基于WorldView-3多光谱和关键环境变量的土壤镉含量反演

李国旭; 耿静; 许选虹; 谭秋园; 郭一帆; 方华军

doi:10.11975/j.issn.1002-6819.2022.12.026

基于WorldView-3多光谱和关键环境变量的土壤镉含量反演

1.
中山大学测绘科学与技术学院，珠海 519082
2.
自然资源部华南热带亚热带自然资源监测重点实验室，珠海 519082
3.
中国科学院地理科学与资源研究所生态系统观测与模拟重点实验室，北京 100101
4.
中科吉安生态环境研究院，吉安 343000

基金项目: 国家自然科学基金项目（32101301）；广东省基础与应用基础研究基金项目（2020A1515110172）；中科吉安生态环境研究院院长基金项目（ZJIEES-2021-01）；井冈山农高区科技计划项目（井农科字〔2021〕51号）

计量
- 文章访问数: 322
- HTML全文浏览量: 8
- PDF下载量: 278
出版历程
- 收稿日期: 2022-04-18
- 修回日期: 2022-06-09
- 发布日期: 2022-06-29

Inversion of soil Cd content using WorldView-3 multispectral and key environmental variables

1.
School of Geospatial Engineering and Science, Sun Yat-sen University, Zhuhai 519082, China
2.
Key Laboratory of Natural Resources Monitoring in Tropical and Subtropical Area of South China, Ministry of Natural Resources, Zhuhai 519082, China
3.
Key Laboratory of Ecosystem Network Observation and Modeling, Institute of Geographical Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
4.
The Zhongke-Ji'an Institute for Eco-Environmental Sciences, Ji'an 343000, China

摘要

摘要: 快速准确获取农田土壤重金属含量对区域土地质量评估和粮食安全至关重要。该研究以江西省仙槎河流域小龙钨矿周边农田土壤为研究对象，采用WorldView-3多光谱影像提取光谱反射率并进行光谱变换处理，同时考虑了地形、人类活动和土壤属性等影响农田土壤镉（Cd）含量空间分布的关键环境因子，将光谱、环境变量、光谱与环境变量分别作为模型的自变量，选取了偏最小二乘（Partial Least Squares Regression，PLSR）、支持向量机（Support Vector Machines，SVM）、BP神经网络（Back Propagation Neural Network，BPNN）和随机森林（Random Forest，RF）4种回归算法构建土壤Cd含量预测模型，并利用精度评价指标优选出最佳反演模型。结果表明：仅输入多光谱特征进行Cd含量反演的模型精度总体偏低，R2低于0.2。相比之下，单独输入环境变量的模型精度结果最为理想，最优模型（RF）精度R2可达0.782。然而，融合光谱信息与环境变量共同建模后并未显著提高模型精度，反而导致较优模型（RF）精度略微降低，R2为0.693。研究结果表明，关键环境协变量是决定研究区农田土壤重金属Cd空间分布反演的重要变量，而利用多光谱信息进行土壤重金属反演的能力有限。此外，随机森林模型是预测土壤重金属空间分布的有效手段。
- 土壤 /
- 污染 /
- 重金属 /
- 机器学习 /
- WorldView-3 /
- 环境变量
Abstract: Abstract: A rapid and accurate detection of heavy metal content in farmland soils is crucial for land quality assessment and food security. In this study, 203 soil samples were collected from the farmland polluted by Xiaolong Tungsten Mine located in the Xiancha River watershed in Jiangxi Province, southern China. Soil Cadmium (Cd) content was measured using inductive coupling plasma mass-spectrometric (ICP-MS). High-resolution WorldView-3 multispectral imagery was used to extract the spectral reflectance and transformations, including the first order differential reflectance (FDR) and reciprocal logarithm spectra. A correlation analysis was performed to select the sensitive bands suitable for the prediction of soil cadmium (Cd) content. Different from previous studies that merely used the spectral information for modeling, the key environmental factors were also considered as the influence factors of the spatial distribution of Cd content, including the terrain factor (DEM), soil attribute factors (soil organic carbon and pH), and anthropogenic factors (distance to mine and residential area). Partial Least Squares Regression (PLSR), Support Vector Machine (SVM), Back Propagation Neural Network (BPNN), and Random Forest (RF) were used to construct the prediction models of soil Cd content. The best inversion model was selected by the accuracy metrics. The results showed that the transformation of WorldView-3 original spectral data by the first-order differential improved the correlation between spectral data and soil Cd content. However, the prediction accuracy remained low using the inversion model only with the spectral characteristic parameters. By contrast, the environmental covariates alone generated the best accuracy (R2=0.782, RMSE=0.384, MAE=0.294) using RF modelling. Surprisingly, the predictive performance was not significantly improved as expected, when integrating environmental covariates and the spectral information transformed by reciprocal logarithm for modelling, which instead resulted in a slight reduction in the accuracy of the optimal RF model (R2=0.693, RMSE=0.448, MAE=0.336). According to the variable importance ranking, it was found that the relative importance of the five key environmental variables was higher than 74%, which was significantly higher than that of the multispectral bands. Moreover, the model driven by the integration of environmental variables and spectral bands produced a similar spatial distribution trend of soil Cd content to that of the model driven by environmental variables alone from the perspective of spatial prediction. Both models showed that the Cd content in the farmland soil in the study area presented a high degree of spatial heterogeneity, both indicating an increasing distribution trend from the northwest to the southeast. In addition, the soil Cd content showed an increasing trend with the decrease of the distance from the mining area and enriched in the densely populated areas. Despite these similarities, the spatial prediction map with environmental variables alone presented the outstanding strip effect in the southeastern region of the study area. Contrastingly, there was better spatial continuity in the soil Cd map generated by integrating spectral information and environmental variables. These findings indicated that the key environmental covariates were important variables to predict the spatial distribution of heavy metals in farmland soil, whereas the capability of soil heavy metal retrieval using multispectral imagery alone was limited. In addition, the random forest was an effective way to predict the spatial distribution of heavy metals in soil.
- soils /
- pollution /
- heavy metals /
- machine learning /
- WorldView-3 /
- environmental variables

HTML全文

参考文献(44)

[1]	赵其国，骆永明. 论我国土壤保护宏观战略[J]. 中国科学院院刊，2015，30(4)：452-458.Zhao Qiguo, Luo Yongming. The macro strategy of soil protection in China[J]. Bulletin of Chinese Academy of Sciences, 2015, 30(4): 452-458. (in Chinese with English abstract)
[2]	宋姿蓉，俄胜哲，袁金华，等. 不同有机物料对灌漠土重金属累积特征及作物效应的影响[J]. 中国农业科学，2019，52(19)：3367-3379.Song Zirong, E Shengzhe, Yuan Jinhua, et al. Heavy metal accumulation in irrigated desert soils and their crop effect after applying different organic materials[J]. Scientia Agricultura Sinica, 2019, 52(19): 3367-3379. (in Chinese with English abstract)
[3]	Zhang R, Chen T, Zhang Y, et al. Health risk assessment of heavy metals in agricultural soils and identification of main influencing factors in a typical industrial park in northwest China[J]. Chemosphere, 2020, 252: 126591.
[4]	全国土壤污染状况调查公报[EB/OL]. https: //www. mee. gov. cn/gkml/sthjbgw/qt/201404/t20140417_270670_wh. Htm, 2014-04-17.
[5]	Rinklebe J, Antoniadis V, Shaheen S M, et al. Health risk assessment of potentially toxic elements in soils along the Central Elbe River, Germany[J]. Environment International, 2019, 126: 76-88.
[6]	张旺，高珍冉，邰粤鹰，等. 基于APCS-MLR受体模型的贵州喀斯特矿区水田土壤重金属源解析[J]. 农业工程学报，2022，38(3)：212-219.Zhang Wang, Gao Zhenran, Tai Yueying, et al. Source analysis of the heavy metals in paddy field soils in Karst mining areas of Guizhou using APCS-MLR receptor model[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2022, 38(3): 212-219. (in Chinese with English abstract)
[7]	Khosravi V, Doulati Ardejani F, Yousefi S, et al. Monitoring soil lead and zinc contents via combination of spectroscopy with extreme learning machine and other data mining methods[J]. Geoderma, 2018, 318: 29-41.
[8]	Tan K, Ma W, Wu F, et al. Random forest-based estimation of heavy metal concentration in agricultural soils with hyperspectral sensor data[J]. Environmental monitoring and assessment, 2019, 191(7): 1-14.
[9]	成永生，周瑶. 土壤重金属高光谱遥感定量监测研究进展与趋势[J]. 中国有色金属学报，2021，31(11)：3450-3467.Cheng Yongsheng, Zhou Yao. Research progress and trend of quantitative monitoring of hyperspectral remote sensing for heavy metals in soil[J]. The Chinese Journal of Nonferrous Metals, 2021, 31(11): 3450-3467. (in Chinese with English abstract)
[10]	Guan Q, Zhao R, Wang F, et al. Prediction of heavy metals in soils of an arid area based on multi-spectral data[J]. Journal of Environmental Management, 2019, 243: 137-143.
[11]	尹芳，封凯，吴朦朦，等. 一种基于分段偏最小二乘模型的土壤重金属遥感反演方法[J]. 遥感技术与应用，2021，36(6)：1321-1328.Yin Fang, Feng Kai, Wu Mengmeng, et al. A remote sensing estimation method for heavy metals in soil based on piecewise partial least squares model[J]. Remote Sensing Technology and Application, 2021, 36(6): 1321-1328. (in Chinese with English abstract)
[12]	王腾军，方珂，杨耘，等. 随机森林回归模型用于土壤重金属含量多光谱遥感反演[J]. 测绘通报，2021(11)：92-95.Wang Tengjun, Fang Ke, Yang Yun, et al. Multi-spectral remote sensing inversion of soil heavy metal content using random forest regression model[J]. Bulletin of Surveying and Mapping, 2021(11): 92-95. (in Chinese with English abstract)
[13]	贺军亮，韩超山，韦锐，等. 基于偏最小二乘的土壤重金属镉间接反演模型[J]. 国土资源遥感，2019，31(4)：96-103.He Junliang, Han Chaoshan, Wei Rui, et al. Research on indirect hyperspectral estimating model of heavy metal Cd based on partial least squares regression[J]. Remote Sensing for Land and Resources, 2019, 31(4): 96-103. (in Chinese with English abstract)
[14]	张霞，丁松滔，岑奕，等. 结合野外光谱与实验室光谱的土壤Pb含量反演方法研究[J/OL]. 武汉大学学报：信息科学版：1-9[2022-04-15]. doi:10.13203/j.whugis20200386.Zhang Xia, Ding Songtao, Cen Yi, et al. Soil heavy metal pb content inversion method by combining field with laboratory spectra[J/OL]. Geomatics and Information Science of Wuhan University: 1-9[2022-04-15]. doi: 10.13203/j.whugis20200386. (in Chinese with English abstract)
[15]	刘彦平，罗晴，程和发. 高光谱遥感技术在土壤重金属含量测定领域的应用与发展[J]. 农业环境科学学报，2020，39(12)：2699-2709.Liu Yanping, Luo Qing, Cheng Hefa. Application and development of hyperspectral remote sensing technology to determine the heavy metal content in soil[J]. Journal of Agro-Environment Science, 2020, 39(12): 2699-2709. (in Chinese with English abstract)
[16]	Wang X, Zhang F, Kung H T, et al. New methods for improving the remote sensing estimation of soil organic matter content (SOMC) in the Ebinur Lake Wetland National Nature Reserve (ELWNNR) in northwest China[J]. Remote Sensing of Environment, 2018, 218: 104-118.
[17]	朱阿兴，杨琳，樊乃卿，等. 数字土壤制图研究综述与展望[J]. 地理科学进展，2018，37(1)：66-78.Zhu Axing, Yang Lin, Fan Naiqing, et al. The review and outlook of digital soil mapping[J]. Progress in Geography, 2018,37(1): 66-78. (in Chinese with English abstract)
[18]	张甘霖，史舟，朱阿兴，等. 土壤时空变化研究的进展与未来[J]. 土壤学报，2020，57(5)：1060-1070.Zhang Ganlin, Shi Zhou, Zhu Axing, et al. Progress and perspective of studies on soils in space and time[J]. Acta Pedologica Sinica, 2020, 57(5): 1060-1070. (in Chinese with English abstract)
[19]	Zhu A X, Yang L, Li B, et al. Purposive Sampling for Digital Soil Mapping for Areas with Limited Data[M]. Dordrecht: Springer, 2008: 233-245.
[20]	周蓓蓓，郭江，陈晓鹏，等. 基于UNMIX模型的安徽大矾山废弃矿区土壤重金属源解析[J]. 农业工程学报，2021，37(24)：240-248.Zhou Beibei, Guo Jiang, Chen Xiaopeng, et al. Source apportionment of soil heavy metals in abandoned mining areas in Dafan Mountain of Anhui Province based on the UNMIX model[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(24): 240-248. (in Chinese with English abstract)
[21]	刘焕军，潘越，窦欣，等. 黑土区田块尺度土壤有机质含量遥感反演模型[J]. 农业工程学报，2018，34(1)：127-133.Liu Huanjun, Pan Yue, Dou Xin, et al. Soil organic matter content inversion model with remote sensing image in field scale of black soil area[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(1): 127-133. (in Chinese with English abstract)
[22]	Liu F, Wu H, Zhao Y, et al. Mapping high resolution national soil information grids of China[J]. Science Bulletin, 2022, 67(3): 328-340.
[23]	江西省土地利用管理局，江西省土壤普查办公室. 江西土壤[M]. 北京: 中国农业科技出版社，1991.
[24]	敖明，柴冠群，范成五，等. 稻田土壤和稻米中重金属潜在污染风险评估与来源解析[Jari兡乢嵩穬孩扴ｹ of ｨeavy ｭetals in the coastal soils under long-term reclamation[J]. Estuarine, Coastal and Shelf Science, 2014, 151: 310-317.
[25]	Peng Y, Kheir R B, Adhikari K, et al. Digital mapping oftoxic metals in Qatari soils using remote sensing and ancillary data[J]. Remote Sensing, 2016, 8(12): 1003.
[26]	Taghizadeh-Mehrjardi R, Fathizad H, Ali Hakimzadeh Ardakani M, et al. Spatio-temporal analysis of heavy metals in arid soils at the catchment scale using digital soil assessment and a random forest model[J]. Remote Sensing, 2021, 13(9): 1698.
[27]	Xiong X, Grunwald S, Myers D B, et al. Holistic environmental soil-landscape modeling of soil organic carbon[J]. Environmental Modelling & Software, 2014, 57: 202-215.
[28]	Keskin H, Grunwald S, Harris W G. Digital mapping of soil carbon fractions with machine learning[J]. Geoderma, 2019, 339: 40-58., 2022, 197: 106923.
[29]	李保杰，王思宇，周生路，等. 田块尺度下农田重金属污染特征及其源汇关系响应解析[J]. 农业工程学报，2018，34(6)：204-209.Li Baojie, Wang Siyu, Zhou Shenglu, et al. Heavy metal pollution characteristics and its response of source-sink relationship in agricultural soil at field scale[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(6): 204-209. (in Chinese with English abstract)
[30]	Zhong X, Chen Z, Li Y, et al. Factors influencing heavy metal availability and risk assessment of soils at typical metal mines in Eastern China[J]. Journal of hazardous materials, 2020, 400: 123289.
[31]	黄亚捷，李菊梅，马义兵. 土壤重金属调查采样数目的确定方法研究进展[J]. 农业工程学报，2019，35(24)：235-245.Huang Yajie, Li Jumei, Ma Yibing. Research progress of methods for determining sampling numbers of soil heavy metals survey[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(24): 235-245. (in Chinese with English abstract)
[32]	Zhou T, Geng Y, Ji C, et al. Prediction of soil organic carbon and the C: N ratio on a national scale using machine learning and satellite data: A comparison between Sentinel-2, Sentinel-3 and Landsat-8 images[J]. Science of The Total Environment, 2021, 755: 142661.
[33]	Fathololoumi S, Vaezi A R, Alavipanah S K, et al. Improved digital soil mapping with multitemporal remotely sensed satellite data fusion: A case study in Iran[J]. Science of The Total Environment, 2020, 721: 137703.
[34]	Wang F, Gao J, Zha Y. Hyperspectral sensing of heavy metals in soil and vegetation: Feasibility and challenges[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2018, 136: 73-84.
[35]	张霞，王一博，孙伟超，等. 基于铁氧化物特征光谱和改进遗传算法反演土壤Pb含量[J]. 农业工程学报，2020，36(16)：103-109.Zhang Xia, Wang Yibo, Sun Weichao, et al. Inversion of Pb content in soil based on iron oxide characteristic spectrum and improved genetic algorithm[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2020, 36(16): 103-109. (in Chinese with English abstract)
[36]	张秋霞，张合兵，刘文锴，等. 高标准基本农田建设区域土壤重金属含量的高光谱反演[J]. 农业工程学报，2017，33(12)：230-239.Zhang Qiuxia, Zhang Hebing, Liu Wenkai, et al. Inversion of heavy metals content with hyperspectral reflectance in soil of well-facilitied capital farmland construction areas[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2017, 33(12): 230-239. (in Chinese with English abstract)
[37]	Li C, Sanchez G M, Wu Z, et al. Spatiotemporal patterns and drivers of soil contamination with heavy metals during an intensive urbanization period (1989-2018) in southern China[J]. Environmental Pollution, 2020, 260: 114075.
[38]	焦文涛，蒋新，余贵芬，等. 土壤有机质对镉在土壤中吸附-解吸行为的影响[J]. 环境化学，2005，24(5)：545-549.Jiao Wentao, Jiang Xin, Yu Guifen, et al. Effects of organic matter on cadmium adsorption-desorption in three soils[J]. Environmental Chemistry, 2005, 24(5): 545-549. (in Chinese with English abstract)
[39]	Chen Y G, Ye W M, Yang X M, et al. Effect of contact time, pH, and ionic strength on Cd (II) adsorption from aqueous solution onto bentonite from Gaomiaozi, China[J]. Environmental Earth Sciences, 2011, 64(2): 329-336.
[40]	Wang L, Coles N A, Wu C, et al. Spatial variability of heavy metals in the coastal soils under long-term reclamation[J]. Estuarine, Coastal and Shelf Science, 2014, 151: 310-317.
[41]	Peng Y, Kheir R B, Adhikari K, et al. Digital mapping of toxic metals in Qatari soils using remote sensing and ancillary data[J]. Remote Sensing, 2016, 8(12): 1003.
[42]	Taghizadeh-Mehrjardi R, Fathizad H, Ali Hakimzadeh Ardakani M, et al. Spatio-temporal analysis of heavy metals in arid soils at the catchment scale using digital soil assessment and a random forest model[J]. Remote Sensing, 2021, 13(9): 1698.
[43]	Xiong X, Grunwald S, Myers D B, et al. Holistic environmental soil-landscape modeling of soil organic carbon[J]. Environmental Modelling & Software, 2014, 57: 202-215.
[44]	Keskin H, Grunwald S, Harris W G. Digital mapping of soil carbon fractions with machine learning[J]. Geoderma, 2019, 339: 40-58.

施引文献(6)

期刊类型引用(5)

1.	杨朗建，雷基林，王东方，王正江，刘康，孙亮. 喷油策略对高寒柴油机冷起动特性与环境适应性的影响. 农业工程学报. 2025(02): 76-84 . 本站查看
2.	徐萌，王俊，文奕钧，史志鹏，王坤. 小型农用柴油机油耗和排放的多目标优化. 中国农机化学报. 2022(02): 112-120 . 百度学术
3.	魏肖，鲍久圣，谭飞，袁晓明，阴妍，张磊. 矿用防爆柴油机瞬态工况特性及参数优化. 工矿自动化. 2022(02): 138-146 . 百度学术
4.	王俊，申立中，毕玉华，雷基林. 不同海拔下基于VNT驱动的EGR对轻型柴油机燃烧与排放的影响. 汽车工程. 2022(07): 1088-1097 . 百度学术
5.	李成越，何超. 高海拔环境下农用柴油机适应性分析与展望. 内燃机与配件. 2021(20): 72-73 . 百度学术