FRANS LANTING, MINT IMAGES / SCIENCE PHOTO LIBRARYScientists familiar with the work of British Marxist historian Eric Hobsbawm, who died last year, could well borrow the title of his book about the political and economic changes in the 20th century, and use it to describe Brazil’s climate outlook for the decades to come. When it comes to climate change, the age of extremes—the title of Hobsbawm’s book—has scarcely begun and, researchers say, it is here to stay for a very long time. Owing to a gradual increase in the concentration of greenhouse gases—last May, the carbon dioxide (CO2) levels reached 400 parts per million (ppm) for the first time in recent human history—and changes in land use, Brazil’s climate at the end of the 21st century will probably be different from what it is today, in line with expectations for other parts of the planet.
Projections indicate that the average temperature in every major region of the country without exception will be 3ºC to 6ºC higher in 2100 than it was at the end of the 20th century, depending on the future pattern of greenhouse gas emissions. Rainfall is expected to present a more complex scenario. In biomes such as Amazonia and the Caatinga, rainfall estimates indicate a possible 40% reduction. In the Pampa, the trend is in the opposite direction, with an increase of about one-third in overall rainfall over the course of this century. For the other areas of the country, climate models also indicate scenarios with worrisome changes, but those projections are less reliable. Nevertheless, there are indications that it could rain significantly more in the southern and southeastern parts of the Atlantic Forest and less in the northeastern portion, the Cerrado, the Caatinga and the Pantanal. “Except for the central and southern coast of Chile, where cooling has been observed in recent decades, we are measuring and have also projected a future increase in temperature in all other areas of South America,” says José Marengo of the National Institute for Space Research (INPE), who is working on future projections based on regional climate models. “The perception is that the seasons have gone half ‘crazy,’ with more frequent manifestations of climate extremes.”
This means that Brazilians will be living through more periods of prolonged drought as well as more periods of heavy rain, sometimes one after the other. And this is without even mentioning the possibility of phenomena with great destructive potential that were once very rare in our country, such as Hurricane Catarina, which hit the coast of the states of Santa Catarina and Rio Grande do Sul in March 2004. In major metropolitan areas, and even in medium-sized cities, the proliferation of concrete and asphalt intensifies the urban heat island effect, making them warmer and changing their rainfall patterns.
This scenario is part of the more complete analysis already produced on the subject of the principal future climate trends in Brazil: the First National Assessment Report (RAN1) by the Brazilian Panel on Climate Change (PBMC), created in 2009 by the Ministry of the Environment (MMA) and the Ministry of Science, Technology and Innovation (MCTI). The report will be released during the first National Conference on Global Climate Change, organized by FAPESP and slated for September 9-13, 2013. Patterned after the Intergovernmental Panel on Climate Change (IPCC) established by the United Nations, which will release the first part of its own fifth report in late September, the PBMC brought together 345 researchers from different areas to prepare an unprecedented overview of the state of the art in Brazilian scientific output on the topic.
RAN1 is divided into three parts, each prepared by a different working group. Part one discusses the principal conclusions of studies done between 2007 and early 2013 which showed that climate changes had occurred in Brazil. Part two provides details of the impacts of climate changes in Brazil, and highlights vulnerabilities and ways to adapt to the new reality. Part three enumerates new ways to reduce greenhouse gas emissions in the national territory (see article on page 22 about parts two and three of the document). “We did a critical compilation of the data produced by the most recent studies,” explains meteorologist Tércio Ambrizzi of the University of São Paulo (USP), a coordinator of the PBMC’s first working group on national scientific output. “There are areas of the country, such as the Central-West region, on which almost no studies have been done. We also have very little research on paleoclimate in Brazil.”
According to the researchers, most studies of the subject analyze the fossilized pollen of plants in the national territory, and the dating is uneven in quality. “Research on what the climate was like in the past around Brazil’s Atlantic coast is even rarer,” says paleoceanographer Cristiano Chiessi of USP Leste, a co-author of the report. “We need to invest in this type of study to find out what is natural climate variation and what happened as a result of human activity.”
A Brazilian climate model
The release of the PBMC report marks the incorporation of a sophisticated tool for improving our understanding of climate and making projections in Brazil. The Brazilian Earth System Model (BESM) is a set of computer programs that simulate the evolution of the principal climate parameters on a global scale. “Brazil is currently the only country in the Southern Hemisphere that has its own model,” says Paulo Nobre of INPE, one of the BESM coordinators. “This will give us considerable autonomy for doing the simulations that are of greatest interest to us.” The BESM will make it possible, for example, to do projections of probable effects on climate in Brazil caused by changes in ocean circulation in the tropical Atlantic and in the various Brazilian biomes. Australia was also creating its own climate model but elected to pool its efforts with the Hadley Centre in the United Kingdom. The Brazilian model has been in development since 2008 through the efforts of researchers from several institutions that participate in the FAPESP Research Program on Global Climate Change (PFPMCG), the Network of Brazilian Research on Global Climate Change (Rede Clima) and the National Institute of Science and Technology for Climate Change (INCT-MC).
Like any computer program, the BESM is a work in progress, meant to be refined on an ongoing basis. It is designed not only to provide the country with a state-of-the-art model to represent the earth system, but also to help train a new generation of scientists capable of handling a powerful instrument dedicated to climate forecasting. The current version of the BESM—which runs on the Tupã supercomputer used by Rede Clima/PFPMCG at the INPE facility in Cachoeira Paulista—is now able to reproduce several global and regional climate phenomena and forecast future scenarios. For example, the model is able to reconstruct the most recent El Niños and estimate when that climate phenomenon will return. An El Niño is an abnormal warming of the surface waters of the Equatorial Pacific, an alteration of the ocean and the atmosphere that affects rainfall patterns on much of the planet. In Brazil it tends to promote droughts in the Amazonian and Northeastern regions and intensify rainfall in the South. Simulations done with the BESM have shown that a hypothetical complete deforestation of Amazonia would raise the intensity of El Niños and reduce annual precipitation in the Northern region by up to 40%.
The climate scenarios generated by the BESM were accepted this year by Phase 5 of the Coupled Model Intercomparison Project (CMIP5), the international initiative that brings together the data produced by the 20 global models developed thus far. They serve as the launchpad for Brazil’s participation in the IPCC as a nation that supplies planetary-scale projections of climate change. The projections generated by the national model will be used in drawing up the IPCC’s fifth report on climate change.
The BESM does not yet provide scenarios as detailed as the ones generated by other global models, or even by INPE’s regional model, which focuses on climate in South America and served as the basis for many of the projections in the first PBMC report. Its spatial resolution is 200 x 200 kilometers, while INPE’s regional model, running for the time being “inside” the Hadley Centre’s global model, is usually 40 x 40 km but can achieve 5 x 5 km. Despite being in its infancy, the BESM is already producing simulations that outline a panorama of climate changes projected to occur in Brazil in the next 30 years. Pesquisa FAPESP is publishing firsthand results of an entirely new simulation that shows how the average annual temperature of the atmosphere could vary in each state of the country between now and 2035, based on the early results of the most recent version of the BESM model. The data point to a Brazil that is warmer at nearly every latitude. “This is the first result showing a future global warming scenario produced entirely in Brazil without relying on simulations obtained by models from other countries,” notes Paulo Nobre, another RAN1 co-author.
If the level of CO2—the principal gas responsible for intensifying the greenhouse effect—holds its current trend and reaches 450 ppm by three decades from now, the average annual temperature in much of Brazil, particularly in areas more removed from the coast, is expected to rise as much as 1ºC. Only in the South and in the northernmost sections of the North will the temperature tend to remain stable or even drop slightly. “This initial result takes into account the impacts of the trends toward long-term changes in global ocean circulation and atmospheric warming that result from a moderate increase in CO2 levels on a planetary scale,” Nobre explains. “The results are preliminary. We need to run the model some more in order to get more reliable results, and then we can talk more specifically about climate trends for a particular state or smaller area.”
The BESM forecasts for the southernmost part of Brazil are the only ones that do not completely agree with the ones produced by INPE’s regional model, which projects a small increase in temperature in the South by 2040. Most of the projections indicate, however, that by the end of the century, Rio Grande do Sul will continue on the same warming trend as the rest of the country. With a continual rise in the CO2 level, the models show a gradual increase in temperature and a greater possibility of more or less rainfall in a given region over time.
The size of a raindrop
The most recent version of the BESM was able to partially circumvent a major limitation of climate modeling: reasonably accurate forecasting of rainfall in the Amazon, a definitive feature of the Northern region without which a tropical forest so dense and lush is not sustainable in the long term. Annual rainfall in the North varies from 2,500 to 3,400 millimeters, more or less double the annual total seen in the Central-West region, with typical Cerrado savannah vegetation of predominantly grasses and scattered small trees. “Every global climate model underestimates rainfall in the Amazon,” says Paulo Nobre.
The forecasting of rainfall over the Amazon Forest has been improved through the introduction of a series of refinements to the atmospheric component of the BESM, in particular the revision of a specific parameter: the average diameter of the raindrops represented within the clouds generated by the model. The raindrops simulated by the BESM, which previously averaged 1 millimeter in diameter, are now calculated at 1.4 millimeters. “The CAM5 climate model from the NCAR (National Center for Atmospheric Research) in the United States was already using that average diameter, but the results of its projections did not adjust the total rainfall over the Amazon as well as our model did,” Nobre asserts. “We are still not perfect at simulating rainfall, but so far no climate model has done that.”
After the modifications, the BESM made a leap in quality, generating improved simulations of the trade winds that carry moisture to the Amazon. It started to more correctly record the ocean temperature variations between Brazil and Africa. It was even able to reproduce an important climate mechanism known as the South Atlantic Convergence Zone, which governs rainfall formation in the Southeast and the southern portion of the Northeast. This zone consists of a band of clouds that can extend up to 5,000 kilometers in a northwest-southeast orientation, crossing over the Brazilian coast at 18 to 25 degrees South latitude.
There is a reasonably simple explanation for the difference in performance. Each model is made up of smaller parts that attempt to reproduce the functioning of the major climate components, such as the atmosphere, the oceans, land use, vegetation, and global ice cover. A series of specific data and equations enables each component to function uniquely and interact with the other parts of the model. Therefore, by tweaking one parameter such as average raindrop diameter in the cloud cover, one model might perform better while another does the opposite or fails to show any significant change. “The models have more trouble making projections for rainfall than for temperature,” comments physicist Alexandre Costa of the State University of Ceará (UECE), a co-author of the chapter on clouds and aerosols (tiny solid or liquid particles suspended in a gas) in the first report by the PBMC. “Depending on the size of the raindrops in a cloud, there may be more or less rainfall.”
In support of an environmental data network
In the opinion of USP physicist Paulo Artaxo, a leading specialist in the process of aerosol formation, the first report by the PBMC will enable Brazil to identify areas that are still deficient in terms of research, as well as provide an overview of studies on climate change. “We have a long way to go,” says Artaxo, who is on the PBMC’s governing board. “The IPCC has been in place for 20 years and is working on its fifth report. We still don’t have a critical mass of scientists, and there are not enough people to cover some important areas.” Artaxo warns that Brazil does not yet have a national network to systematically gather more sophisticated environmental data beyond just temperature and rainfall measurements. In the Amazon there are 12 towers that record carbon and energy exchange between the forest and the atmosphere and measure properties of other biogeochemical cycles—an initiative sustained by the Large Scale Biosphere-Atmosphere Experiment in Amazonia (LBA), a successful partnership that for more than two decades has brought together researchers from Brazil and other countries. Outside the Northern region there are few towers in Brazilian territory. There is one in the Pantanal, another in the Cerrado, a third in the Pampa and another in inland São Paulo State. “This small-scale framework hinders us from having a national radiography of, for example, atmospheric C02 emissions and capture,” Artaxo says. “In Europe and the United States there are hundreds of towers that provide a radiographic account of what is happening with the functioning of the ecosystems as a result of climate change.”
In the view of climatologist Carlos Nobre, secretary of research and development policies and programs at the MCTI and chairman of the PBMC, the data made available by the Brazilian Panel serve to guide public policy on adaptation and mitigation of climate change. “The work of the panel will not end with this first assessment report; it will continue and become increasingly important,” Nobre maintains.
Brazilian model of the Global Climate System (nº 2009/50528-6); Grant mechanism PFPMCG/Pronex FAPESP Thematic Project; Coord. Carlos Nobre/INPE; Investment R$571,200.00.