ISPRS, Vol.34, Part 2W2, “Dynamic and Multi-Dimensional GIS’’, Bangkok, May 23-25, 2001 
16 
LARGE SCALE GIS FOR A SUBURBAN TOWNSHIP OF BEIJING 
TO MODEL STRATEGIES FOR SUSTAINABLE AGRICULTURE ON FIELD LEVEL 
Georg BARETH 12 , Si JIN 23 , Tailai YAN 3 and Reiner DOLUSCHITZ 1 
’University of Hohenheim 
Department of Agricultural Economics (410 A) 
Sub-Group of Applied Computer Technologies in Agriculture 
70593 Stuttgart, GERMANY 
Tel/Fax: ++49-711-459-2841/3481 
E-mail: bareth@uni-hohenheim.de, agrarinf@uni-hohenheim.de 
2 China Agricultural University 
Sino-German Project 
College of Agricultural Resources & Environmental Sciences 
Beijing, 100094, P.R. CHINA 
Tel/Fax: ++86-10-6289-3423/3539 
E-mail: bareth@gmx.de 
3 China Agricultural University 
Department of Information Management 
Section: Information Systems and Remote Sensing 
Beijing, 100094, P.R. CHINA 
Tel/Fax: ++86-10-6289-2119/2332 
E-Mail: zhudg@hns.cjfh.ac.cn 
KEY WORDS: Large Scale, GIS, Sustainable Agriculture, Beijing, China 
ABSTRACT 
The Sino-German Project between the China Agricultural University and the University of Hohenheim, Germany, focuses on sustainable 
agriculture in the North China Plain. In the first phase of the project, an experiment field has been established near Beijing to investigate 
different agricultural practices and their impact on harvest and environment. Researchers from several departments are involved in the 
project: Agricultural Economics, Agricultural Informatics, Vegetable Science, Landscape Ecology, Phytomedicine, Plant Nutrition, Plant 
Production and Soil Science. Apart from the field experiment, the project focuses on regional modeling. For the township of 
Dongbeiwang, where the experiment field is situated, a GIS is established containing base geodata, statistical data, remote sensing 
data and the project database of the field experiment. This GIS enables the modeling of strategies for sustainable agriculture with large 
scale data. This paper reflects the experiences of data collection in China and presents the problems and the method of the set up of a 
large scale GIS for a township in China. 
INTRODUCTION 
The Sino-German Project focuses on the development and 
implementation of sustainable strategies in agriculture. 
Therefore, spatial modeling and regionalization for policy support 
as well as field studies are essential. For a suburban township of 
Beijing, Dongbeiwang, a large scale GIS is necessary to 
extrapolate the results of the field studies to estimate the impact 
of the research. Furthermore, a large scale township GIS 
enables the modeling on field level to simulate sustainable 
scenarios for agriculture. Based on our own experiences in 
China, the set up of a spatial information system in a large scale 
in China for modeling environmental aspects faces three major 
problems: 
• Information about large scale data 
• Access to large scale data 
• Information about map projections and coordinate systems 
The information about large scale data in China is very limited. 
Especially for foreigners or sino-foreign co-operations, such data 
are restricted due to data policy and are hardly accessible even 
so they are available. But also for Chinese it seems very difficult 
to get access to any spatial or statistical large scale data. The 
biggest problem is that there are no central institutions for data 
and information delivery. Therefore, the whole process of “What 
can I get where?” is like a puzzle work and needs an immense 
amount of time for driving around and asking people. After you 
finally acquired some data, there are still major problems to face. 
For spatial data, the most serious problem is that paper maps in 
large scale sometimes come without the information about 
coordinate systems and map projections and this information is 
almost impossible to get. So, is it finally impossible for sino- 
foreign projects to work on a large scale in China or are there 
still possibilities for large scale modeling of environmental 
aspects for sustainable development? The latter is very apparent 
in Chinese press . 
GIS DATABASE 
The acquisition of large scale data in China is, as already 
mentioned, very difficult. Some sort of land register maps exist in 
a scale of 1:2,500 and 1:10,000 including points of elevation and 
general topographical information. The date of the maps differs 
widely. On the basis of the 1:10,000 map, a soil and land use 
map is available as well, but the information quality is rather poor. 
In Fig.1, this digitized soil and land use map for the township of 
Dongbeiwang is shown. For the digitization, ArcView was used. 
The attribute tables of this map are still not finished due to 
translation problems. The different colors of the map indicate the 
different administrative units (villages) of the township. The 
polygons with the black outlines indicate the land use units for 
which it is described if the land use is agriculture, residential etc.. 
These land use units carry vague soil information as well. This 
map represents the spatial base information and is also used for 
the field campaigns. The problem due to the lack of information 
of x,y-coordinates and map projection was solved by a DGPS 
field campaign. The results of the first DGPS campaign are 
represented in Fig.1 as red lines and points. The DGPS 
information which has an accuracy of around 2 m is described in 
more detail in the second contribution of the MDGIS’01 (Bareth 
2001). Additional information of the DGPS mapping is the 
elevation with an accuracy of around 5 m.