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

• Published in: Environmental monitoring and assessment 2018-12, Vol.190 (12), p.1-14, Article 741
• Main Authors: Freitas, Luciana P. S., Lopes, Mara L. M., Carvalho, Leonardo B, Panosso, Alan R., La Scala Júnior, Newton, Freitas, Ricardo L. B., Minussi, Carlos R., Lotufo, Anna D. P.
• Format: Article
• Language: English
• Subjects: Agricultural management; Algorithms; Artificial neural networks; Atmosphere; Atmospheric Protection/Air Quality Control/Air Pollution; Back propagation; Carbon dioxide; Carbon dioxide atmospheric concentrations; Carbon dioxide emissions; Climate change; Data; Dependence; Earth and Environmental Science; Ecology; Ecotoxicology; Emission; Emissions; Environment; Environmental Management; Environmental monitoring; Environmental science; Evaluation; Farm buildings; Gases; Global climate; Green cane; Greenhouse effect; Greenhouse gases; Harvesting; Monitoring/Environmental Analysis; Multilayer perceptrons; Neural networks; Organic chemistry; Organic matter; Soil; Soil chemistry; Soil investigations; Soils; Spatial variability; Spatial variations; Sugarcane; Training
• ISSN: 0167-6369; 1573-2959
• DOI: 10.1007/s10661-018-7118-0

Carbon dioxide (CO 2 ) 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 CO 2 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 CO 2 emissions (FCO 2 ). The objective of the study was to investigate the use of neural networks with backpropagation algorithm to predict the spatial patterns of soil CO 2 emission during short periods in sugarcane areas. FCO 2 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 FCO 2 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 ( R 2 ) of 0.9188. Data obtained from the observed and estimated values of FCO 2 present moderate spatial dependence, and it is observed from the maps of the spatial pattern of the CO 2 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 CO 2 emissions and produce an overestimation of the range of spatial dependence (0.45 m) and an underestimation of the interpolated values in the field ( R 2  = 0.80; MAPE = 12.0591), when compared to the actual soil CO 2 emission values. Therefore, the results indicate that the artificial neural network provides reliable estimates for the evaluation of FCO 2 from data of the soil’s physical and chemical attributes and describes the spatial variability of FCO 2 in sugarcane fields, thereby contributing to the reduction of uncertainties associated with FCO 2 accountings in these areas.