Abstract
This research aims to develop a novel deep learning-based model for predicting soil properties based on visible and near-infrared (VIS–NIR) spectroscopic data. Soil samples were collected from the European topsoil dataset prepared by the LUCAS project provides various soil physicochemical properties analyzed within 28 EU countries (including sand, silt, clay, pH, organic carbon, calcium carbonates (CaCO3), and N). In this study, one-dimensional (1D) convolutional neural network (CNN) models were developed using absorbance spectral data. The performance of feature learning from discrete wavelet transforms as a powerful preprocessing method was tested. Moreover, the results of the proposed CNN model were compared with partial least squares regression (PLSR) with raw absorbance and optimum classical preprocessing (Savitzky–Golay smoothing with first-order derivative). The ratio of percent deviation (RPD) of CNN with absorbance data for prediction of soil OC, CaCO3, pH, N, sand, silt, and clay content were 4.02, 3.89, 2.82, 3.02, 1.63, 1.43, and 2.16, respectively. While the RPD of PLSR with optimal preprocessing of absorbance data for predicting the mentioned parameters were 2.89, 3.00, 2.79, 2.50, 1.37, 1.27, and 1.84, respectively. The study demonstrated the feasibility of using deep learning-based models and VIS–NIR spectral data as a rapid non-destructive tool for the assessment of important soil properties.
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig1_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig2_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig3_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig4_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig5_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig6_HTML.png)
![](http://media.springernature.com/m312/springer-static/image/art%3A10.1007%2Fs12665-023-11073-0/MediaObjects/12665_2023_11073_Fig7_HTML.png)
Similar content being viewed by others
Data availability
The datasets generated during and/or analyzed during the current study are available in the European soil data center (ESDAC) repository, (https://esdac.jrc.ec.europa.eu/resource-type/datasets).
Abbreviations
- X:
-
Current batch
- Μ :
-
Mean
- Σ :
-
Standard deviation
- γ, β, ∈ :
-
Constant parameters
- j :
-
Scale parameter
- k :
-
Shift parameter
- ψ*:
-
Complex conjugate of mother wavelet
- Oi:
-
Measured value
- n :
-
Number of samples
- Pi:
-
Predicted value
- SD:
-
Standard deviation
- R 2 :
-
Coefficient of determination
- RMSE:
-
Root-mean-square error
- MAE:
-
Mean absolute error
- RPD:
-
Ratio of percent deviation
References
Adeline KRM, Gomez C, Gorretta N, Roger JM (2017) Predictive ability of soil properties to spectral degradation from laboratory Vis-NIR spectroscopy data. Geoderma 288:143–153. https://doi.org/10.1016/j.geoderma.2016.11.010
Bai Z, **e M, Hu B, Luo D, Wan C, Peng J, Shi Z (2022) Estimation of soil organic carbon using VIS-NIR spectral data and spectral feature bands selection in southern **njiang China. Sensors 22(16):6124. https://doi.org/10.3390/s22166124
Baumgardner MF, Silva LF, Biehl LL, Stonery ER (1985) Reflectance properties of soils. Adv Agron 38:1–44. https://doi.org/10.1016/S0065-2113(08)60672-0
Bünemann EK, Bongiorno G, Bai Z, Creamer RE, De Deyn G, De Goede R, Fleskens L, Geissen V, Kuyper TW, Mäder P, Pulleman M (2018) Soil quality: a critical review. Soil Biol Biochem 120:105–125. https://doi.org/10.1016/j.soilbio.2018.01.030
Carnieletto A, Simao R, Dalmolin D, Grunwald S (2018) A systematic study on the application of scatter-corrective and spectral- derivative preprocessing for multivariate prediction of soil organic carbon by Vis-NIR spectra. Geoderma 314:262–274. https://doi.org/10.1016/j.geoderma.2017.11.006
Castaldi F, Palombo A, Santini F, Pascucci S, Pignatti S, Casa R (2016) Evaluation of the potential of the current and forthcoming multispectral and hyperspectral imagers to estimate soil texture and organic carbon. Remote Sens Environ 179:54–65. https://doi.org/10.1016/j.rse.2016.03.025
Chang C, Laird DA, Mausbach MJ, Hurburgh CR (2001) Analyses of soil properties. Soil Sci Soc Am J 65(2):480–490. https://doi.org/10.2136/sssaj2001.652480x
Chen S, Li S, Ma W, Ji W, Xu D, Shi Z, Zhang G (2019) Rapid determination of soil classes in soil profiles using VIS–NIR spectroscopy and multiple objectives mixed support vector classification. Eur J Soil Sci 70(1):42–53. https://doi.org/10.1111/ejss.12715
Gholizadeh A, Borůvka L, Saberioon MM, Kozák J, Vašát R, Němeček K (2015) Comparing different data preprocessing methods for monitoring soil heavy metals based on soil spectral features. Soil Water Res 10(4):218–227. https://doi.org/10.17221/113/2015-SWR
Gholizadeh A, Žižala D, Saberioon M, Borůvka L (2018) Soil organic carbon and texture retrieving and map** using proximal, airborne and Sentinel-2 spectral imaging. Remote Sens Environ 218:89–103. https://doi.org/10.1016/j.rse.2018.09.015
Greenberg I, Seidel M, Vohland M, Koch HJ, Ludwig B (2022) Performance of in situ vs laboratory mid-infrared soil spectroscopy using local and regional calibration strategies. Geoderma 409:115614. https://doi.org/10.1016/j.geoderma.2021.115614
Hartemink AE (2006) Assessing soil fertility decline in the tropics using soil chemical data. Adv Agron 89:179–225. https://doi.org/10.1016/S0065-2113(05)89004-2
Islam K, Singh B, McBratney A (2003) Simultaneous estimation of several soil properties by ultra-violet, visible, and near-infrared reflectance spectroscopy. Soil Res 41(6):1101–1114. https://doi.org/10.1071/SR02137
Jones A, Fernandez-Ugalde O, Scarpa S (2020) LUCAS 2015 topsoil survey. Publ off Eur Union EUR. https://doi.org/10.2760/616084
Kuang B, 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. Soil Tillage Res 146:243–252. https://doi.org/10.1016/j.still.2014.11.002
Maia AJ, da Silva YJ, do Nascimento CW, Veras G, Escobar ME, Cunha CS, da Silva YJ, Nascimento RC, de Souza Pereira LH (2020) Near-infrared spectroscopy for the prediction of rare earth elements in soils from the largest uranium-phosphate deposit in Brazil using PLS, iPLS, and iSPA-PLS models. Environ Monit Assess 192:1–14. https://doi.org/10.1007/s10661-020-08642-2
Morellos A, Pantazi XE, Moshou D, Alexandridis T, Whetton R, Tziotzios G, Wiebensohn J, Bill R, Mouazen AM (2016) Machine learning based prediction of soil total nitrogen, organic carbon and moisture content by using VIS-NIR spectroscopy. Biosyst Eng 152:104–116. https://doi.org/10.1016/j.biosystemseng.2016.04.018
Ng W, Minasny B, de Sousa MW, Melo Demattê JA (2020) The influence of training sample size on the accuracy of deep learning models for the prediction of soil properties with near-infrared spectroscopy data. Soil 6(2):565–578. https://doi.org/10.5194/soil-6-565-2020
Nyarko F, Tack FMG, Mouazen AM (2022) Potential of visible and near infrared spectroscopy coupled with machine learning for predicting soil metal concentrations at the regional scale. Sci Total Environ 841:156582. https://doi.org/10.1016/j.scitotenv.2022.156582
Padarian J, Minasny B, McBratney AB (2019) Using deep learning to predict soil properties from regional spectral data. Geoderma Reg 16:e00198. https://doi.org/10.1016/j.geodrs.2018.e00198
Shen Z, Viscarra Rossel RA (2021) Automated spectroscopic modelling with optimised convolutional neural networks. Sci Rep 11(1):208. https://doi.org/10.1038/s41598-020-80486-9
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. ar**v Prepr. https://doi.org/10.48550/ar**v.1409.1556
Singh S, Kasana SS (2019) Estimation of soil properties from the EU spectral library using long short-term memory networks. Geoderma Reg 18:e00233. https://doi.org/10.1016/j.geodrs.2019.e00233
Sorenson PT, Small C, Tappert MC, Quideau SA, Drozdowski B, Underwood A, Janz A (2017) Monitoring organic carbon, total nitrogen, and pH for reclaimed soils using field reflectance spectroscopy. Can J Soil Sci 97(2):241–248. https://doi.org/10.1139/cjss-2016-0116
Stenberg B, Viscarra Rossel RA, Mouazen AM, Wetterlind J (2010) Visible and near infrared spectroscopy in soil science. Adv Agron 107:163–215. https://doi.org/10.1016/S0065-2113(10)07005-7
Tiruneh GA, Alemayehu TY, Allouche FK, Reichert JM (2021a) Spatial variability modeling of soil fertility for improved nutrient management in Northwest Ethiopia. Arab J Geosci 14:2797. https://doi.org/10.1007/s12517-021-08814-5
Tiruneh GA, Alemayehu TY, Meshesha DT, Vogelmann ES, Reichert JM (2021b) Spatial variability of soil chemical properties under different land-uses in Northwest Ethiopia. PLoS ONE 16(6):e0253156. https://doi.org/10.1371/journal.pone.0253156
Tiruneh GA, Meshesha DT, Adgo E, Tsunekawa A, Haregeweyn N, Fenta AA, Belay AW, Tadesse N, Fekadu G, Reichert JM (2022) Use of soil spectral reflectance to estimate texture and fertility affected by land management practices in Ethiopian tropical highland. PLoS ONE 17(7):e0270629. https://doi.org/10.1371/journal.pone.0270629
Tiruneh GA, Alemayehu TY, Meshesha DT, Adgo E, Tiruneh GA, Reichert JM (2023) Variability modeling and map** of soil properties for improved management in Ethiopia. Agrosyst Geosci Environ 6(1):1–12. https://doi.org/10.1002/agg2.20357
Turaga SC, Murray JF, Jain V, Roth F, Helmstaedter M, Briggman K, Denk W, Seung HS (2010) Convolutional networks can learn to generate affinity graphs for image segmentation. Neural Comput 22(2):511–538. https://doi.org/10.1162/neco.2009.10-08-881
Viscarra Rossel RA, Hicks WS (2015) Soil organic carbon and its fractions estimated by visible-near infrared transfer functions. Eur J Soil Sci 66(3):438–450. https://doi.org/10.1111/ejss.12237
Viscarra Rossel RA, Behrens T, Ben-Dor E, Brown DJ, Demattê JA, Shepherd KD, Shi Z, Stenberg B, Stevens A, Adamchuk V, Aïchi H (2016) A global spectral library to characterize the world’s soil. Earth-Science Rev 155:198–230. https://doi.org/10.1016/j.earscirev.2016.01.012
Yamashita R, Nishio M, Kinh Gian Do R, Togashi K (2018) Convolutional neural networks: an overview and application in radiology. Insights Imaging 9:611–629. https://doi.org/10.1007/s13244-018-0639-9
Yang R-M, Wang L-J, Chen L-M, Zhang Z-Q (2021) Assessment of soil quality using VIS–NIR spectra in invaded coastal wetlands. Environ Earth Sci 81:19. https://doi.org/10.1007/s12665-021-10134-6
Zhao X, Zhao D, Wang J, Triantafilis J (2022) Soil organic carbon (SOC) prediction in Australian sugarcane fields using VIS–NIR spectroscopy with different model setting approaches. Geoderma Reg 30:e00566. https://doi.org/10.1016/j.geodrs.2022.e00566
Zhong L, Guo X, Xu Z, Ding M (2021) Soil properties : their prediction and feature extraction from the LUCAS spectral library using deep convolutional neural networks. Geoderma 402:115366. https://doi.org/10.1016/j.geoderma.2021.115366
Acknowledgements
The authors thank the Department of Agricultural Machinery Engineering, University of Tehran, for providing the workspace and other necessary resources to carry out the research. The LUCAS 2015 topsoil dataset used in this work was made available by the European Commission through the European Soil Data Centre managed by the Joint Research Centre (JRC) (http://esdac.jrc.ec.europa.eu/).
Funding
No funding was obtained for this study.
Author information
Authors and Affiliations
Contributions
MH-Z: methodology, software, conceptualization, writing—original draft preparation. MO: software, writing—original draft preparation, supervision, validation, project administration. FS: writing—original draft preparation, supervision, validation. HG-M: writing—reviewing and editing, validation.
Corresponding authors
Ethics declarations
Conflict of interest
The authors declare no competing interests.
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Hosseinpour-Zarnaq, M., Omid, M., Sarmadian, F. et al. A CNN model for predicting soil properties using VIS–NIR spectral data. Environ Earth Sci 82, 382 (2023). https://doi.org/10.1007/s12665-023-11073-0
Received:
Accepted:
Published:
DOI: https://doi.org/10.1007/s12665-023-11073-0