EPICphase, a version of the EPIC model simulating the effects of water andnitrogen stress on biomass and yield, taking account of developmental stages: validation on maize, sunflower, sorghum, soybean and winter wheat
M. Cabelguenne et al., EPICphase, a version of the EPIC model simulating the effects of water andnitrogen stress on biomass and yield, taking account of developmental stages: validation on maize, sunflower, sorghum, soybean and winter wheat, AGR SYST, 60(3), 1999, pp. 175-196
The plant module of the EPIC model (Erosion Productivity Impact Calculator)
has been modified to simulate the effects of water and nitrogen (N) stress
on biomass production and grain yield, taking account of the sensitivity o
f the crops to water and N stress during the course of their developmental
cycle. This French version of the model, EPICphase, was validated with maiz
e (Zea mays L.), sunflower (Helianthus annuus L.), sorghum (Sorghum bicolor
L. Moench), soybean (Glycine max L.), and winter wheat (Triticum aestivum
L.) over 9 years, using experimental data from a long-term cropping system
experiment carried out at three levels of cropping intensity. The results h
ave been compared with those from a normal version of EPIC. The two version
s use the same input files. The results show that EPIC overestimates crop p
roduction by comparison with the measured data, notably under conditions of
severe moisture stress. The additional crop parameters introduced into EPI
Cphase concern the water extraction capacities peculiar to each crop, the d
ivision of the growth period into four phases, with adjustments to the conv
ersion efficiency of intercepted radiation into biomass, and the drought ad
aptation of sunflower and soybean. Finally, the sensitivity of the harvest
index to water and N stress has been introduced for each phase of growth. T
hese refinements have led to simulations very close to the measured values.
Comparison of results from the two versions was done by means of a statist
ical study of mean biomass production and grain yield, standard deviations,
root mean square error, regression lines, and R-2 values. (C) 1999 Elsevie
r Science Ltd. All rights reserved.