How did the evolution of plants, microbial symbionts and terrestrial nutrient cycles change Earth's long-term climate?

显生宙（过去5.4亿年）涵盖了陆地植物的进化史，从最初的陆地殖民到森林和开花植物。在这段时间内，地球气候发生了重大变化，但尚不确定这些变化主要是由陆地生物圈的变化引起的，还是由火山释放二氧化碳等构造因素引起的。这个问题的答案是我们理解地球如何运行的核心，但由于我们的生物地球化学计算机模型中缺乏陆地生物圈的代表，回答这个问题的能力受到了阻碍。为了计算非常长的时间尺度，这些“深时间”模型需要非常简单，这限制了包括降雨地点等空间特征的能力，而这些空间特征对陆地建模至关重要。一个更根本的问题可能是缺乏对植物进化通过改变组织中碳氮磷比例来改变全球化学循环的方式的理解，以及真菌和微生物共生体在提供关键限制性营养物质方面的贡献。该项目汇集了计算机科学、地球化学、生态学和植物共生生理学方面的专业知识，通过一系列有针对性的植物生长实验和大量的文献综述，建立了一个新的深时空间地球系统模型。首先，我们将对早期分化植物和共生固氮树进行实验室实验，在控制大气CO2浓度的微观环境中与真菌和/或固氮共生体合作或不合作。引入同位素标记的碳、氮和磷将使我们能够捕获跨越植物系统发育的不同植物-共生体伙伴关系的碳-氮-磷化学计量比和营养获取途径，填补目前对这些过程的知识的重大空白。这些实验将使我们了解：植物-共生体碳-养分的“成本”和“收益”——从植物固定碳和共生体获得的养分收益来看。生态化学计量学和养分获取途径如何在陆地植物系统发育过程中发生变化。物种、共生体和矿物风化速率之间的关系第二，我们将建立新的地球系统模型。在这里，我们将建立在“COPSE”模型（碳、氧、磷、硫演化）的框架上，这可以说是文献中最完整的预测“深时间”盒子模型，PI Mills在过去十年的发展中发挥了关键作用。利用MATLAB中的矩阵开发了一个快速空间陆面模块的原型，在本项目中，我们将空间陆面模块与COPSE耦合。这将使我们能够在实验室实验和文献植被模型的基础上，建立一个不断演变的陆地生物圈的动态表示。这个模型将在地质时间尺度上绘制出磷、氮和碳在陆地系统中的流动图。将模式输出与多个独立的地球化学代理进行比较，将使我们能够探索(1)植物进化和共生伙伴关系的发展如何反馈到地球气候；(2)演化过程中发生的关键事件及其能否解释奥陶纪和新生代等突出的CO2减少事件；(3)陆地生物圈与构造在控制地球气候历史中的相对作用。除了直接的结果，我们创建的混合模型将弥合全球地球化学盒模型和真正的古气候环流模型之间的差距，为社区进一步扩展和应用提供了一个有用的工具。

The Phanerozoic Eon (the last 540 million years) encompasses the evolutionary history of land plants from the initial colonization of the land through to forests and flowering plants. Earth's climate has undergone major changes over this timeframe, but it remains uncertain whether these changes were primarily driven by revolutions in the terrestrial biosphere, or by tectonic factors such as volcanic degassing of CO2. Resolution of this question lies at the heart of our understanding of how our planet operates, but the ability to answer it has been hampered by a lack of representation of the terrestrial biosphere in our biogeochemical computer models. These 'deep-time' models need to be simple in order to compute very long timescales, and this limits the ability to include spatial features such as locations of rainfall, which are vital to terrestrial modelling. A perhaps more fundamental problem is the lack of understanding of the way that plant evolution has altered global chemical cycling through changes to carbon-nitrogen-phosphorus ratios in tissue, and what the contribution of fungal and microbial symbionts were to supplying key limiting nutrients. This project brings together expertise in computer science, geochemistry, ecology and plant-symbiont physiology to build a new deep-time spatial Earth system model, informed by a targeted suite of plant growth experiments and a robust literature review.Firstly, we will run laboratory experiments with early diverging plants and symbiotic nitrogen-fixing trees, with and without partnership with fungal and/or nitrogen-fixing symbionts in microcosms with controlled atmospheric CO2 concentrations. Introduction of isotopically-labeled carbon, nitrogen and phosphorus will allow us to capture the carbon-nitrogen-phosphorus stoichiometric ratios and nutrient acquisition pathways for diverse plant-symbiont partnerships across the plant phylogeny, filling significant gaps in current knowledge of these processes. These experiments will allow us to understand:a. Plant-symbiont carbon-nutrient "costs" and "benefits" in terms of plant-fixed carbon and symbiont-acquired nutrient gainsb. How ecological stoichiometry and nutrient acquisition pathways vary across the land plant phylogenyc. Relationships between species, symbiont and mineral weathering ratesSecond, we will develop our new Earth system model. Here we will build on the framework of the 'COPSE' model (Carbon Oxygen Phosphorus Sulphur Evolution), which is arguably the most complete predictive 'deep time' box model in the literature, and which PI Mills has had a key role in developing over the last decade. A prototype fast spatial land surface module has been developed utilizing matrices in MATLAB and in this project we will couple the spatial land surface module to COPSE. This will allow us to build a dynamical representation of the evolving terrestrial biosphere, based both on our laboratory experiments and on literature vegetation models. This model will map the flows of phosphorus, nitrogen and carbon through the terrestrial system over geological timescales. Comparison of model outputs with multiple independent geochemical proxies will allow us to explore (1) how plant evolution and the development of symbiotic partnerships feeds back on Earth's climate; (2) the key evolutionary events that occurred through time and whether they can explain prominent CO2 drawdown events, such as during the Ordovician and Cenozoic; and, (3) the relative roles of the terrestrial biosphere vs. tectonics in controlling Earth's climatic history.Beyond the immediate results, the hybrid model we create will bridge the gap between box modelling of global geochemistry and true paleoclimate general circulation modelling, providing a useful tool for the community to further extend and employ.