Dominique Ripoche

An overview of the crop model stics

Abstract stics is a model that has been developed at INRA (France) since 1996. It simulates crop growth as well as soil water and nitrogen balances driven by daily climatic data. It calculates both agricultural variables (yield, input consumption) and environmental variables (water and nitrogen losses). From a conceptual point of view, stics relies essentially on well-known relationships or on simplifications of existing models. One of the key elements of stics is its adaptability to various crops. This is achieved by the use of generic parameters relevant for most crops and on options in the model formalisations concerning both physiology and management, that have to be chosen for each crop. All the users of the model form a group that participates in making the model and the software evolve, because stics is not a fixed model but rather an interactive modelling platform. This article presents version 5.0 by giving details on the model formalisations concerning shoot ecophysiology, soil functioning in interaction with roots, and relationships between crop management and the soil–crop system. The data required to run the model relate to climate, soil (water and nitrogen initial profiles and permanent soil features) and crop management. The species and varietal parameters are provided by the specialists of each species. The data required to validate the model relate to the agronomic or environmental outputs at the end of the cropping season. Some examples of validation and application are given, demonstrating the generality of the stics model and its ability to adapt to a wide range of agro-environmental issues. Finally, the conceptual limits of the model are discussed.

Multimodel ensembles of wheat growth: many models are better than one.

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.

The uncertainty of crop yield projections is reduced by improved temperature response functions

Increasing the accuracy of crop productivity estimates is a key element in planning adaptation strategies to ensure global food security under climate change. Process-based crop models are effective means to project climate impact on crop yield, but have large uncertainty in yield simulations. Here, we show that variations in the mathematical functions currently used to simulate temperature responses of physiological processes in 29 wheat models account for >50% of uncertainty in simulated grain yields for mean growing season temperatures from 14 °C to 33 °C. We derived a set of new temperature response functions that when substituted in four wheat models reduced the error in grain yield simulations across seven global sites with different temperature regimes by 19% to 50% (42% average). We anticipate the improved temperature responses to be a key step to improve modelling of crops under rising temperature and climate change, leading to higher skill of crop yield projections.

Harmonization and translation of crop modeling data to ensure interoperability

The Agricultural Model Intercomparison and Improvement Project (AgMIP) seeks to improve the capability of ecophysiological and economic models to describe the potential impacts of climate change on agricultural systems. AgMIP protocols emphasize the use of multiple models; consequently, data harmonization is essential. This interoperability was achieved by establishing a data exchange mechanism with variables defined in accordance with international standards; implementing a flexibly structured data schema to store experimental data; and designing a method to fill gaps in model-required input data. Researchers and modelers are able to use these tools to run an ensemble of?models on a single, harmonized dataset. This allows them to compare models directly, leading ultimately to model improvements. An important outcome is the development of a platform that facilitates researcher collaboration from many organizations, across many countries. This would have been very difficult to achieve without the AgMIP data interoperability standards described in this paper. Heterogeneous data can be harmonized and translated to multiple model formats.The ICASA data standards provide an extensible data structure and ontology.JSON structures provide a flexible, efficient means of handling heterogeneous data.DOME functions enable a consistent means of providing missing or inadequate data.Data provenance is maintained from data sources through simulated model outputs.

A crop model for land suitability evaluation a case study of the maize crop in France

Abstract A crop model to evaluate land suitability is described. It has been devised to study spatial variation and uses readily available input data. The case study described is for the maize crop and uses a simple growth model for this crop. The model is incorporated within procedures that allow the descrip tion of crop environment variability both in space and time and the model is run under a Geographical Information System. Input data are stored in soil, climate and crop management data bases, for 20 × 20 km areas and constitute the basic information for crop growth simulation. From the network of synoptic meteorological stations, climatic variables are spatially interpolated to give predicted values for each elementary area. The model computes every ten days : i) potential crop productivity in the absence of any stress ; ii) productivity in limited-water situation. The modelling principles for the soilplant-atmosphere system are simple : development depends on thermal time, growth depends on energy use efficiency and the calculated water balance uses a reservoir model. Because of the ten-day time step, particular attention was given to the way in which water stress affects the growth-development functions. A study proved the model to be reliable for estimating maize productivity in various locations although some discrepancies between measurements and simulations can occur for intermediate variables in extreme environmental conditions. As illustrations of the model performance, map outputs of land suitabilities over France for maize growing are presented.

Pesticide fate modeling in soils with the crop model STICS: Feasibility for assessment of agricultural practices.

Numerous pesticide fate models are available, but few of them are able to take into account specific agricultural practices, such as catch crop, mixing crops or tillage in their predictions. In order to better integrate crop management and crop growth in the simulation of diffuse agricultural pollutions, and to manage both pesticide and nitrogen pollution, a pesticide fate module was implemented in the crop model STICS. The objectives of the study were: (i) to implement a pesticide fate module in the crop model STICS; (ii) to evaluate the model performance using experimental data from three sites with different pedoclimatic contexts, one in The Netherlands and two in northern France; (iii) to compare the simulations with several pesticide fate models; and (iv) to test the impact of specific agricultural practices on the transfer of the dissolved fraction of pesticides. The evaluations were carried out with three herbicides: bentazone, isoproturon, and atrazine. The strategy applied in this study relies on a noncalibration approach and sensitivity test to assess the operating limits of the model. To this end, the evaluation was performed with default values found in the literature and completed by sensitivity tests. The extended version of the STICS named STICS-Pest, shows similar results with other pesticide fate models widely used in the literature. Moreover, STICS-Pest was able to estimate realistic crop growth and catch crop dynamic, which thus illustrate agricultural practices leading to a reduction of nitrate and a change in pesticide leaching. The dynamic plot-scale model, STICS-Pest is able to simulate nitrogen and pesticide fluxes, when the hydrologic context is in the validity range of the reservoir (or capacity) model. According to these initial results, the model may be a relevant tool for studying the effect of long-term agricultural practices on pesticide residue dynamics in soil and the associated diffuse pollution transfer.

Multimodel ensembles improve predictions of crop–environment–management interactions

A recent innovation in assessment of climate change impact on agricultural production has been to use crop multimodel ensembles (MMEs). These studies usually find large variability between individual models but that the ensemble mean (e-mean) and median (e-median) often seem to predict quite well. However, few studies have specifically been concerned with the predictive quality of those ensemble predictors. We ask what is the predictive quality of e-mean and e-median, and how does that depend on the ensemble characteristics. Our empirical results are based on five MME studies applied to wheat, using different data sets but the same 25 crop models. We show that the ensemble predictors have quite high skill and are better than most and sometimes all individual models for most groups of environments and most response variables. Mean squared error of e-mean decreases monotonically with the size of the ensemble if models are added at random, but has a minimum at usually 2-6 models if best-fit models are added first. Our theoretical results describe the ensemble using four parameters: average bias, model effect variance, environment effect variance, and interaction variance. We show analytically that mean squared error of prediction (MSEP) of e-mean will always be smaller than MSEP averaged over models and will be less than MSEP of the best model if squared bias is less than the interaction variance. If models are added to the ensemble at random, MSEP of e-mean will decrease as the inverse of ensemble size, with a minimum equal to squared bias plus interaction variance. This minimum value is not necessarily small, and so it is important to evaluate the predictive quality of e-mean for each target population of environments. These results provide new information on the advantages of ensemble predictors, but also show their limitations.

Author Correction: The uncertainty of crop yield projections is reduced by improved temperature response functions

Nature Plants3, 17102 (2017); published online 17 July 2017; corrected online 27 September 2017.