MMIA: Integrated Assessment of African Savannas with Spatial- Dynamic Vegetation and Land Use Modeling

目前全球气候变化对陆地生态系统影响的预估是通过在区域到全球尺度上模拟气候-植被相互作用产生的。然而，目前的模式通常预测新的平衡条件，而不能模拟过程驱动的瞬态条件，这些条件可能持续很长时间，如果气候变率大幅增加，甚至可能无限期地持续下去。这些模型不解决植被对景观结构的响应，也不代表植被对野生动物或家畜的食草性或其他人类土地利用的响应。该项目的一个主要目标是通过使用面向过程的生态系统级模拟，评估人们可以在多大程度上改进对植被、生态系统和人类的气候变化影响的估计。当前的气候模式将被用来驱动三种不同类型的模型（准平衡、热带稀树草原-气候、热带稀树草原-生物），所有这些模型都被参数化，以模拟塞伦盖蒂区域生态系统中现存的植被条件；一个热带稀树草原系统，其中土壤、景观特征、人类土地利用、火灾和草食动物的影响都与气候相互作用，影响植被模式和生态系统状态。将从以下方面评价模式在预测对全球变化的反应方面的差异：(a)模式在当前和加倍二氧化碳气候条件下预测现存和潜在植被的准确性；(b)与单独考虑气候相比，考虑草食、土地利用的影响；(c)空间尺度对预测响应的影响。支持实地和遥感研究将有助于更好地了解当前植被与气候、土壤、草食和土地利用的关系。这项研究将与东非区域的自然资源政策有关。综合建模方法将对政府土地管理和规划、饥荒预警、公园和野生动物保护区管理人员以及牧民具有实用价值。该研究将为全球植被动态模型（DGVM’s）的发展做出贡献。

Coughenour 97-11627 Current projections of the effects of global climate change on terrestrial ecosystems are generated through simulations of climate-vegetation interactions at regional to global scales. However current models generally predict new equilibrium conditions and do not emulate process-driven transient conditions which could last for a long time, or even indefinitely if climate variability increases substantially. The models do not resolve vegetation responses to landscape structure, nor do they represent vegetation responses to herbivory by wild or domestic animals, or other human land uses. A major objective of this project is to evamate how much one can improve estimates or climate change effects on vegetation, ecosystems and people, by using process-oriented ecosystem-level simulauons. Current climate patterns will be used to drive three different types or models (quasi-equilibrium, SAVANNA-climate, SAVANNA-biotic), all parameterized to simulate extant vegetation conditions in the Serengeti Regional Ecosystem; a savanna system where soil, landscape features, human land use, fire and herbivore impacts all interact with climate to influence vegetation patterns and ecosystem state. Differences among models in predictions of responses to global change will be evaluated in terms of: (a) the accuracy of the models at predicting extant and as potential vegetation under current and doubled CO2 climate conditions; (b) the effects of considering herbivory, land-use compared to climate alone; (c) the effects of spatial scale on the predicted responses Supporting field and remote sensing studies will result in a better understanding of current vegetation in relationship to climate, soils, herbivory, and land use. The research will be relevant to natural resources policy in the East African region. The integrated modeling approach will have utility for government land management and planning, for famine early warning, for park and game reserve managers, and for pastoral peoples. The re search will contribute to the development of dynamic global vegetation models (DGVM's).