Abstract: Crop models are increasingly being used to assess the impacts of future climate change on production and food security. High quality, site-specific data on weather, soils, management, and cultivar are needed for those model applications. Also important is that model development, evaluation, improvement, and calibration require additional high quality, site-specific measurements on crop yield, growth, phenology, and ancillary traits. We review the evolution of minimum data set requirements for agroecosystem modeling and then describe the characteristics and ranking of sentinel site data needed for crop model improvement, calibration, and application. We in the Agricultural Model Intercomparison and Improvement Project (AgMIP), propose to rank sentinel site data sets as platinum, gold, silver, and copper, based on the degree of true site-specific measurement of weather, soils, management, crop yield, as well as the quality, comprehensiveness, quantity, accuracy, and value. For example, to be ranked platinum, the weather and soil characterization must be measured on-site, and all management inputs must be known. Dataset ranking will be lower for weather measured off-site or soil traits estimated from soil mapping. Ranking also depends on the intended purposes for data use. If the purpose is to improve a crop model for response to water or N, then additional observations are necessary, such as initial soil water, initial soil inorganic N, and plant N uptake during the growing season to be ranked platinum. Rankings are enhanced by presence of multiple treatments and sites. Examples of platinum-, gold-, and silver-quality data sets for model improvement and calibration uses are illustrated.

Abstract: Crop models of crop growth are increasingly used to quantify the impact of global changes due to climate or crop management. Therefore, accuracy of simulation results is a major concern. Studies with ensembles of crop models can give valuable information about model accuracy and uncertainty, but such studies are difficult to organize and have only recently begun. We report on the largest ensemble study to date, of 27 wheat models tested in four contrasting locations for their accuracy in simulating multiple crop growth and yield variables. The relative error averaged over models was 24-38% for the different end-of-season variables including grain yield (GY) and grain protein concentration (GPC). There was little relation between error of a model for GY or GPC and error for in-season variables. Thus, most models did not arrive at accurate simulations of GY and GPC by accurately simulating preceding growth dynamics. Ensemble simulations, taking either the mean (e-mean) or median (e-median) of simulated values, gave better estimates than any individual model when all variables were considered. Compared to individual models, e-median ranked first in simulating measured GY and third in GPC. The error of e-mean and e-median declined with an increasing number of ensemble members, with little decrease beyond 10 models. We conclude that multimodel ensembles can be used to create new estimators with improved accuracy and consistency in simulating growth dynamics. We argue that these results are applicable to other crop species, and hypothesize that they apply more generally to ecological system models.