Climate records in the countryside

São Paulo State weather stations with longer historical data series first appeared in Campinas and Piracicaba

Record books from 1917, from the Esalq weather station

GERHARD WALLER / ESALQ-USPRecord books from 1917, from the Esalq weather stationGERHARD WALLER / ESALQ-USP

The systematic collection of climate data in Brazil began in the first half of the 19th century at the Imperial Observatory of Rio de Janeiro, today the National Observatory. The Campinas Institute of Agronomy (IAC) installed its first weather station in 1890 in the interior of the state of São Paulo, with an objective beyond that of simply understanding weather behavior: it was intended more specifically for research on the influence of climate on agricultural crops. “On pages measuring one meter wide, the observer/technicians recorded information on temperature, humidity, solar radiation and cloud cover five times a day,” says Orivaldo Brunini, coordinator of the IAC’s Integrated Center for Agrometeorological Information, an agency connected with the São Paulo State Department of Agriculture and Supply.

The pioneering mechanical station was put into operation three years after the establishment of the IAC. Three technicians took turns reading the data every day, without exception. “The instruments—the most modern ones of the era—came from Germany,” Brunini says. The data were subsequently analyzed by agronomists, but it took time for the information to reach farmers. Over time, other cities in the interior received their stations. In 1956, the Institute began to produce annual agrometeorological almanacs and, soon afterwards, monthly bulletins. Today there are 150 stations operating with automatic sensors that eliminate reliance on observers and provide real-time information on the Internet. The IAC receives 120,000 hits per month, or 4,000 per day. The data are used to guide farmers and Civil Defense personnel in responding to floods and fires.

“The cumulative historical material provides data series that help us study climate change, and it is part of the regional history because it contains the record of major droughts and frosts, for example,” Brunini says. He notes that the record shows only two service interruptions. These occurred during the Revolution of 1932, in the month of July when São Paulo State troops stayed at the IAC facility where the station was located, and in October when federal forces were at the same site.

Esalq weather station in the 1930s

IAC ARCHIVESEsalq weather station in the 1930sIAC ARCHIVES

The year 1917 marks the beginning of the observations that led to São Paulo’s second-oldest data series, at the station installed at the former Practical Agriculture School of Piracicaba, today the Luiz de Queiroz College of Agriculture of the University of São Paulo (Esalq-USP). “The readings were taken four times a day, one of which was sent by telegraph to the Office of Meteorology and Astronomy at the National Observatory—today the National Institute of Meteorology (INMET)—to compile the daily meteorological information and weather forecasts for the entire country,” says Luiz Roberto Angelocci, a senior professor in the Department of Biosystems Engineering at Esalq. Beginning in the 1960s, the studies focused more specifically on analysis of the climate-agriculture connection and, in late 1996, automated data collection began, although conventional collection was not eliminated. The data are available on the internet.

Angelocci is working on analyzing the historical data series generated during the 97 years of record-keeping. “The plan is to do a complete reanalysis of the series and improve data exploration,” he says. Long series are essential to achieving an understanding of climate and the causes of climate variability. They are also important for distinguishing between a mere climate anomaly and a potential trend. “The fact is that some anomalies are not as rare as they seem. The current dry spell in the state of São Paulo, for example, has occurred similarly in other years,” Angelocci notes. He observed that, in five years of the series, the first of which occurred in 1935, rainfall was less than 40% of the average for the entire series for the month of January. “You can’t categorically say whether or not the current drought is the result of recent climate change, when we know that similar periods have occurred in the past 97 years,” he concluded.