In COP21 followed by the Paris Agreement, the world is now seriously planning actions to mitigate greenhouse gas emissions toward a “below 2 °C above preindustrial levels” future. Currently, we are still far from identifying the emission pathways to achieve this target because of the various uncertainties in both climate science and the human behavior. As a part of the ICA-RUS project, conducted by Dr. Seita Emori of the National Institute for Environmental Studies we have studied how these uncertainties are eliminated by the accumulation of scientific knowledge and the decision-making processes. We consider the following three questions: first, when and how will the uncertainty range on the global temperature rise be eliminated, second which global emission pathway should be chosen before we get the perfect information, and third how much expenditure is justified in reducing the climate uncertainties. The first question has been investigated by one of the authors. Shiogama et al. (Sci Rep 6:18903, 2016) developed the Allen–Stott–Kettleborough (ASK) method further to estimate how quickly and in what way the uncertainties in future global mean temperature changes can decline when the current observation network of surface air temperature is maintained. Fourteen global climate model results in CMIP5 (CMIP http://cmip-pcmdi.llnl.gov/, 2017) are used as virtual observations of surface air temperature. The purpose of this study is to answer the remaining two questions. Based on the ASK research outcomes, we apply the multi stage decision-making known as Act Then Learn (ATL) process to an integrated assessment model MARIA which includes energy technologies, economic activities, land use changes and a simple climate model block. We reveal how accumulating observations helps to mitigate economic losses by expanding the existing ATL method to deal with the uncertainty eliminating process by ASK. The primary findings are as follows. First, the value of information largely increases as the climate target policy is more stringent. Second, even if the uncertainties in the equilibrium climate sensitivity are not fully resolved, scientific knowledge is still valuable. In other words, the expenditure for scientific researches is rationalized when we really concern the global climate changes.

中文翻译：

---

知识积累对气候敏感性不确定性的价值：完美信息、单阶段与行动学习决策之间的比较。

在 COP21 和《巴黎协定》之后，世界现在正在认真规划采取行动，以减少温室气体排放，以实现“比工业化前水平低 2°C”的未来。目前，由于气候科学和人类行为的各种不确定性，我们距离确定实现这一目标的排放途径还很遥远。作为国家环境研究所 Seita Emori 博士主持的 ICA-RUS 项目的一部分，我们研究了如何通过科学知识的积累和决策过程来消除这些不确定性。我们考虑以下三个问题：第一，全球气温上升的不确定性范围何时以及如何消除；第二，在获得完美信息之前，应该选择哪种全球排放路径；第三，减少气候变化支出多少才是合理的？不确定性。第一个问题已由一位作者进行了调查。盐灶等人。(Sci Rep 6:18903, 2016) 进一步开发了 Allen-Stott-Kettleborough (ASK) 方法，以估计在当前的地表气温观测网络维持不变的情况下，未来全球平均温度变化的不确定性可以以多快的速度和以何种方式下降。CMIP5 中的 14 个全球气候模型结果（CMIP http://cmip-pcmdi.llnl.gov/, 2017）被用作地表空气温度的虚拟观测。本研究的目的是回答剩下的两个问题。基于 ASK 研究成果，我们将被称为“行动然后学习”(ATL) 流程的多阶段决策应用于综合评估模型 MARIA，其中包括能源技术、经济活动、土地利用变化和简单的气候模型块。我们揭示了如何通过扩展现有的 ATL 方法来处理 ASK 的不确定性消除过程，积累观测值有助于减轻经济损失。主要调查结果如下。首先，随着气候目标政策的更加严格，信息的价值大幅增加。其次，即使平衡气候敏感性的不确定性没有完全解决，科学知识仍然有价值。换句话说，当我们真正关注全球气候变化时，科研支出就合理化了。