Synoptic forcing associated with extreme precipitation events over Southeastern South America as depicted by a CORDEX FPS set of convection-permitting RCMs

• Published in: Climate dynamics 2021-05, Vol.56 (9-10), p.3187-3203
• Main Authors: Lavin-Gullon, A., Feijoo, M., Solman, S., Fernandez, J., da Rocha, R. P., Bettolli, M. L.
• Format: Article
• Language: English
• Subjects: Analysis; Atmospheric circulation; Atmospheric models; Climate; Climate models; Climatology; Convection; Convective precipitation; Earth and Environmental Science; Earth Sciences; Extreme weather; Geographical distribution; Geophysics/Geodesy; Maximum precipitation; Moist convection; Moisture flux; Oceanography; Precipitation; Regional climate models; Regional climates; Simulation; Spatial distribution
• ISSN: 0930-7575; 1432-0894
• DOI: 10.1007/s00382-021-05637-8

Southeastern South America (SESA) stands out as a remarkable region of occurrence of deep convection. This is mainly due to the proximity of the Andes, which eventually determine their magnitude and intensity. In this work, we used a set of convection-permitting (4 km horizontal resolution) regional climate models to explore their ability to reproduce the synoptic forcings that trigger deep convection over La Plata basin. The study considered simulating three extreme convective precipitation events in two different timescales. On one hand, a short-term simulation initiated a few hours before the onset of each event, spanning 3–4 days. On the other hand, as regional climate modelling, a 6-month simulation that includes the three selected events. In contrast to parameterized convection, the convection-permitting resolutions not only intensified the events, but also modified the location of the maximum precipitation by modulating the low-level atmospheric circulation. Vertically integrated moisture flux convergence emerged as a noticeable footprint of deep moist convection, regardless of the model and timescale. The performance of the models in reproducing the observed precipitation was also quantitatively analyzed. The skill depends on the spatial scale. The results were case-dependent in the short-term simulations. However, an analysis over multiple events in the long-term simulations revealed that, in general, convection-permitting resolutions better capture the spatial distribution of the extreme precipitation in SESA. The study comprises the first multi-model ensemble of convection-permitting simulations over the region, a seed for a further analysis with a more complete ensemble to better understand the results presented here.