2004 
——— — 
‘aphic 
made 
elécts 
d this 
n the 
d soil 
n the 
18 soil 
user. 
alysis 
dable 
Some 
cision 
ystem 
mount 
ed by 
ledge. 
ledge 
ill be 
spatial 
nulate 
lels in 
'ed its 
lution 
bution 
| from 
ch are 
self or 
nately 
can be 
is the 
search 
vledge 
1aking 
The 
vledge 
F GIS 
| and 
nation" 
ledge? 
3IS 
ace oi 
n area, 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B7. Istanbul 2004 
  
irregular in shape, and  scatted in distribution. The 
overpopulation makes this even worse since a large farm field 
usually has to be divided into bits and pieces to meet all farmers 
need for sharing. This is particularly true in China and many 
overpopulated countries. The mode of digital agriculture that a 
large land evenly partitioned into regular grid is inapplicable in 
those regions. Moreover models are the main component that 
calculates fertilizer, water and pesticide application for different 
grids while expert knowledge is usually fixed in models. 
Knowledge lacks flexibility in maintenance. This also limits the 
extension of GIS use. The approach discussed here for using 
farm fields (grids) variability information and expert knowledge 
for enhancement of yields and reduction of risk in farm field 
management should be applicable over much of those regions. 
To offer an application system accessible to location-distributed 
users, a web-based spatial decision system with the integration 
of GIS and expert knowledge, GZ-AgriGIS is developed. 
Expert knowledge associated with different crops obtained from 
human expert and analysis models can lead to appropriate field 
management to any farm field no matter where the field locates. 
The novelty of GZ-AgriGIS is its integrated knowledge base, 
which contains information on most of agronomic knowledge. 
With the system run, it is possible to tap the complex spatial 
decision-making and gain an insight into the variety of options 
of management practices available to each piece of farm fields. 
It fits with uneven area thus it has more flexibility in practice, 
esp. in mountainous regions with scattered, small area and 
irregular farm fields. 
References 
À GIS application to enhance cell-based information modeling. 
losef Benedikt, Sebastian Reinberg, Leopold Riedl, Information 
Sciences 142 (2002):151-160 
Andre Zergera, David Ingle Smith. Impediments to using GIS 
for real-time disaster decision support. Computers, Environment 
and Urban Systems, 27 (2003) 123-141 
Chinese Academy of Agricultural Science (CAAS), 1993. 
Devel-opment Strategy of Agricultural Fundamental Science 
(Nong Ye Ji Chu Ke Xue De Fa Zhan Zhan Lue). Agricultural 
Science and Technology Publisher. Beijing (in Chinese) 
Geoffrey I. Webb, Jason Wells, Zijan Zheng. An experimental 
evaluation of integrating machine learning with knowledge 
acquisition. Machine Learning, 1999, 35(1): 5-23 
Guo, W.T., 1988. History of Chinese Agricultural Science and 
Technology (Zhong Guo Nong Ye Ke Xue Ji Shu Li Shi) 
Science and Technology Publisher of China (in Chinese) 
Harald Vacik, Manfred J. Lexer. Application of a spatial 
decision support system in managing the protection forests of 
Vienna for sustained yield of water resources. Forest Ecology 
and Management 143 (2001): 65-76 
Mchiael P. J.. A spatial decision support system prototype for 
housing mobility program planning. J. Geograph Syst (2001) 
3:49-67 
Peter B. Keenan. Spatial decision support systems for vehicle 
routing. Decision Support Systems, 1998, 22:65-71 
Shane Runquist, Naiqian Zhang, Randy K. Taylor. 
Development of a field-level geographic information system. 
Computers and Electronics in Agriculture, 31 (2001) 201-209 
T.A. Arentze, H.J.P. Timmermans. A spatial decision support 
system for retail plan generation and impact assessment. 
Transportation Research Part C 8 (2000) 361-380 
Acknowledgements 
This research was partially supported by Guangzhou municipal 
government.