Remote Sensing Inversion Of Spatial And Temporal Dynamics Of Saltmarsh Wetlands Along Southern Coast Of China Based On Google Earth Engine

Coastal wetland is an important ecological function area.It has important ecosystem service functions Among them,Spartina alterniflora is also an important saltmarsh vegetation.S.alterniflora has high adaptability,high productivity,and the ability to reproduce quickly.It is widely distributed in coastal intertidal zones in many countries around the world.S.alterniflora has spread rapidly in the eastern coastal zone of China in the past forty years and is now widely distributed in the intertidal zone of coastal provinces from north(Liaoning province)to south(Hainan province).S.alterniflora invasion brings a variety of negative ecological effects,such as changing the original ecological environment of the coastal zone,reducing biodiversity,and increasing the vulnerability of ecosystems.Monitoring the growth of saltmarsh vegetation and the influence of environmental factors is very important to prevent and control the negative ecological impact of alien species S.alterniflora invasion and maintain coastal wetland ecosystems.In the southern provinces of China,S.alterniflora invasion has brought great harm to the conservation of native vegetation mangroves.At present,many satellite remote sensing change detection studies have applied time series-based remote sensing classification algorithms to farmland,mangroves and other ecosystems,and have achieved good classification results.The phenology of saltmarsh has obvious seasonal dynamics.Considering the phenology saltmarsh can improve the classification accuracy of intertidal vegetation,especially it can be well distinguished from evergreen vegetation mangroves in the same intertidal zone.In addition,the time span of historical Landsat satellite imagery in the United States is basically consistent with the time span of saltmarsh invasion and spread in China,which provides a good source of satellite data for historical dynamic remote sensing monitoring of S.alterniflora.The effects of environmental factors on the diffusion of saltmarsh were futher studied based on the spatial and temporal changes of saltmarsh.Because environmental factors can affect not only the survival and reproduction of individual organisms,but also the entire community.The frequency of flooding is an important environmental factor in the intertidal zone.Studies have shown that S.alterniflora has a strong ability to adapt to habitats,especially the tolerance to flooding,saltmarsh is distributed at different elevation gradients.Therefore,studying the response of the frequency of flooding to suitable habitats helps to grasp the factors and thresholds that affect its distribution,and plays an important role in predicting the potential distribution of saltmarsh in the intertidal zone.The phenology of saltmarsh can also directly reflect the influence of environmental factors on the growth of S.alterniflora.Many studies have used remote sensing data to monitor the growth status of vegetation and estimate key phenological parameters of vegetation.Field experiments have shown that the phenological traits of saltmarsh adapt to evolution of latitude gradients,and form a geographic gradient pattern that changes with latitude.However,the long-term trends and interannual variability of saltmarsh growth in large-scale and long-term are still unclear.It is necessary to focus on exploring the long-term trends and inter-annual variability of saltmarsh growth in large-scale and long-term.Based on the Continuous Change Detection and Classification(CCDC)algorithm and the Google Earth Engine(GEE)cloud computing platform,this study aims to obtain the vegetation cover maps of different years in the intertidal regions of the southern provinces of China and to analyze the spatio-temporal distribution of saltmarsh.The inundation frequency of the intertidal zone was obtained indirectly from historical remote sensing images to analyze the effect of the inundation frequency on the growth of saltmarsh.Then TIMESAT program was used to extract the phenological parameters of S.alterniflora at different locations and under flooding,and analyze the interannual variability of the phenological parameters of saltmarsh on a large scale.The main findings are as follows:(1)GEE is a tool that can solve large-scale vegetation classification.Combining CCDC algorithms can extract S.alterniflora from the intertidal zone.The results showed that the distribution area of saltmarsh decreased briefly in 2000.After that,the distribution area of saltmarsh has been increasing.In this study area,Zhejiang and Fujian have the most distribution and continue to increase.(2)The growth area of saltmarsh is approximately hump-shaped with the frequency of flooding.The area between 60-90%of flooding frequency is the largest,but the NDVI value is the largest at 20-30%.The annual NDVI of saltmarsh with time under low flooding frequency showed an increasing trend,and the annual NDVI of saltmarsh under extremely high flooding frequency showed a decreasing trend.(3)We used the NDVI time series and the Trend model in the TIMESAT program to determine key phenological parameters,from 1986 to 2018,and the results showed that the starting season of saltmarsh and the time at the peak of NDVI at different locations and different flooding frequencies have a tendency to move backward,and the end of growth has a tendency to move forward.The growing season is shortened.(4)This research had the following two major uncertainties.First,due to the existence of mixed pixels,there is a certain error in the area of the saltmarsh vegetation finally calculated.In the future,it may be possible to use Sentinel data or consider the method of mixed pixel decomposition to obtain a more accurate area.The second is that the flooding frequency in this study is only a static flooding frequency map.It does not consider the change of time series,and ignores the effects of sea level rise,surface decline,and sediment deposition.In the future,it may be possible to combine the tidal data to obtain the flooding frequency of coastal wetlands in different years,and to further study the impact of flooding frequency on salt marsh vegetation.