多粒度时空对象支持下湿地时空行为认知与表达初探

湿地作为一种活的空间实体，其时空行为错综复杂，认知和表达一直是该领域研究的重点和难点。全空间信息系统以多粒度时空对象直接对现实世界进行抽象和描述，比传统GIS更符合人的认知特点。因此，本文基于多粒度时空对象建模思想，分别对湿地数据组织、时空行为认知和表达进行了初探。结果表明：湿地时空行为可按对象粒度、数量和特征属性进行分类，且行为表达可通过动态展示，分析建模和模型预测层层递进，较以往表征方式具有动态关联近实时等优势，可为湿地资源的动态监测提供一种新的研究思路。

The cognition and representation of spatio-temporal behaviors of wetland based on spatio-temporal objects of multi-granularity

Wetland, as a kind of living spatial entity, has complex spatio-temporal behaviors. The cognition and presentation of its behaviors have always been the focus and difficulty in this field. The pan-spatial information system could abstract and describe the real world directly based on spatio-temporal objects of multi-granularity, which is more convenient than traditional GIS for people's cognitive habits. Therefore, this paper, using the method of spatio-temporal objects of multi-granularity modeling, initially explores the cognition and presentation of its spatio-temporal behaviors by analyzing and summarizing the wetland data organization method, spatio-temporal behavioral cognition, classification, representation respectively. It turns out that wetland's spatio-temporal behaviors could be classified by the granularity, number, characteristic attributes of wetland respectively. And the representation of behaviors could be progressed by three levels, including dynamic display, analytical modeling and model prediction. Furthermore, this method, compared with the previous representation methods, has many advantages, such as dynamic, correlation and real-time. Finally, it could provide a new method for the dynamic monitoring of wetland resources.