Grains of sand extracted from riverbanks are revealing information about changes in the relief and pointing to possible climate oscillations in Brazil’s Southeast and Northeast regions over the past 20,000 years. Two papers published in April 2015 in the journal Revista Brasileira de Geomorfologia document different ages obtained through the analysis of quartz crystals from sand deposits along the banks of two rivers in São Paulo State—the Mogi Guaçu and the Corumbataí—and suggest periods of intense rain alternating with periods of scant rainfall. These are not isolated cases. Geographers at the University of Campinas (Unicamp), authors of the papers, have found grains of quartz ranging in age between 200 and 3,500 years in nine of the state’s rivers and have delineated how their movements built and rebuilt the fluvial plains that, along with other types of terrain, form what French geographer Jean Tricart called the Earth’s epidermis.
The findings of this and other studies were obtained through a technique called optically stimulated luminescence, which reveals when a quartz crystal was last exposed to sunlight before it was covered by the most recent sediments. The results lend support to the hypotheses about the evolution of the Brazilian landscape that were proposed 50 years ago by geographer Aziz Ab’Saber and zoologist Paulo Vanzolini. They maintained that dry/moist climate alternation was a decisive factor in building and sculpting the topography and in determining the soil formation and expansion/contraction of forests and the Cerrado savannah throughout Brazil.
The analyses of the quartz crystals—which are to be confirmed or adjusted by other methods—also challenge established notions. “The data we obtained show that the tropical landscape is fragile and recent, contrary to prior assumptions,” says geographer Archimedes Perez Filho, who is a professor at the Institute of Geosciences at Unicamp. Perez Filho and his team traveled 8,610 kilometers and collected 93 samples of sand from nine São Paulo rivers that discharge into the Paraná River (see map). “The possible climate oscillations over the past 20,000 years are not just regional,” he says, on the basis of additional observations carried out along the Itapicuru River and at the mouth of the Jequitinhonha River in the state of Bahia.
“The idea that the landscape, the forests and the soil are very old—hundreds of thousands or millions of years—needs to be revisited. That’s not what we are seeing,” says geographer Antonio Carlos de Barros Correa, a professor at the Federal University of Pernambuco (UFPE). He says that optically stimulated luminescence can date materials back to about one million years, but the ages obtained so far have not exceeded 100,000 years. Correa himself found a maximum age of 40,000 years in surveys he did in the states of Piauí, Pernambuco, Paraíba, Rio Grande do Norte and Alagoas. “Older sediments had already been carried out to sea,” he says.
The different ages of the quartz crystals indicate that the intensity of the erosive processes may have varied by region. “Each region has its own climate history that is more or less connected to the adjacent region, and even vice-versa. When it rained in the Southeast, there was drought in the Northeast, in alternating fashion,” says Correa, who has seen signs that the climate in the Northeast was much more dynamic than scientists had thought. “We identified signs of torrential rains on the slopes of the Borborema Plateau in Pernambuco, 17,000 years ago. During a drought period in the Northeast, however, there were periods of heavy rain that may have lasted for decades.” Analysis of the movements of sediments transported by the rains led him to think that “the landscape undergoes transformation gradually, by means of highly intense climatic pulses without definitive cycles, rather than in continuous fashion.”
Formation of the rivers
Perez Filho identified variations in river discharge when he examined quartz crystals collected from depths of 80 centimeters to 1 meter on the high and low terraces. The high terraces, located 30 to 50 meters above the present-day level of the rivers, constitute previously flooded areas containing pebbles from the riverbed, while the low terraces are three to five meters above the current river level. He notes that the volume of water caused the plains to expand in drier periods, and sculpted them into terraces during rainy periods when the river overflowed its bed. “The climate has to be hot and dry with torrential rains in order for the terraces to expand,” he says, “while riverbed carving occurs when there are continuous rains predominating in a hot, moist climate.”
The analyses of the quartz crystals delineated four periods in which sediments accumulated on the river plains—therefore probably in a dry climate—in the past 2,000 years amongst the nine rivers that the researchers examined. The first period was about 200 to 300 years ago, the second dates back 600 to 700 years, the third 1,100 to 1,200 years, and the fourth 1,900 to 2,000 years. “These kinds of pulses are also being identified in Argentina by teams of geomorphologists who are studying glacial processes,” says Perez Filho. Climatologists have not yet been able to delineate cycles with that level of precision, and they have noted that the formation of plains should not be considered a direct indicator of climate oscillations, although other studies have shown accelerated erosive processes in the past 4,000 years as a result of more intense and frequent rains.
“With the current climatological data, much of which consists of records spanning periods of less than 100 years, it is very difficult to identify or evaluate climate cycles on a scale of hundreds of years,” says climatologist José Marengo, a researcher at the National Center for Natural Disaster Monitoring and Alerts (CEMADEN), who is studying interannual climate variability over the past century. “Paleoclimate indicators can be useful for filling in the gaps in our knowledge about the climate on a scale of thousands of years, and for comparing the present-day mechanisms of climate variability against those of the past.”
Each river has its own history. Drier periods that occurred around 5,060 years, 2,570 years and 1,070 years ago likely promoted sediment deposition in the high and low terraces of the Corumbataí River, while the low terraces of the Mogi Guaçu River appear to have been formed at the same time or in more recent dry periods 1,900, 1,150 and 630 years ago. The dating results, Perez Filho notes, have a margin of error of plus or minus 10%.
In 2012, Fred Teixeira Trivellato, from the same group, repeated and compared the measurements taken in 1906 by members of the São Paulo Geography and Geology Commission in the Peixe River, a tributary of the Paraná River. He confirmed that, contrary to the comments of residents of that region, the width, channel depth, velocity and discharge of the river have increased due to the expansion of urban and agricultural land areas and the removal of native vegetation. “Before, with the forests, there was more water infiltration into the soil,” says Perez Filho. “Now, when it rains, the water drains off faster into the rivers.” Another change is that, as a result of construction of hydroelectric reservoirs, the forest corridors have disappeared.
The Cerrado is recent too
Each region or segment studied can present a mosaic of areas having different ages. In a doctoral dissertation to be defended in late 2015, geographer Gizelle Prado da Fonseca confirms that the plains in a region north of the Pantanal wetlands in the state of Mato Grosso range in age from less than 10,000 to 70,000 years. “These studies show how the relief is produced and sculpted and how the vegetation becomes established and gets remodeled,” comments geographer Jurandyr Ross, a professor at the University of São Paulo (USP), who guided her research on the Pantanal.
The team from Unicamp found that the age of the terrain now occupied by the Cerrado savannah in São Paulo State ranges from 12,000 to 15,000 years—much younger than expected. Therefore, Perez Filho concluded, the different physiognomies of the Cerrado today are likely to be approximately that old, since vegetation relies on soil formation to keep growing. This finding is consistent with other research, such as that of Luiz Carlos Pessenda of the Center for Nuclear Energy at USP, who identified records of Cerrado at least 15,000 years old in coal fragments naturally buried in the soil in the vicinity of Jaguariúna and Campinas.
Biologist Marcelo Simon of Embrapa Genetic Resources and Biotechnology (Cenargen) observes that “Soil age is not necessarily linked to the vegetation associated with it. Fairly recent vegetation may be established in very old terrain.” Before they formed a specific vegetation cover, the trees that characterize the Cerrado today were probably scattered among other trees, such as araucárias (Brazilian pine), which were better adapted to the cold climate that is thought to have predominated in the Southeast region around 20,000 years ago.
Geologist Francisco Cruz, a professor at the USP Geosciences Institute, coauthored a paper published in 2012 that showed intense climate variations, inferred based on an analysis of the ratio of forms of oxygen in cave minerals and lake sediments over the past 2,000 years in the state of São Paulo. Now, in turn, the Unicamp team is detecting signs of climate oscillations around the year 1100, lending support to the idea that the Southern Hemisphere may have had a hot, dry climate with heavy rainfall that contrasted with the so-called Little Ice Age, known to have occurred in the Northern Hemisphere during that same period. To expand on his findings, Perez Filho has begun to collect samples of soil material on the river plains of the Uberlândia-Uberaba Plateau in the state of Minas Gerais. “I would love to be 20 years old and have the equipment I have today,” says the 67-year-old geographer.
Landscape evolution and geochronology of Depressão Periférica and Planalto Ocidental Paulista relief (nº 2012/00145-6); Grant Mechanism: Regular Research Grant; Principal Investigator: Archimedes Perez Filho (IG/Unicamp); Investment: R$258,247.58.
STORANI, D. L. and PEREZ FILHO, A. Novas informações sobre geocronologia em níveis de baixo terraço fluvial do rio Mogi Guaçu, SP, Brasil. Revista Brasileira de Geomorfologia. V. 16, No. 2, p. 191-9. 2015.
DIAS, R. L. and PEREZ FILHO, A. Geocronologia de terraços fluviais na bacia hidrográfica do rio Corumbataí-SP a partir de luminescência opticamente estimulada (LOE). Revista Brasileira de Geomorfologia. V. 16, No. 2, p. 341-9. 2015.
VUILLE, M. et al. A review of the South American monsoon history as recorded in stable isotopic proxies over the past two millennia. Climate of the past. V. 8, p. 1309-21. 2012.