Publicações

Understanding the relationships between climate and carbon exchange by terrestrial ecosystems is critical to predict future levels of atmospheric carbon dioxide because of the potential accelerating effects of positive climate-carbon cycle feedbacks. However, directly observed relationships between climate and terrestrial CO2 exchange with the atmosphere across biomes and continents are lacking. Here we present data describing the relationships between net ecosystem exchange of carbon (NEE) and climate factors as measured using the eddy covariance method at 125 unique sites in various ecosystems over six continents with a total of 559 site-years. We find that NEE observed at eddy covariance sites is (1) a strong function of mean annual temperature at mid-and high-latitudes, (2) a strong function of dryness at mid-and low-latitudes, and (3) a function of both temperature and dryness around the mid-latitudinal belt (45 degrees N). The sensitivity of NEE to mean annual temperature breaks down at similar to 16 degrees C (a threshold value of mean annual temperature), above which no further increase of CO2 uptake with temperature was observed and dryness influence overrules temperature influence.

Neste trabalho são analisadas as mudanças no padrão de circulação que possam vir a acontecer no clima da América do Sul (AS), como conseqüência do aumento nas concentrações dos gases de efeito estufa. Para isto utilizam-se cinco modelos globais do IPCC AR4 (CCCMA, GFDL, HadCM3, MIROC e GISS), para o clima do século XX (1961-1990 – 20C3M) e para o cenário futuro SRES_A2 (2011-2100). As características em comum que os modelos apresentaram (a exceção do MIROC) para as três climatologias futuras (2011-2040, 2041-2070 e 2071-2100), principalmente, no verão e na primavera, foram o deslocamento da baixa continental (associada à baixa do Chaco) para o sudoeste da sua posição climatológica (1961-1990), e da Alta da Bolívia para o noroeste. Além disso, os cinco modelos simularam, para o clima presente, uma Alta do Pacífico Sul (APS) menos intensa em relação à Reanálise do NCEP, sugerindo menor subsidência sobre a sua região de atuação. Para cenários futuros os modelos GISS e HadCM3 simularam a APS menos intensa. Por outro lado, para a alta do Atlântico Sul, não existiu um consenso nos modelos. Em geral foi simulada mais intensa (a exceção do GISS), sobretudo no outono e no inverno. O modelo HadCM3 simulou a circulação de verão e primavera mais próxima à Reanálise, com uma ZCAS melhor definida, e uma área menor de anomalias negativas de chuva sobre a Amazônia, em relação aos outros modelos. Já para o cenário futuro este modelo modificou seu padrão de chuvas, e anomalias positivas, sobre a costa norte do Peru e Equador, e negativas sobre o Nordeste e leste da Amazônia, foram observadas, associadas a uma APS enfraquecida e deslocada para o sul, o que reforçou a ZCIT do Pacífico sobre 5°S. Uma diminuição da convergência de umidade sobre a Amazônia também foi observada.

Conserving natural vegetation cover is of critical importance for maintaining the ecological integrity and hydrological properties of large river basins (more than 100 000km2). Recent estimates indicate that more than 700 000km2 of Brazilian Amazon have already been deforested, and to reduce further losses and preserve the important natural and cultural resources in this region, large conservation areas have been created by the Brazilian government. The present study analysed land cover and land use change in the major watersheds of the Brazilian Amazon, in order to evaluate the current balance between deforestation and conservation of natural areas in the region. The results show that watersheds draining the southern part of the basin have suffered the highest deforestation rates, with the largest losses (8.320% of total basin area) occurring in the Madeira, Tapaj´os, Xingu, Araguaia and Tocantins river basins. Most largewatersheds already have significant deforestation in their headwaters, which can affect hydrological functions and ecological sustainability. The greatest allocation of land for conservation was encountered in the Trombetas, Xingu and Negro watersheds, where conservation areas occupied 92.5, 56.9 and 50.6% of the total basin, respectively. While extensive areas of the Amazon biome have been deforested, on the scale of large watersheds there is a positive balance between conservation areas and deforestation, and on average the area delimited by conservation areas is more than three times larger than the deforested areas. An analysis by subwatersheds, however, indicates that certain regions have achieved more critical levels of deforestation, in some situations affecting more than 80% of the subwatersheds.

Hydroelectric reservoirs generate energy without significant combustion of fossil fuels. However, these systems can, potentially, emit greenhouse gases (GHGs) at a rate which may be significant at the global scale, and, possible, co-equal, per kilowatt-hour, to that from conventional coal or oil-fired systems. Although much of the new construction of hydroelectric reservoirs is in the tropics, most of the data on GHG emissions comes from temperate regions. Further, much of the existing data on reservoir gas emissions comes from single sites, usually near the terminal dams. Large tropical reservoirs often involve the impoundments of river systems with complex morphology which in turn can cause spatial heterogeneity in gas flux. We evaluated spatial and seasonal variability in CO2 concentrations and gas flux for five large (50 1,400 km2) reservoirs in the Cerrado region of Brazil. Most of data set (87% of all measurements) showed CO2 supersaturation and net efflux to the atmosphere. There was as much or more variation in pCO2 over space and among seasons. The large studied reservoirs showed different zones in terms of CO2 emission because those fluxes are dependent on flooded biomass, watershed input of organic matter and dam operation regime. Here we demonstrate that the reservoirs in the Brazilian Cerrado have low rates of CO2 emissions compared to existing global comparisons. Our results suggest that ignoring the spatial variability can lead to more than 25% error in total system gas flux.

It is well known that land use and land-cover changes (LUCC), particularly deforestation, have the potential to modify the hydrological response. Although those signals are relatively well documented in worldwide microcatchment studies, conflicting results reported in literature indicate that those signals can be sometimes difficult to detect and isolate in basins at larger scales. In order to detect signals in the hydrological response potentially, related to LUCC, streamflow records from Ji-Paran´a Basin located in SW Amazonia are analysed in conjunction with deforestation maps derived from remote sensors. The basin has a drainage area greater than 30 000 km2 and has been through severe LUCC in the last decades. Statistical descriptors of daily streamflow series were correlated with landscape indices using non-parametric methodologies. To take into account scale effects, statistical analyses were repeated in different sub-basins. Results showed that the impact of LUCC on the hydrological response is time lagged at larger scales. The flow paths are clearly affected, depending on basin characteristics such as topography. In general, LUCC impacts lead to higher peak streamflows, the reduction of minimal values and the increment of stormflow. In agreement with previous studies, the detection of signals associated with LUCC was clearly detected at the smallest basin, but proved to be difficult at larger scales, suggesting the existence of non-linear effects, which aggregate across scale compensating small scale effects. Such behaviour indicates a challenge for mathematical models, which are usually developed to represent immediate hydrological response to basin wide LUCC.

A new version of ETA WS (workstation) forecast model destined for long-term climate change simulation (ETA CCS) was designed. Numerous modifications and corrections have been made in the original code of the ETA WS forecast model. As a first step in the ETA CCS validation program, we have integrated the model over South America with a horizontal resolution of 40 km for the period 1960-1990. We forced it at its lateral boundaries by the outputs of HadAM3P, which provides a simulation of modern climate with a resolution of about 150 km. The climate ETA model was run on the supercomputer SX-6. Here, we present and compare the output fields of the ETA model and HadAM3P and analyze the geopotential, temperature, and wind fields of both models. For evaluating the similarities of the model outputs, we used a Fourier analysis of time series, the consistency index from linear regression coefficients, the time mean and space mean models’ arithmetic difference and root mean square difference. The results of the study demonstrate that there are no significant differences in behavior and spatial arrangement of large-scale structures of the two models. In addition, the regional model characteristics do not have major positive or negative trends during the integration in relation to the global model. Our analysis shows that the descriptions of large-scale climate structures by these two models are consistent. This means that the ETA CCS model can be used for downscaling HadAM3P output fields. Our proposed technique can be used to evaluate the consistency of any regional model and its driving global model.

This study shows the results from a regional climate simulation of the present-day climate, corresponding to the period 1961-1970 over South America, using the regional Eta Model nested within the HadAM3P model from the UK Hadley Centre. The simulation analysis is focused on assessing the capability of the nested regional model in representing spatial patterns of seasonal mean climate and the annual cycle of precipitation and temperature. The goals of this 10-year run for South America are to verify if the Eta Model can be used in climate-change scenarios and to verify if this model has the ability to generate added value for the South American continent. The Eta Model was chosen because there are few investigations using the Eta Model for long integrations over South America and because the vertical coordinate system used in this model is recommended for use over South America due to the presence of the Andes range. In the present 10-year simulation, the regional model reproduced many of the South American mesoscale climate features and together added new value to the driver model. Value was also added to the driver model by reducing seasonal biases in austral winter relative to austral summer. The regional model also exhibits better performance in the representation of low-level circulation, such as the topographically induced northwesterly flux.

A melhor ferramenta para projetar cenários prováveis de alterações climáticas para o futuro são os modelos matemáticos do sistema climático global (MCG), que levam em conta de forma quantitativa (numérica) o comportamento dos compartimentos climáticos (atmosfera, oceanos, criosfera (áreas com gelo e neve), vegetação, ciclos biogeoquímicos, etc.) e de suas interações. Esses modelos permitem que se simulem prováveis cenários de evolução do clima para vários cenários de emissões dos Gases de Efeito Estufa (GEE). Porém, há duas grandes fontes de incertezas ao utilizar estes modelos. A primeira, é que não sabemos precisamente a trajetória futura das emissões dos GEE e de aerossóis atmosféricos, que depende de decisões humanas sobre o caminho socioeconômico-ambiental desejado e que venha a ser efetivamente implementado. A segunda fonte de incerteza advém do fato que os modelos matemáticos são representações imperfeitas da natureza e diferentes modelos climáticos diferem substancialmente em suas projeções para o clima do futuro, dado o mesmo cenário de evolução das concentrações de GEE e de aerossóis na atmosfera. A maneira de abordar essas duas incertezas é utilizar vários cenários de emissões de GEE e diferentes modelos climáticos.