Abstract
Choosing appropriate multivariate calibration and preprocessing transformation techniques is important in the determination of soil organic carbon (SOC) content based on visible and near-infrared (Vis–NIR) spectroscopy. The performance levels of partial least-squares regression (PLSR), support vector machine regression (SVMR), and wavelet neural network (WNN) calibration methods coupled with different preprocessing approaches were compared using three kinds of criteria, including the coefficient of determination (R2), root mean square error (RMSE), and residual prediction deviation (RPD). A total of 328 soil samples collected from the south bank of Hangzhou Bay were used as the dataset for the calibration–validation procedure and SOC content inversion. The effects of spectra preprocessing transformation methods were evaluated for raw spectra, Savitzky–Golay smoothing with the first derivatives of reflectance (FDR) and Savitzky–Golay smoothing with logarithm of reciprocal of the reflectance (log R−1). The results indicate that the SVMR is superior to the PLSR, and WNN models for SOC content prediction. The combination of the SVMR model with FDR provided the best prediction results for the SOC content, with R2p = 0.92, RPDP = 2.82, RMSEP = 0.36%, and a kappa correlation coefficient of interpolation as high as 0.97. The FDR of Vis–NIR spectroscopy combined with the SVMR model is recommended over the PLSR and WNN modeling techniques for the high-accuracy determination of the SOC content.
Similar content being viewed by others
References
Abdi D, Tremblay GF, Ziadi N et al (2012) Predicting soil phosphorus-related properties using near-infrared reflectance spectroscopy [J]. Soil Sci Soc Am J 76:2318–2326
Annea NJP, Abd-Elrahmana AH, Lewis DB et al (2014) Modeling soil parameters using hyperspectral image reflectance in subtropical coastal wetlands [J]. Int J Appl Earth Obs Geoinf 33:47–56
Askari MS, Cui JF, O’Rourke SM et al (2015) Evaluation of soil structural quality using VIS-NIR spectra [J]. Soil Till Res 146:108–117
Bartholomeus HM, Schaepman ME, Kooistra L et al (2008) Spectral reflectance based indices for soil organic carbon quantification[J]. Geoderma 145(1–2):28–36
Ben-Dor E, Inbar Y, Chen Y (1997) The reflectance spectra of organic matter in the visible near-infrared and short wave infrared region (400–2500) during a controlled decomposition process[J]. Remote Sense Environ 61(1):1–15
Brown DJ, Bricklemyer RS, Miller PR (2005) Validation requirement for diffuse reflectance soil characterization models with a case study of VNIR soil C prediction in Montana [J]. Geoderma 129(3–4):215–267
Cambou A, Cardinael R, Kouakoua E et al (2016) Prediction of soil organic carbon stock using visible and near infrared reflectance spectroscopy (VNIRS) in the field [J]. Geoderma 261(2):151–159
Cisty M, Bajtek Z, Bezak J (2011) Support vector machine based model for water content in soil interpolation [J]. Geophys Res Abstr 13:1–2
Clairotte M, Grinand C, Kouakoua E et al (2016) National calibration of soil organic carbon concentration using diffuse infrared reflectance spectroscopy [J]. Geoderma 276:41–52
Erzin Y, Rao BH, Singh DN (2008) Artificial neural network models for predicting soil thermal resistivity [J]. Int J Therm Sci 47(10):1347–1358
Guo L, Zhang HT, Shi TZ et al (2019) Prediction of soil organic carbon sock by laboratory spectral data and airborne hyperspectral images [J]. Geoderma 337:32–41
Hazama K, Kano M (2015) Covariance-based locally weighted partial least squares for high performance adaptive modeling [J]. Chemometr Intell Lab Sys 146:55–62
Heinze S, Vohland M, Joergensen RG et al (2013) Usefulness of near-infrared spectroscopy for the prediction of chemical and biological soil properties in different long-term experiments[J]. J Plant Nutr Soil Sci 176(4):520–528
Hong YS, Liu YL, Chen YY et al (2019) Application of fractional-order derivative in the quantitative estimation of soil organic matter content through visible and near-infrared spectroscopy [J]. Geoderma 337:758–769
Jafarzadeh AA, Pal M, Servati M et al (2016) Comparative analysis of support vactor machine and artificial nenural network models for soil cation exchange capacity predicition [J]. Int J Environ Sci Technol 13(1):87–96
Janik LJ, Cozzolino D, Dambergs R et al (2007) The prediction of total anthocyanin concentration in red-grape homogenates using vis–NIR spectroscopy and artificial neural networks. Anal Chim Acta 594(1):107–118
Jiang QH, Li QX, Wang XG et al (2017) Estimation of soil organic carbon and total nitrogen in different soil layers using VNIR spectroscopy: effects of spiking on model applicability [J]. Geoderma 293:54–63
Kennard RW, Stone LA (1969) Computer aided design of experiments [J]. Technometrics 11(1):137–148
Knox NM, Grunwald S (2018) Total soil carbon assessment: linking field, lab, and landscape through VNIR modeling [J]. Landsc Ecol 33:2137–2152
Kuang B, Mouazen AM (2011) Calibration of visible and near infrared spectroscopy for soil analysis at the field scale on three European farms[J]. Eur J Soil Sci 62(4):629–636
Kuang BY, Tekin Y, Mouazen AM (2015) Comparison between artificial neural network and partial least squares for on-line visible and near infrared spectroscopy measurement of soil organic carbon, pH and clay content [J]. Soil Till Res 146:243–252
Kusumo BH, Hedley MJ, Hedley CB et al (2010) Predicting pasture root density from soil spectral reflectance: field measurement [J]. Eur J Soil Sci 61:1–13
Kweon G, Maxton C (2013) Soil organic matter sensing with an on-the-go optical sensor [J]. Biosys Eng 115(1):66–81
Li Y, DemetriadesShash TH, Kanemasu ET et al (1993) Use of Second derivation canopy reflectance for monitoring prairie vegetation over different soil backgrounds [J]. Remote Sens Environ 44:81–87
Li S, Shi Z, Chen S et al (2015a) In situ measurements of organic carbon in soil profiles using vis-NIR spectroscopy on the Qinghai-Tibet plateau [J]. Environ Sci Technol 49:4980–4987
Li YL, Pan C, Meng X et al (2015b) Haar wavelet based implementation method of the non–integer order differentiation and its application to signal enhancement [J]. Meas Sci Rev 15(3):101–106
Lucà F, Conforti M, Castrignanò A et al (2017) Effect of calibration set size on prediction at local scale of soil carbon by Vis-NIR spectroscopy [J]. Geoderma 288:175–183
Malley DF, Williams PC (1997) Use of near-infrared reflectance spectroscopy in prediction of heavy metals in freshwater sediment by their association with organic matter[J]. Environ Sci Technol 31:3461–3467
Martens H, Næs T (1989) Multivariate calibration [M]. Wiley, New York
Mcdowell ML, Bruland GL, Deenik JL et al (2012) Soil total carbon analysis in Hawaiian soils with visible, near-infrared and mid-infrared diffuse reflectance spectroscopy [J]. Geoderma 189–190(4):312–320
Mouazen AM, Kuang B, Baerdemaeker JD et al (2010) Comparison among principal component, partial least squares and back propagation neural network analyses for accuracy of measurement of selected soil properties with visible and near infrared spectroscopy [J]. Geoderma 158(1–2):23–31
Mutanga O, Skidmore AK, Prins HHT (2004) Predicting in situ pasture quality in the Kruger National Park, South Africa, using continuum removed absorption features [J]. Remote Sens Environ 89:393–408
Nelson DW, Sommers L (1974) A rapid and accurate procedure for estimation of organic carbon in soils[J]. In: Proceedings of the Indiana academy of sciences, pp 456–462
Palacios-Orueta A, Ustin SL (1998) Remote sensing of soil properties in the Santa Monica mountains I. spectral analysis [J]. Remote Sens Environ 65:170–183
Peng XT, Shi TZ, Song A et al (2014) Estimating soil organic carbon using Vis/NIR spectroscopy with SVMR and SPA Methods[J]. Remote Sens 6(4):2699–2717
Rozenstein O, Paz-Kagan T, Salbach C et al (2015) Comparing the effect of preprocessing transformations on methods of land-use classification derived from spectral soil measurements [J]. IEEE J Sel Top Appl Earth Obs Remote Sens 8:2393–2404
Samadianfard S, Asadi E, Jarhan S et al (2018) Wavelet neural networks and gene expression programming models to predict short-term soil temperature at different depths [J]. Soil Till Res 175:37–50
Sarkhot DV, Grunwald S, Ge Y et al (2011) Comparison and detection of total and available soil carbon fractions using visible/near infrared diffuse reflectance spectroscopy [J]. Geoderma 164(1):22–32
Schlerf M, Atzberger C, Joachim H et al (2010) Retrieval of chlorophyll and nitrogen in Norway spruce (Piceaabies L. Karst.) using imaging spectroscopy[J]. Int J Appl Earth Observ Geoinf 12:17–26
Schlesinger WH, Andrews JA (2000) Soil respiration and the global carbon cycle[J]. Biogeochemistry 48:7–20
Shao YN, He Y (2011) Nitrogen, phosphorus, and potassium prediction in soils, using infrared spectroscopy[J]. Soil Res 49:166–172
Shi TZ, Cui LJ, Wang JJ et al (2013) Comparison of multivariate methods for estimating soil total nitrogen with visible/near-infrared spectroscopy [J]. Plant Soil 366(1–2):363–375
Stenberg B, Viscarra Rossel RA, Mouazen AM et al (2010) Visible and near infrared spectroscopy in soil science [J]. Adv Agron 107:163–215
Stevens A, Udelhoven T, Denis A et al (2010) Measuring soil organic carbon in croplands at regional scale using airborne imaging spectroscopy[J]. Geoderma 158:32–45
Stockmann U, Adams MA, Crawford JW et al (2013) The knowns, known unknowns and unknowns of sequestration of soil organic carbon. Agr Ecosyst Environ 164:80–99
Thissen U, Pepers M, Üstün B et al (2004) Comparing support vector machines to PLS for spectral regression applications[J]. Chemometr Intell Lab Syst 73(2):169–179
Viscarra Rossel RA, Behrens T (2010) Using data mining to model and interpret soil diffuse reflectance spectra [J]. Geoderma 158:46–54
Viscarra Rossel RA, Hicks WS (2015) Soil organic carbon and its fractions estimated by visible-near infrared transfer functions [J]. Eur J Soil Sci 66(3):438–450
Viscarra Rossel RA, Lark RM (2009) Improved analysis and modelling of soil diffuse reflectance spectra using wavelets [J]. Eur J Soil Sci 60:453–464
Viscarra Rossel RA, Mcglynn RN, Mcbratney AB (2006a) Determining the composition of mineral-organic mixes using UV-vis-NIR diffuse reflectance spectroscopy [J]. Geoderma 137(1):70–82
Viscarra Rossel RA, Walvoort DJJ, Mcbratney AB et al (2006b) Visible, near infrared, mid infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties[J]. Geoderma 131(1–2):59–75
Viscarra Rossel RA, Behrens T, Ben-Dor E et al (2016) A global spectral library to characterize the world’s soil [J]. Earth Sci Rev 155:198–230
Vohland M, Besold J, Hill J (2011) Comparing different multivariate calibration methods for the determination of soil organic carbon pools with visible to near infrared spectroscopy [J]. Geoderma 166:198–205
Wang WS, Ding J (2003) Wavelet network model and its application to the prediction of hydrology [J]. Nat Sci 1:67–71
Wilding LP (1985) Spatial variability: it’s documentation, accommodation and implication to soil surveys [M]. In: Nielsen DR, Bouma J (eds) Soil spatial variability. Pudoc, Wageningen, pp 166–194
Xu SX, Zhao YC, Wang MY et al (2018) Comparison of multivariate methods for estimating selected soil properties from intact soil cores of paddy fields by Vis–NIR spectroscopy [J]. Geoderma 310:29–43
Yang H, Kuang B, Mouazen AM (2012) Quantitative analysis of soil nitrogen and carbon at a farm scale using visible and near infrared spectroscopy coupled with wavelength reduction [J]. Eur J Soil Sci 63:410–420
Yu Y, Liu Q, Wang YB et al (2016) Evaluation of MLSR and PLSR for estimating soil element contents using visible/near-infrared spectroscopy in apple orchards on the Jiaodong peninsula [J]. CATENA 137:340–349
Zou P, Yang JS, Fu JR et al (2010) Artificial neural network and time series models for predicting soil salt and water content [J]. Agric Water Manag 97(12):2009–2019
Acknowledgements
This research was funded by geological survey project of China Geological Survey (Grant numbers 12120115048501 DD20160320), Zhejiang Geological Exploration Fund Project (Grant number 201312), Key project of Zhejiang Gongshang University (Grant number X13-03), and Science and Technology Project of Department of Natural Resources of Zhejiang Province (Grant number 2020-33). The spectral reflectance of soil sample was measured by the Institute of Agricultural Remote Sensing and Information Technology of Zhejiang University, P. R. China. We sincerely appreciated the editors and the reviewers for their constructive suggestions and insightful comments which helped us greatly to improve this manuscript.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Xu, M., Chu, X., Fu, Y. et al. Improving the accuracy of soil organic carbon content prediction based on visible and near-infrared spectroscopy and machine learning. Environ Earth Sci 80, 326 (2021). https://doi.org/10.1007/s12665-021-09582-x
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12665-021-09582-x