Validação do modelo Agro-ibis para um sítio experimental de soja no Rio Grande do Sul

Abstract

The Agro-IBIS model is validated for a soybean experimental site in the state of Rio Grande do Sul. This experiment is conducted in order to evaluate the impact of different soil management systems on different areas of science. In the site, we obtained micrometeorological measures of energy, mass and water exchanges between the soybean ecosystem and the environment during the 2009/2010 soybean growing season. We used two management systems: (1) no-tillage system (NT) and (2) conventional tillage system (CT). The model was forced with meteorological data from the Brazilian National Institute of Meteorology (INMET). The results of the model were compared with the observed data in order to evaluate the capacity of the model in simulating the interaction of the soybean ecosystem with environment and also to verify the effects of surface coverage by crop residues in NT. Since the model does not have the representation of crop residues, i.e., of NT, its effects are not expected in the simulation results. The results indicate that the model simulates the radiative balance and flows of energy in a satisfactory way. Besides this, soil moisture in the 0-20 cm and 20-50 cm layers and the 0-50 cm integration as well as soil temperature at 2 cm are simulated in a suitable way. For soil moisture, the simulated results represent better the observed data in CT, indicating that the effect of soil coverage by crop residues is not simulated. In the variables related to CO2, the model has problems concerning the simulation of the ecosystem respiration, which is strongly underestimated. This leads to an inconsistency in the simulation of the others variable related to CO2, mainly of those which consider the night-time data, since at night the net CO2 exchange of the ecosystem (NEE) is the respiration. For the daytime CO2 the model shows good performance, except in the period of leaf senescence, when the model indicates CO2 uptake by the ecosystem and the experimental data indicate emission.