Tropical Biomes in Transition

热带雨林和稀树草原构成了热带地区的两个主要生物群系。它们占世界植物种类的70%以上。热带雨林和热带稀树草原面积大，潜热和感热交换率高，对全球气候也产生了巨大而不同的影响。我们对它们当前和未来对全球变化的反应了解有限。热带雨林是一个主要的陆地碳储存库，目前估计约占全球陆地碳汇的一半。虽然降雨是影响其相对分布的关键决定因素，但土壤条件、火灾和放牧等干扰以及人类影响等其他因素也涉及其中。我们对这些因素以及它们如何相互作用影响植被类型的认识仍然很差——目前所有的全球植被模型都错误地指定了这些生物群落的分布。这些知识是理解和预测在地方、区域和全球尺度上热带植被变化的基础。未来几十年亚马逊流域的干旱可能导致热带雨林被稀树草原取代，这将作为地表能量和物质平衡变化的结果反馈给气候系统，从而加速全球变暖和热带干旱。以热带草原植被的增减为中心的重大转变并不局限于南美洲。对导致从森林向热带草原转变的因素的改进预测是理解这一点的基础，并将取决于对全球范围内热带植被分布的环境和土壤决定因素的显著提高的理解。因此，我们的主要目标是获得这些决定因素潜在生理基础的新知识，更好地理解森林和稀树草原之间地表能量和二氧化碳交换差异的基础，并将这些信息整合起来，以便在全球植被模型中比目前可能的更准确地表示这些过程。这将通过一项“模型数据开发计划”来实现，该计划涉及与当地合作者进行实地活动，以检查关键“热点”雨林/稀树草原过渡地带的气候/土壤/干扰关联。综合测量包括植物关键生理过程（光合作用与水分关系）、植被林分结构与组成、土壤理化性质等。还将开发新的高分辨率气候学，并提出新的方法，以便从空间确定热带植被结构和火灾频率。这些将与实地数据一起得到验证和测试，然后用于在全球范围内发展对热带生物群落分布的新的定量理解。将把推导出的关系与目前用来描述热带植被/气候关系的更原始的关系进行比较，并对过去和未来植被变化的预测结果进行评价。将利用实地观测所得的资料与从项目外吸收的数据结合起来，发展热带植被分布和功能的新模式。最后，我们将把我们对热带植被分布和过程的新的和改进的认识与新的全球气候模式（GCM）结合起来，为即将发生的热带植被变化、相关的气候反馈和未来的全球气候提供新的见解。我们预计，这将从根本上提高我们预测21世纪全球气候的能力。

Rainforest and savanna constitute the dominant two biomes of the tropical zone. They account for over 70% of the world's plant species. With massive areas and with high rates of latent and sensible heat exchange, rain forest and savanna also exert large, yet different, effects on the global climate. We have a limited understanding of their contemporary and future responses to global change. Tropical rainforests are a major terrestrial carbon store and are currently estimated to account for around half of the global terrestrial carbon sink. Although rainfall is a key determinant affecting their relative distributions, other factors such as soil conditions, fire and disturbances such as grazing and human influence are also involved. Our knowledge of these factors and how they interact in influencing vegetation type is still poor - all global vegetation models currently misspecify the distributions of these biomes. Such knowledge is fundamental for understanding and predicting transitions in tropical vegetation at local, regional and global scales. A drying of the Amazon Basin in coming decades could lead to the irreplaceable replacement of tropical forest with savanna, this then feeding back on the climate system as a consequence of changes in surface energy and mass balances - thereby accelerating global warming and tropical drying. Significant transitions centered around the gain or loss of savanna vegetation are not restricted to South America Improved predictions of the factors causing a transition from forest to savanna are fundamental to understanding this, and will depend on a significantly improved understanding of the environmental and edaphic determinants of the distribution of tropical vegetation at a global scale. Our main objective is to thus obtain a new knowledge of the underlying physiological basis of these determinants, to better understand the basis of differences in surface energy and CO2 exchange differences between forest and savanna, and to integrate this information for a much more accurate representation of these processes in global vegetation models than is currently possible. This will be achieved by a 'model-data development program involving field campaigns with local collaborators to examine climate/soil/disturbance associations in key 'hot spot' rainforest/savanna transition zones. Comprehensive measurements will be made including key plant physiological processes (photosynthesis water relations), vegetation stand structure and composition, and soil physical and chemical properties. New high resolution climatologies will also be developed and novel methodologies advanced to allow determination of tropical vegetation structure and fire frequency from space. These will be validated and tested and along with the field based data and then used to develop a new quantitative understanding of tropical biome distributions at a global scale. Derived relationships will be compared against more primitive ones currently used to describe tropical vegetation/climate relationships and consequences for predictions of past and future vegetation change evaluated Information from field observations will be used in conjunction with data assimilated from outside the project to develop new models of tropical vegetation distribution and function. Models derived will be as mechanistic as possible with predicted distributions and structure/flux of vegetation tested using larger scale distributional and structural data from remote sensing data Finally we will incorporate our new and improved understanding of tropical vegetation distribution and processes with new global climate model (GCM) runs to provide new insights into impending tropical vegetation change, associated climatic feedbacks and the future global climate. We anticipate this will lead to a fundamental improvement in our ability to predict the global climate of the 21st century.