USE OF LANDSAT-SERIES DATA IN NATIONAL GEOGRAPHIC CONDITION MONITORING IN CHINA

To fully grasp the nature and human geography situation information, solve the problem of ecological environment, economic and social development of the country, monitoring the state of geographic condition by uniform index system has great significance. By collecting the existing standard documents, our paper established a suit of index system considering the characteristics of long time series remote sensing data. The index system includes basic, subject, composite statistical indexes, and statistical indexes based on basic geographic element. The spatial and temporal distribution of geographic condition with Landsat TM image in Haidian district of Beijing from 1983 to 2013 are studies. Results show that farmland decreases by 28.60%, build-up land increases by 38.95% in this period. The amount of land resources in different elevation/slope shows that, with the increase of elevation/slope, farmland and build-up land is gradually reduced, while grassland area is gradually increasing. In plains areas of elevation less than 50m and within the scope of the 0 to 3° slope, farmland and build-up land are the main land cover types, and both show the characteristic of tradeoffs. Urban area extended to the west and the north, meanwhile mass center of Haidian also moves to the northwest. The urban compactness decreases and the fractal index increased gradually, reflecting the city saturation degree become reduced, the city boundary becomes complicated gradually. The comprehensive land cover dynamic degree after the first decrease and then increases. Finally, based on the above statistic results, the spatial distribution of land cover in 2015 is predicted.


INTRODUCTION
Promoted by social needs and technology development, the national geographic condition monitoring becomes an important mission in better serving the country's sustainable economic and social development in China.A supplement to the short history of high spatial resolution data, Landsat data supplies a much longer record of our earth, which also has the ability to detect the variation of geographic condition.Monitoring index system and related technical method suit for Landsat-series data are strongly required, which are help to realize the data sharing mechanism, and serve to the national strategic decision making and planning (Shi, 2013).

STUDY AREA AND DATA
The study area is located in HaiDian District, Beijing, covers an area of ~430km 2 .The elevation increases from about 50m for the eastern plain area to 1300m in the western mountain area (Yu, 2010).Landsat-series data were obtained from the Landsat archive * Zhaoyali.Email:880215zyl@163.comThe theory values of city compact ratio and fractal dimension are between 0 and 1(Figure.9),the larger the index, the city range is more regular, which is the ideal city extension model.
Both the two indexes are consistent with the shape variation in

(
http://glovis.usgs.gov/) between 1983 and 2013, besides the year of 1985, 2008 and 2012 for cloudy.After the geometric processing, all Landsat data were classified into six different land cover types, including crop, forest, grass, build-up area, water and undeveloped land.3. RESEARCH METHOD By collecting the standard documents used in land resources investigation, ecology, environment, meteorology and such kinds of related regions, the indexes can be measured by satellite data are extracted and analyzed.Combined with the spatial, temporal and spectral characteristics of Landsat data, a systematic index system suitable for National geographic condition monitoring in China has been constructed, including the following four kinds of statistical indexes: basic, subject, composite statistical indexes, and statistical indexes based on basic geographic element.The basic statistical indexes make a fundamental statistics of landform, the ratio of different land cover type, as well as the scope and circumference of build-up area.The calculation of the basic statistical indexes helps to detect the temporal-spatial The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XL-7/W3, 2015 36th International Symposium on Remote Sensing of Environment, 11-15 May 2015, Berlin, Germany This contribution has been peer-reviewed.doi:10.5194/isprsarchives-XL-7-W3-1339-20151339 variation of all the land cover types and fundamental characteristics.The subject statistical indexes are aimed at the three mainly land covers types directly relating to sustainable economic and social development, which are vegetation, build-up area and wetland.For vegetation subject, the indexes mainly concentrated in descripting the spatial distribution and dynamic expanding characteristics.For build-up area, the dynamic expanding and its reasonability are quantified.While for wetland, the indexes focus on the variation of ecological construction and spatial distribution.Composite statistical indexes are used to analyze the change rules and variation trend of six different land cover types based on landsat-series data, which can be used to serve the national strategic decision making and planning.While the CA-MARKOV (Cellular Automata-Markov) model is used for prediction.Statistical indexes based on basic geographic element point at the ratio of different land cover types, which is used to recognize the dynamic variation for the specific region, benefiting for local distinctive development.

Figure
Figure.1 Index system for national condition monitoring

Figure. 2
Figure.2 Landform information in Haidian 4.1.2Ratio of Different land cover types From Figure.3, the ratio of crop land shows a decreasing trend from 44.57% to 15.97% from 1983 to 2013, which were mainly replaced by build-up land.The ratio of forest and grass increases slowly, which is guided by series of policy, like conversion of cropland to forest and urban greening.Meanwhile, the ratios of water and other land cover type have no obvious change.

Figure. 3
Figure.3 Temporal variation of different land cover ratio byLandsat TM between 1983 and 2013, Haidian District

Figure. 4
Figure.4 Temporal variation of city area and circumferences by Landsat TM between 1983 and 2013, Haidian District

Figure. 6 .
Figure.6.During the period between 1992 and 2007, the city extents towards west and north, the city compact ratio and fractal dimension varied less evident.Around 2010, the city extent greatly towards north, which induces the city range are much more irregular than ever.

Figure. 6
Figure.6 Temporal-spatial variation of city range by Landsat TM between 1986 and 2010, Haidian District

Figure
Figure.11 spatial distribution of prediction land cover in 2015, Haidian District

Figure. 12
Figure.12 Temporal variation of different land cover ratio by Landsat TM between 1983 and 2013, Malianwa Street

Table 1
Basic Statistical of landform in Haidian District