology adopted here). Given that the COA estimate does not rely on polarimetric variables, this finding can help nowcasting applications by providing a simple way to infer rain cells internal structure. Overall, the COA-VIL approach seems to be adequate to study rain cells in both research and operational frameworks given the relatively high amount of information it provides from simple calculations.