黄河三角洲典型盐沼植被时空分布数据集（1999-2020）

以盐地碱蓬、芦苇、互花米草为代表的黄河三角洲典型盐沼植被，在维持生物多样性、提供重要栖息地、降低暴雨径流、调节气候等方面提供多种生态服务功能。作者以黄河三角洲河口湿地为研究区，基于Google Earth Engine(GEE)大数据平台，利用1999-2020年2068景Landsat TM/ETM/OLI、Sentinel-2 MSI光学数据和Sentinel-1 SAR雷达数据，结合盐沼植被的物候季相特征，利用特征优选算法筛选出最佳特征组合，利用随机森林算法进行植被分类，得到黄河三角洲典型盐沼植被时空分布数据集（1999-2000）。该数据集包括：（1）研究区范围数据；（2）1999-2020年13期盐沼植被类型空间分布数据；（3）1999-2020年互花米草、盐地碱蓬、芦苇的分布频率数据；（4）92个调查样点数据；（5）实测光谱数据。其中，栅格数据的空间分辨率为10 m。数据集存储为.shp、.tif和.xlsx格式，一共由64个数据文件组成，数据量为172 MB（压缩为1个文件，31.7 MB）。

Typical saltmarsh vegetation in the Yellow River Delta, represented by Suaeda salsa, Phragmites australis and Spartina alterniflora, provides multiple ecosystem services including maintaining biodiversity, providing critical habitats, reducing storm runoff and regulating climate. Taking the estuarine wetland of the Yellow River Delta as the study area and based on the Google Earth Engine (GEE) big data platform, the authors utilized 2068 scenes of Landsat TM/ETM/OLI, Sentinel-2 MSI optical data and Sentinel-1 SAR radar data from 1999 to 2020. Combined with the phenological and seasonal characteristics of saltmarsh vegetation, the optimal feature combination was screened out via feature selection algorithms, and vegetation classification was conducted using the random forest algorithm, thus generating the spatio-temporal distribution dataset of typical saltmarsh vegetation in the Yellow River Delta (1999–2000). This dataset includes five parts: (1) Study area boundary data; (2) Spatial distribution data of 13-period saltmarsh vegetation types from 1999 to 2020; (3) Distribution frequency data of Spartina alterniflora, Suaeda salsa and Phragmites australis from 1999 to 2020; (4) Data of 92 field survey sampling plots; (5) Field-measured spectral data. The spatial resolution of all raster data is 10 m. The dataset is stored in .shp, .tif and .xlsx formats, consisting of 64 data files in total, with an uncompressed total size of 172 MB (compressed into a single file of 31.7 MB).

搜集汇总

背景与挑战

背景概述

该数据集提供了1999年至2020年黄河三角洲典型盐沼植被（如碱蓬、芦苇和互花米草）的时空分布信息，基于Google Earth Engine平台融合多源遥感数据，采用特征选择算法和随机森林分类方法生成。数据集包含13期空间分布数据、分布频率数据、样本点和光谱数据，空间分辨率为10米，分类整体精度达86%-92.4%，适用于长期生态监测和湿地研究。