At present, there is a lack of research on the soil environmental pollution caused by the mining gold mines in Mojiang, Yunnan Province, China. This study focuses on the polluted farmland soil in Mojiang Hani Autonomous County, Yunnan Province, and 61 soil samples from the region. Continuous Wavelet Transform (CWT) analysis was performed on raw soil hyperspectral reflectance (R) near infrared spectra and after first-order derivative transformation (R′) using seven decomposition scales. Four models (Ridge, Lasso, Elastic net, and partial least squares (PLS) regression) were used for quantitative prediction of soil arsenic (As) content. The simulation results show that: (1) appropriately increasing the CWT decomposition scale can effectively extract feature bands and amplify the differences between features. In most cases, the modeling accuracy of the model, initially increases and then decreases as the CWT decomposition scale increases, with scale five being the optimal decomposition scale. (2) The R′-CWT method, improves the correlation coefficient between hyperspectral data and arsenic content. The modeling results for predicting soil As content using R′-CWT are superior to those obtained with R-CWT. (3) Regularized modeling methods that incorporate a regular term in the loss function can improve the prediction accuracy and generalization ability of the model. (4) The performance of predicting soil As content based on R′-CWT transformation is ranked as follows: Ridge regression>Elastic net regression>Lasso regression>PLS regression. The ridge regression model based on first derivative spectra using continous wavelet transformation with a scale factor of 5 (R′-CWT-25) has the highest prediction accuracy, with validation sets of root mean square error (RMSE), coefficient of determination (r2), residual prediction deviation (RPD) and the ratio of performance to the interquartile range (RPIQ) of 0.07 g∙kg-1, 0.94, 3.9, and 3.7, respectively. This study can provide a certain scientific reference for the ecological environment management of polluted soil around the gold mining area in Mojiang County.
SevakPPushkarB. Arsenic pollution cycle, toxicity and sustainable remediation technologies: a comprehensive review and bibliometric analysis. J Environ Manag2024; 349: 119504.
2.
AbdElgawadHSchoenaersSZintaG, et al.Soil arsenic toxicity differentially impacts C3 (barley) and C4 (maize) crops under future climate atmospheric CO2. J Hazard Mater2021; 414: 125331.
3.
SunYLeiSZhaoY, et al.Spatial distribution prediction of soil heavy metals based on sparse sampling and multi-source environmental data. J Hazard Mater2024; 465: 133114.
4.
ZhangYZhuYZhaoS, et al.Arsenic in soils contaminated by arsenic-containing chemical weapons in a site of Jilin, China: fraction and bioaccessibility. Environ Sci Pollut Res2022; 29(19): 28957–28972.
5.
NikolaosLTChristosGCJohnBT, et al.A three-level multiple-kernel learning approach for soil spectral analysis. Neurocomputing2020; 389: 27–41.
6.
FuCGanSXiongH, et al.A new method to estimate soil organic matter using the combination model based on short memory fractional order derivative and machine learning model. Infrared Phys Technol2023; 134: 104922.
7.
TarnikMGGhafariSBahrainiT, et al.Minimum variance based-Bayes combination for prediction of soil properties on Vis-NIR reflectance spectroscopy. Chemometr Intell Lab2020; 207: 104194.
8.
HongYLiuYChenY, et al.Application of fractional-order derivative in the quantitative estimation of soil organic matter content through visible and near-infrared spectroscopy. Geoderma2019; 337: 758–769.
9.
YangCFengMSongL, et al.Hyperspectral monitoring of soil urease activity under different water regulation. Plant Soil2022; 477(1-2): 779–792.
10.
TiecherTMoura-BuenoJMCanerL, et al. Improving the quantification of sediment source contributions using different mathematical models and spectral preprocessing techniques for individual or combined spectra of ultraviolet–visible, near-and middle-infrared spectroscopy. Geoderma. 2021; 384: 114–815.
11.
ZhangJLiuTWangJ, et al.A simple but effective evaluation criterion for parameters optimization of EPO and its application to moisture insensitive prediction of soil organic matter. Chemometr Intell Lab2023; 236: 104794.
12.
BarraIHaefeleSMSakrabaniR, et al.Soil spectroscopy with the use of chemometrics, machine learning and pre-processing techniques in soil diagnosis: recent advances–A review. TrAC, Trends Anal Chem2021; 135: 116166.
13.
SeemaGhoshAKDasBS, et al. Application of VIS-NIR spectroscopy for estimation of soil organic carbon using different spectral preprocessing techniques and multivariate methods in the middle indo-gangetic plains of India. Geoderma Regional. 2020; 23: e00-e349.
14.
ZhangJSunHGaoD, et al.Detection of canopy chlorophyll content of corn based on continuous wavelet transform analysis. Remote Sens2020; 12(17): 2741.
15.
HongYChenSChenY, et al.Comparing laboratory and airborne hyperspectral data for the estimation and mapping of topsoil organic carbon: feature selection coupled with random forest. Soil Tillage Res2020; 199: 104589.
16.
LinDLiGZhuY, et al.Predicting copper content in chicory leaves using hyperspectral data with continuous wavelet transforms and partial least squares. Comput Electron Agric2021; 187: 106293.
17.
HongYChenYChenS, et al.Improving spectral estimation of soil inorganic carbon in urban and suburban areas by coupling continuous wavelet transform with geographical stratification. Geoderma2023; 430: 116284.
18.
LiMSunHQiaoJ. Deep learning assisted continuous wavelet transform-based spectrogram for the detection of chlorophyll content in potato leaves. Comput Electron Agric2022; 195: 106–802.
19.
DeissLMargenotAJCulmanSW, et al.Tuning support vector machines regression models improves prediction accuracy of soil properties in MIR spectroscopy. Geoderma2020; 365: 114227.
20.
LiuJDongZXiaJ, et al.Estimation of soil organic matter content based on CARS algorithm coupled with random forest. Spectrochim Acta2021; 258: 119823.
21.
ZhouPSudduthKAVeumKS, et al.Extraction of reflectance spectra features for estimation of surface, subsurface, and profile soil properties. Comput Electron Agric2022; 196: 10684.
22.
QuasdaneMRamchounHMasrourT. Sparse smooth group L0∘ L1/2 regularization method for convolutional neural networks. Knowl Base Syst2024; 284: 111327.
23.
MokhtiaMEftekhariMSaberi-MovahedF. Feature selection based on regularization of sparsity based regression models by hesitant fuzzy correlation. Appl Soft Comput2020; 91: 106255.
24.
ZhangRLiYGuiY. Prediction of rock blasting induced air overpressure using a self-adaptive weighted kernel ridge regression. Appl Soft Comput2023; 148: 110851.
25.
ShafieeSLiedLMBurudI, et al.Sequential forward selection and support vector regression in comparison to LASSO regression for spring wheat yield prediction based on UAV imagery. Comput Electron Agric2021; 183: 106036.
26.
ZhaoHTangJZhuQ, et al.Associations of prenatal heavy metals exposure with placental characteristics and birth weight in hangzhou birth cohort: multi-Pollutant models based on elastic net regression. Sci Total Environ2020; 742: 140613.
27.
XiaoDHuangJLiJ, et al.Inversion study of soil organic matter content based on reflectance spectroscopy and the improved hybrid extreme learning machine. Infrared Phys Technol2023; 128: 104488.
28.
LiLGengSLinD, et al.Accurate modeling of vertical leaf nitrogen distribution in summer maize using in situ leaf spectroscopy via CWT and PLS-Based approaches. Eur J Agron2022; 140: 126607.
29.
WangGFangQTengY, et al.Determination of the factors governing soil erodibility using hyperspectral visible and near-infrared reflectance spectroscopy. Int J Appl Earth Obs2016; 53: 48–63.
YıldırımHÖzkaleMR. A combination of ridge and liu regressions for extreme learning machine. Soft Comput2023; 27(5): 2493–2508.
32.
TibshiraniR. Regression shrinkage and selection via the lasso. J Roy Stat Soc B1996; 58(1): 267–288.
33.
ZhangLTeddeAHoP, et al.Mining data from milk mid-infrared spectroscopy and animal characteristics to improve the prediction of dairy cow's liveweight using feature selection algorithms based on partial least squares and elastic net regressions. Comput Electron Agric2021; 184: 106106.
34.
YuBChenCWangX, et al.Prediction of protein–protein interactions based on elastic net and deep forest. Expert Syst Appl2021; 176: 114876.
35.
YangCFengMSongL, et al.Study on hyperspectral monitoring model of soil total nitrogen content based on fractional-order derivative. Comput Electron Agric2022; 201: 2107307.
36.
SarstedtMHairJFPickM, et al.Progress in partial least squares structural equation modeling use in marketing research in the last decade. Psychol Market2022; 39(5): 1035–1064.
WangJDingJYuD, et al.Machine learning-based detection of soil salinity in an arid desert region, Northwest China: a comparison between Landsat-8 OLI and Sentinel-2 MSI. Sci Total Environ2020; 707: 136092.
39.
AlloMTodoroffPJameuxM, et al.Prediction of tropical volcanic soil organic carbon stocks by visible-near- and mid-infrared spectroscopy. Catena2020; 189: 104452.
40.
ZhangZDingJWangJ, et al.Prediction of soil organic matter in northwestern China using fractional order derivative spectroscopy and modified normalized difference indices. Catena2020; 185: 104257.
41.
DingPHuangJTangJ. Multi-peak FBG reflection spectrum segmentation based on continuous wavelet transformation. Opt Fiber Technol2019; 50: 250–255.
42.
FuPYangKMengF, et al.A new three-band spectral and metal element index for estimating soil arsenic content around the mining area. Process Saf Environ2022; 157: 27–36.
43.
HuangXWangXBaishanK, et al.Hyperspectral estimation of soil organic carbon content based on continuous wavelet transform and successive projection algorithm in arid area of Xinjiang, China. Sustainability2023; 15(3): 2587.
44.
GuXWangYSunQ, et al.Hyperspectral inversion of soil organic matter content in cultivated land based on wavelet transform. Comput Electron Agric2019; 167: 105053.
45.
LiDChengTZhouK, et al.WREP: a wavelet-based technique for extracting the red edge position from reflectance spectra for estimating leaf and canopy chlorophyll contents of cereal crops. Isprs J Photogramm2017; 129: 103–117.
46.
MangalathuSJeonJSDesRochesR. Critical uncertainty parameters influencing seismic performance of bridges using lasso regression. Earthq Eng Struct Dynam2018; 47(3): 784–801.
