Dear Bhusal,
Yield as you know is an integrated result of crop growth and development. There's a large list of potential causes for yield inaccuracies. Also there's a number of ways to "fix" your yield simulations, but producing results that have little to do with the system you are trying to represent. Of course, you don't want to do that and sacrificing a good learning experience. Yield underestimation may be the result of two sources of error: 1) Errors in your inputs, 2) Errors in the model.
You need to check first your inputs: weather (you may have some missing rains, the radiation sensor may have produced some unnoticed numbers), soil as Matthias suggested (DUL, LL, SAT), genetic coefficients (is phenology being simulated correctly?), X file (fertilizer amount and date, Irrigation, planting date?). Sometimes, searching for a clue you may want to look at specific things. Is LAI being simulated correctly? Biomass? Kernel number? Water and N stress look reasonably?
Once you are confident with the accuracy of your inputs you look into the model. The current model may not be accurate enough to simulate your particular environment. Your local cultivars may be adapted to your conditions and produce different responses than other cultivars. Many model parameters are now external in species and ecotype files.
However, I would recommend to focus on your inputs and make sure there's nothing wrong there. Please, let me know if I can be of further assistance, and good luck.
Jon
Jon I. Lizaso
Dpto. Producción Vegetal. Fitotecnia
ETS Ingenieros Agrónomos
Senda del Rey s/n
28040 Madrid
Tel: 34 91 549 1122
[log in to unmask]
________________________________________
From: DSSAT - Crop Models and Applications [[log in to unmask]] On Behalf Of T.N. Bhusal [[log in to unmask]]
Sent: Monday, June 09, 2008 10:38 PM
To: [log in to unmask]
Subject: underestimation of yield
Dear all
I am working with ceres maize model. Now i am encounter with a problem. The
simulated yield is underestimated than observed. what would be the problem?
I have make all possible solution regarding changing in genetic coefficients
but still it is low estimation of yied. What would be the other possible
causes/
TN Bhusal
IAAS, Rampur, Nepal
|
