气候变化对粮食产量影响的研究方法综述

IMPACTS OF CLIMATE CHANGE ON GRAIN YIELD: A REVIEW OF RESEARCH METHODS

关键词:  气候变化粮食产量农业生产研究方法

Key words:  climate change  grain  yield  agricultural production  research method

摘要:

［目的］受气候变化的影响，全球粮食安全面临严峻挑战，及时准确地评估气候变化对粮食产量的影响是应对挑战、制定农业适应性对策的关键。相关研究已产生了不少方法，通过综述对方法进行分类，明晰各种方法的优缺点和适用性，以期扬长避短，促进研究方法的综合、发展与完善。［方法］利用文献法、归纳法和比较法，从方法的原理和应用、存在的问题、发展的趋势3个方面进行探讨。［结果］产量分解法可用于分析粮食产量及其构成要素与不同生育期气候变化的关系，实验比较法一般用于粮食产量对单个气候因子或若干气候因子变化的敏感性分析，生产函数法适用于在农业生产系统中分析气候变化对粮食产量影响的边际效应，气候生产潜力模型法侧重于农业生产环境发展评估，作物生长模型法便于结合气候情景预测未来气候变化对粮食产量的影响。在不同研究方向上得以运用的同时，各方法也暴露了一些问题：产量分解法的技术产量难以拟合，实验比较法的数据获取难、模型稳定性较差，生产函数法容易遗漏重要变量、函数构造困难，气候生产潜力模型法的结论难以验证，作物生长模型法参数标定难、模型应用存在尺度错位。［结论］研究方法将逐渐形成一套综合的气候—水文—作物—经济模型法，多源数据融合和多目标模式已经成为方法发展的驱动力。

Abstract:

Global food security faces serious challenges of climate change. Accurately assessing the impacts of climate change on grain yield is the key to the formulation of agricultural adaptability. Many research methods have been used in practice, so the advantages, disadvantages and applicability of these methods need to be clarified in order to promote their development, integration and improvement. Using the methods of literature, induction and comparative law, this article classified the related research methods into five types and summarized their applicability. The result shows that the yield division methods can be used to analyze the relationships between grain yield and climatic factors at different growth stages; the experimental methods are generally used for the sensitivity analysis of grain yield to climatic factors; the production function methods are suitable for analyzing the marginal effect of climate change on food production; climatic productivity models focus on the assessment of changes in agricultural production environment, while crop growth models can predict the impact of climate change on food production by incorporating climate scenarios. The defects of these methods include the difficulties of simulating technical yields, weak anti interference ability, omitted variables, overly complex functions, non verifiability of conclusion, problems in calibrating parameters, scale effect of models, etc. With the development of research on mechanism, the forecasting models of grain yield will experience a transition from the crop models to the climate hydrology crop economic models. In addition, multi source data fusion and multi objective models have provided important clues to the innovation of research methods.