Calibration and validation of EPIC for crop rotations in southern France

Abstract

Management of complex crop rotations in southern France requires accurate assessment of the effects of irrigation, nitrogen fertilization, and the previous crop on crop growth and yield. The Erosion-Productivity Impact Calculator (EPIC) Cropping Systems model simulates the effects of weather, soil characteristics, tillage, fertilization, irrigation and other management practices on crops grown in complex rotations. The present study was conducted to evaluate EPIC's ability to simulate growth and yield of corn ( Zea mays L.), grain sorghum ( Sorghum bicolor (L.) Moench), sunflower ( Helianthus annuus L.), soybean ( Glycine max (L.) Merr.) and wheat ( Triticum aestivum L.) when grown in rotations at three levels of management inputs, including three levels each of fertilizer, irrigation and tillage, over a 5-year period. After calibration, the root mean square error of simulated model grain yields ranged from 0.4 Mg ha −1 for sunflower to 1.6 Mg ha −1 for corn. Only in the case of winter wheat were mean simulated yields significantly different from mean measured yields. Simulations were more accurate for summer crops than for winter wheat, primarily because EPIC overestimated the effects of cool spring temperatures on wheat development in 1986. We conclude that EPIC can simulate yields of summer crops grown in complex rotations in southern France with accuracy acceptable for many applications. The model could be improved by more accurate simulation of disease effects, like those suffered by wheat in 1988, and by more accurate simulation of the effects of cool temperatures on winter wheat phenology.