A New Remote Sensing Vegetation Index For Coastal Saltmarsh Wetland

As a consequence of human activities and climate change,saltmarsh in coastal wetlands of China is facing with serious threats with its distribution,area and ecological services changing rapidly.It is of great theoretical and practical guidance to study how to monitor the coastal wetland saltmarsh quickly and accurately,in order to maintain the structure and function of saltmarsh ecosystem and sustainable utilization of natural resources of saltmarsh and to promote the sustained and healthy development of regional society.Coastal saltmarsh is flooded by tide periodically.Generally speaking,the spectral characteristic of saltmarsh with low coverage in satellite images is showed as a mixture of characteristics of saltmarsh and underlying wet soil.It is hard to extract vegetation from satellite images by using the currently prevailing vegetation indices and moisture indices efficiently,especially with low coverage(<30%).In this study,based on the measurement and analysis of saltmarsh vegetation(Phragmites australis?Scirpus mariqueter?Spartina alterniflora)spectrum in the Chongming Dongtan Bird National Nature Reserve,a vegetation index(Coastal Salt Mashes Vegetation Index,CSMVI)which can minimize the influence of underlaying moisture significantly was developed to map the distribution of saltmarsh in coastal zone.The accuracy and applicability of CSMVI in extracting saltmarsh vegetation from remote sensing images with Landsat-8 OOLI?GF1 WFV and SPOT-7 NAOMI had been verified.It results in dramatic improvement of accuracy in distinguishing coastal halophytes from other land covers in satellite images,which provides great potential for coastal wetland mapping.The results show that:(1)The spectrum differences of saltmarsh vegetation are mainly affected by vegetation type,vegetation coverage and underlying soil.The spectral reflectance of typical saltmarsh vegetation(Phragmites australis?Scirpus mariqueter?Spartina alterniflora)is significantly different in 533?560 nm and 683?751 nm.In the wavelength range of 350?730 nm,spectral reflectance of saltmarsh decreases with increasing coverage,but increases in the near infrared band after 730 nm.In the wavelength range of 350?950 nm,the spectral reflectance of saltmarsh increases with the tide level.When the tide level continues to rise,the spectral reflectance decreases sharply after the plants are gradually submerged.(2)According to the spectrum of remote sensing image and spectral data measured in the field,it shows that the spectral response of difference underlaying moisture to saltmarsh is mainly manifested in the visible blue band,green band,red band and near infrared band.This paper constructs a new vegetation index CSMVI,based on the analysis of commonly-used vegetation indices and moisture indices,and through multiplication between ARVI(Atmospherically Resistant Vegetation Index)and NDWI(Normalized Difference Water Index),the new index will weaken the influence of underlaying moisture.Field verification results show that the accuracy of CSMVI is 84.44%,which is higher than that other vegetation indices in the tidal area with high moisture.(3)The application results of CSMVI with GF1 WFV(spatial resolution of 16 m)and SPOT-7 NAOMI(spatial resolution of 6 m)images show that,in the Yangtze estuary and Hangzhou Bay,CSMVI is suitable for extraction the vegetation distribution using remote sensing images with different spatial resolutions in different coastal areas.CSMVI applied with higher spatial resolution will achieve higher accuracy.(4)This study only adjusts underlaying moisture,but other factors such as soil particle size,soil salinity and saltmarsh type are not considered.The influence of underlying soil will study in further study.In addition,for mixed-pixel problems,remote sensing images with higher spatial resolution will be tested.