239 
this study using the automatic classification and artificial 
interpretation as a method to translate the 2000 TM image of 
the study area (3) Data storage : Results of the interpretation 
were imported into ARC / INFO with vector format, getting the 
land-use data of the study area in 2000 with 1:100,000 scale; 
Based on the land-use data of this study area comparing the TM 
images of 2000 and 1986, getting the Data Dynamics of land- 
use in the 1986 -2000 and the land use / land cover data in 1986; 
Similarly comparing the MSS images of 1972 and the use of 
land use / land cover data of 1986,getting dynamic change of 
land use data and 1972 land use / land cover data. ©Accuracy 
evaluation : In view of the data interpretation process, in the 
principle of testing while interpreting, organizing experts to test 
the accuracy of interpretation at random for ensuring the results 
of the accuracy of interpretation; while the 2000 interpretation 
of the six types of land use types were selected for the 150 spots 
in a field verification, according to the Statistics of the types of 
land use the classification accuracy were about 94%, ensuring 
the accuracy of the work.. 
A +A„ 
1 
t2-/l 
- x 100% 
(3 ) 
Formula 3 is considered from the initial study to the end of 
research the bidirectional changes of the land use type i , and 
measures the dynamic changes, which may be known as the 
land use of a single type of comprehensive dynamic model. 
However, this model can not be compared to the output speed 
of land-use type i. In order to better calculate the types of land 
use changes, we propose the following model: 
V. m = 
A, n 
S(un 
x 
1 
/2-/1 
x 100% 
(4) 
2. RESEARCH METHODS 
2.1 The rate of land-use Changes 
The rate of land use / cover refers to the rate of areas changes 
of land use / cover types, which reflects the different types of 
land use change in the overall situation and the structure of land 
use changes. The mathematical expression of land use and land 
cover changes range is 
Formula 4 is described the speed of turn in from the initial 
study to the end of research type i of land use. In order to 
compare the type i of land use output speed and input speed, 
reflect the trends and changes in the state of land use and land 
cover types, We carried out under a deal with the state of Di 
index: 
V -V. 
_ out in 
V +v. 
r out 1 r in 
(-!<£><!) 
(5) 
* = (S W 2)-W/Swo x 100% 
(1) 
Where S tl) , S (i) g) are the areas of initial studies tl and 
end t2 of a certain type of land-use (GSD). 
2.2 Analyzing the Land Use / Cover Change speed and trend 
Land use / cover change could have quantitative description 
of the speed of LUCC , which have positive significance to 
compare the regional differences of LUCC and predict the 
future trends of the LUCC. The change of Land use is the 
different types of land use conversion. On a particular land use 
type i, its spatial pattern changes can be divided into three types: 
©The part of turn out (Août), i change for other types of land 
use types. ©The part of Turn in (Ain), other types of land use 
change to land use type i. ©The part of no changes (USi), land 
use type i have no change. Currently the most widely used of a 
single type of land use dynamic model: 
V = s (i M) US, x x 10Q% = A, 
S„. n /2-/1 S„ 
/2-/1 
- x 100% = F 
(2) 
These models just consider the changes of type i of land use 
to other types of land-use in one-way process. We may call the 
output speed of type i of land use, recorded as V out . This model 
exclusion changes of other types of land use, Therefore, the 
model for the transfer of those small, fast-growing land use 
types, especially in the urban land for the construction of the 
dynamic change is not applicable. Then scholars have proposed 
the following model [7,8] : 
At that time -1<£)<0, from the initial study to the end of 
research input speed of type i of land use is bigger than the 
output speed. Type i of land use is in the direction of increasing 
the scale of development. Di closer -1, input speed of the type i 
of land use is bigger than the output speed extremely, the 
situation is unbalanced, area increased substantially. Di closer 0, 
there is two existences, ©the input speed and the output speed 
of type i of land use are both very small, the input speed of type 
i of land use is slightly bigger than input speed, the increase of 
area is small, showing a balance situation; ©the input speed 
and the output speed of type i of land use are both big, input 
speed slightly bigger than the output speed, the bidirectional 
conversion is obvious, though the area increase rarely, the 
situation is balanceable, but this is a state of dynamic 
equilibrium based on the high speed of change. 
At that time0<D<l, from the initial study to the end of 
research input speed of type i of land use is smaller than the 
output speed, the area reduce, in the "downsizing" of the state. 
Di closer 1, the output speed is bigger than input speed slightly, 
a significant reduction in area, in a very unbalanced state. Di 
closer 0, the existence of two, ©the input speed and output 
speed of type i of are very small, however, the output speed is 
slightly bigger than the input speed and reduce the area is small, 
showing a balance situation; ©the input speed and output speed 
of type i of are enormous, the output speed is slightly bigger 
than input speed, the bidirectional conversion is obvious, 
although small area reduction, it is a state of equilibrium, But it 
based on the transfer of high input speed and high output speed, 
the circumstances of state is a dynamic equilibrium. 
From the right formula 2, 3, 4, 5 analysis, we can see that 
used alone with any one model can not be good for a single 
land use types and the changes described. Therefore, use the