Artificial neural networks and adaptive neuro-fuzzy inference systems for prediction of soil respiration in forested areas southern Brazil.

Autor: Vicentini ME; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil. mevicentini@gmail.com., da Silva PA; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil., Canteral KFF; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil., De Lucena WB; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil., de Moraes MLT; Department of Phytotecnics, Faculty of Engineer (FEIS/UNESP), Avenida Brasil-Centro, Ilha Solteira, São Paulo, 15385-000, Brazil., Montanari R; Department of Phytosanity, Rural Engineering and Soils, Faculty of Engineer (FEIS/UNESP), Avenida Brasil-Centro, Ilha Solteira, São Paulo, 15385-000, Brazil., Filho MCMT; Department of Phytosanity, Rural Engineering and Soils, Faculty of Engineer (FEIS/UNESP), Avenida Brasil-Centro, Ilha Solteira, São Paulo, 15385-000, Brazil., Peruzzi NJ; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil., La Scala N Jr; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil., De Souza Rolim G; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil., Panosso AR; Department Engineering and Exact Sciences, School of Agricultural and Veterinarian Sciences, São Paulo State University (FCAV/UNESP), Via de Acesso Prof. Paulo Donato Castellane S/N, Jaboticabal, São Paulo, 14884-900, Brazil.
Jazyk: angličtina
Zdroj: Environmental monitoring and assessment [Environ Monit Assess] 2023 Aug 24; Vol. 195 (9), pp. 1074. Date of Electronic Publication: 2023 Aug 24.
DOI: 10.1007/s10661-023-11679-8
Abstrakt: The purpose of this study was to estimate the temporal variability of CO 2 emission (FCO 2 ) from O 2 influx into the soil (FO 2 ) in a reforested area with native vegetation in the Brazilian Cerrado, as well as to understand the dynamics of soil respiration in this ecosystem. The database is composed of soil respiration data, agroclimatic variables, improved vegetation index (EVI), and soil attributes used to train machine learning algorithms: artificial neural network (ANN) and an adaptive neuro-fuzzy inference system (ANFIS). The predictive performance was evaluated based on the mean absolute error (MEA), root mean square error (RMSE), mean absolute percentage error (MAPE), agreement index (d), confidence coefficient (c), and coefficient of determination (R 2 ). The best estimation results for validation were FCO 2 with multilayer perceptron neural network (MLP) (R 2  = 0.53, RMSE = 0.967 µmol m -2  s -1 ) and radial basis function neural network (RBF) (R 2  = 0.54, RMSE = 0.884 µmol m -2  s -1 ) and FO 2 with MLP (R 2  = 0.45, RMSE = 0.093 mg m -2  s -1 ) and RBF (R 2  = 0.74, 0.079 mg m -2  s -1 ). Soil temperature and macroporosity are important predictors of FCO 2 and FO 2 . The best combination of variables for training the ANFIS was selected based on trial and error. The results were as follows: FCO 2 (R 2  = 16) and FO 2 (R 2  = 29). In all models, FCO 2 outperformed FO 2 . A primary factor analysis was performed, and FCO 2 and FO 2 correlated best with the weather and soil attributes, respectively.
(© 2023. The Author(s), under exclusive licence to Springer Nature Switzerland AG.)
Databáze: MEDLINE
Externí odkaz: