https://digitalcommons.unl.edu/cgi/viewcontent.cgi?article=1449&context=natrespapers

The Agricultural Model Intercomparison and Improvement Project (AgMIP): Protocols and Pilot Studies

Adaptation of crop management to water-limited environments

The complexity of risks posed by climate change and possible adaptations for crop production has called for integrated assessment and modelling (IAM) approaches linking biophysical and economic models. This paper attempts to provide an overview of the present state of crop modelling to assess climate change risks to food production and to which extent crop models comply with IAM demands. Considerable progress has been made in modelling effects of climate variables, where crop models best satisfy IAM demands. Demands are partly satisfied for simulating commonly required assessment variables. However, progress on the number of simulated crops, uncertainty propagation related to model parameters and structure, adaptations and scaling are less advanced and lagging behind IAM demands. The limitations are considered substantial and apply to a different extent to all crop models. Overcoming these limitations will require joint efforts, and consideration of novel modelling approaches. Extreme events and future climate uncertainty represent risk for food production.Crop models are largely able to simulate crop response to climate factors.Adaptations are best evaluated in integrated assessment models (IAM).Key limitations for crop models in IAM are low data availability and integration.Cross-scale nature of IAM suggests novel modelling approaches are needed.

/pdf/crop-modelling-for-integrated-assessment-of-risk-to-food-4webrvi34c.pdf

Crop modelling for integrated assessment of risk to food production from climate change

Combining the simulation crop model AquaCrop with an economic model for the optimization of irrigation management at farm level

The procedure for applying phosphorus (P) fertilizer to soil can be divided into three consecutive steps: (i) Measurement of soil-P availability, (ii) calibration of the soil-P fertility level and (iii) estimation of the recommended P dose. Information on each of these steps was obtained for 18 European countries and regions with the aim of comparing P fertilizer recommendation systems at the European scale. We collected information on P fertilizer recommendations through conventional or grey literature, and personal contacts with researchers, laboratories and advisory services. We found much variation between countries for each of the three steps: There are more than 10 soil-P tests currently in use, apparent contradictions in the interpretation of soil-P test values and more than 3-fold differences in the P fertilizer recommendations for similar soil-crop situations. This last result was confirmed by conducting a simple experimental inter-laboratory comparison. Moreover, soil properties (pH, clay content) and crop species characteristics (P responsiveness) are used in some countries in the calibration and recommendation steps, but in different ways. However, there are also common characteristics: soil-P availability is determined in all countries by extraction with chemical reagents and the calibration of the soil-P test values, and the fertilizer recommendations are based on the results from empirical field trials. Moreover, the fertilizer recommendations are nearly all based on the amount of P exported in the crops. As long as rational scientific and theoretical backgrounds are lacking, there is no point in trying to synchronize the different chemical methods used. We therefore call for a mechanistic approach in which the processes involved in plant P nutrition are truly reproduced by a single standard method or simulated by sorption-desorption models.

An overview of fertilizer-P recommendations in Europe: soil testing, calibration and fertilizer recommendations

MODERATO: an object-oriented decision tool for designing maize irrigation schedules

Improving irrigation schedules by using a biophysical and a decisional model

Published in Agron. J. 104:1274–1283 (2012) Posted online 22 June 2012 doi:10.2134/agronj2012.0038 Copyright © 2012 by the American Society of Agronomy, 5585 Guilford Road, Madison, WI 53711. All rights reserved. No part of this periodical may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. A important objective in crop modeling is to use models to evaluate and compare management strategies (Boote et al., 1996, 2010; Stephens and Middleton, 2002; Bergez et al., 2010). Experimental comparisons are, of course, essential but are necessarily limited to a relatively small number of locations, years, and management practices. Models make it possible to generalize the experimental results to new conditions and also to compare strategies other than those tested experimentally. The specific aspect of crop management of interest in this study was irrigation management. Several studies have used models to explore the consequences of different irrigation strategies (Baumhardt et al., 2009; Garcia-Vila et al., 2009). Bergez et al. (2001) proposed a simulation tool specifically adapted for evaluating irrigation strategies. The power of crop models as a generalization tool comes, however, with an important drawback: the model is only an approximation of reality and so there is uncertainty in the results calculated using a model. An essential question then is the extent of this uncertainty and, as a result, the usefulness of a model for determining management strategy. Evett and Tolk (2009) specifically posed the question of whether crop models are sufficiently accurate to aid in deciding irrigation practices. The common procedure for dealing with model error is to first do a general evaluation of the model, which involves comparing the model calculations with experimental data. If the model is deemed sufficiently accurate, it is then used to evaluate various management strategies (for example, Garcia-Vila et al., 2009). Common criteria for comparing calculated and observed values are the mean square error (MSE) and modeling efficiency (EF). This approach is not fully satisfactory. First of all, it furnishes an estimate of the quality of the model averaged across the population represented by the data. When we use a model to evaluate strategies, however, we want the quality of predictions for those specific strategies. Second, it would be more informative to have a full probability distribution of uncertainty for each calculated quantity rather than an average error. Third, we may be interested in various criteria for evaluating a strategy other than the criteria related to the quality of prediction year by year. For example, an important criterion for judging irrigation strategies could be yield averaged across years, so we would want to evaluate the uncertainty in predicting that criterion. This suggests that it would be useful to do a detailed uncertainty analysis. In such a study, the sources of uncertainty are identified and quantified, and then how those uncertainties translate into uncertainties in the predictions of interest is evaluated (de Rocquigny et al., 2008). The result is an indicator of uncertainty (for example, a probability distribution) for each specific prediction. The purpose of this study was to develop a methodology for applying uncertainty analysis to a comparison of irrigation strategies for corn. We illustrate using a crop model coupled with algorithms that translate irrigation strategies into ABStrAct

Bernard Lacroix

Papers

MODERATO: an object-oriented decision tool for designing maize irrigation schedules

Improving irrigation schedules by using a biophysical and a decisional model

Assessing the Uncertainty when Using a Model to Compare Irrigation Strategies

A conceptual model of farmers' decision-making process for nitrogen fertilization and irrigation of durum wheat