Quantitative Inversion of Lithium Content and Exploration of Clay-Type Lithium Deposits Using Satellite Hyperspectral Remote Sensing in Tongchuan, Shaanxi Province
-
摘要:
为实现基于卫星高光谱影像的锂含量空间分布制图以支持找矿工作，以富锂黏土岩的实测光谱、XRD（X-ray Diffraction）、锂含量及ZY1-02D高光谱影像为基础，采用多算法结合的技术路线，构建锂含量反演模型.通过ETNN（Ensemble Transformer Neural Network）回归算法建立定量反演模型，结合SSEE-SAM（Spatial Spectral Endmember Extraction-Spectral Angle Mapper）识别矿化露头，并采用CORAL（CORrelation Alignment）进行光谱域校正，然后应用最优模型生成锂含量空间分布图.训练集R2=0.93、RPD=3.91、RMSE=110.13，测试集R2=0.89、RPD=3.08、RMSE=183.04，表明模型具有较高准确性和较强拟合能力；野外验证集R2=0.75、RPD=2.02、RMSE=263.86，表明模型具有很强的泛化能力.相关系数表明锂与蒙脱石、绿泥石关系最为密切，重要性分析表明2 132~2 350 nm波段为反演模型的关键波段.本研究构建了一套从光谱到锂含量的反演技术体系，可为铜川地区及滇中盆地、黔中盆地黏土型锂矿找矿勘查提供技术支撑.
Abstract:To achieve satellite-based hyperspectral mapping of lithium spatial distribution to support mineral exploration efforts, a lithium content inversion model was developed using a multi-algorithm approach, based on spectral data, XRD analysis, lithium content measurements, and hyperspectral imagery from the ZY1-02D site. The quantitative inversion model was established using the Ensemble Transformer Neural Network (ETNN) regression algorithm. This model was integrated with Spatial Spectral Endmember Extraction-Spectral Angle Mapper (SSEE-SAM) for mineralized outcrop identification and supplemented with the CORrelation Alignment (CORAL) algorithm for spectral domain correction. The quantitative inversion model was then applied to generate a spatial distribution map of lithium content. Training set R2=0.93, RPD=3.91, RMSE=110.13; validation set R2=0.89, RPD=3.08, RMSE=183.04, indicate high accuracy and strong fitting capability; field test set R2=0.75, RPD=2.02, RMSE=263.86, demonstrate robust generalization ability. Correlation coefficients indicate that lithium exhibits the strongest association with montmorillonite and chlorite. Importance analysis reveals the 2 132-2 350 nm wavelength band as critical for the inversion model. This study establishes a comprehensive inversion methodology linking spectral data to lithium content, providing technical support for exploration of clay-type lithium deposits in the Tongchuan region, Central Yunnan basin, and Central Guizhou basin.
-
图 1 地质背景：鄂尔多斯盆地大地构造位置（a；据何发岐等，2022修改），铜川地区地质简图（b）
Fig. 1. Geological background: Tectonic position of the Ordos basin (a; modified by He et al., 2022), geological schematic map of the Tongchuan area (b)
图 3 实测光谱数据重采样至ZY1-02D影像波段范围
a.样品光谱曲线；b.样品连续统去除光谱曲线
Fig. 3. Resampled measured spectral data to the ZY1-02D image band range
图 4 典型样品XRD矿物成分分析结果
Fig. 4. XRD mineral composition analysis results for typical samples
图 5 锂含量定量分析结果
a.训练集结果；b.验证集结果
Fig. 5. Quantitative analysis results for lithium content
图 6 模型应用与验证
a.铜川地区锂含量反演结果；b.东部细节图；c.西部细节图；d.野外验证组的定量反演结果对比；e.TC-16点野外露头照片；f. PC-05点野外露头照片
Fig. 6. Model application and validation
图 7 多元线性回归：散点图(a)和回归系数柱状图(b)
Fig. 7. Multiple linear regression: Scatter plot (a); bar chart of regression coefficients (b) 20.55CHL+185.91MGC+25.87DIA.(5)
图 8 最优模型的SHAP分析
Pos表示特征位置；SAI表示光谱吸收指数；Ratio_表示波段比值；_CR表示为连续统去除之后
Fig. 8. SHAP analysis of the optimal model
表 1 特征信息表
Table 1. Feature information table
波段范围(nm) 敏感波段比值(nm) 特征吸收峰(nm) 395~1 290(94个波段) 722/799 619~808 782/1 290 808~1 005 979/1 290 808~1 223 2 098~2 434(21个波段) 2 199/2 132 2 132~2 266 2 317/2 132 2 165~2 233 2 266~2 366 2 283~2 350 注：波段范围为395~902指的是这个范围内的波段(94个波段)；敏感波段比722/799指的是722波段反射率比上799波段反射率；特征吸收峰，以619~808为例，指的是这个范围内的Pos、Dep、Are和SAI四个参数.对连续统去除之后的光谱同样取这些参数，获得115个波段反射率，5个敏感波段比值，28个特征吸收峰参数. 
下载: 导出CSV
表 2 部分样品矿物含量(%)及锂含量(μg/g)
Table 2. Mineral composition (%) and lithium content (μg/g) of selected samples
ID Li MON CHL DIA ILL KAO MGC NAC NAM 22-BS1-1-1 2 150 8.5 9.6 9.5 56.0 2.0 1.4 0.0 8.5 22-BS1-2-3 850 0.0 0.9 0.2 0.0 96.2 0.7 0.0 0.0 22-BS1-2-4 3 290 8.5 6.9 28.5 0.6 35.9 2.7 0.0 0.0 22-CC2-2 1 840 5.9 5.4 3.9 0.0 82.0 1.6 0.0 3.6 22-FG-01 620 0.9 0.7 0.0 6.2 80.2 0.7 0.0 0.0 22-FG-01-1 530 0.0 8.4 0.0 9.0 77.8 0.4 4.0 0.0 22-FG-02 850 9.3 10.3 10.7 6.5 48.5 0.0 0.0 9.3 22-FG-03 154 2.9 9.2 0.5 8.4 55.5 0.5 0.0 2.9 22-FG-03-1 285 0.0 1.0 0.0 6.0 84.9 1.0 1.9 0.0 22-FG-03-2 570 0.0 8.4 0.0 4.4 79.5 0.5 3.7 0.0 22-FG-04 287 1.2 8.1 0.3 9.6 61.8 0.0 1.0 1.2 22-FG-05 550 0.0 3.5 0.6 7.0 77.8 0.4 0.0 0.0 22-FG-05-1 940 11.2 14.2 22.0 4.1 15.6 0.0 0.0 11.2 22-FG-06 490 1.9 5.7 0.4 7.3 67.4 0.7 0.0 1.9 22-FG-07 87.4 0.1 3.9 0.0 1.9 83.6 0.3 0.5 0.1 22-FG-08 325 1.1 3.1 0.0 4.4 72.7 0.3 0.0 1.1 22-FG-09 175.5 4.9 7.2 0.0 3.0 67.0 0.0 0.0 4.6 22-HC2-2 1 180 1.5 3.4 28.5 5.5 49.9 1.4 0.0 0.0 22-HC2-3 1 030 2.2 6.7 0.0 7.6 79.0 1.1 0.6 0.0 注：MON代表钙蒙脱石；NAM代表钠蒙脱石；ILL代表伊利石；KAO代表高岭石；NAC代表珍珠石；CHL代表绿泥石；MGC代表镁绿泥石；DIA代表水铝石.
下载: 导出CSV
表 3 RF、LGBM和XGB模型参数与结果
Table 3. Parameters and results for RF, LGBM, and XGB models
模型 随机森林（RF） LightGBM（LGBM） XGBoost（XGB） 模型参数 n_estimators=297 n_estimators=83 n_estimators=300 max_depth=4 max_depth=14 max_depth=8 min_samples_split=6 num_leaves=89 learning_rate=0.103 4 min_samples_leaf=3 learning_rate=0.150 2 subsample=0.24 max_features=log2 subsample=0.780 5 colsample_bytree=0.92 random_state=5 colsample_bytree=0.533 4 random_state=5 n_jobs=-1 random_state=5 n_jobs=-1 训练集 R2=0.75, RPD=2.00 R2=0.76, RPD=2.05 R2=0.99, RPD=24.18 验证集 R2=0.22, RPD=1.13 R2=0.42, RPD=1.32 R2=0.71, RPD=1.87 运行时间 1 444.66 s 511.69 s 1 428.39 s
下载: 导出CSV
-
Amer, R., El Mezayen, A., Hasanein, M., 2016. ASTER Spectral Analysis for Alteration Minerals Associated with Gold Mineralization. Ore Geology Reviews, 75: 239-251. https://doi.org/10.1016/j.oregeorev.2015.12.008 Balaram, V., Santosh, M., Satyanarayanan, M., et al., 2024. Lithium: A Review of Applications, Occurrence, Exploration, Extraction, Recycling, Analysis, and Environmental Impact. Geoscience Frontiers, 15(5): 101868. https://doi.org/10.1016/j.gsf.2024.101868 Bedini, E., 2019. Application of WorldView-3 Imagery and ASTER TIR Data to Map Alteration Minerals Associated with the Rodalquilar Gold Deposits, Southeast Spain. Advances in Space Research, 63(10): 3346-3357. https://doi.org/10.1016/j.asr.2019.01.047 Benson, T. R., Coble, M. A., Dilles, J. H., 2023. Hydrothermal Enrichment of Lithium in Intracaldera Illite-Bearing Claystones. Science Advances, 9(35): eadh8183. https://doi.org/10.1126/sciadv.adh8183 Benson, T. R., Coble, M. A., Rytuba, J. J., et al., 2017. Lithium Enrichment in Intracontinental Rhyolite Magmas Leads to Li Deposits in Caldera Basins. Nature Communications, 8: 270. https://doi.org/10.1038/s41467-017-00234-y Butler, B. M., Hillier, S., 2021. Automated Full-Pattern Summation of X-Ray Powder Diffraction Data for High-Throughput Quantification of Clay-Bearing Mixtures. Clays and Clay Minerals, 69(1): 38-51. https://doi.org/10.1007/s42860-020-00105-6 Chen, Z. X., Wang, Z. H., Bu, H. J., et al., 2026. A Geologically Information and Enhanced Graph Convolutional Network Method for Extracting Information on Lithium and Beryllium-Rich Pegmatites from Hyperspectral Imagery. Earth Science, 53(3): 1058-1065 (in Chinese with English abstract). Clark, R. N., Roush, T. L., 1984. Reflectance Spectroscopy: Quantitative Analysis Techniques for Remote Sensing Applications. Journal of Geophysical Research: Solid Earth, 89(B7): 6329-6340. https://doi.org/10.1029/JB089iB07p06329 Clark, R. N., Swayze, G. A., Livo, K. E., et al., 2003. Imaging Spectroscopy: Earth and Planetary Remote Sensing with the USGS Tetracorder and Expert Systems. Journal of Geophysical Research: Planets, 108(E12): 2002JE001847. https://doi.org/10.1029/2002JE001847 Cui, Y., Wen, H. J., Yu, W. X., et al., 2022. Study on the Occurrence State and Enrichment Mechanism of Lithium in Lithium-Rich Clay Rock Series of the Daoshitou Formation of Lower Permian in Central Yunnan. Acta Petrologica Sinica, 38(7): 2080-2094 (in Chinese with English abstract). doi: 10.18654/1000-0569/2022.07.16 Dai, J. J., Wang, D. H., Ling, T. Y., 2019. Quantitative Estimation of Content of Lithium Using Reflectance Spectroscopy. Remote Sensing Technology and Application, 34(5): 992-997 (in Chinese with English abstract). Di, S. J., Zhang, G. S., Wen, H. J., et al., 2025. Geochemical Characteristics of the Lithium-Rich Clayston in the Benxi Formation in the Southeastern Margin of the Ordos Basin and Its Sedimentary Environment. Bulletin of Mineralogy, Petrology and Geochemistry, 44(5): 1081-1095 (in Chinese with English abstract). doi: 10.3724/j.issn.1007-2802.20250033 Dufréchou, G., Grandjean, G., Bourguignon, A., 2015. Geometrical Analysis of Laboratory Soil Spectra in the Short-Wave Infrared Domain: Clay Composition and Estimation of the Swelling Potential. Geoderma, 243/244: 92-107. https://doi.org/10.1016/j.geoderma.2014.12.014 Feng, L. X., Zhao, Z. F., Yang, H. Y., et al., 2025. Clay-Hosted Lithium Exploration in the Wenshan Region of Southeastern Yunnan Province, China, Using Multi-Source Remote Sensing and Structural Interpretation. Minerals, 15(8): 826. https://doi.org/10.3390/min15080826 Geng, X. L., Zhang, X. R., Wang, K. B., et al., 2025. Discovery of Lithium-Rich Sedimentary Rocks in Mesoproterozoic Wumishan Formation in Eastern Hebei Province and Implications for Mechanism of Lithium Enrichment and Mineralization. Earth Science, 50(7): 2689-2706 (in Chinese with English abstract). Guo, C. L., Zhang, B. W., Zheng, Y., et al., 2024. Granite-Type Lithium Deposits in China: Important Characteristics, Metallogenic Conditions, and Genetic Mechanism. Acta Petrologica Sinica, 40(2): 347-403 (in Chinese with English abstract). doi: 10.18654/1000-0569/2024.02.02 Guo, Y., Zhang, H. D., Kong, H. L., et al., 2026. Deep Learning for Hyperspectral Prediction of Rare Earth Oxide Grades in Drill Cores: A 1D CNN-BiLSTM Framework Applied to the Weishan Deposit, China. Ore Geology Reviews, 188: 107025. https://doi.org/10.1016/j.oregeorev.2025.107025 He, F. Q., Wang, J., Zhao, Y. Q., et al., 2022. Accumulation Characteristics of Dongsheng Helium-Rich Gas Field in Ordos Basin and Its Tectonic Background. Journal of Palaeogeography (Chinese Edition), 24(5): 937-950 (in Chinese with English abstract). Hu, P., Yu, X. H., Wang, J., et al., 2017. Chemical Characteristics of Benxi Formation Water and Its Significance for Natural Gas Accumulation in Southeast Ordos Basin. Journal of Northwest University (Natural Science Edition), 47(1): 92-100, 109 (in Chinese with English abstract). Kesler, S. E., Gruber, P. W., Medina, P. A., et al., 2012. Global Lithium Resources: Relative Importance of Pegmatite, Brine and Other Deposits. Ore Geology Reviews, 48: 55-69. https://doi.org/10.1016/j.oregeorev.2012.05.006 Kruse, F. A., Heidebrecht, K. B., Shapiro, A. T., et al., 1993. The Spectral Image Processing System (SIPS) Interactive Visualization and Analysis of Imaging Spectrometer Data. Remote Sensing of Environment, 44(2-3): 145-163. https://doi.org/10.1016/0034-4257(93)90013-N Li, Y., Zhang, J. W., Li, J., et al., 2014. A Study on Sedimentary Microfacies of Benxi Formation and Its Controlling Effect on Gas Enrichment in Yanchang District of Ordos Basin. Northwestern Geology, 47(2): 216-222 (in Chinese with English abstract). Ling, K. Y., Wen, H. J., Han, T., et al., 2024. Lithium-Rich Claystone in Pingguo Area, Guangxi, Southwest China: Precursor Kaolinite Controls Lithium Enrichment. Mineralium Deposita, 59(2): 329-340. https://doi.org/10.1007/s00126-023-01210-x Liu, L., Mei, J. C., Wen, H. J., et al., 2025. Clay Minerals and Lithium Content Quantitative Analysis Based on SWIR Spectra for Carbonate-Hosted Clay-Type Lithium Deposits, Central Yunnan. Acta Geologica Sinica, 99(9): 3072-3083 (in Chinese with English abstract). Liu, L., Yin, C. T., Shaheen Khalil, Y., et al., 2024a. Alteration Mapping for Porphyry Cu Targeting in the Western Chagai Belt, Pakistan, Using ZY1-02D Spaceborne Hyperspectral Data. Economic Geology, 119(2): 331-353. https://doi.org/10.5382/econgeo.5045 Liu, L., Yin, C. T., Wen, H. J., et al., 2026. Mapping Lithium Hosting Claystone Using WorldView-3 Image and Deep Learning Method. Gondwana Research, 151: 18-35. https://doi.org/10.1016/j.gr.2025.10.012 Liu, T. J., Duan, S. B., Liu, N. T., et al., 2024b. Estimation of Crop Leaf Area Index Based on Sentinel-2 Images and PROSAIL-Transformer Coupling Model. Computers and Electronics in Agriculture, 227: 109663. https://doi.org/10.1016/j.compag.2024.109663 Mei, J. C., Wang, X., Zhang, H. R., et al., 2025. Rapid Estimation of Cobalt Content in Lateritic Cobalt Ores: A Quantitative Inversion Study of VNIR-SWIR Spectra. Spectroscopy and Spectral Analysis, 45(9): 2578-2584 (in Chinese with English abstract). Rogge, D. M., Rivard, B., Zhang, J., et al., 2007. Integration of Spatial-Spectral Information for the Improved Extraction of Endmembers. Remote Sensing of Environment, 110(3): 287-303. https://doi.org/10.1016/j.rse.2007.02.019 Rogge, D., Bachmann, M., Rivard, B., et al., 2012. Spatial Sub-Sampling Using Local Endmembers for Adapting OSP and SSEE for Large-Scale Hyperspectral Surveys. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 5(1): 183-195. https://doi.org/10.1109/JSTARS.2011.2168513 Rowan, L. C., Schmidt, R. G., Mars, J. C., 2006. Distribution of Hydrothermally Altered Rocks in the Reko Diq, Pakistan Mineralized Area Based on Spectral Analysis of ASTER Data. Remote Sensing of Environment, 104(1): 74-87. https://doi.org/10.1016/j.rse.2006.05.014 Sun, B. C., Feng, J. S., Saenko, K., 2016. Return of Frustratingly Easy Domain Adaptation. Proceedings of the AAAI Conference on Artificial Intelligence, 30(1): 2058-2065. https://doi.org/10.1609/aaai.v30i1.10306 Sun, B. C., Saenko, K., 2016. Deep CORAL: Correlation Alignment for Deep Domain Adaptation. Computer Vision-ECCV 2016 Workshops. Springer, Cham. https://doi.org/10.1007/978-3-319-49409-8_35 Wang, J. N., Zheng, L. F., Tong, Q. X., 1996. The Spectral Absorption Identification Model and Mineral Mapping by Imaging Spectrometer Data. Remote Sensing of Environment, 11(1): 21-31 (in Chinese with English abstract). Wen, H. J., Luo, N., Du, S. J., et al., 2020. Carbonate-Hosted Clay-Type Lithium Deposit and Its Prospecting Significance. Chinese Science Bulletin, 65(1): 53-59 (in Chinese with English abstract). doi: 10.1360/TB-2019-0179 Yuan, C. J., Zhao, J., Liu, Z. B., et al., 2025. Detecting Mineralization via White Mica Spectral Footprints: From Field-Based Sampling to Satellite Hyperspectral Remote Sensing. Ore Geology Reviews, 181: 106588. https://doi.org/10.1016/j.oregeorev.2025.106588 Zhang, X. L., Zhao, Z. F., Chen, Q., et al., 2023. Mapping Hydrothermal Alteration of the Pulang Porphyry Copper Deposit, SW China, Using ASTER and ZY1-02D Satellite Data. Ore Geology Reviews, 161: 105605. https://doi.org/10.1016/j.oregeorev.2023.105605 Zhang, Z. F., Li, N., Fan, X. L., et al., 2025. Preliminary Study of the Occurrence of Lithium in the Ertanggou Deposit, East Tianshan. Earth Science, 50(2): 585-595 (in Chinese with English abstract). Zhao, H., Wu, Z. X., Zhang, S. Q., et al., 2023. Geochemical Features of Lithium-Rich Bauxite from the Benxi Formation in Qinyuan County, Shanxi, China: Insights into Their Depositional Environment and Lithium Enrichment. Ore Geology Reviews, 163: 105780. https://doi.org/10.1016/j.oregeorev.2023.105780 Zuo, L., Wang, G. W., Carranza, E. J. M., et al., 2022. Short-Wavelength Infrared Spectral Analysis and 3D Vector Modeling for Deep Exploration in the Weilasituo Magmatic-Hydrothermal Li-Sn Polymetallic Deposit, Inner Mongolia, NE China. Natural Resources Research, 31(6): 3121-3153. https://doi.org/10.1007/s11053-022-10111-1 陈志行, 王正海, 卜浩坚, 等, 2026. 基于地质信息的改进GCN高光谱富锂铍伟晶岩信息提取方法. 地球科学, 53(3): 1058-1065. 崔燚, 温汉捷, 于文修, 等, 2022. 滇中下二叠统倒石头组富锂黏土岩系锂的赋存状态及富集机制研究. 岩石学报, 38(7): 2080-2094. 代晶晶, 王登红, 令天宇, 2019. 基于地面反射波谱技术的锂含量定量反演研究. 遥感技术与应用, 34(5): 992-997. 邸思佳, 张贵山, 温汉捷, 等, 2025. 鄂尔多斯盆地东南缘本溪组富锂黏土岩地球化学特征及其沉积环境. 矿物岩石地球化学通报, 44(5): 1081-1095. 耿晓磊, 张晓容, 王克冰, 等, 2025. 冀东中元古代雾迷山组富锂岩系的发现及其对锂富集成矿机制的启示. 地球科学, 50(7): 2689-2706. doi: 10.3799/dqkx.2024.125 郭春丽, 张斌武, 郑义, 等, 2024. 中国花岗岩型锂矿床: 重要特征、成矿条件及形成机制. 岩石学报, 40(2): 347-403. 何发岐, 王杰, 赵永强, 等, 2022. 鄂尔多斯盆地东胜富氦气田成藏特征及其大地构造背景. 古地理学报, 24(5): 937-950. 胡鹏, 于兴河, 王娇, 等, 2017. 鄂尔多斯盆地东南部本溪组地层水化学特征与天然气成藏意义. 西北大学学报(自然科学版), 47(1): 92-100, 109. 李云, 张建伍, 李晶, 等, 2014. 鄂尔多斯盆地东南部石炭系本溪组沉积微相特征及其对天然气富集控制作用. 西北地质, 47(2): 216-222. 刘磊, 梅佳成, 温汉捷, 等, 2025. 基于SWIR光谱滇中地区碳酸盐黏土型锂矿黏土矿物与锂含量定量分析. 地质学报, 99(9): 3072-3083. 梅佳成, 王雪, 张洪瑞, 等, 2025. 快速估算红土型钴矿中的钴含量: VNIR-SWIR光谱定量反演研究. 光谱学与光谱分析, 45(9): 2578-2584. 王晋年, 郑兰芬, 童庆禧, 1996. 成像光谱图象光谱吸收鉴别模型与矿物填图研究. 环境遥感(1): 20-31. 温汉捷, 罗重光, 杜胜江, 等, 2020. 碳酸盐黏土型锂资源的发现及意义. 科学通报, 65(1): 53-59. 张征峰, 李诺, 范香莲, 等, 2025. 东天山二塘沟黏土型锂矿锂的赋存状态初探. 地球科学, 50(2): 585-595. doi: 10.3799/dqkx.2024.045 -
点击查看大图
计量
- 文章访问数: 111
- HTML全文浏览量: 26
- PDF下载量: 44
- 被引次数: 0
