Observationally-constrained climate projections following the Paris Agreement

最近生效的《巴黎协定》要求签署国将全球平均气温升幅控制在远低于工业化前2摄氏度的水平，并努力将气温升幅控制在比工业化前水平高1.5摄氏度的范围内。由于二氧化碳是主要的温室气体，而且无论一次性排放还是长期排放，一定总量的二氧化碳排放都会造成相同程度的变暖，因此1.5摄氏度的变暖阈值可以与特定的二氧化碳排放总量或碳预算相关联。了解在很有可能将全球平均气温限制在1.5摄氏度和2.0摄氏度以下的情况下可以排放的最大二氧化碳量，是规划气候减缓战略和履行《巴黎协定》承诺的关键。联合国政府间气候变化专门委员会（IPCC）在上一份报告中提出的碳预算直接基于所有可用气候模型的模拟排放，并指出，以目前的排放水平，在不到10年的时间里，将超过1.5摄氏度预算的最佳估计，而且有可能已经超过了这个预算。然而，我们知道，历史上的二氧化碳排放，以及其他气候驱动因素，迄今为止只导致了大约1摄氏度的全球变暖。此外，我们知道，IPCC上一份报告中考虑的气候模型平均比观测值更暖。我们能否利用对过去气候变化的观测来更好地预测未来的气候变化？已经提出了许多使用观测来约束预估的方法，包括将模拟的气候变化放大或缩小，以最佳地匹配观测到的变化，根据模型与观测的一致程度对模型进行加权，或者使用观测来筛选气候模型。该项目将探索一系列观测受限气候预测的方法，重点是获得更好的观测受限碳预算和全球平均温度预测。还将考虑观测约束对其他变量的应用，如区域温度变化、降雨变化和海冰变化。

The Paris Agreement, which recently came into force, commits signatories to holding the increase in global average temperature to well below 2C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5C above pre-industrial levels'. Since carbon dioxide is the dominant greenhouse gas, and since a certain total amount of carbon dioxide emissions gives the same amount of warming irrespective of whether it is emitted all at once or over a long period, a warming threshold of 1.5C can be associated with a particular total amount of carbon dioxide emissions, or carbon budget. Knowing the maximum amount of carbon dioxide which can be emitted while having a good chance of limiting global mean temperature to below the 1.5C and 2.0C thresholds is key to planning climate mitigation strategies and fulfilling the commitments made in the Paris Agreement. Carbon budgets presented in the last report of the United Nations Intergovernmental Panel on Climate Change (IPCC) were based directly on simulated emissions in all available climate models, and indicated that the best estimate of the 1.5C budget would be exceeded in less than ten years at current emissions levels, with a chance that this budget has already been exceeded. However, we know that historical carbon dioxide emissions, in conjunction with other climate drivers, have led to only about 1C global warming to date. Further, we know that on average the climate models considered in the last IPCC report on average warm more than observations. Can we use observations of past climate change to obtain better projections of future climate change? A number of ways of using observations to constrain projections have been proposed, including scaling simulated climate change up or down to best match observed changes, weighting models based on how well they agree with observations, or using observations to screen climate models in or out. This project will explore a range of approaches to observationally-constraining climate projections, with a focus on obtaining improved observationally-constrained carbon budgets and projections of global mean temperature. The application of observational constraints to other variables, such as regional temperature changes, rainfall changes and sea ice changes, will also be considered.