Improving Projections of Regional Climate Change for the United States Using Detection and Attribution Studies

政府间气候变化专门委员会（IPCC）第三次评估综合报告得出结论，区域气候变化的关键不确定性之一是“将区域趋势与人为气候变化联系起来”。最近的研究表明，在大陆尺度的地表温度变化中可以探测到人为的气候变化信号。本项目将利用气候变化的探测和归因方法来估计美国在区域尺度上观测到的可归因于人为强迫和其他气候强迫因子的温度变化的比例。这些方法将适用于年和季节平均值的平均、最低和最高温度，以及极端温度。过去100年温度变化的最优指纹分析将用于约束许多不同气候模式对未来全球和区域温度变化的估计，并发展区域温度变化的概率分布。通过与英国climateprediction.net小组的合作，他们的瞬态气候变化模拟的超大扰动参数化集合将用于更好地估计美国当前和未来区域气候变化预估的不确定性。这项研究的许多方面都具有创新性。气候变化的区域归因和人为气候变化模拟信号的标度以前没有进行过研究，但它们是最近研究的延伸。将在美国地区应用一种新的方法，通过去除与降雨相关的温度变化来增强相对于自然变率的人为温度变化信号，初步结果表明，它可能会提高信噪比。将climateprediction.net集合模拟应用于区域气候变化预估是他们早期工作的延伸。更广泛的影响：该项目通过量化不同因素对美国观测到的气候变化的贡献和估计美国未来气候变化预测的概率分布，直接解决了美国气候变化科学计划（CCSP）战略计划的五个目标中的两个。这项研究的预期结果可能会在国际和美国产生重大影响。量化温室强迫对区域气候变化的贡献将有助于决策者就温室气体减排或适应温室气候变化的政策作出决定。不同区域气候变化的概率分布对于更好地量化区域气候变化的可能影响和估计区域气候变化的成本或收益非常有用。这些区域气候变化的概率分布可以用作经济模型、生态系统模型或农业模型的改进输入。这些结果将作为投入提供给相关的CCSP综合产品和IPCC的下一次评估。除了科学成果和对科学评估的投入外，该项目还将产生重要的培训成果。参与该项目的研究生和研究员都将获得气候变化科学方面的宝贵培训和经验，这将使他们能够更好地为未来气候变化及其对美国的影响的研究和发展做出贡献。

The Synthesis Report of the Intergovernmental Panel on Climate Change (IPCC) Third Assessment concluded that one of the key uncertainties regarding regional climate change was "relating regional trends to anthropogenic climate change". Recent studies have shown that an anthropogenic climate change signal can be detected in surface temperature changes at continental scales. This project will use methods from detection and attribution of climate change to estimate the fraction of the observed temperature changes at regional scales in the United States (US) that can be attributed to anthropogenic forcing and other climate forcing factors. These methods will be applied to mean, minimum and maximum temperatures for both annual and seasonal averages, as well as to temperature extremes. Optimal fingerprint analysis of the temperature changes over the last 100 years will be used to constrain estimates of future global and regional temperature changes from a number of different climate models and to develop probability distributions of the regional temperature changes. Through collaboration with the climateprediction.net group in England, their very large perturbed parameterization ensemble of transient climate change simulations will be used to better estimate the uncertainties in current and future regional climate change projections for the US. Many aspects of the research are innovative. The regional attribution of climate change and the scaling of modeled signals of anthropogenic climate change have not been undertaken previously, but are extensions from recent studies. A novel approach to enhancing the anthropogenic temperature change signal relative to natural variability by removing the rainfall related temperature variations will be applied in the US region, with preliminary results indicating that it is likely to enhance the signal-to-noise ratio. The application of the climateprediction.net ensemble simulations to regional climate change projections is an extension of their earlier work. Broader Impacts: This project directly addresses two of the five goals of the Strategic Plan of the US Climate Change Science Program (CCSP) through quantification of the contribution of different factors to observed climate change in the US and estimation of probability distributions for projections of future climate change in the US. The expected outcomes of this research are likely to have significant impact internationally and in the US. Quantifying the contribution to regional climate change due to greenhouse forcing will assist policymakers in reaching decisions about policies on greenhouse gas emission reductions or adaptation to greenhouse climate change. The probability distributions of different regional climate changes due to greenhouse forcing will be very useful in better quantifying the possible impacts of regional climate change and estimating the costs or benefits of regional climate change. These probability distributions of regional climate changes can be used as improved inputs to economic models, ecosystem models, or agricultural models. The results will be provided as inputs to the relevant CCSP synthesis products and to the next IPCC assessment. In addition to the scientific outcomes and inputs to scientific assessments, there will be important training outcomes from this project. Both the graduate student and research fellow employed on this project will gain valuable training and experience in climate change science, which will allow them to contribute better to future research and development in climate change and its impacts in the US.