基于模型-遥感整合的人工林应对干扰及气候变化的响应规律研究

项目名称： 基于模型-遥感整合的人工林应对干扰及气候变化的响应规律研究

项目编号： No.31270587

项目类型： 面上项目

立项/批准年度： 2013

项目学科： 农业科学

项目作者： 李明诗

作者单位： 南京林业大学

项目金额： 80万元

中文摘要： 因生态系统组成、结构及干扰模式改变而导致的气候变化是国际社会长期关注的最紧迫的环境问题，它严重威胁着人类可持续发展。此建议的研究旨在通过整合遥感、气象及野外调查数据和植被变化自动检测模型，发展新方法来探索1982-2012年间人工林应对干扰及气候变化的响应机制。我们将整合多源遥感及野外数据，利用基于数据挖掘工具Cubist的回归树算法发展一个30米分辨率、2-3年间隔期的人工林结构变化数据库（包括树冠密度、树冠高及生物量）；同时分析由植被变化自动追踪模型导出的森林干扰时空模式与气象因子、森林结构变化之间的统计关联效应；预测气候变化背景下人工林将来可能遭遇的干扰及结构变化，并提出人工林适应或减缓气候变化效应的经营策略和措施。成功执行此研究将生成一个在气候变化背景下量化区域尺度森林结构及干扰模式变化的区域自适应的新方法,并为评价后续区域碳循环模式奠定坚实的方法和数据基础。

中文关键词： 森林干扰；森林恢复；结构参数；气候变化；建模与制图

英文摘要： Climate change induced by variations in composition, structure and disturbance patterns of ecosystems has long been recognized as the most pressing environmental concern of the international society, which creates a horrible threat to the sustainability of human being. This proposed research aims to understand the response mechanisms of the plantations to distrubances and climate change from 1982 to 2012, via applying an integrated and regionally tuned method that synthesizes multisource remote sensing, meteorological and field inventory data, as well as a vegetation change tracker model. The investigation will develop a 30 m spatial resolution forest change database with a temporal interval of 2-3 years, for the period of 1982-2012 including changes in forest canopy density, canopy height, and biomass. The database will be developed from using primarily NASA-EOS Landsat TM/ETM+ data, topographic derivatives, Light Detection and Ranging (LiDAR),Quickbird and the field inventory data, coupled with the regression trees algorithm built upon a data mining tool of Cubist. The remote sensing-based forest disturbances products developed from applying a vegetation change tracker model to the imagery stacks consisting of a time series of Landsat TM/ETM+ will be analyzed in concert with long-term meteorological data and

英文关键词： Forest disturbance；Forest recovery；structural parameters；climate change；Modeling and mapping