Uso do modelo 3-PG para estimar componentes do balanço de C em cafezal

Abstract

An alternative to the empiric models used to determine the yield and estimate the mass flux in coffee is the utilization of process-based models. Among many process-based models applied, one of the most used in Brazil, mainly in forestry, is the 3-PG model (Physiological Principles in Prediction Growth), a simplified process-based model. This model has shown great results to eucalyptus in Brazil, but has not been used in agricultural crops yet. The present study sought to evaluate the adaptability of 3-PG model to the coffee crop. This study was made in Lavras, Minas Gerais, using Coffea arabica L., Mundo Novo cultivar, evaluated from 3 to 6 years old, 4 m x 1 m spacing between rows and plants. The evaluated area was 0.4 hectares, where the leaf area index (LAI) of 30 plants were measured monthly during the time of the study. The litter fall rate was monitored trough 10 collectors of 0.3 m x 0.4 m distributed in the area who were monthly evaluated. At the end of the study, eight plants were harvested randomly to compare the results obtained by the model with the measured in field. A non-linear model was adjusted to describe the relationship between minimum temperature and leaf fall, since the 3-PG original leaf fall model was note capable to determine the great variation of the coffee leaf fall rate through the year. The results obtained by the model were compared to the field data and the literature data in order to determine whether the 3-PG model could be used in coffee culture. The results obtained by the model corroborate with the field and literature data, leading to believe that the 3-PG model can be calibrated and validated to the coffee crop.