Investigating state and timescale-dependency of climate variability using an isotope-enabled coupled general circulation model

全球和区域气候变率是否会随着全球平均气温的上升而沿着增加或减少，这一点从根本上说还不清楚。对社会而言，了解变异性变化至关重要，而为之做好准备的代价也很高，因为它们会导致极端事件发生频率的变化。有鉴于此，本项目的目的是了解和量化的平均变化和温度的变化之间的联系。地球的状态在最后一次盛冰期（LGM，21000年前）和中全新世温暖期（6000年前）是非常适合这个目的，因为许多代理和建模研究有助于表征其平均状态的气候。 然而，即使是最先进的，协调的建模工作，到目前为止，不能用于面向变量的模型数据比较。首先，已发表的模拟结果通常只有几个世纪的时间，接近单个极地冰芯数据点通常代表的时间。其次，自然强迫在这些气候状态中造成温度变率的作用是未知的，因为最先进的气候模拟是在没有变化的太阳和火山强迫的情况下进行的。第三，过去温度变化的大多数高分辨率证据来自极地冰芯中的稳定同位素比率，其温度校准在很大程度上不受限制，特别是在南极洲，为了克服这些挑战，并系统地确定全球平均温度变化和变化之间的关系，将使用同位素启用的大气环流模式模拟地球系统四个状态的集合轨迹。一个冰川状态将根据末次冰期，一个间冰期状态作为中全新世建模。对于一个超级冰川和超级间冰期的状态，全球温度进一步降低和增加，分别通过改变大气中的CO2浓度和大陆冰盖的大小。将进行两套千年模拟，一套是常数，另一套是变化的太阳和火山强迫。集合方法允许对千年时间尺度进行时间序列分析，弥合代理和模式分辨率之间的差距。由于该模型明确地模拟了水的同位素特征，因此可以研究代理校准的平均状态依赖性，并将其与大气和海洋动力学的变化联系起来。此外，对有强迫和无强迫的实验进行比较将首次阐明自然强迫在不同气候状态下造成气候变率的作用，模型实验和代用数据将共同提供关于平均温度和温度变率变化之间关系的全面看法，从而有助于更好地了解未来气候变率的潜在变化。

It is fundamentally unclear whether global and regional climate variability will increase or decrease along with global mean temperature rise. For societies, variability changes are crucial to understand, and expensive to prepare for, as they entail changes in the frequency of extreme events. In light of this, the aim of this project is to understand and quantify links between mean changes and variability in temperature.The Earth's state during the last Glacial Maximum (LGM, 21000 years ago) and the mid-Holocene warm period (6000 years ago) is well suited for this purpose, as many proxy and modeling studies have helped to characterize their mean state climates. However, even the most advanced, coordinated modeling efforts so far cannot be used for variability-oriented model-data comparison. Firstly, published simulations are often only a few centuries long, close to what a single polar ice core data point typically represents. Secondly, the role of natural forcing in creating temperature variability in these climate states is unknown, as the state-of-the-art climate simulations were run without varying solar and volcanic forcing. Thirdly, most high-resolution evidence for past temperature variability is derived from stable isotope ratios in polar ice cores, whose calibration to temperature is largely unconstrained, particularly for Antarctica.To overcome these challenges, and to identify relationships between global mean temperature changes and variability changes systematically, ensemble trajectories of four Earth system states will be simulated using an isotope-enabled general circulation model. A glacial state will be modeled according to the LGM, an interglacial state as the mid-Holocene. For a super-glacial and a super-interglacial state, global temperature is further reduced and increased, respectively, by changing the atmospheric CO2 concentrations and the continental ice-sheet size. Two suites of millennial simulations will be performed, one with constant, a second with varying solar and volcanic forcing.The ensemble approach allows time-series analyses on up to millennial timescales, bridging the gap between proxy and model resolution. Since the model explicitly simulates the isotopic signature of water, the mean-state dependence of the proxy calibration can be investigated, and linked to changes in atmospheric and oceanic dynamics. Furthermore, comparing the experiments with and without forcing will, for the first time, shed light on the role of natural forcing in creating climate variability in different climate states.Together, model experiments and proxy data will provide a comprehensive view on the relationship between mean temperature and temperature variability changes, thus contributing to a better understanding of potential changes to future climate variability.