Aplicações de GNSS meteorologia: estudos de caso de eventos extremos de precipitação no Rio de Janeiro e Belém

Abstract

Extreme rainfall events, concurrently triggered with floods, waterlogging and landslides, are temporally related to variations in zenith total delay (ZTD) and integrated water vapor (IWV) from the Global Navigation Satellite System (GNSS) meteorology. The relationship between ZTD/IWV variations and precipitation was investigated in this work, using means, time series, and case studies of extreme events in Rio de Janeiro between 2015 and 2018, and in Belém between 2010 and 2022. The GNSS data are from the International GNSS Monitoring and Assessment System (iGMAS) and the Brazilian Continuous Monitoring Network (RBMC), while the precipitation data come from the Rio de Janeiro City Hall Alert System (Alerta Rio), the National Institute of Meteorology (INMET), and the Vale Technology Institute (ITV). In the composition of the ZTD/IWV and precipitation time series in Rio de Janeiro, quasi-linear ramps followed by nonlinear ramps were identified, with the highest rates of change and ZTD/IWV peaks occurring, respectively, 1–2 hours and 0.5 hours before the precipitation maxima, predominantly between 18:00 and 00:00 local time (LT). Case studies of extreme precipitation events during the rainy season revealed configurations in ZTD/IWV curves in the shape of: (i) oscillations named ‘asymmetric semisinusoid’ lasting 3–5 hours, formed by rapid vespertine growth in non-linear ramps, with an average rate of change at the inflection point of +11 (1.4) mm [15min]–1; (ii) a jump with an average rate of change of +17.3 (+2.66) mm [15min]–1 and a maximum of +21.3 (+3.33) mm [15min]–1; (iii) elongated ramps and oscillations named ‘bumps’ that were sequentially staggered on the ascending ramps lasting 1–2 hours; and (iv) ‘bumps’ or ascending ramps in already elevated values of ZTD/IWV due to the influence of meteorological systems such as the Humidity Convergence Zone (in Portuguese Zona de Convergência de Umidade, ZCOU) or the South Atlantic Convergence Zone (SACZ) (in Portuguese Zona de Convergência do Atlântico Sul, ZCAS). Events with large volumes of rainfall were identified, resulting in the elevation of river, igarapé, and canal levels, causing flooding in various locations in Belém. Three events occurred during the less rainy season with very heavy rainfall from Mesoscale Convective Systems (MCSs), and others during the rainy season under the main influence of the Intertropical Convergence Zone (ITCZ). The ZTD curves showed non-linear variations before and after significant rainfall events, with quasi-linear ascending ramps and peaks followed by descending ramps. As the events approached, the curves exhibited a ‘ZTD jumps’ configuration, with sudden increases preceding heavy rainfall or occurring in multiple jumps during intense rain events over several hours. However, some ZTD/IWV curves showed semisinusoid, bump, or jump configurations without resulting in precipitation, highlighting the existence of false alarms. Installing meteorological stations alongside GNSS stations to measure at least precipitation, pressure, temperature, and relative humidity with a 5-minute resolution is essential for monitoring extreme weather events. It is recommended to expand the analyses to longer periods, identify significant configurations in ZTD and IWV time series, define critical thresholds, and use advanced and more complex techniques such as neural networks, wavelet analysis for ZTD/IWV time series, or tropospheric tomography. These approaches are essential to improve the prediction of severe precipitation events, prevent and mitigate the impacts of adverse meteorological phenomena, ensure safety, and provide adequate infrastructure in affected areas.