基于PCA的盐地碱蓬植被信息提取新指数研究

盐地碱蓬是海岸带一种重要的先锋植被,其时空分布信息是海岸带湿地生态系统保护及湿地修复工程的基础数据.为准确地获取碱蓬群落时空分布信息,构建一种盐地碱蓬提取指数十分必要.该文以辽东湾北部海岸带湿地为研究区域,基于Landsat8 OLI数据,采用主成分分析方法(PCA, principal components analysis)构建盐地碱蓬指数(suaeda principal components analysis index, SPCAI),并将其应用到多源遥感数据(Landsat5 TM、Landsat7 ETM+、Landsat8 OLI和Sentinel-2 MSI)中,最后利用SPCAI获取近21年辽东湾北部海岸带湿地碱蓬群落的时空分布信息.结果表明：1)经主成分分析的Landsat数据的第4波段(PCA₄)对盐地碱蓬的响应最为明显,且Landsat8 OLI数据的主成分分析结果明显优于TM和ETM+数据;2)基于Landsat8 OLI的PCA₄波段构建SPCAI,SPCAI对碱蓬的响应明显优于NDVI、SAVI、MSAVI,并...

Study on new index of Suaeda salsa vegetation information extraction based on PCA

Suaeda salsa is an important pioneer vegetation in the coastal zone, and the spatial and temporal distribution information of Suaeda salsa is the basic data of coastal wetland ecosystem protection and wetland restoration project. In order to accurately obtain the information on the temporal and spatial distribution of the Suaeda salsa community, it is necessary to construct a Suaeda salsa extraction index. Taking the northern coastal wetland of Liaodong Bay as the study area, Suaeda Principal Components Analysis Index (SPCAI) was constructed using PCA (Principal Components Analysis) method based on Landsat8 OLI. SPCAI was applied to multi-source remote sensing data (Landsat5 TM, Landsat7 ETM＋, Landsat8 OLI, and Sentinel-2 MSI), and finally performed application to obtain the temporal and spatial distribution information of Suaeda salsa in the northern coastal wetland of Liaodong Bay in the past 21 years. The results are shown as follows. 1) the 4th band (PCA4) response of Landsat data processed by PCA was the most obvious, and the principal component analysis results of Landsat8 OLI satellite data was significantly better than TM and ETM＋data. 2) SPCAI's response to Suaeda salsa was significantly better than NDVI, SAVI, MSAVI, and SPCAI has high extraction accuracy in the four types of data, which can effectively extract Suaeda in the intertidal zone, indicating that SPCAI has good applicability to multi-source remote sensing data. 3) The Suaeda salsa community showed a trend of degradation, and the community distribution mainly experienced three transitions from large patches to small patches scattered distribution(2000-2002, 2002-2011, 2011-2020).