Remote Sensing Monitoring And Analysis Of Invasion Plant Spartina Alterniflora In Modern Yellow River Delta

The coastal wetland is in the intertwined area between and land.The landscape structure is complex and the ecosystem is diversified.It plays an important role in improving the climate,maintaining the biodiversity and maintaining the regional ecological balance.Shandong Yellow River Delta National Nature Reserve is located in the youngest and most important coastal wetland distribution area——the modern Yellow River Delta,is a nature reserve to protect the rare waterfowl red-crowned crane and black-headed gulls,and the rare birds often inhabit reed,Suaeda and Tamarix.As a kind of plant species which has strong invasions in the coastal zone of China,Spartina alterniflora has spread in the modern Yellow River Delta,especially the Yellow River Delta National Nature Reserve in recent years,and has been introduced since 1990.It has brought bad influence to the development and utilization of coastal resources and the breeding habitat of rare birds in coastal wetlands.There are three salient features of Spartina alterniflora in the modern Yellow River Delta region,namely,the particularity of the distribution area determined by the growth habit,the spectral separability determined by the physical and chemical composition,and the high dynamic of temporal variation determined by the rate of growth and propagation.Based on the above characteristics,this paper is based on multi-source remote sensing data,including domestic GF data and PROBA CHRIS data,using GIS analysis technology,combined with field survey data and measured spectral data,carrying out the research on remote sensing monitoring methods of invasive plant Spartina alterniflora,providing scientific basis for the practice of beach conservation,and it is of great significance to protect the wetland and wetland ecology and maintain the ecological balance in the reserve.The main contents and conclusions of this paper are as follows:(1)According to the particularity of its growth habit,presenting a automatic extraction method of Spartina alterniflora based on GF data.According to the characteristics of Spartina alterniflora suitable for growing in the lower part of the climax to the lower part of mid-tide zone,developing a extraction method which combined the spatial position with decision tree classification.Firstly,using wetness component of tasseled cap transform of Landsat TM remote sensing image acquired in climax to extract growth landward boundary of Spartina alterniflora,and then mask the Spartina alterniflora growth area of the GF images;Finally,using decision tree classification method to extract growth area of Spartina alterniflora.Through the statistics and analysis,we found the whole modern Yellow River Delta total area of Spartina alterniflora about 3278.11 ha as of September 2015,mainly distributed in the west of previous flow path of the Yellow River,Wuhaozhuang,the southeast side of Gudong oilfield and both sides of the current Yellow River estuary,of which the Spartina alterniflora area of both sides of the current Yellow River estuary is widest,and the west of previous flow path of the Yellow River was found for the first time.(2)According to the spectral separability determined by the physical and chemical composition,proposing a method of the characteristic spectral band analysis and hyperspectral classification.Firstly,the continuum removal method was used to deal with the vegetation data of the field after quality control,and carrying out the spectral separability analysis of the wetland vegetation.Then,based on the separable characteristic band distribution of Spartina alterniflora,using artificial neural network classification method to extract the Spartina alterniflora information based on 2012 PROBA CHRIS image.Finally,using the confusion matrix to evaluate the accuracy of the extraction of Spartina alterniflora,the results show that the overall classification accuracy of the extracted result based on the selected feature band is 98.66%,which is higher than that of all bands.Through GIS analysis and statistics,it is found that the total area of Spartina alterniflora in the study area is about 544.66 ha in 2012.It is mainly distributed in the southeast side of the Gudong oilfield and both sides of the current Yellow River estuary.(3)According to the high dynamic of temporal variation,putting forward a method of automatic detection and extraction of Spartina alterniflora change information.Firstly,using change detection method of the multi-band principal component based on domestic GF data to obtain the information change of the Spartina alterniflora and Spartina alterniflora changes associated with vegetation types during the monitoring period.Then,using the SVM method to classify change information,and efficiently update 2016 the reserve of the terrain information based on the 2014original vector layer,to rapidly and effectively monitor information changes of Spartina alterniflora.Through the statistics and analysis,we found that compared with other vegetation types,Spartina alterniflora changed most violently,from 2547.73hm~2 in 2014 increase to 3332.4 hm~2 in 2016,and the average annual growth rate reached 392.55 hm~2.Spartina alterniflora has occupied almost the intertidal zone of the Yellow River Delta National Nature Reserve,and it competed with native vegetations.