Land Use Change In Tao River And Driving Force Analysis

Taohe River,a tributary of Danjiang,is situated at Qinling Mountains south foot,with a total length of 155 km and a total drainage area of 1210 km2.It successively flows through Shangnan County,Shangluo City,Yunyang District,Shiyan City,and Xichuan County,Nanyang City.In the past decade,the social economy,policies and regulations of the three counties have developed to varying degrees,then the land use change has been deepening.This paper studies the land-use change and driving force of Tao River in the past ten years,which is helpful to reveal the effect of ecological environment management,to provide a reliable theoretical basis for scientific development planning and land management measures,and to provide guidance for the future study of the area similar to high mountains,dense forests and deep valleys.In this paper,the remote sensing image of the study area is taken as the data source.Firstly,three different methods are used to extract land use information,including unsupervised classification,minimum distance method and object-oriented method in the supervised classification.The most accurate classification method is selected to classify the three phases of remote sensing image of the study area in 2008,2013 and 2018.Secondly,by calculating land use change variation degree,land use dynamic degree,land use degree and other indicators,the paper analyzes land use area structure change,degree change and spatial change,studies the source and direction of land transfer in different time and space,and summarizes the law of land evolution.Finally,combined with the social statistics data,the driving factor index system is established,and the driving force of land use change in the study area is analyzed qualitatively and quantitatively.The main conclusion are as follows:(1)Unsupervised classification,minimum distance method and object-oriented method in supervised classification respectively have the overall classification accuracy of 83.01%,85.07% and 84.6%.The average classification accuracy is more than 80%,and the interpretation effect is good.The kappa coefficients(KIA)of the three classification methods are 0.646,0.781 and 0.653 respectively,and kappa coefficients are all above 0.6.Among them,the overall classification accuracy and kappa coefficient of the supervised classification are slightly higher than the other two.(2)In 2008,2013 and 2018,the forest land area accounted for more than 3 / 4 of the total land-use area,accounting for 80.08%,79.30% and 81.88%,with an average of 80%.In the research period,the structure of land use type is stable,and the change of land use type is largely unused land,which shows that the unused land has been effectively used.Land use transfer mainly occurs between forest land and cultivated land.In the early stage of the study(2008-2013),the area of forest land to cultivated land is larger than that of cultivated land to forestland,and the loss of forestland is more serious.In the later period of the study(2013-2018),the conversion area of forest land to cultivated land is less than that of cultivated land to forestland,which indicates that the implementation area of forestation is large and the implementation effect is good.(3)Through qualitative analysis,it is found that population density is positively correlated with cultivated land area,GDP is positively correlated with building area,and the proportion of output value of agriculture,forestry,animal husbandry and fishery is positively correlated with forest land area.That is to say,the larger the population density,the greater the proportion of cultivated land;the higher the GDP,the bigger proportion of urban construction land;the higher the proportion of output value of agriculture,forestry,animal husbandry and fishery,the higher the proportion of forest land.At the same time,the fluctuation of output value is basically consistent with that of forest land.(4)Through quantitative analysis,it is found that policy factors and social factors are the main driving factors.From 2008 to 2013,the top three driving factors are: forestation,per capita GDP and distance to residential areas.From 2013 to 2018,the top three driving factors are: forestation,altitude and population density.Generally speaking,policy factors and social factors(population,GDP)are the main driving factors,and the acting force and influence of these two factors are on average stronger than natural factors and spatial factors.In addition,the study shows that social factors will increase the probability of change,and the forestation,GDP per capita and population density exp(b)are greater than 1.It shows that the implementation of the policy,the increase of GDP per capita and the increase of population density may increase the probability of change.