Dynamics of the Earth System and the Ice-Core Record (DESIRE)

大气成分与气候密切相关，因为二氧化碳和甲烷等化合物是温室气体：它们浓度的增加预计会使大气变暖。这样的增长发生在过去的两个世纪，预计在未来几十年将加速。然而，这些浓度和气候究竟将如何共同演变，取决于将它们在所谓的地球系统内联系起来的过程。我们对这些过程的理解表现在模型中，这些模型代表并连接了系统的各个部分，如植被生长、海洋环流、大气环流和化学等。然而，我们验证这些模型是否正确地代表了地球的最好方法是回顾过去。各种各样的古气候记录为我们提供了过去气候变化的观点。冰芯记录特别有价值，因为它显示了气候和大气成分在过去80万年中是如何演变的。在这段时间里，地球多次进入和退出冰川状态，事实证明，主要的温室气体和气候在这段时间里一起发生了变化。二氧化碳和甲烷在温暖的间冰期浓度高，在寒冷的间冰期浓度低。因此，它们提供了许多关于气候如何对大气成分变化作出反应的例子，以及关于二氧化碳和甲烷的来源和汇如何对气候变化作出反应的有力线索。我们目前的理解是，当气候变暖时，由于湿地来源的扩大和大气汇的减少，甲烷会增加。然而，我们缺乏关于这些源和汇的许多细节，也没有明确的证据来区分它们各自的作用。对于二氧化碳，这些变化被认为主要源于南大洋的过程，但在这一观点中有许多相互竞争的假设。该提案将结合英国和法国相关观测和建模社区的最强元素。首先，我们将检查冰芯和其他数据集，以提供尽可能多的关于二氧化碳和甲烷浓度变化原因的约束。这将特别涉及用于诊断碳来源的碳同位素的新测量，以及可以限制碳循环变化机制的南大洋海洋沉积物的新测量。将审议甲烷和二氧化碳的排放和加工的特定方面，以便对模式作出必要的改进。然后，我们将使用各种不同复杂程度的模型来探索冰记录中的主要变化：在冷冰期和暖间冰期之间，在不同的间冰期之间，以及在过去80万年的其他特定时间，这些变化可能使我们能够区分某些机制的运作。详细的模型，包括新的QUEST地球系统模型，将用于评估记录中特定时间甲烷的产生和损失。复杂性较低的模式将在较长的时间内运行，以确定二氧化碳变化的各种机制的不同古气候档案中的预期信号，以期缩小每种机制重要性的不确定性。模型还将用于测试我们是否能够理解在过去的间冰期看到的不同气候，知道来自太阳的能量输入和冰芯中看到的温室气体的浓度。该项目的最终结果将是提高模拟过去的能力，更好地理解控制大气成分、气候和碳循环的过程，并作为最终结果，改进用于预测地球系统未来演变的模型中所有相关过程的表示。

Atmospheric composition and climate are closely linked because compounds such as carbon dioxide and methane are greenhouse gases: increases in their concentration are expected to warm the atmosphere. Such increases have occurred in the last two centuries, and are expected to accelerate in the next few decades. However, exactly how these concentrations and climate will evolve together depends on processes that link them within the so-called Earth System. Our understanding of these processes is expressed in models that represent and connect parts of the system such as the growth of vegetation, ocean circulation, atmospheric circulation and chemistry, etc. However, the best way we have of validating whether these models are correctly representing the Earth is by looking at the past. Various palaeoclimate records provide us with a view of how climate has behaved in the past. The ice core record is particularly valuable because it shows how both climate and atmospheric composition have evolved over the last 800,000 years. During this time, the Earth has passed into and out of glacial states many times, and it turns out that the principal greenhouse gases and climate have varied together during this period. Carbon dioxide and methane have high concentrations during warm interglacials and low concentrations in cold glacials. They thus offer numerous examples of how climate reacts to changes in atmospheric composition, and strong clues about how the sources and sinks of carbon dioxide and methane react to climate change. Our current understanding is that methane increases when climate warms because of a combination of expanded wetland sources and diminished atmospheric sinks. However, we lack many details about these sources and sinks, and have no clear evidence to differentiate their respective roles. For carbon dioxide, the changes are believed to stem mainly from processes in the Southern Ocean, but within this view there are a number of competing hypotheses. This proposal will combine the strongest elements of the relevant observational and modelling communities in the UK and France. Firstly, we will examine both the ice core and other datasets to provide as many constraints as possible on the causes of change in concentration of carbon dioxide and methane. This will involve particularly new measurements of isotopes of carbon that are diagnostic of sources, and new measurements of marine sediments in the Southern Ocean that can constrain mechanisms for changes in the carbon cycle. Particular aspects of the emission and processing of methane and carbon dioxide will be considered in order to make necessary improvements in models. We will then use a variety of models of different levels of complexity to explore the major changes seen in the ice record: between cold glacials and warm interglacials, between different interglacials, and at other particular times in the last 800,000 years that may allow us to differentiate the operation of certain mechanisms. Detailed models, including the new QUEST Earth System Model, will be used to assess the production and loss of methane at particular times in the record. Models of lower complexity will be run over longer time periods to determine the expected signal in different palaeoclimate archives of various mechanisms for changes in carbon dioxide, with a view to narrowing the uncertainties on the importance of each mechanism. Models will also be used to test whether we can understand the different climates seen in past interglacials knowing the energy input from the Sun and the concentrations of greenhouse gases seen in ice cores. The end result of this project will be an improved ability to simulate the past, a better understanding of the processes that control atmospheric composition, climate and the carbon cycle and, as an end result, an improved representation of all relevant processes in models used to predict the future evolution of the Earth System.