Publicações

Soil moisture is being monitored over the Brazilian semiarid since 2014 using a network of 595 stations. The network was implemented by Cemaden, the National Center for Monitoring and Early Warning of Natural Disasters, with the objective of monitoring of spatial and temporal patterns of soil water and potential impacts on the regions agriculture activities. The Brazilian semiarid is the poorest in the country and hosts most of the family owned farms in Brazil. Crops such as maize and beans are mostly planted over small properties with no irrigation. In this work, agricultural risk is assessed using measurements of soil moisture which are presented in formats that highlight spatial, temporal and short-term variability. The Soil Moisture Index (SMI), based on a normalization of soil moisture by field capacity and wilting point, is used to characterize the current soil moisture conditions in a scale proportional to the water available to plants. The temporal evolution of SMI is used to visualize trends in short-term drought and response to rainfall events at daily time steps or recent weeks. A novel index based on continuous exposure to critical SMI is developed to help bring awareness of real time risk of water stress over the region: the Index of Stress in Agriculture (ISA). The index takes into account continuous exposure to a critical value of SMI (assumed as 0.4), below which water stress is expected to be triggered in plants. The index was tested during a short-term drought over the region and successfully identified locations under water stress for periods of three days or more. The monitoring tools presented here help to describe the real time conditions of drought over the region using daily observations. These tools are expected to be implemented over an online platform hosted at Cemadens website.

The largest single block of seasonally dry forest in the world is the Caatinga biome found in northeastern Brazil. It covers 844,453 km2 and houses a high level of endemic plant species adapted to the low precipitation (400-800 mm year-1), high mean air temperature (26-28 oC), and elevated potential evapotranspiration (over 2,500 mm year-1). The Caatinga biome has about 55% of its original area preserved, and 45% with some degree of degradation related to anthropogenic impacts such as unsustainable natural resources management and land cover clearing for agricultural and grazing activities. Since 2012 the region has been experiencing the most severe drought of the last 50 years. Under this scenario, the caatinga vegetation performs a crucial role due to its capacity to fix atmosphere carbon dioxide even when water is scarce. Quantifying radiation, energy, CO2 and evapotranspiration in different caatinga areas is essential for a better understanding of their functioning and for evaluating possible changes in their interaction with the atmosphere due to dynamic climate and environmental change. Eddy covariance is a widely accepted method to evaluate energy and carbon fluxes over ecosystem scale, and was used to study two sites in the Pernambuco state of Brazil within the Caatinga biome. The first study site is a pristine caatinga (PC) located in Petrolina municipality, Brazil (09.03S, 40.32W). This site has an area of 600 ha and has been preserved for over 40 years. Its vegetation comprises a mix of shrubby-arboreal hyperxerophilic species plants with average height of 5 m. The second study site, characterized by 5 ha of degraded caatinga (DC), is located in Araripina (07.45S, 40.42W), about 176 km north from PC. It was used as a rainfed agricultural area during the 1980s and 1990s and has slowly been reclaimed by caatinga species since the beginning of the last decade. Micrometeorological measurements were conducted at both sites for an average rainfall year (2011) and an unusually extremely dry year (2012). The towers monitor turbulent fluxes by measuring three-dimensional wind velocity and temperature with an ultrasonic anemometer, and CO2 and H2O concentrations with an open path infrared gas analyzer at a frequency of 10 Hz. The collected high-frequency data were filtered, despiked, rotated, and corrected using standard eddy covariance (EC) processing methods. In addition, air temperature, air humidity, net radiation, soil heat flux, soil temperature, and photosynthetically active radiation were measured. Data were processed and analysed on daily and seasonal time scales. The dynamics of net ecosystem exchange and of latent heat flux (evapotranspiration) over these two contrasting land use of caatinga were compared for the two years with very different water availability, revealing the importance of long term ecosystem studies across the Caatinga biome and its various physiognomies.

Understanding how land use change can affect natural dynamics of water resources in Amazon and Brazilian Cerrado is of importance due to the constant pressure for deforestation suffered by these biomes. In this work, the dynamic vegetation model with river routing and flood dynamics INLAND-THMB (Integrated LAND Surface Model – Terrestrial Hydrology Model with Biogeochemistry) was used to simulate how land use change (LUC) can affect the water balance, river discharge and flood dynamics in the Amazon Basin and in Brazilian Cerrado. To achieve this goal, we performed simulations following the recent Global Carbon Budget Land modelling protocol (Trendy-v7) and compared the S2 (historical CO2 and climate and time-invariant 1700 land use mask) and S3 (historical CO2, climate and land use). Thus we can evaluate the effects of LUC keeping the information of dry and wet events during the period. To better represent the land use state in S3 simulation, we compiled deforestation information from the PRODES/INPE project in the Brazilian Amazon and Cerrado and produced a map with remaining natural area for each year. The evaluation was performed to each biome separately and also in some of their sub-basins. In the Amazon, we looked at the upstream Amazon Basin (Alto Solimões), which includes an area with very low LUC, and at the Madeira, Tapajós and Xingu all in the arc of deforestation and in Cerrado at Tocantins Basin. Preliminary results show higher river discharge related to LUC, as expected, and also less available water, since the ET decreases when we include anthropic changes in land cover in the model. It is important to highlight that this study did not include atmospheric feedbacks in the precipitation, which can mask some of the results, since considerable amount of the Amazon precipitation is provided by recycling as shown in previous studies.

Soil moisture variations and soil water potentials are crucial to understand the plant behavior under water stress conditions. Such kind of data are also valuable for modeling efforts in hydrology, ecophysiology, soil-plantatmosphere interactions and earth system models. Here we present soil moisture variation along the hydrological year of 2015 for 4 different soil profiles along a topographic gradient in the Brazilian caatinga, Sao Joao site, northeast Brazil, Pernambuco state. The data were collected using Sentek probes, Enviroscan type, in sandy soil profiles of 0.4 m (P1), 0.5 m (P2), 0.9 m (P3) and 1.10 m (P4), from the bottom to the top of an instrumented hillslope of around 1 km length. Along this topographic gradient, the maximum depth of probes installation occurred due to the existence of an impenetrable layer, commonly found in some areas of the Caatinga biome. Our results show that, during the rainy season, in the deepest soil profile (P4) from the upper part of the slope, soil moisture reached the maximum of 16% at the top layer (0.1-0.4 m soil depths), while below 0.4 m depth, the maximum soil moisture registered during rainy season were 6%. However, during the dry period, soil moisture at the top (0.1-0.4 m soil depths) decreased to values below 1 %, while the deeper soil layers (below 0.70 m) maintained moisture content at values around 4% for the whole dry period. In the shallower soil profile (P4), soil moisture also reached the maximum of 16% in the wet season along the whole soil profile. In the dry period, however, the top layer presented maximum soil content of 1% while the bottom (0.4 m) maintain moisture content of around 4%, similar to the deepest layer of the P4 profile. Similar behavior was found for the intermediate soil profiles (P2 and P3), suggesting that, independently of the depth of the soil profile, soil moisture at the interface soil/atmosphere present values below 1% in the dry season while at the interface sandy soil/impenetrable layer, moisture is maintained at around 3-5%. For the next steps, these results will be analyzed in terms of rainfall rates, soil physical properties and water storage in order to understand the water balance dynamic for different soil depths above the impenetrable soil layer.

O Pantanal, localizado na Bacia do Alto Paraguai (BAP), abrange porções do Brasil, Bolívia e Paraguai e é uma das mais importantes áreas úmidas da Terra. No Brasil ele possui uma população de aproximadamente 1,1 milhões de habitantes e é considerado de importância nacional por sua biodiversidade e serviços ambientais relevantes (JUNK e CUNHA, 2012). O ciclo hidrológico no Pantanal Sul afeta a área de produção extensiva de gado de corte, que é historicamente dependente e adaptada ao uso do pasto nativo das planícies, que é renovado naturalmente pela inundação sazonal. A variabilidade da precipitação e vazões dos principais rios da BAP pode ser influenciada por processos complexos das configurações ambientais da planície, inerentes à sua geomorfologia, topografia e pedologia (SILVA et al., 2016; CLARKE et al., 2003; GALDINO et al., 1997) e por forçantes como mudanças no uso da terra, O Pantanal, localizado na Bacia do Alto Paraguai (BAP), abrange porções do Brasil, Bolívia e Paraguai e é uma das mais importantes áreas úmidas da Terra. No Brasil ele possui uma população de aproximadamente 1,1 milhões de habitantes e é considerado de importância nacional por sua biodiversidade e serviços ambientais relevantes (JUNK e CUNHA, 2012). O ciclo hidrológico no Pantanal Sul afeta a área de produção extensiva de gado de corte, que é historicamente dependente e adaptada ao uso do pasto nativo das planícies, que é renovado naturalmente pela inundação sazonal. A variabilidade da precipitação e vazões dos principais rios da BAP pode ser influenciada por processos complexos das configurações ambientais da planície, inerentes à sua geomorfologia, topografia e pedologia (SILVA et al., 2016; CLARKE et al., 2003; GALDINO et al., 1997) e por forçantes como mudanças no uso da terra.

O Pantanal é a planície de inundação da Bacia do Alto Paraguai (BAP), uma das mais importantes áreas úmidas da Terra. Ao ritmo dos pulsos de inundação (JUNK, 2004) do Rio Paraguai e seus inúmeros afluentes, a dinâmica hidrológica desta região reflete padrões em múltiplas escalas temporais, desde a intranual até multidecadais (CLARKE et al., 2003). O Pantanal Sul possui expressiva produção pecuária bovina de corte. A atividade é suscetível e dependente da sazonalidade natural das inundações, contudo é considerada historicamente adaptada, com sistema extensivo desenvolvido sob a lógica de articulação de áreas e ocupação temporária das terras baixas para o aproveitamento dos campos nativos, que são sempre renovados com a passagem das águas. O presente trabalho visa analisar a evolução dos níveis dos rios, que possui forte influência nas atividades pecuárias do Pantanal Sul.

A pecuária bovina de corte é um dos principais empreendimentos agrícolas do sul do Pantanal, estado do Mato Grosso do Sul. A natureza dessa atividade varia dentro da região, juntamente com a variabilidade espacial e temporal das pastagens nativas (SILVA et al., 2016a), a suscetibilidade de diferentes áreas às cheias anuais e as formas de manejo técnico e administrativo da criação (SANTOS et al., 2002; ABREU et al., 2010). Corumbá, um município localizado na porção oeste da região, ocupa 69,9% da área sul do Pantanal e se destaca como importante produtor de bezerros para a cadeia de suprimentos do estado; em 2014 teve o segundo maior rebanho do Brasil, com 1.761.574 animais (BRASIL, 2014). Enquanto municípios situados na transição da planície com o planalto de borda, como Aquidauana, Coxim, Rio Verde de Mato Grosso e Porto Murtinho apresentam uma pecuária mais intensiva e modernizada, se situam mais próximos à indústria frigorífica de Mato Grosso do Sul e atuam como importantes canais de comercialização dos rebanhos para recria e engorda (ARAUJO e BICALHO, 2010). Esses municípios recebem gado de regiões com terras mais baixas, mais propensas a secas ou inundações prolongadas, e por isso representam a integração de Corumbá com outas localidades do estado.