Modelagem da interação superfície- atmosfera: calibração dos modelos sib2 e sib3 em uma área de pastagem no sul do Brasil

Abstract

The Simple Biosphere Model (SiB) modelwas developed to represent and estimate the interactions between soil, vegetation and atmosphere, which describe the transfer processes of energy, mass and momentum. In this sense, this study goal is the optimization of the fluxes of energy, water and carbon dioxide, estimated by the second and third versions of the SiB model (SiB2 and SiB3), using experimental data obtained by the Eddy Covariance method. The study area is located in the county of Santa Maria under the Pampa biome witch has pasture fields as characteristic vegetation. In this area experimental measurements of atmospheric forcing of model initialization and surface fluxes are obtained. The study period comprises the years of 2014 to 2016. First, a soil respiration equation was inserted into the SiB2 model, which is calibrated from observed night data of CO2 flux, soil and air temperature, and soil moisture. Then the soil and vegetation parameters were calibrated by the sequential modules method in the SiB2 and SiB3 models. This method consists of the calibration of each parameter following the execution sequence of the calculations in the models. The the optimal parameter was obtained through the Nash-Suticliffe efficiency (NSE). The calibration performed sought to optimize all three fluxes simultaneously. The same atmospheric variables that strongly influence the observed fluxes, also influence in equal proportion, the modeled fluxes. Therefore, it is inferred that the representation of the vegetation by the models is insufficient for the study area. The results obtained in this research were satisfactory. Thus, the parameters calibrated in this work can be used as representative parameters of the Pampa biome in surface models as well to evaluate the changes of land use in this ecosystem.