Artificial Neural Network Model of Soil Heat Flux over Multiple Land Covers in South America

Código

XXIV-SBRH0792

Autores

Bruno César Comini de Andrade, ANDERSON LUIS RUHOFF, Olavo Correa Pedrollo, Adriana Aparecida Moreira, LEONARDO LAIPELT DOS SANTOS, Rafael Kayser, Marcelo Sacardi Biudes, Carlos Antonio Costa dos Santos, DEBORA REGINA ROBERTI, Antônio Celso Dantas Antonino, EDUARDO SOARES DE SOUZA, HIGO JOSÉ DALMAGRO, Nadja Machado, RODOLFO MARCONDES DE SOUZA, José Romualdo de Sousa Lima

Tema

SE07.B - Sensoriamento remoto da água: de avanços técnicos-científicos a aplicações na nova era de disponibilidade de informação

Resumo

Soil heat flux (G) is an important component for the closure of the surface energy balance (SEB) and the estimation of evapotranspiration (ET) by remote sensing algorithms. Over the last decades, efforts have been focused on parameterizing empirical models for G prediction, based on biophysical parameters estimated by remote sensing. However, due to the existing models? empirical nature and the restricted conditions in which they were developed, large-scale applications may lead to significant errors. Thus, our objective was to assess the ability of the artificial neural network (ANN) to predict mid-morning G using remote sensing and meteorological data over a broad range of climates and land covers in South America. Surface temperature (Ts), albedo (?), and enhanced vegetation index (EVI), from the moderate resolution imaging spectroradiometer (MODIS), and net radiation (Rn) from the global land data assimilation system 2.1 (GLDAS 2.1), were used as inputs. The ANN?s predictions were validated against measurements from 23 flux towers over multiple land cover in South America. Their performance was compared to that of existing and commonly used models, the Jackson et al. (1987) and Bastiaanssen (1995), which were priorly calibrated for quadratic error minimization. The ANN's outperformed existing models, with mean absolute error (MAE) reductions of 43% and 36%, respectively. Additionally, the inclusion of land cover information as an input in the ANN reduced MAE by 22%. This study indicates that the ANN?s structure is more suited for large-scale G prediction than existing models, which can potentially refine SEB fluxes and ET estimates.