DROUGHT TOLERANCE OF BRAZILIAN SOYBEAN CULTIVARS SIMULATED BY A SIMPLE AGROMETEOROLOGICAL YIELD MODEL

Abstract

SUMMARYThe objective of this study was to calibrate and evaluate a simple crop yield model for 101 Brazilian soybean cultivars, and through the calibrated water deficit sensitivity index (Ky) to classify groups of cultivars in relation to drought tolerance. The cultivars’ actual yield was obtained from field experiments conducted by Pro-Seeds Foundation in 17 locations in southern Brazil from 2008 to 2011. Daily weather data were obtained from the government weather networks and rainfall was recorded at each experimental location. The crop yield model FAO–Agroecological zone was used to estimate potential yield (Yp), while the water deficit yield depletion model was used to estimate actual yield (Ya) and to determine Ky. Calibrated Ky values were used in a cluster analysis to determine groups of soybean cultivars with the same degree of drought tolerance. The crop yield model performed very well with lower values of mean absolute error (284 kg ha−1) and mean error (7 kg ha−1). The Ky values of 0.97, 0.90, 0.88 and 0.78 were obtained for the most sensitive soybean phenological phase to water deficit (flowering/yield formation), and were used to identify the groups of low, medium-low, medium-high and high drought tolerance respectively. In spite of Ky differences in cultivar groups, harvest index (CH) also varied, ranging from 0.31 to 0.35 for the group of high to low drought tolerance. The crop yield model proved to be an efficient tool for identifying drought tolerance of Brazilian soybean cultivars and for choosing the best cultivar for a given environment.