certain mathematical model for such a quantitative correlation 
between simulation to the estimated on the basis of inference 
and prediction, the offender usually this process is known as 
regression analysis. It can take advantage of certain 
mathematical model for such a quantitative correlation between 
simulation to the estimated on the basis of inference and 
prediction, the offender usually is known as regression analysis. 
In this paper, we use regression analysis of multivariate linear 
regression model to determine the area of land that the major 
driving factor; promote regional land-use sustainable 
development. 
income X3) Disaster category (collapsed area X4), and 
construction sites X5. 
Use each type of factor and cultivated areas for stepwise 
regression analysis, to determine the coefficient of such factors 
contribution rate arable land area by regression equations of the 
variable in order to determine changes in the main driving 
factors. Process to take into account various indicators 
dimensional data volume is relatively varied more, so first of 
all make raw data standardization by formula. 
4.2.2 Establish driving model 
We make experiment based on the statistical of the study area 
during 1976-2005, take a land type for example, select an area 
of arable land due to variable Y, and choose five categories, 
eight indicators as independent variables to analyze its driving 
mechanism: population category (the local population XI1, 
agricultural population XI2, the number of total households 
X13). Economic output (total agricultural output X21, 
industrial output X22), the standard of living (per capita 
Arable 
land area 
>;i; i¡ 
Total 
population 
end of year 
Agricultura 
1 population 
Total 
home end 
of year 
Agriculture 
output 
Industrial 
output 
Per capita 
income 
Collapse 
area 
Constructio 
n area 
-.12 
.21 
.26 
.15 
.37 
.45 
.47 
.54 
.56 
-.23 
.18 
.21 
.18 
.27 
.43 
.53 
.60 
.49 
-.17 
.22 
.24 
.11 
.18 
.37 
.39 
.46 
.41 
v=- 
y.-y 
(a-l H) 
(4-3) 
Ya is the dimensionless data for non-dimensional data, the 
average variable Syy is the deviation of the square and square 
root. The following table by the above formula standardized 
data: 
Fig.4-4.Standardized statistical data table 
ani rn 
By the use of standardized statistical data table stepwise 
regression analysis, obtained the regression equation: 
Y=0.350+0.453X11+0.369X21-0.789X4-0.537X5 
We can see that total population end of year; construction land 
and collapse area and agriculture have very important relation 
with the change of arable land, population end of year, the total 
output value of agriculture and farmland area is a positive 
correlation, and the collapse area, Construction sites, with the 
area of cultivated land is a negative correlation, it note that the 
increase in population and the increase in agricultural output 
led to the increase in the amount of arable land, and the 
collapse area and the construction area will inevitably lead to 
an increase in the decrease in the amount of arable land, And 
these two factors on changes in land area is greater than that of 
the local population and total output value of agriculture, Total 
causing the result of the decrease in arable land area. However, 
we still see a positive side, that the population and agricultural 
output value of the two drivers factor also plays an important 
role, This shows that the state of agricultural and rural 
investment is increasing, and especially for mine land 
reclamation and reconstruction increased the intensity of the 
treatment, it is gratifying. In addition, we also note that mining 
subsidence factor is cultivated acreage change the dominant 
factor, it is the need to reduce the coal mining subsidence and 
subsidence areas have land structure, this is the significance of 
the study lies. 
asumvB adtio aoridmonsriq vrit ol boom iso items 1 ism 
land use drivers analysis found: Mine as a special geographical 
area, due to resource development and the cumulative affects 
of ongoing, facing a serious ecological damage to the 
environment. Which mine land resources and the evolution 
triggered by a series of negative effects is the most serious 
problem impact of the mining area of sustainable development. 
Shenyang mine district need to reduce the coal mining 
subsidence and make land rectification for subsidence areas. 
Reference 
[1] Yin Zhang-cai ,Li Lin,AI Zi-xing, 2003. A study of spatio- 
temporal data model based on graph theory [Jj.ACTA 
GEODAETICA et CARTOGRAPHICAL SINICA,pp.32(2): 
168 
rf/v j .. ,, • 7 r, r ,;r r - t-j, ri i rvrlt l'i trr •f-rf'' iri* '7 
[2] Shu Hong,Chen jun,Du Daosheng., 1997.definition of spatio- 
temporal topological relationships and description of temporal 
topological relationships^]. .ACTA GEODAETICA et 
CARTOGRAPHICAL SINICA., pp.26 (4)299 
[3] LI Xiao-juan,YIN Lian-wang, 2002.CUI Wei-hong.spatio- 
temporal data model for landuse monitoring[J], Journal of 
remote sensing,pp.6(5): 370. 
[4] ZHANG Shan-shan, BIAN Fu-ling, 2003. Object-oriented 
Three-leveled spatiotemporal Data model[J].Computer 
Applications,pp.23(11):29-35. 
5 conclusions 
Using remote sensing and geographic information system as 
the core technology, After the Shenyang mine the land and 
[5]Lu Yangsheng, Qin chuan, Tang Bo, 2003. A new object- 
oriented spatio-temporal data model based on attribute-