ApsimBot
commented
5 years ago Author: fai04d Date: 2014-06-20 10:23:00 +0000 UTC
J. Exp. Bot.-2013-Zheng-3747-61 Supp.pdf Size: 360953
Open ApsimBot opened 5 years ago
ApsimBot
commented
5 years ago Author: fai04d Date: 2014-06-20 10:23:00 +0000 UTC
J. Exp. Bot.-2013-Zheng-3747-61 Supp.pdf Size: 360953
ApsimBot
commented
5 years ago Author: fai04d Date: 2014-06-20 10:23:00 +0000 UTC
J. Exp. Bot.-2013-Zheng-3747-61.pdf Size: 2458768
ApsimBot
commented
5 years ago Author: murphych Date: 2014-08-29 09:44:00 +0000 UTC
Approved by RP at 8 August 2014 meeting to proceed with Bangyou Zheng as the key contact.
Legacy Bug ID: 1904 Author: Justin Bangyou Zheng Date: 2014-06-20 10:21:39 +0000 UTC
Submission on behalf of Dr Bangyou Zheng (CSIRO Plant Industry):
A gene-based phenology model has been developed to improve the performance of APSIM-Wheat to predict heading or flowering times of wheat across Australian wheatbelt by Bangyou Zheng and Scott Chapman. A field experiment at South Perth was used to estimate the gene-effects of Vrn-A1, Vrn-B1, Vrn-D1 and Ppd-D1, and then the generic parameters of vernalisation and photoperiod sensitivities in APSIM-Wheat for 210 cultivars or breeding lines. In the new model, we only need one genotypic parameter to predict wheat heading or flowering time (target thermal time from floral initial to flowering).
We collected 4475 observed heading or flowering times from different sources to validate the gene-based model. These datasets covered 172 lines at 72 sites across Australian wheatbelt with sowing times between 30 March and 11 September from 2005 to 2011. The results indicated the high correlation between observed and simulated heading times for validation datasets using the parameters derived only from the South Perth trial (RMSE = 4.3 d). Meanwhile we refitted the parameters in the original APSIM with our calibration and validation datasets. The RMSE was reduced from 7.2 d with the default parameters in APSIM to 3.9 d after re-fitting. We have published the parameters of 141 commercial cultivars as the supporting information of our paper (Zheng et al. 2013 Journal of Experimental Botany).
The gene-based model and the refitted parameters will be of benefit to APSIM communities through providing more accuracies of flowering time predictions.