Table 1.
RMSEP and percentages of variance explained (R2) of the relationships between observed and predicted heading dates simulated with a modified version of the Weir et al. (1984) phenological model for the calibration and validation data sets grown in contrasting location × sowing date combinations using optimized (calibration data set only) or QTL-based parameters (calibration and validation data sets)a
| Data set | Location | Sowing date | n | Mean (min.; max.) | Optimized parameters | QTL-based parameters | ||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2p | 3p | 2p | 3p | |||||||||
| RMSEP | R2 | RMSEP | R2 | RMSEP | R2 | RMSEP | R2 | |||||
| Calibration | Clermont- Ferrand | 27/10/2004 | 206 | 139 (126; 154) | 2.3 (5.2) | 0.93 (0.87) | 2.0 | 0.94 | 3.6 | 0.68 | 5.0 | 0.34 |
| 08/11/2005 | 62 | 143 (133; 157) | 3.7 (2.4) | 0.93 (0.89) | 3.0 | 0.94 | 4.5 | 0.68 | 5.3 | 0.31 | ||
| 23/02/2006 | 75 | 149 (140; 165) | 5.5 (8.1) | 0.73 (0.56) | 3.5 | 0.84 | 8.2 | 0.41 | 8.7 | 0.27 | ||
| 20/11/2008 | 210 | 143 (130; 159) | 2.4 (3.6) | 0.92 (0.84) | 1.9 | 0.94 | 4.3 | 0.62 | 5.5 | 0.32 | ||
| Le Moulon | 20/10/2003 | 171 | 146 (125; 161) | 2.0 (10.0) | 0.97 (0.95) | 1.4 | 0.98 | 4.2 | 0.76 | 6.3 | 0.42 | |
| 05/04/2004 | 114 | 181 (160; 226) | 5.4 (7.9) | 0.95 (0.91) | 5.0 | 0.95 | 17.6 | 0.63 | 14.6 | 0.68 | ||
| 20/10/2004 | 164 | 143 (121; 156) | 1.9 (8.1) | 0.96 (0.90) | 1.5 | 0.97 | 4.1 | 0.73 | 5.9 | 0.41 | ||
| 04/04/2005 | 122 | 178 (160; 223) | 4.0 (6.2) | 0.94 (0.93) | 4.1 | 0.94 | 13.4 | 0.56 | 10.6 | 0.62 | ||
| 20/10/2005 | 165 | 146 (132; 162) | 1.5 (8.6) | 0.96 (0.91) | 1.6 | 0.96 | 4.3 | 0.69 | 6.5 | 0.34 | ||
| 07/04/2006 | 131 | 179 (160; 220) | 2.6 (4.8) | 0.97 (0.90) | 2.5 | 0.97 | 15.0 | 0.54 | 15.4 | 0.48 | ||
| Validation | Le Moulon | 26/10/2006 | 88 | 127 (100; 141) | - | - | - | - | 5.6 | 0.59 | 6.6 | 0.38 |
| 23/10/2007 | 88 | 142 (115; 157) | - | - | - | - | 5.0 | 0.61 | 7.0 | 0.37 | ||
| Joze | 29/10/2006 | 88 | 133 (112; 146) | - | - | - | - | 5.7 | 0.57 | 7.7 | 0.31 | |
| 25/10/2007 | 88 | 147 (134; 160) | - | - | - | - | 8.6 | 0.48 | 10.9 | 0.30 | ||
| Estrées- Mons | 17/10/2006 | 88 | 135 (104; 151) | - | - | - | - | 5.6 | 0.63 | 8.0 | 0.41 | |
| 22/10/2007 | 88 | 148 [129; 160) | - | - | - | - | 6.7 | 0.58 | 9.0 | 0.31 | ||
a Unusual spring sowings are highlighted in bold. Results obtained with optimized parameters on the calibration data set with the original model from Weir et al. (1984) are shown in parentheses. Results obtained after optimization of two (2p) or three (3p) parameters are shown. The number of wheat genotypes (n), the mean and the range of variation of heading dates (in days) are indicated. Genotypes and location × sowing date combinations of the validation data set were not used to optimize parameters or calibrate the QTL-based model and are therefore totally independent.