Fluxes of CO2, CH4, and N2O in tundra-covered and Nothofagus forest soils in the Argentinian Patagonia.

While most soils in periglacial environments present high fluxes of CO2 (FCO2), CH4 (FCH4), and N2O (FN2O), few of them have a tendency to drain greenhouse gases from the atmosphere. This study aimed to assess greenhouse gas fluxes at different sub-Antarctic sites and time periods (at the beginning of thaw and height of summer). To investigate the time of year effect on greenhouse gas emissions, FCO2, FCH4, and FN2O were measured at two sites tundra-covered (Ti and Th) and Nothofagus forest soil (Nf) on Monte Martial, at the southernmost tip of South America, Tierra del Fuego, Argentina. FCO2 ranged from 96.33 to 225.72 µg CO2 m-2 s-1 across all sites and periods, showing a positive correlation with soil temperature (Ts) (4.1 and 8.2 °C, respectively) (r2 > 0.7; p < 0.05). The highest values of FCO2 were found at Ti and Th (728.2 and 662.64 µg CO2 m-2 s-1, respectively), which were related to higher temperatures (8.2 and 8.6 °C, respectively) when compared to those of Nf. For FCH4, the capture (drain) occurred during both periods at Nf (-26 and -79 µg C-CH4 m-2 h-1) as well as Ti and Th (-21 and 12 µg C-CH4 m-2 h-1, respectively). FN2O also presented low values during both periods and showed a tendency to drain N2O from the atmosphere, especially at Nf (-2 µg N-N2O m-2 h-1). In addition, FN2O was slightly positive for Ti and Th (0.3 and 0.55 µg N-N2O m-2 h-1, respectively). Soil moisture did not show a correlation (p > 0.05) with the measured greenhouse gas fluxes. A scenario of increased temperatures might result in changes in the balance between the emissions and drains of these gases from soils, leading to higher emission values of CH4 and N2O, especially for tundra covered soils (Ti and Th), where the highest average fluxes and thermohydric variations were observed over the year.

Forecasting the spatiotemporal variability of soil CO2 emissions in sugarcane areas in southeastern Brazil using artificial neural networks.

Carbon dioxide (CO2) is considered one of the main greenhouse effect gases and contributes significantly to global climate change. In Brazil, the agricultural areas offer an opportunity to mitigate this effect, especially with the sugarcane crop, since, depending on the management system, sugarcane stores large amounts of carbon, thereby removing it from the atmosphere. The CO2 production in soil and its transport to the atmosphere are the results of biochemical processes such as the decomposition of organic matter and roots and the respiration of soil organisms, a phenomenon called soil CO2 emissions (FCO2). The objective of the study was to investigate the use of neural networks with backpropagation algorithm to predict the spatial patterns of soil CO2 emission during short periods in sugarcane areas. FCO2 values were collected in three commercial crop areas in the São Paulo state, southeastern Brazil, registered through the LI-8100 system during the years 2008 (Motuca), 2010 (Guariba city), and 2012 (Pradópolis), in the period after the mechanical harvesting (green cane). A neural network multilayer perceptron with a backpropagation algorithm was applied to estimate the FCO2 in 2012, using data from 2008 and 2010 as training for the neural network. The neural network initially presented a mean absolute percentage error (MAPE) of 18.3852 and a coefficient of determination (R2) of 0.9188. Data obtained from the observed and estimated values of FCO2 present moderate spatial dependence, and it is observed from the maps of the spatial pattern of the CO2 flow that the results from the neural network show considerable similarity to the observed data. The model results identify the higher and lower characteristics in sample points of CO2 emissions and produce an overestimation of the range of spatial dependence (0.45 m) and an underestimation of the interpolated values in the field (R2 = 0.80; MAPE = 12.0591), when compared to the actual soil CO2 emission values. Therefore, the results indicate that the artificial neural network provides reliable estimates for the evaluation of FCO2 from data of the soil's physical and chemical attributes and describes the spatial variability of FCO2 in sugarcane fields, thereby contributing to the reduction of uncertainties associated with FCO2 accountings in these areas.

Modeling sugarcane ripening as a function of accumulated rainfall in Southern Brazil.

The effect of weather variables on sugarcane ripening is a process still not completely understood, despite its huge impact on the quality of raw material for the sugar energy industry. The aim of the present study was to evaluate the influence of weather variables on sugarcane ripening in southern Brazil, propose empirical models for estimating total recoverable sugar (TRS) content, and evaluate the performance of these models with experimental and commercial independent data from different regions. A field experiment was carried out in Piracicaba, in the state of São Paulo, Brazil, considering eight sugarcane cultivars planted monthly, from March to October 2002. In 2003, at the harvest, 12 months later, samples were collected to evaluate TRS (kg t(-1)). TRS and weather variables (air temperature, solar radiation, relative humidity, and rainfall) were analyzed using descriptive and multivariate statistical analysis to understand their interactions. From these correlations, variables were selected to generate empirical models for estimating TRS, according to the cultivar groups and their ripening characteristics (early, mid, and late). These models were evaluated by residual analysis and regression analysis with independent experimental data from two other locations in the same years and with independent commercial data from six different locations from 2005 to 2010. The best performances were found with exponential models which considered cumulative rainfall during the 120 days before harvest as an independent variable (R (2) adj ranging from 0.92 to 0.95). Independent evaluations revealed that our models were capable of estimating TRS with reasonable to high precision (R (2) adj ranging from 0.66 to 0.99) and accuracy (D index ranging from 0.90 to 0.99), and with low mean absolute percentage errors (MAPE ≤ 5 %), even in regions with different climatic conditions.