Objective To address the inefficiencies of traditional approaches for monitoring the soil moisture content (SMC) and soil organic matter content (SOMC) in saline-alkaline farmlands, an estimation method that integrates hyperspectral data with interpretable machine learning was investigated. The goal was to establish a theoretical foundation for the rapid acquisition of soil information and quality assessments of the Hetao Plain, China. Methods Ground-based hyperspectral reflectance data and field-measured SMC and SOMC were used as the primary data sources. Spectral data were processed using a fractional-order differential (FOD) transformation, and various spectral indices were constructed. The models were developed using partial least squares regression (PLSR), support vector machines (SVM) and random forest (RF). To enhance interpretability, the Shapley additive explanations (SHAP) method was employed to evaluate the relative contribution of each variable to model predictions. Results ① The spectral indices derived from the 1.25-order differential transformation showed the highest correlation with the SMC and SOMC. In particular, the generalized difference index (GDI) and optimal spectral index (OSI) exhibited the strongest correlations, with coefficients of 0.505 4 and 0.682 5, respectively. ② The RF model significantly outperformed the PLSR and SVM models in estimating both the SMC and SOMC. For the validation datasets, the RF models achieved R2 values of 0.734 and 0.870, root mean square errors of 3.28 and 1.53, and recognition-primed decisions of 2.07 and 2.43 for SMC and SOMC, respectively. ③ The SHAP analysis indicated that the normalized plane domain index (NPDI) and ratio index (RI) were the most influential variables for estimating the SMC and SOMC, respectively. The combined contributions of the NPDI, OSI and difference index (DI) to SMC modeling reached 68.58%, whereas RI, GDI and NPDI collectively contributed 61.86% to SOMC modeling. Conclusion The integration of FOD and spectral indices enhanced the utility of hyperspectral data. The RF model demonstrated superior accuracy and robustness in estimating soil properties, whereas the SHAP analysis effectively elucidated the contribution of individual variables. Spectral indices (such as NPDI, RI, OSI and DI) played significant roles in modeling SMC and SOMC in saline-alkaline farmland.
文献参数: 丁启东, 黄华雨, 张俊华, 等.基于可解释机器学习的河套平原盐碱农田土壤水分和有机质含量估算[J].水土保持通报,2025,45(4):184-197. Citation:Ding Qidong, Huang Huayu, Zhang Junhua, et al. Estimation of soil moisture and organic matter content in saline-alkaline farmland of Hetao Plain based on interpretable machine learning [J]. Bulletin of Soil and Water Conservation,2025,45(4):184-197.
LuBaojin, TianShengchang, ZuoZhong, et al. Review and prospect on sustainable utilization of salinized land [J]. Journal of Ningxia University (Natural Science Edition), 2023,44(1):79-88.
DingQidong, WangYijing, ZhangJunhua, et al. Estimation of soil moisture and organic matter content in saline alkali farmland by using CARS algorithm combined with covariates [J]. Chinese Journal of Applied Ecology, 2024,35(5):1321-1330.
[5]
GengJing, TanQiuyuan, LvJunwei, et al. Assessing spatial variations in soil organic carbon and C:N ratio in northeast China’s black soil region:Insights from Landsat-9 satellite and crop growth information [J]. Soil and Tillage Research, 2024,235:105897.
[6]
ZhaoMingsong, GaoYingfeng, LuYuanyuan, et al. Hyperspectral modeling of soil organic matter based on characteristic wavelength in East China [J]. Sustainability, 2022,14(14):8455.
DingQidong, WangYijing, ZhangJunhua, et al. Estimation of soil organic matter content in Ningxia based on hyperspectral information [J]. Chinese Journal of Ecology, 2024,43(8):2531-2538.
LiJiaqi, FengYuhua, ChenShuhuang, et al. Estimation of soil organic matter and total nitrogen based on hyperspectral technology [J]. Xinjiang Agricultural Sciences, 2024,61(10):2491-2499.
DingQidong, WangYijing, ZhangJunhua, et al. Estimation of soil water and organic matter content in medium and low yield fields of Ningxia Yellow River irrigation area based on hyperspectral information [J]. Chinese Journal of Applied Ecology, 2023,34(11):3011-3020.
ShangTianhao, ChenRuihua, ZhangJunhua, et al. Estimation of soil organic matter content in Yinchuan Plain based on fractional derivative combined with spectral indices [J]. Chinese Journal of Applied Ecology, 2023,34(3):717-725.
WangYijing, ChenRuihua, ZhangJunhua, et al. Hyperspectral inversion of soil water and salt information based on fractional order derivative technology [J]. Chinese Journal of Applied Ecology, 2023,34(5):1384-1394.
[17]
Ben AbbesA, JarrayN, FarahI R. Advances in remote sensing based soil moisture retrieval:Applications, techniques, scales and challenges for combining machine learning and physical models [J]. Artificial Intelligence Review, 2024,57(9):224.
QiuHaolei, WangHaiyan. Application of machine learning to the prediction of soil properties: A review [J]. Chinese Journal of Ecology, 2025,44(1):283-294.
LiTianle, ZhaoQuanhua, JiaShuhan, et al. Soil organic matter content inversion methods: Differential-instantaneous frequency transform of soil hyperspectral data [J]. Journal of Geo-Information Science, 2024,26(7):1733-1744.
ShangTianhao, JiaPingping, SunYuan, et al. Spectral characteristics of soil moisture in salinized soil and model fitting accuracy in northern Yinchua City, Ningxia Hui Autonomous Region [J]. Bulletin of Soil and Water Conservation, 2020,40(4):183-189.
RohmerJ, BelbezeS, GuyonnetD. Insights into the prediction uncertainty of machine-learning-based digital soil mapping through a local attribution approach [J]. Soil, 2024,10(2):679-697.
[27]
PanMengyao, XiaBisheng, HuangWenbo, et al. PM2.5 concentration prediction model based on random forest and SHAP [J]. International Journal of Pattern Recognition and Artificial Intelligence, 2024,38(5):2452012.
YeMiao, ZhuLin, LiuXudong, et al. Hyperspectral inversion of soil organic matter content based on continuous wavelet transform, SHAP, and XGBoost [J]. Environmental Science, 2024,45(4):2280-2291.
FuPingfan, YangXiaojing, SuZhicheng, et al. Prediction of soil moisture content based on ensemble learning: A case study of western Liaoning Province [J]. Soils, 2023,55(3):671-681.
[32]
LaoCongcong, ChenJunying, ZhangZhitao, et al. Predicting the contents of soil salt and major water-soluble ions with fractional-order derivative spectral indices and variable selection [J]. Computers and Electronics in Agriculture, 2021,182:106031.
[33]
DharumarajanS, LalithaM, KalaiselviB, et al. Remote sensing of soils:Spectral signatures and spectral indices [M]. Remote Sensing of Soils, 2024.
ShiGuangyao, YangSiqi, ZhangJinsong, et al. Importance analysis of ozone influencing factors in high-altitude regions based on machine learning algorithms [J]. Journal of Ningxia University (Natural Science Edition), 2024,45(2):196-202.
HuangHuayu, DingQidong, ZhangJunhua, et al. Ground-based hyperspectral inversion of salinization and alkalinization of different soil layers in farmland in Yinbei area, Ningxia, China [J]. Chinese Journal of Applied Ecology, 2024,35(11):3073-3084.
[38]
LetifM, BaharR, MezouarN. The use of machine learning models and SHAP interaction values to predict the soil swelling index [J]. Periodica Polytechnica Civil Engineering, 2025,69(1):239-250.
HouHuagang, WangDanyang, MaSiqi, et al. Hyperspectral prediction of organic matter in soils of different salinity levels in the Yellow River delta [J]. Scientia Agricultura Sinica, 2023,56(10):1905-1919.
[41]
HongYongsheng, LiuYaolin, ChenYiyun, et al. Application of fractional-order derivative in the quantitative estimation of soil organic matter content through visible and near-infrared spectroscopy [J]. Geoderma, 2019,337:758-769.
[42]
LiuYa, ZhangYuewen, LuHuizhong, et al. Application of fractional-order differential and ensemble learning to predict soil organic matter from hyperspectra [J]. Journal of Soils and Sediments, 2024,24(1):361-372.
[43]
GeXiangyu, DingJianli, TengDexiong, et al. Exploring the capability of Gaofen-5 hyperspectral data for assessing soil salinity risks [J]. International Journal of Applied Earth Observation and Geoinformation, 2022,112:102969.
[44]
LiWangyang, XiangYouzhen, LiuXiaochi, et al. Estimation of soil moisture content based on fractional differential and optimal spectral index [J]. Agronomy, 2024,14(1):184.
[45]
YuanJing, GaoJichao, YuBo, et al. Estimation of soil organic matter content based on spectral indices constructed by improved Hapke model [J]. Geoderma, 2024,443:116823.
[46]
WuDongqing, LeiLei, XieMeiting, et al. High-performance metamaterial light absorption from visible to near-infrared assisted by anti-reflection coating [J]. Photonics, 2023,10(9):998.
[47]
ZhangXinle, DongChang, LiuHuanjun, et al. Methodology for regional soil organic matter prediction with spectroscopy: Optimal sample grouping, input variables, and prediction model [J]. Remote Sensing, 2024,16(3):565.
LiuYing, FanKaixuan, PeiWeihao, et al. Hyperspectral inversion of soil moisture based on PLSR and LSSVM models [J]. Mining Safety & Environmental Protection, 2024,51(5):147-153.
ChenRuihua, ShangTianhao, ZhangJunhua, et al. Effects of different spectra types on the accuracy and correction of soil salt content inversion in Yinchuan Plain, China [J]. Chinese Journal of Applied Ecology, 2022,33(4):922-930.
[52]
JiaPingping, HeWei, HuYi, et al. Inversion of coastal cultivated soil salt content based on multi-source spectra and environmental variables [J]. Soil and Tillage Research, 2024,241:106124.