Anrong, Dang 
  
Step 1: Supported by statistical function of GIS, classification map of single factor is obtained according to the 
factor, index, method, and model of first class regionalization. 
Step 2: Supported by spatial analysis function of GIS, integrated classification scheme is acquired by overlaying the 
multi-factors classification map obtained above. 
Step 3: Supported by synthesized assessment method, the integrated classification scheme is evaluated quantitative 
and qualitative criterions (He Guangwen, 1987). If the integrated classification scheme is perfect, the first 
class regionalization is determined. Otherwise, the factor, index, method, and model of first class 
regionalization must be adjusted at first, and then, repeat step one and step two again, until the result 
becoming perfect. 
Based on the first class regionalization obtained above, the subclass regionalization of China’s grain production system 
is determined in the following four steps by applying the uniting method from small region to large one. 
Step 1: In order to avoid the influence of data value and unit, some index data in the integrated database of spatial 
and attribute is standardized by a series of data processing to create synthesized index. 
Step 2: Based on the integrated database and synthesized index data, clustering map of both single factor and multi- 
factors are achieved by dynamic clustering analysis (Yao Jianxu, 1988). 
Step 3: Supported by spatial analysis function of GIS, integrated clustering scheme is acquired by overlaying the 
single factor and multi-factors clustering map achieved above. 
Step 4: Supported by synthesized assessment method, the integrated clustering scheme is evaluated by quantitative 
and qualitative criterions. If the integrated clustering scheme is perfect, the subclass regionalization scheme 
is determined. Otherwise, the factor, index, method, and model of subclass regionalization must be adjusted 
at first, and then, repeat step one to step three again, until the clustering result becoming perfect. 
Finally, the regionalization scheme of China’s grain production system can be determined by integrating the first class 
regionalization scheme with the subclass regionalization scheme. The final regionalization scheme of China’s grain 
production system contains nine first class regions and thirty four subclass regions. 
3.3 Achievement Output 
Generally, there are three kinds of achievement output: table listing, map layout, and text reporting. First of all, taking 
the comprehensive agriculture regionalization of China (Zhou Lisan, 1981, 1993) and the characteristic of agricultural 
natural resources and crop cultivation distribution in China (Zhu Zhongyu, 1992) as reference, the code system and 
name of each region can be defined and listed in table 2. Then, supported by polygon dissolving function of GIS, the 
regionalization map of China’s grain production system can be compiled as shown in figure 2 by using the coding 
system and pattern system at the same time. 
Bf ez 7 
94 I. 
Lu 22 
A A 
| 7 
Z 
E Northeast Region 7 
2 
[IT] Inner Mongolia Region 7 
Ex] Loess Plateau Region 7 ^ 
[SSI Huang-huai-hai Region 7 Co 
7 7 
ES Yanatse River Region ? Z 
7 à gl 77 
o 7 
7 
7 7 
7 
Southwest Region A 
[III South China Region 
[7] Gansu-Xinjiang Region 
[1 Ginghai-Tibet Region 
250 0 250 500 KM 
— 
  
Figure 2. Regionalization map of China's grain production system 
  
74 International Archives of Photogrammetry and Remote Sensing. Vol. XXXIII, Part B7. Amsterdam 2000.