The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing 2008 
236 
was. Pastoralists were rarely in low-elevation areas in study area, 
and the soil and moisture conditions were good in sunny slope, 
Thus, vegetation capacity was higher, and anti-jamming 
capability was strong. 
H M0CW50U § 
I996-2W6 
lnrrei$e ratio 
§ WMmÊê 
hkwiNblacibcasc Brat 
Aspect 
© im 
10200 
©.»» 
ÛAW 
©500 
i>i№ 
0.W) 
orno 
Figure 6 Vegetation increase rate in different slope aspects and 
elevation in study area from 1996 to 2006 
Flat Vmì So«thô8St^trttiS<mîw««W8s.t.H<»tlm«{sj. 
Aspect 
Figure 7 Vegetation increase rate in different slope aspects and 
elevation in study area from 1996 to 2006 
4.2.2 The relationship between vegetation coverage change 
and the roads 
During 1977-1996, on the 19 buffer zones with a space 
between of 1km, the vegetation increase rate tended to rise first 
and then fall as the distance to road went farther. From 0 to 4km, 
it increased in a linear way, while at farther than 4km, it went 
down, especially when the distance was more than 8km. It is 
because during this period of time, the Longyangxia reservoir 
was set up, which is located on out of 8km to the roads, 
resulting in the decrease of the vegetation increase rate. As a 
whole, the vegetation decrease rate tended to drop as the 
distance to road increased, which decreased linearly with the 
correlation coefficient up to -0.928 ( for a level of significance 
0.05 ) during 0~4km, but change trend was irregular out of 4km, 
showing that there was no obvious relationship. 
to road km 
197? “ 19% Hm Increase mo EHI Deere«» rate 
Figure 8 Relationship of vegetation change rate and distance to 
road in study area from 1977 to 1996 
During 1996-2006, the vegetation increase rate, as a rule, rose 
gradually as the distance to road went farther, while the 
vegetation decrease rate tended to fall. From 0 to 4km, it 
decreased in a linear way with the correlation coefficient up to 
-0.912 (for a level of significance 0.05). Analysis on the density 
of residential spots showed that in a distance of 0~4km to road, 
there was 93.2% of the total residential spots, which confirmed, 
to some extent, the effect of human activities (mainly grazing) 
on vegetation. 
Figure 9 Relationship of vegetation change rate and distance to 
road in study area from 1996 to 2006 
4.2.3 The relationship between vegetation coverage change 
and the residents 
On the 14 buffer zones with a space between of 1km, the 
vegetation coverage change and the change rate had a clear 
relationship with the distance to residents. During 1977-1996, 
the vegetation increase rate was inclined to rise first and then 
fall as the distance to road goes farther. It is because during this 
period of time, the Longyangxia reservoir was set up and stored 
water, and then the vegetation coverage decreased inside the 
reservoir. Furthermore, the reservoir is located in a distance of 
8-19km to residents, resulting in the decrease of the vegetation 
increase rate. However, during 0~4km, the decrease rate of 
grasslands defined linearly, while it changed irregularly out of 
5km, which showed that there was no obvious relationship, and 
human activities had made slight effect on it. 
The closer the vegetation to the residents was, the more it 
degraded. Speaking of the two periods of 1977-1996 and 
1996-2006, the vegetation decrease rate tended to go down as 
the distance to resident increased. Nevertheless, during the