Parameter Estimation for Crop Models
A New Approach and Application to a Corn Model
Daniel Wallach*,a,
Bruno Goffinetb,
Jacques-Eric Bergeza,
Philippe Debaekea,
Delphine Leenhardta and
Jean-Noël Aubertotc
a Unité d'Agronomie, INRA, BP 27, 31326 Castanet Tolosan Cedex, France
b Unité de Biométrie et d'Intelligence Artificielle, INRA, BP 27, 31326 Castanet Tolosan Cedex, France
c Laboratoire d'Agronomie, INRA INA P-G, 78850 Thiverval-Grignon, France

View larger version (25K):
[in a new window]
|
Fig. 1. Calculated vs. observed yields, for (A) initial parameter values and for (B) the best model as given in Table 4. The 1:1 line is shown. For the model with no adjusted parameter values, Cy = MSEPy by definition.
|
|

View larger version (25K):
[in a new window]
|
Fig. 2. Calculated vs. observed yield losses, for (A) initial parameter values and for (B) the best model as given in Table 4. The 1:1 line is shown. For the model with no adjusted parameter values, Cyloss = MSEPyloss by definition.
|
|
Copyright © 2001 by the American Society of Agronomy.