Analysis Of RS And GIS Technology Under The Support Of Gannan Citrus Orchard Area Dynamic Changes

Ganzhou City in Jiangxi Province is an important orange production base in China.Xinfeng County of Ganzhou City,as the birthplace of navel orange in Gannan Province,is known as "the hometown of Chinese Citrus".In recent years,the planting area and yield of citrus in south Jiangxi have been greatly reduced due to the effects of Huanglong disease,drought and freezing.According to the statistics of the fruit industry department of Xinfeng County,in 2011,the citrus area was 205.13km2.In 2014,the area of citrus was reduced to 182.4km2,and by 2015,the area of citrus was only 174km2.Gannan citrus was regarded as the pillar industry of Xinfeng County.The reduction of its output has caused huge economic losses to the fruit farmers in the whole county.Therefore,using scientific methods to grasp the dynamic process of citrus orchard area change in south Jiangxi in time and accurately,and to analyze the driving factors of citrus orchard yield reduction in an allround way,will have a profound impact on the rapid development of fruit industry and even agricultural economy in south Jiangxi.This paper takes Xinfeng County of Ganzhou City as the research object,selects HJ-1 and Landsat TM/OLI satellite multi-phase remote sensing image,and uses the object-oriented classification method to extract citrus orchard information.At the same time,GIS spatial analysis technique and multiple regression analysis model were used to analyze the change of citrus orchard area and its causes in 2009-2016,so as to realize remote sensing dynamic monitoring of citrus orchard area change in south Jiangxi.The main contents of this paper are as follows:(1)Study on the optimal temporal Phase of Information extraction in Citrus Orchard.Based on the HJ-1 images of four phases in the growing period of citrus in south Jiangxi,the best phase of citrus orchard identification was determined by phrenological analysis and spectral characteristics of main crops.The results show that the best time for identification and extraction of citrus orchards by remote sensing was citrus maturity period(middle of October to Late November).(2)Study on object oriented extraction of citrus orchard area.Based on the remote sensing images of Landsat on 2009 to 2016,the hierarchical extraction model of citrus orchards was constructed by using the object-oriented classification method.Combining the data of confusion matrix and statistical yearbook,the classification accuracy and the precision of citrus orchard extraction are tested.The results show that the precision of object oriented classification extraction of citrus orchard area is the lowest 85.79.The result of extraction accords with the actual situation.The classification accuracy of citrus orchard is improved by using object-oriented classification,which can be applied to citrus orchard extraction.(3)Analysis of dynamic change of citrus orchard area in study area.According to the objectoriented classification map and the transfer matrix of various land use types,the dynamic changes of land use in Xinfeng County are analyzed from a macro perspective.Combined with ArcGIS spatial superposition and statistical analysis model,the dynamic changes of citrus orchard area in different villages and towns in the study area were quantitatively analyzed.From 2009 to 2016,the change trend of citrus orchard planting area increased first and then decreased.From 2010 to 2016,the change of orchard area in Datangbu Town,Xinfeng County,was the biggest,the area of Jiading town orchard is the least.The change of citrus orchard area is closely related to natural disasters and agricultural development in recent years.(4)Analysis on driving factors of area change of Citrus Orchard in study area.Based on the statistical yearbooks of 2009 and 2016 of Xinfeng County and the data of Fruit and Tea Bureau,7 driving force indexes(annual mean temperature,topographic slope,number of Huanglong disease plants,population density,per capita garden area)were selected.A stepwise regression model of the change rate of citrus orchard area was established,and the main driving factors affecting the dynamic change of citrus orchard area were analyzed.The results showed that Huanglong disease was the main cause of the significant change of citrus orchard area.The annual mean temperature and per capita garden area were driving factors,among which the outbreak of citrus yellow dragon disease had a significant effect on citrus planting.