UTILIZANDO DADOS DE QUALIDADE DA ÁGUA E DE SENSORIAMENTO REMOTO NA CALIBRAÇÃO AUTOMÁTICA E VALIDAÇÃO DE UM MODELO HIDROSSEDIMENTOLÓGICO

Código

PAP014

Autores

Hugo de Oliveira Fagundes, Rodrigo Cauduro Dias de Paiva, Fernando Mainardi Fan

Tema

MEDIÇÃO, MODELAGEM, TÉCNICAS E TECNOLOGIAS NO CONTEXTO DA HIDROSSEDIMENTOLOGIA

Resumo

Observed data of suspended sediment concentration(SSC) are essentialfor studies that use sediment modelling,mainly becauseof thecalibration and validation needs.Because the high cost in sediment traditional monitoring, alternativemonitoring techniques have emerged in recent decades presenting good results in estimating of sediment. In this context, the aim of this study istoinvestigate the use not only of SSC, but also turbidity, total suspended matter(TSM) and of surface reflectance (SR) data to calibrate and validate an erosion and sediment transport model for large scales.The Doce river basin (Brazil) was chosenas test caseandwereused observed data of SSC, water quality and remote sensingfrom 25, 63 and 21 stations,respectively. Six experiments for automatic calibration of the MGB-SED model and its validationwere performed. Results show that increasingspatial discretization provide better results as well as the use of the logarithm of SSC instead of just SSC and the use of abackgroundconcentration. For Fazenda Ouro Fino gauge, the calibration improved the ENScoefficient from -0.44 to 0.44. This experiment highlights the importance of use of data measure in situ with a high sampling frequency. Experiments show that turbidity, TSM and SR data can be used for verification and improvement of sediment models.