Remote Sensing Monitoring Research On The Succession And Degradation Of Typical Saline Vegetation Community

Wetlands are multifunctional ecosystems between terrestrial ecosystems and aquatic ecosystems,and enjoy the reputation of "kidneys of the earth".The halophytes growing in wetlands often have high salt tolerance,which can improve the saline soil of coastal wetlands,and at the same time resist the impact of waves and prevent coastal erosion.Due to the disturbance of natural and human factors,the halophyte vegetation in wetlands has undergone succession and degradation.Taking the coastal wetland of Liaohe Estuary as the research area,this paper used the advantages of remote sensing technology such as timeliness and large-scale observation to monitor the temporal and spatial changes of halophyte vegetation in the Liaohe Estuary wetland,and analyzed its driving mechanism,which is helpful for understanding and deepening the protection and restoration of wetland ecosystems.In this paper,Landsat,GF-1,Sentinel-2 and MOD09Q1 data were used as remote sensing data sources,the downscaled ESTARFM spatiotemporal fusion algorithm and SG filtering algorithm were used to reconstruct the high spatiotemporal NDVI data from 2001 to 2020,On this basis,the NDVI time series characteristics of typical wetland vegetation and NDVI spatial distribution characteristics of coastal wetlands in Liaohe Estuary in recent 20 years were obtained.The optimal feature combination was selected by calculating the J-M distance,and the object-oriented SVM classification method was used to obtain the distribution of coastal wetlands in Liaohe Estuary from 2001 to 2020.On this basis,the information on the succession and degradation of halophyte vegetation in the coastal wetland of Liaohe Estuary was extracted.Finally,correlation analysis was used to discuss the driving mechanism of the evolution of halophytic vegetation in the coastal wetland of Liaohe Estuary.The results show that:(1) The accuracy of fusion results obtained by the ESTARFM spatiotemporal fusion algorithm based on downscaling fusion is better than the traditional ESTARFM fusion algorithm.After multiple iterations of S-G filtering,the NDVI time series curve can truly reflect the vegetation growth situation.Based on the time series,the NDVI time series characteristics of typical wetland vegetation were obtained,and it was found that there were obvious spatial differences in the NDVI of coastal wetlands in Liaohe Estuary in the past 20 years.(2) After calculating the J-M distance,the best classification combination was determined to be NDVI from March to October,MNDWI,NDBI,SOS,EOS and LOS.The object-oriented SVM method was used for classification,and the classification results obtained in the three cases without introducing phenological parameters and multispectral indices,and the pixel-oriented SVM method were compared,and it is found that the object-oriented SVM method has the highest accuracy.(3) By obtaining the distribution map of coastal wetlands in Liaohe Estuary for 20 years and research on NDVI time series monitoring of wetlands,it is found that the succession and degradation of halophyte vegetation in coastal wetlands of Liaohe Estuary are distributed on both sides of the coast.The main manifestations are the alternating changes of tidal flat and Suaeda salsa from 2001 to 2005,the continuous expansion of Suaeda salsa from 2005 to 2010,the highest area of Suaeda salsa from 2010 to 2014,and the succession of Suaeda salsa to reed in area A and the tidal flats to reed in Area C.From 2014 to 2019,a large area of Suaeda degraded,and by 2020,the area of Suaeda has recovered.Through the NDVI time series analysis of Suaeda salsa and Reed,it is found that the overall trend of Suaeda degraded,and Reed shows an increasing trend.(4) Climate change and runoff affect the change of halophyte vegetation in Liaohe Estuary wetland.Among them,climatic factors have little effect on Suaeda salsa and Phragmites australis.Runoff has a greater impact on the growth of Suaeda salsa.