Artificial intelligence associated with satellite data in predicting energy potential in the Brazilian savanna woodland area

Abstract

The use of artificial intelligence to generate information of the savanna's energy capacity may support sustainable management of those areas. We assessed the efficacy of artificial neural networks (ANNs) combined with satellite data to estimate the energy potential (Pe) for cerradão, a dense savannah-like vegetation type in Brazil. We conducted a forest inventory for measuring dendrometric variables and sampling woody materials and barks in a cerradão area in the state of Tocantins, Brazil. The Pe of cerradão biomass was estimated based on the observed higher calorific power and drier biomass values. Six vegetation indices were retrieved from a RapidEye image and tested for correlation to choose the optimum vegetation index for biomass modeling. The basal area and the Normalized Difference Vegetation Index were used as predictors in the Pe modeling. We estimated an average of 19.234 ± 0.411 GJ ton-1 and 19.878 ± 1.090 GJ ton-1 for higher heating values of the wood species and barks, respectively, and an average Pe of 1022.660 GJ ha-1. The best ANN showed an error of 11.3% by using a structure of two, eight, and one neurons in the input layer, in the hidden layer, and in the output layer, respectively, as well as activation functions of the tangential and sigmoidal types. The validation tests showed no significant difference between the observed and ANNpredicted values. Based on our results, we concluded that Pe can be efficiently predicted by combining ANNs and remotely sensed data, which ultimately is a promising tool for forest sustainable management of the cerrado ecosystems.

Artificial intelligence tools and vegetation indices combined to estimate aboveground biomass in tropical forests

“…This study results corroborate with several authors who have applied remotely sensed data and artificial neural networks to estimate variables in tropical 19,21,26,44,[78][79][80] and subtropical forests. 81…”

“…21,26,27,75,76 In addition, the ANNs and remotely sensed data combined can provide higher modeling precision in forest sites than the regression models, as they are able to assimilate a high complexity and variety of vegetation, environment, and climatic aspects. 21,24 Although developing prediction models using ANNs is a high-complexity task that requires skilled labor and high computational capacity for training them, the ANN has achieved higher estimate precisions when modeling complex relationships among different variables. 22,75,77 More specifically, in this study, it was required approximately 50 working hours of one highly skilled person to train, validate, and apply the ANN model using our datasets and a classical personal computer.…”

“…The Broyden-Fletcher-Goldfarb-Shanno (BFGS) algorithm was used for training the neural network, [45][46][47][48] composed by a quasi-Newton algorithm, which has been widely and efficiently applied for solving problems and optimizing predictions. 21,26,49 For training the ANNs, 170 (70%) of the biomass data collected during the forest inventory using 255 sample plots. The remaining 30% of the biomass data were used for validating the ANNs, as suggested by Refs.…”

“…19,20 Similarily, the artificial neural networks (ANNs) have been applied to predict forest production. [21][22][23][24] The ANNs are computing systems interconnected with nodes (types of functions) showing learning capacity to perform a task. 25,26 The ANNs and remotely sensed data combined have been successfully applied to estimate dendrometric variables in tropical forests.…”

The effect of carbonization on the wood anatomy of Sclerolobium paniculatum Vogel

“…This biome has undergone major changes in the last 50 years due to agricultural expansion and considerable changes in its landscape, generating irreversible environmental impacts and damage to this ecosystem (Dutra e Silva and Barbosa 2020). However, this biome is an important energy source, primarily to the state of Tocantins, and its knowledge is important for sustainable forest management of the ecosystems (Carrijo et al 2020).…”