A case study with extreme low-level jet in southern Brazil

Abstract

This study aimed to analyze a period of intense Low-Level Jets (LLJ) in southern Brazil that occurred in a year of intense El Nino-South Oscillation signal. Observational data were collected by radiosonde data carried out at Salgado Filho airport, which measured values of the order of 30 ms-1 in intensity at 700/750 meters in height and a predominant direction of 270 degrees. The LLJ’s are horizontal vortices in the low-level turbulence process, optimizing the transport of energy and water vapor at the mesoscale; they are low-cost energy structures with maximum transport efficiency. In general, their modeling by numerical models at high resolution is difficult because they have a thin layer with a small thickness of the order of 50 meters and have a non-linear nature in the turbulence processes. However, global reanalysis models, in general, cannot represent the optimization effects in the turbulent process performed by LLJ. However, there is a tendency to represent the magnitude of the wind modulus, an approximation of this value by the current state of the art in global reanalysis modeling in models, in this case, the maximum wind. This implies that global models generally underestimate flows at low levels, in general, as they do not represent the optimization effects associated with LLJ. In this case study, we tried to use an extreme situation of occurrence of LLJ for analysis, therefore, the regional climate model RegCM-4.9 in high spatial resolution was used to simulate the period from May to June 1997. The analyses were concentrated in June 1997. The initialization and boundary conditions were used from the EC-EARTH historical of the Earth system model based on the operational seasonal forecast system of the European Center for Medium-Range Weather Forecasts (ECMWF). It was observed that the simulation result could significantly underestimate the wind intensity and the processes associated with the LLJ with the initialization model, with low information resolution, the global initialization models do not have adequate resolution for this type of simulation.