A STRUCTURED PROCEDURE FOR ASSESSING HOW CROP MODELS RESPOND TO TEMPERATURE

Authors: White, Jeffrey; HOOGENBOOM, G - UNIVERSITY OF GEORGIA; HUNT, L - UNIV OF GUELPH CANADA

Submitted to: Agronomy Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/4/2004
Publication Date: N/A
Citation: N/A

Interpretive Summary: Both extreme high and low temperatures can dramatically reduce crop yields. Unfortunately, temperature responses of crops are often very complex. Crop simulation models are widely used to analyze how temperature affects crop performance, but it often is difficult to judge whether a model is appropriate for a specific research problem. This study describes a procedure for assessing how models respond to temperature. The procedure covers major processes in a balanced fashion, does not require access to software source code, is easily interpretable by non-modelers, and is readily documented and transferable to different models. The approach involves running models for conditions of non-limiting water and nitrogen and constant regimes of daily temperatures for a range of mean temperatures from 3° to 40°C. Daily model outputs such as dry weights and developmental stages define model responses grouped in seven categories: overall growth, phenology, reproductive growth, canopy growth, root growth, resource use efficiency, and water balance. In comparisons of models for sorghum and common bean, the procedure served both to highlight expected differences in overall temperature adaptation of the two crops, but also identified aspects of the models that require further testing or research. The evaluation process proved robust and easily applied to different models. Use of this procedure should lead to better matching of models to research problems, as well as help identify aspects of models that require improvement. The net result for growers, policy makers and other stake holders should be more reliable predictions of how crops respond to management or to external factors such as changes in climatic conditions.

Technical Abstract: Crop simulation models are widely used to analyze temperature effects on crop development and yield. Unfortunately, temperature responses of models seldom are examined critically to ensure that a model is appropriate for a research question. The objective of this study was to develop a procedure for assessing how models respond to temperature that would cover major processes in a balanced fashion, not require access to source code, be easily interpretable by non-modelers, and be readily documented and employed with different models. The approach involves running a sensitivity analysis for conditions of non-limiting water and nitrogen and constant regimes of daily temperatures for a range of mean temperatures from 3° to 40°C. Daily model outputs such as dry weights and developmental stages define model responses grouped in seven categories: overall growth, phenology, reproductive growth, canopy growth, root growth, resource use efficiency, and water balance. To avoid confounding effects of delayed flowering and maturation at low temperatures, several responses are assessed at a reference date chosen prior to when partitioning to reproductive growth affects crop growth. Emphasis is on graphical analysis of individual variables vs. mean temperature, but cardinal temperatures and response stability indices are also estimated. When applied to the CSM-CERES-Sorghum and CSM-CROPGRO-Drybean models, the procedure readily identified differences in assumed temperature adaptation of the two crops, but the procedure also revealed problems in both models that merit further examination. The proposed procedure would require adjustments for specific crops or research questions, but it offers a foundation for assessing modeled responses to temperature in a structured and reproducible fashion.