Dear colleagues:

I write to you from Bolivia, I'm developing a climate change impact on crop yields assessment, for this, I'm analyzing different zones of my country, which have different climate, soil and other characteristics. I´m kind of new on the DSSAT software, but I´ve already created the meteorological stations, soil files and experiment files. To validate my models I used the validation procedure mentioned by K. J. Boote (2007) on one of his conferences at the University of Florida, to make a summary it consists on the following:

 Step 1: Calibrate the crop development using real climate and comparing the phenological results with the observed ones: flowering, maturity.
 Step 2: Calibrate the accumulation of dry matter, by comparing the observed and simulated results: biomass, LAI (leaf area index)
 Step 3: Calibrate the yields and its components

I've followed all of those steps. I´ve chosen 7 varieties of the 52 available on CERES- Rice of the DSSAT v4., this choice was based on the phenological stages, the simulated should be the same as the observed on field experiments.

My results were great following the phenological stages, but they overpredicted the yield of rice at different sites.

The question is the following: Should I make this kind of choice and later reduce the yield? Or should my choice of the variety should be based on the comparison of yields?

If I have to reduce the yield, which parameters do you suggest to vary?

Well, that's all for now, I hope you can help me with this issue which is really important for our work.

Thanks in advance,

Sincerely yours,

Ernesto Viscarra
Environmental Economist