Doctorate Thesis Defense
Student: Andressa Andrade Cardoso
Programme: Meteorology
Title: “Assessing and classifying synoptic-scale cyclones over South America in high-resolution simulations in the present and future climate projections”
Advisor: Profa. Dra. Rosmeri Porfirio da Rocha - IAG/USP
Examining Committee:
- Profa. Dra. Rosmeri Porfirio da Rocha - IAG/USP
- Prof. Dr. Ricardo de Camargo - IAG/USP
- Profa. Dra. Erika Coppola – ICTP/Italy - on videoconference
- Profa. Dra. Claudine Pereira Dereczynski – UFRJ - on videoconference
- Profa. Dra. Natália Machado Crespo - Charles University/Czech Republic - on videoconference
Abstract: Synoptic-scale cyclones influence weather and climate with precipitation and intense winds, which can cause natural disasters and deaths. They are also crucial systems for the meridional transport of heat, moisture, and momentum, playing a fundamental role in the Earth's climate. Thus, investigating the future trends of these cyclones is crucial from a scientific point of view to understand climate change and associated extreme events, while from a social point of view this information is useful for planning mitigation and adaptation actions. Therefore, the main objective of this work is to evaluate the ability of Global Climate Models (GCMs) and Regional Climate Models (RCMs) with refined resolutions to simulate the current climatology (1979-2014) of cyclones (all types and subtropical) and future trends until mid-century (2015-2050). These analyses are conducted for South America and the South Atlantic Ocean focusing on cyclone characteristics (frequency, lifetime, intensity, associated circulation patterns) in three main cyclogenetic regions (SEBrazil, Uruguay and Argentina). For this purpose, simulations from three RegCM4.7 and six HighResMIP-CMIP6 GCMs with medium (MR) and high (HR) resolutions, and two reanalyses (ERA5 and CFSR) are used. In general terms, the simulations (GCMs and RegCM4.7) clearly represent the three main cyclogenetic regions provided by the reanalyses, with some specific differences: a) in Uruguay they present a negative bias, which is associated with deficiencies in the simulation of local processes and the influence of the Andes Mountains on the development of cyclogenesis; b) no SEBrazil RegCM4.7 simulations nested in the GCMs indicate cyclones forming further north than in the reanalysis. The simulations also adequately reproduce the main circulation patterns at low (winds, sea level pressure, and precipitation) and upper (upper-level jets and troughs at 500 hPa) levels associated with cyclones compared to ERA5. Projections indicate a future (2015-2050) synoptic environment in SEBrazil characterized by intensification of warm advection in the eastern and cold ones in the south parts of cyclone, while in Uruguay the moisture flux and its convergence should be more intense. In both regions, the upper-level jet is expected to intensify near the center of the cyclone, reinforcing the upper levels divergence and, consequently, the convergence at lower levels. Additionally, the cyclone classification selected subtropical cyclogenesis using ERA5 and the GCM ECEarth3P-HR, being validated with previous works that used ERAInterim. ERA5 clearly identifies the highest density of subtropical cyclones in the east of southeastern Brazil and their higher frequency in summer-autumn, as shown by ERAInterim. However, ERA5 has a higher frequency of these cyclones than ERAInterim, probably due to both the higher resolution of ERA5 and the inclusion of initially weaker cyclones in the classification process. The ECEarth3P-HR GCM simulates more subtropical cyclones than ERA5, but reproduces the main characteristics of these events shown in ERA5: annual and seasonal frequency, spatial distribution, as well as the characteristics of warm core at low levels and cold core at upper levels, in addition to the symmetry along the trajectories over the South Atlantic. Although the high-resolution GCMs show notable improvements in representing cyclones, the intermediate-resolution GCM also provides reliable information on cyclones. Furthermore, regional models are better able to represent synoptic cyclones compared to global models.
Keywords: HighResMIP-CMIP6, RegCM4.7, synoptic features, cyclogenesis, future projections, South America, subtropical cyclones