气候变化综合评估模型的损失函数研究进展

Progress in damage function of integrated assessment models for climate change

该研究从综合评估模型(IAM)的模型耦合视角出发,介绍了当前损失函数的研究进展,主要从损失函数的构建方法、损失函数与IAM气候模块和经济模块的耦合以及IAM与气候模式的耦合角度分析了损失函数的耦合功能及其存在的科学问题,探讨了损失函数的改进方向。通过文献梳理发现,损失函数的构建方法上,主要采用专家判断法、元分析法和统计学方法,但各有优缺点;与气候模式的耦合功能上,损失函数多以温升为气候变化因素,降水等气候变化信息无法表达,且由全球尺度的年平均值进行标定,不能体现区域的差异和季节的变化,无法直接描述极端气候事件造成的巨大损失;与经济模块的耦合功能上,基于生产部门的损失函数缺乏间接损失评估功能,缺乏对经济增长的动态影响机制。针对上述IAM中气候变化对经济影响的反馈机制的不足,需重点从细化区域气候变化因素影响和细分经济产业部门两个方向重构损失函数,紧密连接气候模式与经济模块,全面评估气候变化经济损失,并需要从技术上解决损失函数在耦合经济模块与气候模式时出现的时空尺度不匹配问题,最终为IAM与气候模式甚至地球系统模式的耦合提供重要的解决方案。

From the perspective of model coupling, this paper introduces the research progress of the current damage function of integrated assessment model （IAM）. It mainly analyzes the coupling function and existing scientifc problems of the damage function from the point of view of the construction method of the damage function, the coupling of the damage function with the climate model and the economic model, and the coupling IAM with climate model, then discusses its improvement direction. On the construction method of the damage function, it mainly adopts the expert judgement, meta-analysis and statistical methods. However, each of them has both advantages and disadvantages. While coupling with the climate model, most of damage functions only contain the temperature rise as the climate change factor, which can’t express the precipitation and other climate change information. And it is calibrated by the annual average temperature at the global scale, which can’t refect the regional differences and the seasonal changes and can’t directly describe the huge damage caused by extreme climate events. On the coupling with the economic model, the damage function based on the production sectors are lacking in the assessment function of indirect damages and the dynamic infuence mechanism of economic growth. In view of above shortcomings of feedback mechanism of the economic impacts of climate change in IAM, the damage function mainly needs to be reconstructed from two directions, namely, the refnement of regional climate change factors and the subdivision of the economic sectors. Thus, it can closely link the climate module with economic module and comprehensively assess the economic damages from climate change. In addition, it is necessary to solve the problem of spatial and temporal scale mismatching in the coupling economic model and climate model to fnally provide an important solution for the coupling of IAM model with the climate model and even the earth system model.