Remote Sensing Phenology Monitoring Ommunity Of Sphagnum Community

Peatlands store more than 30%of terrestrial organic carbon.Sphagnum,as its most important vegetation component,plays a key role in absorbing atmospheric carbon.However,sphagnum is susceptible to climate.Changes in peatlands temperature and seasonal precipitation will alter the phenological cycle of the sphagnum community,but these phenological change rhyme are still unclear.Remote sensing provides effective technical support for regional vegetation phenology monitoring.This study focused on the globally important sphagnum bogs,Mer Bleue bog in Canada,as study area.In the study,spectral analysis methods were used to compare the canopy hyperspectral of bogs representative vegetation to obtain the spectral characteristic bands of sphagnum,and the ability of multispectral satellite sensors and remote sensing vegetation indexes in sphagnum identification were tested to assess the potential of remote sensing in sphagnum monitoring.Further,the adaptability of extracting sphagnum community remote sensed phenology based on remote sensing indexes time series was analyzed to obtain the remote sensing phenology of the sphagnum community and its temporal change trend.This study has carried out a study on the spectral characteristics of sphagnum and remote sensing phenology of the sphagnum community,in order to supplement the theoretical basis of monitoring bog vegetation with remote sensing,and to lay a solid foundation for the application of remote sensing technology to explore the feedback of climate change in bog ecosystem processes.The main conclusions of the study were as follows:?1?Hyperspectral analysis among S.magellanicum,and three vascular vegetation,S.americanus,P.engelmannii,and P.contorta in the sphagnum community in Mer Bleue bog found that their spectral differences were greatest in NIR?780-1350nm?.This study used derivative,logarithmic,and envelope removal spectral methods to achieve hyperspectral dimensionality reduction,and utilized the Mahalanobis distance obtained the identification characteristic spectral bands of S.magellanicum,which were737-1163 nm and 1235-1302 nm.?2?Spectral analysis of S.magellanicum at four multispectral satellite sensors showed that the spectral characteristics of S.magellanicum in NIR were stable and could be used for satellite vegetation identification.The ability of these four sensors to recognize S.magellanicum from strong to weak were:MSI?Sentinel-2A?>ALI?EO-1?>OLI?Lnadsat-8?>ASTER?Terra?.?3?In terms of remote sensing indexes,VI-NIR based vegetation indexes,e.g.NDVI?Normalized difference vegetation index?and EVI?Enhanced vegetation index?cannot effectively distinguish S.magellanicum from vascular vegetation.In the water indexes that rely on vegetation the spectral characteristics in NIR-SWIR?near infrared region-shortwave infrared region?,such as MSI?Moisture stress index?and NDII?Normalized difference infrared index?,S.magellanicum and vascular vegetation have obvious divisions,so they can realize the recognition of S.magellanicum.?4?The DL?Double logistic?reconstruction method effectively eliminated noise from the original MOD13Q1 DNVI and EVI time series data.Because the NDVI-DL value was easy to saturate during the vigorous vegetation growth season,and its ability to represent the turning point of the bog vegetation growth cycle was weak with large interannual scatters.Therefore,the EVI-DL time series curve was more suitable for extracting remote sensing phenology of the sphagnum community.In this study,three remote sensing phenological parameters of the sphagnum community in Mer Bleue bog were extracted with a threshold of 0.5:its SOS was at 93^th,EOS at 281^th and LOS was188d.Using the gross ecosystem productivity phonologies from the flux tower in Mer Bleue bog to assess the accuracy of remote sensing phenologies,and got the root mean square errors of SOS,EOS,and LOS were respectively:8.00 d,8.34 d,and 7.76 d.From 2001 to 2019,the SOS and EOS of the sphagnum community in Mer Bleue bog showed significant advance trends,with the rates of 1.01 d/a?R²=0.68,p<0.01?and0.76 d/a?R²=0.68,p<0.01?,respectively;LOS has a tendency to prolong,with a rate of 0.26 d/a?R²=0.10?.