基于GIS的时空复合体-土地利用变化图谱模型研究方法

本文以地学信息图谱理论、地球空间信息认知理论、地球系统科学与地理综合体理论、地理信息系统中的时空复合体模型为依据 ,提出了研究土地利用 /土地覆被变化的基本单元 ,即“图谱单元”,它是记录时空复合信息的基本单元 ,由“相对均质”的地理单元和“相对均质”的时序单元复合而成 ,并提出了研究土地利用 /土地覆被变化图谱的概念模型。

Research of TUPU on Land Use/ Land Cover Change Based on GIS

This paper studies the urban grade size change of the Pearl River Delta (PRD) based on remote sensing. We use the built up area (rather than population) to analysis the urban size. TM imageries taken in 1988 and 1998 are used to get the built up area theme information, and we set up a spatial database of the Pearl River Delta towns in Arc/Info 8 02. The Maximum likelihood classifier (MLC) is selected to extract urban built up area theme information. The classified result accuracy by MLC is only 0.6559 (kappa index), and it is not enough to analyze the urban size. So we improve the computer classification results according to expert knowledge. The general spatial temporal structure, grade size change features, urban built up area change by administrative grade, and general urban system change features are discussed. From 1988 to 1998, the inner circle and outer circle spatial structure of the urban distribution kept relatively stable, and most of the large urbans concentrated on the estuary area of the PRD. The urban number of each grade size changed greatly during the study period. The urban number of grade size less than 0 1 km 2 decreased to 13 in 1998, but this number in 1988 was 112. And the urban number of grade size over 20 km 2 increased from 4 to 9. Urban area of town grade increased by 290 1%, urban area of county grade grew by 158 0%, and urban area of grade higher than county increased only by 108 4% during 1988 to 1998. The unbalance index of grade system decreased from 0 6011 to 0 5011, and the four largest urban indexes also decreased, but the standard deviation and the mean size of urban increased greatly. The results shows that the grade size based on remote sensing of the Pearl River Delta changed greatly in the end of the 20 century and this study is complementarity for traditional urban size study based on population.