The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part Bl. Beijing 2008 
1033 
(Figure 7). As we can see, the environment in most part of 
Ordos city is really tough, sub-watershed with good ecosystem 
health only counts 18% of the whole area (Table 6). These 
sub-watersheds mainly locate on the south bank plain of Yellow 
River or hill and gully area, and reducing soil erosion is an 
effective way to improve the ecosystem health in these areas. 
The area of assessment units under classIII takes 82% of the 
whole part, which locates mainly on Mu Us Desert, Hobq desert 
and plateau. Actions such as artificial sand precautions, 
afforestation, and soil and water conservation should be taken in 
these areas. Furthermore, with economic data from 2003 to 
2006, we can see, the study area is rich in coal resources and the 
relating industry accounts for a large proportion in the whole 
regional government revenue. However, during the process of 
coal mining, vegetation cover and landscape consistence are 
destroyed, the change of this situation needs good coordination 
of resource use, environment protection and population growth. 
Class 
Area Ratio 
I 
13% 
II 
5% 
III 
22% 
IV 
38% 
V 
22% 
Table 6. Area ratio of ecosystem health assessment result in Ordos city 
N 
e jg too m 
Figure 7. Classification map of ecosystem health in Ordos city 
4. CONCLUSIONS 
As we can see from this study, ecosystem healthy can be 
evaluated by Pressures-State-Effect-Response (PSER) model. In 
the process of assessment, easiness to quantify and being 
independent are the standard of indicator selection. Remote 
sensing and GIS technology provide excellent method to 
quantify indicators. Although parameters in process could be 
got from data inversion by remote sensing, ground monitoring 
data is also needed to improve inversion accuracy and 
availability. 
The indicators selection and system establishment varies in time 
and space, so the selected indicator should be time-varying, 
meanwhile, the difference in ecosystem types should result in 
the different selected key indicators and standards. The index 
system and a series of indicators established in this paper can 
just guide the ecosystem management in grassland area in 
temperate inland areas. Among the process of indicators 
selection and quantification, we can see, for some indicators, 
such as vigor and organization structure, it is easy to measure. 
But other indicators, such as resilience, nature response are hard 
to grasp, which needs further discussion and research. 
Besides the practicability of conceptual model, standard 
selection and indicator system building is also key to the 
success of ecosystem health assessment. This study only ranks 
the internal ecosystem health of study area, but does not build 
the correlation between ecological level and health values. A 
uniform assessment standard and relating indicator system must 
be built quickly with the help of remote sensing and GIS. 
This study adopted watershed as assessment unit, because each 
watershed has unique geomorphology and consistent ecology 
features, it is important in ecosystem health assessment. 
However, sub-watershed division would be different according 
to the scale of DEM data, so the method of grid assessment unit 
could be introduced to implement the spatial feature of 
ecosystem health assessment in further study. 
Overall, ecosystem health assessment is critical to regional 
environment protection and sustainable development, as a new 
research topic, combing traditional ecology principals with 
remote sensing, GIS technology, landscape ecology and 
ecosystem service evaluation, it would has great development. 
REFEREENCES 
Chen, Z.H., Wang, J., 2005. Establishing a ecosystem health 
model in arid and semi-arid area by using Remote sensing data. 
Remote Sensing Technology and Application, 20(6), pp. 
558-562. 
Chi, J., Wang, Z., Xu, X.J., 2007. Study on health Remote 
sensing Monitoring of ecosystem of Baxian Nature Reserve in 
Tianjin. Forest Inventory and Planning, 32(1), pp. 51-54. 
Costanza, R., Mageau, M., 1999. What is a healthy ecosystem? 
Aquatic Ecology, 33,pp. 105-115. 
EEA, 2005. EEA Core Set of Indicators Guide. EE A Technical 
Report, pp. 13-15. 
Liu, M.H., Dong, GH., 2006. Ecosystem health assessment and 
driving force analysis in Qinhuangdao area based on remote 
sensing. Geographical Research, 25(5), pp. 930-938. 
Mai, S.Z., Xu, S.J., Pan, Y.J., 2005. Application of the PSR 
model to the evaluation of wetland ecosystem health. Tropical 
Geography, 25(4), pp. 317-321. 
OEDC, 1993. OEDC core set of indicators for environmental 
performance reviews. Environment monographs, pp.5-10. 
Rapport, D.J., 1989. What constitute ecosystem health? 
Perspectives in Biology and Medicine, 33, pp. 120-132. 
Rapport, D.J., G, Bohm., 1999. Ecosystem health: the concept, 
the ISEH, and the important tasks. Ahead Ecosystem Health, 5, 
pp. 82-90. 
Tamara, S.Galloway., Rebecca, J.Brown., Mark, A.Browne.,