2022年克鲁伦河流域10米分辨率植被覆盖度月度数据集

doi:10.19788/j.issn.2096-6369.100032

摘要/Abstract

摘要：

精确获取流域范围内的植被覆盖度（Fractional Vegetable Cover, FVC），对于深入研究流域生态环境、湿地健康状况及其生态保护策略具有至关重要的意义。克鲁伦河流域是横跨中蒙边境的重要生态区，具有很高的生物多样性价值，对支撑并维护该区域生态系统平衡具有重要作用，鉴于此，本数据集以克鲁伦河流域作为研究区，基于10米空间分辨率的Sentinel-2多光谱遥感影像获取高精度植被覆盖度数据集，为流域生态环境保护提供数据支撑。为克服像元二分法、线性回归方法、随机森林回归模型等传统植被覆盖度反演方法在光谱特征间细微差异挖掘有效性与高维特征间复杂非线性关系发现不足等问题。为更精准估算该流域植被覆盖度，论文比较了基于深度学习的双向长短时记忆网络（Bidirectional Long Short-Term memory, BiLSTM）模型、随机森林回归、多层感知机与LSTM四个模型的性能以确定数据处理方法。所用特征数据以Sentinel-2多光谱数据为基础，综合光谱指数与高程数据，所反映植被相关信息包括叶绿素含量、水分状况以及地形地貌等。该特征数据集将进一步划分为训练集和测试集。比较结果表明，BiLSTM的R2和RMSE分别为0.716和0.103，综合性能最优。论文基于该模型生成了2022年克鲁伦流域的月度植被覆盖度数据集，包括12个月的克鲁伦河流域植被覆盖度反演结果组成，全部数据已经完成拼接和掩膜提取等操作。该数据集可用于评价克鲁伦河流域地表植被生长状况和生态系统健康状况，并为相关流域的生态保护研究提供支持。

数据摘要：

项目	描述
数据集名称	2022年克鲁伦河流域10米分辨率植被覆盖度月度数据集
所属学科	土地资源与信息技术
研究主题	植被覆盖度反演
数据时间范围	2022年度
数据地理空间覆盖	克鲁伦河流域，蒙古，中国
空间分辨率	10 m
数据类型与技术格式	.tif
数据库（集）组成	数据集包含12张图片；内容为2022年每月克鲁伦河流域植被覆盖度数据图片
数据量	148 GB
主要数据指标	植被覆盖度
数据可用性	https://cstr.cn/17058.11.sciencedb.agriculture.00026 https://doi.org/10.57760/sciencedb.agriculture.00026
经费支持	国家重点研发计划项目克鲁伦河流域面源污染遥感监测与评估技术研发 2021YFE0102300)

关键词: 克鲁伦河流域, 机器学习, 深度学习, BiLSTM, 植被覆盖度

Abstract:

Precisely obtaining the Fractional Vegetable Cover (FVC) at the river basin scale is of immense importance for delving into the ecological environment, wetland health, and ecological conservation strategies within watersheds. The Kherlen River Basin is an important ecological area across the border between China and Mongolia. It has high biodiversity and is essential for supporting and maintaining the balance of ecosystems in the region. Thus, this dataset focuses on the Kherlen River Basin, leveraging Sentinel-2 multispectral remote sensing imagery with a spatial resolution of 10 m to derive FVC with high precision. The dataset provides vegetable cover data to support the ecological protection of the Kherlen River Basin. In order to overcome the problem, traditional vegetation coverage inversion methods, such as pixel dichotomy, linear regression, and random forest regression models, could be more effective in mining subtle differences between spectral features and finding complex nonlinear relationships between high-dimensional features. To estimate the vegetation coverage more accurately in the watershed, this paper compares the performance of four models: the Bidirectional Long Short-Term Memory (BiLSTM) model based on deep learning, Random Forest Regression, Multilayer Perceptron, and LSTM, to determine the optimal data processing method. The feature data used are based on Sentinel-2 multispectral data, integrating spectral indices and elevation data. The vegetation-related information reflected includes chlorophyll content, moisture status, and topography. This feature dataset is further divided into training and testing sets. The comparison results show that BiLSTM achieved an R2 of 0.716 and an RMSE of 0.103, indicating the best overall performance. This model generated a monthly vegetation coverage dataset for the Kherlen River Basin in 2022, comprising vegetation coverage inversion results for 12 months. All data have undergone operations such as mosaicking and mask extraction. This dataset can assess the vegetation growth status and ecosystem health of the Kherlen River Basin and support ecological protection research in related watersheds.

Data summary:

Item	Description
Dataset name	A 10-m Fractional Vegetation Cover Monthly Dataset of the Kherlen River Basin in 2022
Specific subject area	Land resources and information technology
Research topic	Fractional Vegetation Cover
Time range	2022
Geographical scope	Kherlen River Basin，Mengolia
Spatial resolution	10 m
Data types and technical formats	.tif
Dataset structure	The dataset contains 12 images; the content is monthly images of vegetation cover data in the Kherlen River Basin for the year 2022.
Volume of dataset	148 GB
Key index in dataset	Fractional Vegetable Cover Index
Data accessibility	https://cstr.cn/17058.11.sciencedb.agriculture.00026 https://doi.org/10.57760/sciencedb.agriculture.00026
Financial support	Research and development on remote sensing monitoring and assessment technology of non-point source pollution in Kherlen River Basin under the National Key Research and Development Program (2021YFE0102300)

Key words: Kherlen River Basin, machine learning, deep learning, BiLSTM, fractional vegetable cover

牛博文, 冯权泷, 张毓, 高秉博, SUKHBAATAR Chinzorig, 冯爱萍, 杨建宇. 2022年克鲁伦河流域10米分辨率植被覆盖度月度数据集[J]. 农业大数据学报, 2025, 7(1): 59-68.

NIU BoWen, FENG QuanLong, ZHANG Yu, GAO BingBo, SUKHBAATAR Chinzorig, FENG AiPing, YANG JianYu. A 10-m Fractional Vegetation Cover Monthly Dataset of the Kherlen River Basin in 2022[J]. Journal of Agricultural Big Data, 2025, 7(1): 59-68.

图/表 8

图1

图2

图3

表1

图4

图5

图6

图7

参考文献 17

[1]	赵文慧, 陈妮, 闫瑞, 等. 近20年来北洛河流域植被覆盖度随地形因子变化特征探究. 水土保持研究, 2016, 23(4): 10-14.
[2]	陈妮, 李谭宝, 张晓萍, 等. 北洛河流域植被覆盖度时空变化的遥感动态分析. 水土保持通报, 2013, 33(3): 206-210.
[3]	程佳蕊, 毛德华, 玉山, 等. 1980—2020年克鲁伦河流域草地的时空变化. 水土保持通报, 2022, 42(3):296-303.
[4]	冯权泷, 陈泊安, 李国庆, 等. 遥感影像样本数据集研究综述. 遥感学报, 2022, 26(4): 589-605.
[5]	冯权泷, 牛博文, 朱德海, 等. 土地利用/覆被深度学习遥感分类研究综述. 农业机械学报, 2022, 53(3): 1-17.
[6]	冯权泷, 牛博文, 朱德海, 等. 2019年全国农业塑料大棚遥感分类数据集. 中国科学数据, 2021, 6(4): 153-170.
[7]	VERGER A, BARET F, WEISS M.GEOV2/VGT: Near real time estimation of global biophysical variables from VEGETATION-P data[C]//MultiTemp 2013:7th International Workshop on the Analysis of Multi-temporal Remote Sensing Images, Banff, AB, Canada,. 2013. DOI: 10.1109/Multi-Temp.2013.6866023.
[8]	LIANG S, CHENG J, JIA K, et al. The global land surface satellite (GLASS) product suite. Bulletin of the American Meteorological Society. 2021, 102: E323-37. DOI: 10.1175/BAMS-D-18-0341.1.
[9]	XIAO J, MOODY A. A comparison of methods for estimating fractional green vegetation cover within a desert-to-upland transition zone in central New Mexico, USA. Remote Sensing of Environment. 2005, 98: 237-250. DOI: 10.1016/j.rse.2005.07.011.
[10]	BRIAN J, TATEISHI R, KOBAYASHI T. Remote sensing of fractional green vegetation cover using spatially-interpolated endmembers. Remote Sensing, 2012, 4:2619-2634. DOI:10.3390/rs4092619.
[11]	LIU D, YANG L, JIA K, et al. Global fractional vegetation cover estimation algorithm for VIIRS reflectance data based on machine learning methods. Remote Sensing. 2018, 10:1648. DOI: 10.3390/rs10101648.
[12]	TU Y, JIA K, LIANG S, et al. Fractional vegetation cover estimation in heterogeneous areas by combining a radiative transfer model and a dynamic vegetation model. International Journal of Digital Earth, 2018, 13: 487-503. DOI: 10.1080/17538947.2018.1531438.
[13]	YU R, LI S, ZHANG B, et al. A deep transfer learning method for estimating fractional vegetation cover of sentinel-2 multispectral images. IEEE Geoscience and Remote Sensing Letters, 2021, 19: 1-5. DOI: 10.1109/LGRS.2021.3125429.
[14]	LIN X, CHEN J, LOU P, et al. Improving the estimation of alpine grassland fractional vegetation cover using optimized algorithms and multi-dimensional features. Plant Methods. 2021, 17:1-18. DOI: https://doi.org/10.1186/s13007-021-00796-5.
[15]	CHEN J, HUANG R, YANG Y, et al. Multi-scale validation and uncertainty analysis of GEOV3 and MuSyQ FVC products: a case study of an alpine grassland ecosystem. Remote Sensing. 2022: 5800. DOI: 10.3390/rs14225800.
[16]	ZHONG G, CHEN J, HUANG R, et al. High spatial resolution fractional vegetation coverage inversion based on UAV and Sentinel-2 data: A case study of Alpine Grassland. Remote Sensing, 2023, 15: 4266. DOI:10.3390/rs15174266.
[17]	LIU D, JIA K, XIA M, et al. Fractional vegetation cover estimation algorithm based on recurrent neural network for MODIS 250 m reflectance data. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing. 2021, 14: 6532-6543. DOI: 10.1109/JSTARS.2021.3075624.

特征指数	计算公式
NDVI	NDVI = (NIR−RED)/(NIR+RED)
EVI	EVI = 2.5×(NIR−RED)/(NIR+6×RED−7.5×BLUE+1)
SAVI	SAVI = (1+L)×(NIR−RED)/(NIR+RED+L)
RVI	RVI=NIR/RED
DVI	DVI=NIR−RED
GVI	GVI = (-0.2848×im_B2) + (-0.2435×im_B2) + (-0.5436×im_B4) + (0.7243×im_B8) + (0.0840×im_B11) + (-0.1800×im_B12)
LSWI	LSWI = (NIR − SWIR)/(NIR + SWIR)
GNDVI	GNDVI = (NIR − G)/(NIR + G)
GRVI	GRVI = NIR/G
GDVI	GDVI = NIR − G
NDBI	NDBI = (SWIR−NIR)/(SWIR+NIR)