GISdevelopment.net ---> AARS ---> ACRS 1990 ---> Poster Session Q

Research on the National Land Information System in China

Jiang Jingtong
Research Institute of Surveying and Mapping
National Bareau of Sureveying and mapping
Beijing, China


Abstract
The National Land Information System is being developed by the Research Institute of Surveying and Mapping, the National Bureau of Surveying and Mapping since 1984. It is one of the largest nation-wide geographic information systems in China. This paper shows the background, hardware and software environment, components and logical structure, data base design, applications and future development of the system.

Introduction
With the science and technology development, geographic information system (GIS) has increasingly become one of the indispensable technical tools for management of resources and environment, national and regional planning or decision-making in the world. A lot of systems have been developed, or are being developed in China as well. Among them are some nationwide or regional systems, such as the National Economic Information System, the Loess Plateau Information System, the Territorial Resources Information System, etc. All of these systems need common basic data as the basis for locating their special data, to ensure all data can be shared and systems can be compatible with each other. The National lal Information System (NLIS) will play an important and hopeful role in realizing these purposes.

According to statistics, more than 1,000,000 sheets of maps at the scales from 1:10,000 to 1:1,000,000 are used each year in almost every government department, national economy organization, institutes, universities and other organs in China. They need to obtain the current basic national land information with high quality very quickly. Only the NLIS could satisfy their needs.

By now, information captured from nearly all technical fields of surveying and mapping, such as geodesy, photogrammetry, remote sensing, cartography, tends to be in the digital form.

The NLIS will be the good linkage and end-result of these technic developing trends.

The NLIS, as one kind of GIS, is being developed by the Research Institute of surveying and mapping, the National Bureau of Surveying and Mapping (RISM , NBSM) since 1984. it is the major information system of the NBSM and one of the sub-systems of the national Economic Information System in China. Actually it is a common basic information system in China.

The NLIS is divided into two levels: the national Center system in Beijing is at the national level (the first level), and the Subsystems in each province are at the provincial level (the second level) . (see Fig.1)


Figure 1 The levels of the NLIS

Hardware and Software Environment
The NLIS is supported bythe VAX II series computers, the ARC/INFO GIS management system and some self-developed software. Table 1 is a hardware configuration list of the NLIS.

Table 1 Hardware Configuration of the NLIS
Computer : VAX 11/785+11/780+11/750+Micro VAX II , linked by the ether Net
Disk Driver : 12 sets, about 6000 MB altogether
Tape driver : 5 sets with 800, 1600 and 6250 bpi
Line printer : 4 sets
Digitizer : Calcomp 9100 (9 sets) and Altek (4 sets)
Scanner : Tektronix 4991 (one color) Graphic display
Unit : Tektronix 4125 , 4115B, 4114B, 4111, 4109 (17 sets) Electrostatic
Plotter : Calcomp 5845 (color)
Plotter : Ta 10, Calcomp1044, 1077

Because the software ARC/INFO is developed for general purposes and it is mainly used to process vector data, and has been expanded. Some other self-developed software, such as Grid DBMS, data format exchanging package, PC GIS software and so on, has been developed to form an important part of the NLIS.

Components and Logical Structure
Besides hardware, the NLIS consists of following components:
  1. DBMS for both vector and raster data;
  2. Data bases including the Map data bases at the different scales, the Geographic Names Data base, the Geodetic data base and the Gravity Data Base;
  3. Application model
The scale series of Map Data Bases are 1:1,000,000, 1:250,000, 1:50,000, 1:10,000. Fig. 2 is the logical structure of the NLIS.


Figure 2 The Logical structure of NLIS

For the first step in developing the NLIS, the main task is to build up the 1:1,000,000- scale Map Data Base (MDB). At the same time, the 1:1,000,000-scale Geographic Names data Base (GNDB) and the testing Gravity Data Base are being built up as well.

The design of the Map Data Base

1. Data Category and Attribute Coding
The main contents of the NLIS map data bases are: surveying control point, hydrography, residence, transportation, boundary , topography, vegetation, and others. The data are fully topologically structured and attribute coded. The data category and attribute codes (or feature codes) are used to describe the features represented by a point, line or area. The data category and codes are based on the Cartographic Symbolizations of the NBSM topographic map series.

The category of the features has four classes. Table 2 shows the catalog of the first class used in the NLIS data bases.

Table 2 The catalog of the first feature class used in the NLIS map data bases.
1-------Surveying Control Point
2-------Hydrography
3-------residence
4-------transportation
5-------Pipe Line and City wall
6-------Topography (Terrain)
7-------Boundary
8------Vegetation
9------Others

The data in the NLIS map data bases are divided into several layers according to different classes and different geometric characteristics. An attribute code in the NLIS map data base is composed of three distinct subfields: a belongingness code, a category code and an identify code.

The belongingness code presented as five characters identifies the organization to which the data belongs.

The category code presented as six alphanumeric characters identifies the category to which the element belongs. It is structured as follows:


The identify code presented as six alphanumeric characters is only used to describe major features , such as cities and boundaries with political level equal to and higher than country, rail-way main road main river and lake. The tag in the category code indicates if there is an identify code for a certain element.

2. Data description Forms
To make the data base develop in an orderly way and to provide detailed information about data to users to evaluate the fitness for a particular use, the two kinds of data description forms are designed.

One is the overall form of the data source. It consists of the basic information of the map sheet, such as the name and number of the map, starting and ending latitude and longitude, the organization and dates of the map compilation and publication, the source material used to

Compile this map, the parameters of the projection, the size of the neat and diagonal lines, the number of layers of this map, and so on. another one is the form of data lineage in a layer. it should be filled one form for each layer of a map sheet. each one consists of the name and number of the layer, the data contents, the operator name and date for each data structuring, attribute assigning, data quality controlling, edge matching, coordinate converting, data updating, the quality evaluation, the coordinate converting coefficients, and other descriptions. These forms should be filled from the beginning to the end in the map data base development.

3. The design of the Digital Elevation Model (DEM)
Based on more than 10,000 sheets of topographic maps at the scales of 1:50,000 and 1:100,000, elevations of those points at the intersection of latitude and longitude lines are captured, to form a DEM for the whole country with 1:1,000,000-scale accuracy. The cell size in 28” . 125 * 18” 750 (latitude difference * longitude difference) . It approximates to 600m * 500m near 55 degrees north, 600m* 650m near 40 degrees north and 600m*800m near 20 degrees north. The elevation error allowed is less than 10m on the ground . The mean square error of elevation interpolation from the DEM is 75m in high mountains, 20m in lower mountains or hills, 1m in plains . The total number of points is about 25,000,000. the application of the DEM will be mentioned later in the paper.

4. Data Capturing and Editing
The positive films which are directly copied from the safety copies of 1:1,000,000-scale map sheets are used as data sources right now. The data captured from field surveying,photogrammetry and image processing will be sources as well.

The data are either manually digitized using the Calcomp 9100 digitizer with a resolution of 0.025mm and an absolute accuracy of 0.2m, or are scanned on the Textronix 4991 scanner. Normally hydrography and contours are scanned, while other features are manually digitized.

Data capturing has to meet following requirements:
  1. Keep the location of the symbol center for a point feature;
  2. Keep continuity for linear features;
  3. Form completely closed polygons for area features;
  4. When an object has two or more attributecodes, for instance, a boundary on a section of river, it is only digitizedonce with a special attribute value;
  5. For manually digitizing, the positional error should be less than 0.2mm.The distance between adjacent two vertex point on curve lines should be less than 0.5- 0.8 mm and the bigger the curvature is, the shorter the distance.
  6. For scanning data, the noises and some useless data, such as too narrow double lines, too small polygons have to be removed scanned data. Then additional manual digitizing for a part of features is usually needed.
Based on data testing and correction, the data topology is structured and some editing is made including pseudo nodes removing, polygon labeling and attribute coding.

For different features, the different attribute tables have been designed. Each table includes several attribute items and each item consist of the item name, input-width, output-width and type. The following three items are the same on all of tables : inner –ID , category code and identify code. Tables 3 and 4 are examples of such tables.

Table 3 Attribute table of rivers (including canals and ditches)
item name inner ID category code identify code name subordinate basin code navigable situation
input width 6 6 6 20 6 1
out put width 6 6 6 20 6 1
Type 1 C C C C C

Table 4 Attribute table of roads
item name inner ID Category code identfy code management level technical level Surface quality
input width 6 6 6 1 1 1
out put width 6 6 6 1 1 1
Type 1 C C C 1 C

The items for every feature or layer can be increased depending on requirements, and existing statistic data and other source materials.

There are descriptions about attribute value for each item. For instance, the description of item navigable situation is as follows:

0-----non-navigable river
1-----wooden ship navigable river
2----steamship (<100 tn.) navigable river
3----steamship (>100 tn.) navigable river

Technical interface between the MDB and the GNDB is designed, so that the geographic names can be retrieved from the GNDB according to the category code and the identify code which are the same in both attribute tables of MDB and GNDB for the same object. The Chinese codes and positional data of geographic name retrieved from GNDB and the other control data of the same name retrieved from MDB are used for automatic annotation on output map or on screen in different Chinese character styles and different sizes. Of course, it is necessary to do some modification interactively. In this case, the items about feature names in attribute tables of the MDB can be missed to save the storage space.

5. Data Quality Control
After data editing and attribute coding, all data should seriously pass the data quality control, to ensure that data are accurate and reliable for the NLIS users. For this purpose, two kinds of verified plots are made : the plots of layers separately and a plot of all layers overlapping

The data quality control consists of following points:
  1. Check Data Description forms;
  2. Check the logical consistency and completness;
  3. Test the Position Accuracy including the registration accuracy and the relationship between different features;
  4. Test the attribute accuracy;
  5. Sum up and make an evaluation about the data quality.
Before entering data base, it is still necessary to make further data processing. The coordinates of the ground coordinate system in meters need to be converted in to degrees of the Geographic Coordinate System. Then edge Matching and data compression should be done.

Other Data Bases

1. The Geographic Names Data Base (GNDB)
The GNDB is one of data base of NLIS. It is also a special data bases of Geographic Names. It is linked with the MDB by attribute code and user-ID, so a geographic name can be retrieved from the GNDB and annotated on a map in Chinese according to its postional information retrieved from the MDB . It can be used for analyzing, statistic and query geographic names, creating geo names dictionary , etc.

The attribute items includes attribute code, user-ID, name on maps, standard name, pinyin, position (latitude and longitude), and so on.

2. The Gravity Data Base
The Gravity data Base consists of the observed gravity data base and a DEM. The observed gravity data base is used to store gravity value and certain attribute values of all gravity points in China. The DEM is created by interpolation from the 1:1,000,000-scale DEM. Its cell size is 30* * 30* (latitude difference X longitude difference).

Applictions Of The NLIS
Some application models are designed to satisfy various requirements of users. The main application fields of the NLIS are:
  1. Macroscopic retrieving and analyzing nationally;
  2. As common basis data for other system to locate their special data;
  3. Reducing the time cycle for comiling and updating maps at 1:1,000,000 and smaller scales;
  4. Supporting remote sensing image processing
The data from the NLIS data bases have been used in following fields:
  1. As basic data of the Real-time Monitoring System of Flood Danger and Disaster;
  2. As basic for long-term earthquake forecasting research
  3. Management of radio frequency ranges
  4. Geographic quantity analysis for compiling the geomorphic Map of China at the scale of 1:4,000,000;
  5. Others.
The NLIS data will be applied to a wide range of fields by many national economy organizations in China.

The future of the NLIS
The NLIS will be founded in the RISM, NBSM in about 5 years. The subsystems in each province of China will be built up one after another. The data updating and its applications will be made continuously. A nation-wide data communication network will be formed gradually. Joint research on GIS will be taken on along with all colleagues in China and in the world.

References
  • ARC/INFO User’s Manual
  • USGS : Data Users Guide 1-7,1986
  • DCDSTF : The Proposed Standard for Digitial Cartographic Data, The American Cartog-rapher, Vol 15, No.1 , 1988
  • I. Wilski: Problems in Establishing and updating a Digital Cartographic Data base at the Millionth Scale, Nachichten as dem Karten-and vermessungswesen, SeriesII, No. 44, 1986
  • D.P. Bickmore : World Digital database for Environmental Science, Nachrichten aus dem karten-and Vermessungs-wesen , Serirs II No.44, 1986
  • W.Weber: Mass Digitization of Topographic – Cartographic Information, Nachichten aus dem Karten –and Vermessungswesen, SerirsII No.44,1986
  • W. Weber : Mass Digitization of Topographic-Cartographic Information, Nachrichten aus dem Karten-and Vermessungswesen, SerirsII, No.46,1987
  • Jiang Jingtong , : The Present Situation and Future of the National Land Information System of China, Proceedings of Asian-Pacific Conference, International Geography Union, Beijing , 1990
  • Jiang Jingtong , Su Shanwu, Liu Deqin etc.: The Design and Charcteristics of the 1:1m-scale map Data Base of the National land Information System in China, Proceedings of the 2nd International workshop on Geographical Information System, Beijing , 1990
  • Liu Deqin : The Architecture and Enchancement of ARC/INFO , Remote sensing Information, No.2, 1990