摘要
地理信息系统(GIS)就其实质而言,是现实世界的一种数据抽象模型。但现有的GIS数据模型缺乏跟踪现实世界实体变化的能力,因为它只保存实体的现势数据,而对于各种历史状态数据则无法保存,是一种快照数据模型。随着地理信息系统在行业应用的普及和深入,人们开始认识到地理信息系统应该具有描述现实世界中各种时空变化的能力,既可以分析其过去,又可以预测其未来(发展趋势)。这种想法导致了时空地理信息系统和时空数据模型概念的提出。本文分析了时空数据建模方面的技术问题;并讨论了主要的建模方法。认为历史接续关系和时序刻画是时空建模的主要内容。
Most of the geographical information systems are based on the static data model but with restrictions when they operate with dynamic information. Only the current status of an entity will be kept and the historical data will be lost during updating because the current GIS data model is lacking ability in tracking the evolution of an entity. We only know what it currently is, we know nothing about its past. The current data models that nowadays GIS use are nothing but snapshot data models.With the widespread use of GIS in many related geographic fields, people begin to realize that GIS should have ability to describe the spatiotemporal changes that are taking place every day in our world, so historical analysis could be achieved and even future evolution trend of phenomenon could be predicted. This idea brings forth the concepts of Spatiotemporal Geographical Information System(STGIS) and spatiotemporal data model(STDM).This paper reviews the progress that has been made in spatiotemporal data modeling. Related concepts have been introduced in details, technical issues in spatiotemporal data modeling and STGIS have been analyzed. Two main kinds of spatiotemporal data models are discussed, namely version-based spatiotemporal data models and object-oriented spatiotemporal data models. This paper points out that lack of a perfect spatiotemporal data model is a main obstacle in STGIS development and the basic requirements for a spatiotemporal data model has been investigated from system aspect and design aspect. Historical retrospect and time sequence description ability are the most essential ones among these requirements. This paper concludes that in spatiotemporal data modeling, object-oriented modeling method should be used and the implementation of such a spatiotemporal data model should be based on state-of-the-art database technologies.
出处
《地球信息科学》
CSCD
2004年第1期49-52,共4页
Geo-information Science