Smartphone app for site-specific early warning of floods and thunderstorms

SOS Chuva, developed with FAPESP’s support, provides nowcasting with 1 km precision and can be downloaded free of charge. Researchers plan more apps for Civil Defense, farmers and other communities.

By Maria Fernanda Ziegler  |  Agência FAPESP – The advent of summer in the southern hemisphere heralds the rainy season in the Southeast region of Brazil. The difference this year is that the free smartphone app SOS Chuva will provide site-specific early warnings of heavy rainfall and thunderstorms. 

Unlike conventional weather forecasting, these immediate forecasts predict rain, hail and thunderstorms with 1 km precision and 30 minutes to six hours in advance. The app has been available since October 2017 and has been downloaded more than 60,000 times. 

The app was developed by scientists at Brazil’s National Space Research Institute (INPE) in collaboration with colleagues at the University of Campinas (UNICAMP), the University of São Paulo’s Luiz de Queiroz College of Agriculture (ESALQ-USP), and the same university’s Institute of Astronomy, Geophysics & Atmospheric Sciences (IAG-USP).

“The weather forecasts broadcast on television and radio are mostly well-established. The theory underlying these forecasts was developed in the 1950s. Immediate forecasting is a new challenge, with entirely different functions, equipment and mathematic models. Telling people in a region or city it’s going to rain tomorrow is quite different from telling you it’s going to rain precisely where you are in two hours’ time,” said Luiz Augusto Toledo Machado, a researcher at INPE’s Center for Weather Forecasting & Climate Studies (CPTEC-INPE) and principal investigator for the project.

With FAPESP’s support, the Thematic Project “SOS Chuva”, which began in 2016, will develop two more apps – one for agriculture and the other for Civil Defense. The researchers also plan to deepen their understanding of cloud dynamics and upgrade the mathematical models used in weather forecasting.

“The project has a science dimension – refining the models used in immediate forecasting – but it’s also about outreach via the development of apps and more-sophisticated early warning systems for farmers and the Civil Defense community,” Machado said.

In November, the project team held a course to train Civil Defense technicians in and around Campinas, São Paulo State, and CPTEC staff working in the Paraíba Valley and on the north coast of the state. The aim was to equip regional meteorology centers to do immediate forecasting, also known as nowcasting, in the first initiative of its kind in Brazil.

“We’re also developing an app for technical personnel to do nowcasting and issue early warnings based on our mathematical models,” Machado said. “Immediate forecasts have to be done regionally because of the high level of detail. For this reason, we’re developing the tool so that our regional weather centers can do nowcasting in the future.”

Precision agrometeorology

The group of researchers at CPTEC-INPE and ESALQ-USP are also developing an app specifically for use by farmers. “Besides showing where it’s raining, the agricultural app will also store rainfall data for a period, for use by farmers in predicting possible yield variations,” said Felipe Pilau, a scientist in ESALQ’s Biosystems Engineering Department. Pilau is responsible for the agricultural part of the project.

Pilau added that the app will serve as a tool with which to establish strategies for what he called precision agrometeorology, a combination of precision agriculture – which analyzes the variability of production based on factors such as soil fertility and water resources – with weather forecasting.

“Including meteorology in precision agriculture will enable farmers to see where rainfall is heaviest and predict the impact of this variability on crop yields,” he said. “Weather forecasting has so far been overlooked in precision agriculture.”

To do nowcasting for ordinary users, farmers and Civil Defense, the project has a dual-polarization (dual-pol) weather radar acquired with support from FAPESP and installed at UNICAMP’s Center for Meteorological & Climate Research Applied to Agriculture (CEPAGRI).

Conventional weather forecasting requires data obtained from satellite images and weather stations, as well as interpolation of these data. To obtain data with 1 km precision, the dual-pol radar uses wavelength emission and reflection.

The radar sends out pulses of horizontally and vertically polarized electromagnetic energy, which bounce off particles in the atmosphere and return back to the radar dish. A computer processes the reflected signals and draws conclusions about the particles, such as raindrops. This is the basis for precision nowcasting of precipitation. 

The SOS Chuva project produces immediate forecasts for São Paulo State using four other radars, located in the cities of Bauru, Presidente Prudente and São Paulo, as well as Rio de Janeiro State.

The dual-pol radar provides a two-dimensional cross-section that gives forecasters a measure of the size and shape of the objects it detects. They can also use it to track the speed of propagation and other parameters such as the accumulation of ice crystals inside clouds or in-cloud lightning that may indicate the existence of hail.

“The dual-pol radar can tell us all about these cloud ice crystals, for example, and we use these data to do calculations and forecasting,” Machado said.

The researchers measure the sizes of the ice crystals to see how they grow or shrink, which helps forecast weather severity, tornado formation etc. “We can also obtain wind data to detect closed circulation and electrical discharges. All this together helps us produce forecasts,” he explained.

Understanding extreme weather

The experience acquired by the SOS Chuva researchers in the development of mathematical models and calculations for nowcasting will be used in a new collaborative project with colleagues in Argentina, Chile and the US.

“We’ll continue to collect data in Campinas and refine our models until August 2018,” Machado said. “Then, we’ll take our instrumentation to São Borja in Rio Grande do Sul State for a new measurement campaign by the international collaboration.”

The São Borja region has the planet’s most severe thunderstorms, he explained. Phenomena known as mesoscale convective complexes occur in the South of Brazil in response to a link between the Amazon region and the Andes.

“Humidity travels from the Amazon to the Andes and is channeled to southern Brazil, forming intense cloud systems there and in Argentina. Low pressure accelerates the flow of humidity from the Amazon and forms very large thunderstorms,” Machado explained.

The new project is called RELAMPAGO (which means “lightning” in Portuguese and Spanish, but is also short for Remote sensing of Electrification, Lightning, And Mesoscale/microscale Processes with Adaptive Ground Observations) and is funded by the US National Science Foundation (NSF). Other agencies that are collaborating include NASA and NOAA (USA), CONICET (Argentina), and CONICYT (Chile), as well as FAPESP and INPE (Brazil).

“It will be a very large experiment, and SOS Chuva will participate in this effort to understand the severe thunderstorms that enter Brazil, sometimes forming tornados,” Machado said.

The SOS Chuva app can be downloaded from the App Store (iOS) and the Google Play Store (Android). More information: soschuva.cptec.inpe.br/soschuva.