ABSTRACT:
Simulation models of crop growth enable estimating crop yield and water balance components with low cost and high precision. This study aimed to calibrate and evaluate the performance of the CS-CROPGRO model for common beans under the conditions of Jaboticabal city (SP), Brazil. The model was calibrated and tested with data previously collected from two experiments conducted in 2002 and 2003. In these experiments, treatments corresponded to two irrigation scheduling methods (tensiometers and class A pan evaporation) and two cropping systems (conventional and no-till). For calibration, data from the experiment carried out in 2002 were used. For testing, we used data from the experiment conducted in 2003. By using the genetic coefficients calibrated for cultivar IAC-Carioca, the model could correctly simulate phenology, leaf area, dry matter and grain yield during both, calibration and testing, under no-till and conventional systems, combined with tensiometer and class A pan irrigation scheduling methods. Conversely, the model did not show the same performance for soil moisture simulations.
KEYWORDS:
simulation; DSSAT; Phaseolus vulgaris; irrigation; planting system
