中国科学院机构知识库网格系统: A CNN-LSTM Model for Soil Organic Carbon Content Prediction with Long Time Series of MODIS-Based Phenological Variables

A CNN-LSTM Model for Soil Organic Carbon Content Prediction with Long Time Series of MODIS-Based Phenological Variables

文献类型：期刊论文


作者	Zhang, Lei 1; Cai, Yanyan 1; Huang, Haili 1; Li, Anqi 1; Yang, Lin 1; Zhou, Chenghu 1,2
刊名	REMOTE SENSING
出版日期	2022-09-01
卷号	14 期号:18 页码:18
关键词	soil organic carbon land surface phenology deep learning predictive mapping convolutional neural network (CNN) long short-term memory (LSTM)
DOI	10.3390/rs14184441
通讯作者	Yang, Lin(yanglin@nju.edu.cn)
英文摘要	The spatial distribution of soil organic carbon (SOC) serves as critical geographic information for assessing ecosystem services, climate change mitigation, and optimal agriculture management. Digital mapping of SOC is challenging due to the complex relationships between the soil and its environment. Except for the well-known terrain and climate environmental covariates, vegetation that interacts with soils influences SOC significantly over long periods. Although several remote-sensing-based vegetation indices have been widely adopted in digital soil mapping, variables indicating long term vegetation growth have been less used. Vegetation phenology, an indicator of vegetation growth characteristics, can be used as a potential time series environmental covariate for SOC prediction. A CNN-LSTM model was developed for SOC prediction with inputs of static and dynamic environmental variables in Xuancheng City, China. The spatially contextual features in static variables (e.g., topographic variables) were extracted by the convolutional neural network (CNN), while the temporal features in dynamic variables (e.g., vegetation phenology over a long period of time) were extracted by a long short-term memory (LSTM) network. The ten-year phenological variables derived from moderate-resolution imaging spectroradiometer (MODIS) observations were adopted as predictors with historical temporal changes in vegetation in addition to the commonly used static variables. The random forest (RF) model was used as a reference model for comparison. Our results indicate that adding phenological variables can produce a more accurate map, as tested by the five-fold cross-validation, and demonstrate that CNN-LSTM is a potentially effective model for predicting SOC at a regional spatial scale with long-term historical vegetation phenology information as an extra input. We highlight the great potential of hybrid deep learning models, which can simultaneously extract spatial and temporal features from different types of environmental variables, for future applications in digital soil mapping.
WOS关键词	LAND-SURFACE PHENOLOGY ; CONVOLUTIONAL NEURAL-NETWORKS ; VEGETATION PHENOLOGY ; COVER DYNAMICS ; REGRESSION ; VARIABILITY ; SYSTEM
资助项目	National Natural Science Foundation of China[41971054] ; Leading Funds for First-Class Universities[020914912203] ; Leading Funds for First-Class Universities[020914902302] ; Postgraduate Research and Practice Innovation Program of Jiangsu Province[KYCX22_0109]
WOS研究方向	Environmental Sciences & Ecology ; Geology ; Remote Sensing ; Imaging Science & Photographic Technology
语种	英语
WOS记录号	WOS:000856760700001
出版者	MDPI
资助机构	National Natural Science Foundation of China ; Leading Funds for First-Class Universities ; Postgraduate Research and Practice Innovation Program of Jiangsu Province
源URL	[http://ir.igsnrr.ac.cn/handle/311030/184803]
专题	中国科学院地理科学与资源研究所
通讯作者	Yang, Lin
作者单位	1.Nanjing Univ, Sch Geog & Ocean Sci, Nanjing 210023, Peoples R China 2.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, State Key Lab Resources & Environm Informat Syst, Beijing 100101, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Lei,Cai, Yanyan,Huang, Haili,et al. A CNN-LSTM Model for Soil Organic Carbon Content Prediction with Long Time Series of MODIS-Based Phenological Variables[J]. REMOTE SENSING,2022,14(18):18.
APA	Zhang, Lei,Cai, Yanyan,Huang, Haili,Li, Anqi,Yang, Lin,&Zhou, Chenghu.(2022).A CNN-LSTM Model for Soil Organic Carbon Content Prediction with Long Time Series of MODIS-Based Phenological Variables.REMOTE SENSING,14(18),18.
MLA	Zhang, Lei,et al."A CNN-LSTM Model for Soil Organic Carbon Content Prediction with Long Time Series of MODIS-Based Phenological Variables".REMOTE SENSING 14.18(2022):18.

入库方式： OAI收割

来源：地理科学与资源研究所

下载0

A CNN-LSTM Model for Soil Organic Carbon Content Prediction with Long Time Series of MODIS-Based Phenological Variables

其他版本