If you have been responsible for calibration of a crop model, we are asking you to take the time to fill out this survey.
Calibration, or parameter estimation, is a difficult but critical aspect of a crop modeling project: Predictions from a model are heavily dependent on the parameter values used in the model. Despite its importance little attention has been paid to the approaches to calibration used by different crop modelers. Progress towards improving methods of calibration and setting out guidelines of good practices could make a real difference to the usefulness of crop models. That is the goal of the calibration activity of the AgMIP/MACSUR projects.
This survey is the first step of the calibration activity. It aims at establishing a baseline; recording the current practices in crop model calibration across the crop modeling community. Following the survey we will collate the results in a report on current methods, which will be shared with all respondents, with a view of possibly extending it to a scientific paper. In future steps of the activity we will compare various methods of calibration, accompanied by discussion as to the ways to evaluate them. These activities will be open to all who want to participate (respondents to this survey will be kept informed), in order to tap into the wealth of experience accumulated by crop model developers and users around the world.
If you have been responsible for a crop model calibration activity, please take the time (about 20 minutes) to respond to this survey. As thanks, respondents will receive a complimentary electronic copy of the chapter “Parameter Estimation with Classical methods (Model Calibration)” of the book Working with Dynamic Crop Models, 2nd edition.
The survey is open until midnight November 30, 2016. If you have any questions, please contact Sabine Seidel (AgMIPcalibrationsurvey@uni-bonn.de).
Thank you for your participation, the crop model calibration project co-leaders:
- Daniel Wallach (INRA, France)
- Taru Palosuo (Luke, Finland)
- Sabine Seidel (INRES, University of Bonn, Germany)
- Peter Thorburn (CSIRO, Australia)
P.S. If you have been responsible for multiple calibration studies, please answer this survey for just one specific study (one data set, one model, one set of calibrated parameters), the one that you feel could be most useful to others.