中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
首页 机构 成果 学者
  
  1. CAS IR Grid
  2. 地理科学与资源研究所
  3. 中国科学院地理科学与资源研究所
  4. 千烟洲站森林生态系统研究中心
  5. 外文论文
Enhanced mapping of soil organic carbon in China's Black Soil Region using spectral-temporal-spatial fusion of multi-sensor satellite data

文献类型：期刊论文

作者Lv, Junwei3; Geng, Jing3,4; Wu, Yi3; Fang, Huajun1,5; Cheng, Shulan2; Pei, Jie3,4; Wang, Tianxing3,4
刊名SOIL & TILLAGE RESEARCH
出版日期2026-03-01
卷号257页码:106965
关键词Soil organic carbon Hyperspectral imagery Multispectral imagery Data fusion
ISSN号0167-1987
DOI10.1016/j.still.2025.106965
产权排序3
文献子类Article
英文摘要Soil organic carbon (SOC) is a critical determinant of soil health and agricultural sustainability in China's Black Soil Region, a vital grain production base facing severe SOC depletion due to intensive farming. However, largescale SOC monitoring remains challenging, as single-sensor satellite data fail to simultaneously capture fine spectral features and dynamic temporal changes in croplands. To address this, we propose a novel spectraltemporal-spatial (STS) fusion framework for SOC mapping, which integrates hyperspectral satellite imagery (166 bands, ZY-1 02D) with multi-temporal Landsat-8 multispectral data. We introduce a semi-supervised spectral-temporal-spatial fusion network based on regional energy-weighted discrete wavelet transform (RDWT-SSTSFN) to optimize the fusion process. This method first extracts temporal features from multi-temporal multispectral data using RDWT, then combines them with hyperspectral data in a dual-branch interactive learning model that captures complex spectral, temporal, and spatial relationships. Our results demonstrate that the RDWT-SSTSFN framework significantly reduces spectral uncertainty by approximately 20 % compared to using hyperspectral imagery alone (mean standard deviation reduced from 0.0588 to 0.0468). Meanwhile, the spectral features of the STS fusion data show stronger correlations with SOC, compared to single-source data. By evaluating four predictive models, including Random Forest (RF), Extreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost), and 1D Convolutional Neural Network (1D CNN), we demonstrate that RF models constructed with STS-fused data achieve the optimal SOC prediction and mapping performance. The framework enables high-resolution SOC mapping at 30 m, revealing spatially heterogeneous carbon patterns critical for targeted soil management. This study provides a scalable, satellite-based solution for monitoring SOC dynamics in intensive agricultural systems, directly supporting precision soil conservation and climate-smart farming practices.
URL标识查看原文
WOS关键词IMAGE FUSION ; PREDICTION ; AIRBORNE ; TEXTURE ; STOCK
WOS研究方向Agriculture
语种英语
WOS记录号WOS:001621761500001
出版者ELSEVIER
源URL[http://ir.igsnrr.ac.cn/handle/311030/217658]  
专题千烟洲站森林生态系统研究中心_外文论文
通讯作者Geng, Jing
作者单位1.Zhongke Jian Inst Ecoenvironm Sci, Jian 343000, Peoples R China;
2.Univ Chinese Acad Sci, Coll Resources & Environm, Beijing 100049, Peoples R China
3.Sun Yat Sen Univ, Sch Geospatial Engn & Sci, Zhuhai 519082, Peoples R China;
4.Sun Yat Sen Univ, Key Lab Comprehens Observat Polar Environm, Minist Educ, Zhuhai 519082, Peoples R China;
5.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Key Lab Ecosyst Network Observat & Modeling, Beijing 100101, Peoples R China;
推荐引用方式
GB/T 7714		 Lv, Junwei,Geng, Jing,Wu, Yi,et al. Enhanced mapping of soil organic carbon in China's Black Soil Region using spectral-temporal-spatial fusion of multi-sensor satellite data[J]. SOIL & TILLAGE RESEARCH,2026,257:106965.
APA Lv, Junwei.,Geng, Jing.,Wu, Yi.,Fang, Huajun.,Cheng, Shulan.,...&Wang, Tianxing.(2026).Enhanced mapping of soil organic carbon in China's Black Soil Region using spectral-temporal-spatial fusion of multi-sensor satellite data.SOIL & TILLAGE RESEARCH,257,106965.
MLA Lv, Junwei,et al."Enhanced mapping of soil organic carbon in China's Black Soil Region using spectral-temporal-spatial fusion of multi-sensor satellite data".SOIL & TILLAGE RESEARCH 257(2026):106965.

入库方式： OAI收割

来源：地理科学与资源研究所

浏览0
下载0
收藏0
其他版本

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。