Evaluation of Optimized Preprocessing and Modeling Algorithms for Prediction of Soil Properties Using VIS-NIR Spectroscopy
Abstract
:1. Introduction
2. Materials and Methods
2.1. Study Area and Sample Collection
2.2. Laboratory Methods
2.3. Spectral Collection
2.4. Optimization of Data Processing
3. Results
3.1. Descriptive Statistics of Selected Soil Properties
3.2. Optimization of Spectral Preprocessing and Modelling
3.2.1. Preprocessing Performance Evaluation
3.2.2. Spectral Prediction for All Soil Properties
4. Discussion
5. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- FAO. The Multi-Faced rôle of Soil in the Near East. and North. Africa Region. Policy Brief. Rome. Available online: https://www.fao.org/3/ca3803en/CA3803EN.pdf (accessed on 13 August 2021).
- Zhang, Y.; Biswas, A.; Ji, W.; Adamchuk, V.I. Depth-specific prediction of soil properties in situ using vis-NIR spectroscopy. Soil Sci. Soc. Am. J. 2017, 81, 993–1004. [Google Scholar] [CrossRef]
- Zinck, J.; Berroterán, J.; Farshad, A.; Moameni, A.; Wokabi, S.; Ranst, E.V. Approaches to assessing sustainable agriculture. J. Sustain. Agric. 2004, 23, 87–109. [Google Scholar] [CrossRef]
- Wetterlind, J.; Stenberg, B.; Söderström, M. The use of near infrared (NIR) spectroscopy to improve soil mapping at the farm scale. Precis. Agric. 2008, 9, 57–69. [Google Scholar] [CrossRef] [Green Version]
- Viscarra Rossel, R.; Webster, R. Discrimination of Australian soil horizons and classes from their visible–near infrared spectra. Eur. J. Soil Sci. 2011, 62, 637–647. [Google Scholar] [CrossRef]
- Stenberg, B.; Rossel, R.V. Diffuse reflectance spectroscopy for high-resolution soil sensing. In Proximal Soil Sensing; Springer: Berlin/Heidelberg, Germany, 2010; pp. 29–47. [Google Scholar]
- Johnson, J.-M.; Vandamme, E.; Senthilkumar, K.; Sila, A.; Shepherd, K.D.; Saito, K. Near-infrared, mid-infrared or combined diffuse reflectance spectroscopy for assessing soil fertility in rice fields in sub-Saharan Africa. Geoderma 2019, 354, 113840. [Google Scholar] [CrossRef]
- Gupta, A.; Vasava, H.B.; Das, B.S.; Choubey, A.K. Local modeling approaches for estimating soil properties in selected Indian soils using diffuse reflectance data over visible to near-infrared region. Geoderma 2018, 325, 59–71. [Google Scholar] [CrossRef]
- Conforti, M.; Matteucci, G.; Buttafuoco, G. Using laboratory Vis-NIR spectroscopy for monitoring some forest soil properties. J. Soils Sediments 2018, 18, 1009–1019. [Google Scholar] [CrossRef]
- Terra, F.S.; Demattê, J.A.; Rossel, R.A.V. Spectral libraries for quantitative analyses of tropical Brazilian soils: Comparing vis–NIR and mid-IR reflectance data. Geoderma 2015, 255, 81–93. [Google Scholar] [CrossRef]
- Gholizade, A.; Soom, M.A.M.; Saberioon, M.M.; BorůvkaP, L. Visible and near infrared reflectance spectroscopy to determine chemical properties of paddy soils. J. Food Agric. Environ. 2013, 11, 859–866. [Google Scholar]
- Leone, A.P.; Viscarra-Rossel, R.A.; Amenta, P.; Buondonno, A. Prediction of soil properties with PLSR and vis-NIR spectroscopy: Application to mediterranean soils from Southern Italy. Curr. Anal. Chem. 2012, 8, 283–299. [Google Scholar] [CrossRef]
- Lee, K.; Lee, D.; Sudduth, K.; Chung, S.; Kitchen, N.; Drummond, S. Wavelength identification and diffuse reflectance estimation for surface and profile soil properties. Trans. ASABE 2009, 52, 683–695. [Google Scholar] [CrossRef]
- Rossel, R.V.; Walvoort, D.; McBratney, A.; Janik, L.J.; Skjemstad, J. Visible, near infrared, mid infrared or combined diffuse reflectance spectroscopy for simultaneous assessment of various soil properties. Geoderma 2006, 131, 59–75. [Google Scholar] [CrossRef]
- Islam, K.; Singh, B.; McBratney, A. Simultaneous estimation of several soil properties by ultra-violet, visible, and near-infrared reflectance spectroscopy. Soil Res. 2003, 41, 1101–1114. [Google Scholar] [CrossRef]
- Douglas, R.K.; Nawar, S.; Alamar, M.C.; Mouazen, A.; Coulon, F. Rapid prediction of total petroleum hydrocarbons concentration in contaminated soil using vis-NIR spectroscopy and regression techniques. Sci. Total Environ. 2018, 616, 147–155. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Minasny, B.; McBratney, A.B. Regression rules as a tool for predicting soil properties from infrared reflectance spectroscopy. Chemom. Intellig. Lab. Syst. 2008, 94, 72–79. [Google Scholar] [CrossRef]
- Hong, Y.; Chen, S.; Zhang, Y.; Chen, Y.; Yu, L.; Liu, Y.; Liu, Y.; Cheng, H.; Liu, Y. Rapid identification of soil organic matter level via visible and near-infrared spectroscopy: Effects of two-dimensional correlation coefficient and extreme learning machine. Sci. Total Environ. 2018, 644, 1232–1243. [Google Scholar] [CrossRef] [PubMed]
- Wu, X.; Wu, B.; Sun, J.; Yang, N. Classification of apple varieties using near infrared reflectance spectroscopy and fuzzy discriminant c-means clustering model. J. Food Process. Eng. 2017, 40, e12355. [Google Scholar] [CrossRef]
- Stevens, A.; Ramirez-Lopez, L.; Vignette, R. An Introduction to the Prospectr Package; 2013 R Package Version 0.1. 2015, Volume 3. Available online: https://mran.microsoft.com/snapshot/2017-08-06/web/packages/prospectr/vignettes/prospectr-intro.pdf (accessed on 13 August 2021).
- Barra, I.; Haefele, S.M.; Sakrabani, R.; Kebede, F. Soil spectroscopy with the use of chemometrics, machine learning and pre-processing techniques in soil diagnosis: Recent advances-A review. TrAC Trends Anal. Chem. 2020, 135, 116166. [Google Scholar] [CrossRef]
- Vasques, G.; Grunwald, S.; Sickman, J. Comparison of multivariate methods for inferential modeling of soil carbon using visible/near-infrared spectra. Geoderma 2008, 146, 14–25. [Google Scholar] [CrossRef]
- Dotto, A.C.; Dalmolin, R.S.D.; ten Caten, A.; Grunwald, S. 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 2018, 314, 262–274. [Google Scholar] [CrossRef]
- Rossel, R.V.; Behrens, T. Using data mining to model and interpret soil diffuse reflectance spectra. Geoderma 2010, 158, 46–54. [Google Scholar] [CrossRef]
- Nawar, S.; Buddenbaum, H.; Hill, J.; Kozak, J.; Mouazen, A.M. Estimating the soil clay content and organic matter by means of different calibration methods of vis-NIR diffuse reflectance spectroscopy. Soil Tillage Res. 2016, 155, 510–522. [Google Scholar] [CrossRef] [Green Version]
- Gholizadeh, A.; Borůvka, L.; Saberioon, M.M.; Kozak, J.; Vašát, R.; Němeček, K. Comparing different data preprocessing methods for monitoring soil heavy metals based on soil spectral features. Soil Water Res. 2015, 10, 218–227. [Google Scholar] [CrossRef] [Green Version]
- Breiman, L. Statistical modeling: The two cultures (with comments and a rejoinder by the author). Stat. Sci. 2001, 16, 199–231. [Google Scholar] [CrossRef]
- Legget, R. Soils in Canada: Geological, Pedological and Engineering Studies; University of Toronto Press: Toronto, ON, Canada, 1961. [Google Scholar]
- Hoffman, D.W.; Matthews, B.; Wicklund, R. Soil Survey of Wellington County, Ontario; Research Branch, Canada Department of Agriculture and the Ontario: Ontario, ON, Canada, 1963. [Google Scholar]
- Hoffman, D.W.; Matthews, B.; Wickland, R. Soil Survey of Dufferin County, Ontario; Research Branch, Canada Department of Agriculture: Ontario, ON, Canada, 1964. [Google Scholar]
- Canadian Climate Normals. Available online: https://climate.weather.gc.ca/climate_normals/ (accessed on 13 August 2021).
- Canadian Agricultural Services Coordinating Committee; Soil Classification Working Group; Soil Classification Working Group; National Research Council Canada, Canada; Agriculture and Agri-Food Canada; Research Branch. The Canadian System of Soil Classification; NRC Research Press: Ottawa, ON, Canada, 1998. [Google Scholar]
- Evaluation, O.C.f.S.R.; Irvine, D.; Schut, L.; Denholm, K.A. Field Manual for Describing Soils in Ontario; Ontario Institute of Pedology: Ontario, ON, Canada, 1982. [Google Scholar]
- Thomas, G.W. Soil pH and soil acidity. Methods Soil Anal. Part 3 Chem. Methods 1996, 5, 475–490. [Google Scholar]
- Rhoades, J.; Oster, J. Solute content. Methods Soil Anal. Part 1 Phys. Mineral. Methods 1986, 5, 985–1006. [Google Scholar]
- Vereș, D.Ș. A comparative study between loss on ignition and total carbon analysis on mineralogenic sediments. Studia UBB Geol. 2002, 47, 171–182. [Google Scholar] [CrossRef]
- Gee, G.; Bauder, J. Particle-size analysis. In Methods of Soil Analysis. Part 1. Agron. Monogr. 9; Klute, A., Ed.; ASA and SSSA: Madison, WI, USA, 1986; pp. 383–411. [Google Scholar]
- Vasava, H.B. Spectral Reflectance of Bulk Soil Samples and Their Aggregate Size Fractions for Estimating soil properties; Indian Institute of Technology Kharagpur: West Bengal, India, 2019. [Google Scholar]
- Ji, W.; Adamchuk, V.I.; Biswas, A.; Dhawale, N.M.; Sudarsan, B.; Zhang, Y.; Rossel, R.A.V.; Shi, Z. Assessment of soil properties in situ using a prototype portable MIR spectrometer in two agricultural fields. Biosyst. Eng. 2016, 152, 14–27. [Google Scholar] [CrossRef]
- Nawar, S.; Buddenbaum, H.; Hill, J.; Kozak, J. Modeling and mapping of soil salinity with reflectance spectroscopy and landsat data using two quantitative methods (PLSR and MARS). Remote Sens. 2014, 6, 10813–10834. [Google Scholar] [CrossRef] [Green Version]
- Wold, S.; Martens, H.; Wold, H. The multivariate calibration problem in chemistry solved by the PLS method. In Matrix Pencils; Kågström, B., Ruhe, A., Eds.; Springer: Berlin/Heidelberg, Germany, 1983; pp. 286–293. [Google Scholar]
- Quinlan, J.R. Improved use of continuous attributes in C4. 5. J. Artif. Intell. Res. 1996, 4, 77–90. [Google Scholar] [CrossRef] [Green Version]
- Rossel, R.A.V.; Webster, R. Predicting soil properties from the Australian soil visible–near infrared spectroscopic database. Eur. J. Soil Sci. 2012, 63, 848–860. [Google Scholar] [CrossRef]
- Breiman, L. Random Forests. Mach. Learn. 2001, 45, 5–32. [Google Scholar] [CrossRef] [Green Version]
- Package ‘elmNN’. Available online: https://mran.microsoft.com/snapshot/2018-04-23/web/packages/elmNN/elmNN.pdf (accessed on 13 August 2021).
- Yang, M.; Xu, D.; Chen, S.; Li, H.; Shi, Z. Evaluation of Machine Learning Approaches to Predict Soil Organic Matter and pH Using vis-NIR Spectra. Sensors 2019, 19, 263. [Google Scholar] [CrossRef] [PubMed] [Green Version]
- Team, R.C. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna. Available online: https://www.yumpu.com/en/document/read/6853895/r-a-language-and-environment-for-statistical-computing (accessed on 13 August 2021).
- Kuhn, M.; Weston, S.; Keefer, C.; Coulter, N. Cubist: Rule-and Instance-Based Regression Modeling (Version 0.2. 2). Available online: https://cran.r-project.org/web/packages/Cubist/index.html (accessed on 13 August 2021).
- Kuhn, M.; Wing, J.; Weston, S.; Williams, A.; Keefer, C.; Engelhardt, A.; Cooper, T.; Mayer, Z. Caret: Classification and Regression Training. R Package Version 6.0-84. Available online: https://CRAN.R-project.org/package=caret.2019 (accessed on 13 August 2021).
- Mevik, B.-H.; Wehrens, R.; Liland, K.H. pls: Partial least squares and principal component regression. R Package Version 2011, 2. Available online: https://cran.r-project.org/web/packages/pls/index.html (accessed on 13 August 2021).
- Kennard, R.W.; Stone, L.A. Computer aided design of experiments. Technometrics 1969, 11, 137–148. [Google Scholar] [CrossRef]
- Stevens, A.; Ramirez-Lopez, L. An introduction to the prospectr package. R Package VignetteRep. No. R Package Version 0.1 2014, 3. Available online: http://cran.nexr.com/web/packages/prospectr/index.html (accessed on 13 August 2021).
- Reeves, J.; McCarty, G. Quantitative analysis of agricultural soils using near infrared reflectance spectroscopy and a fibre-optic probe. J. Near Infrared Spectrosc. 2001, 9, 25–34. [Google Scholar] [CrossRef]
- Reeves, J.; McCarty, G.; Meisinger, J. Near infrared reflectance spectroscopy for the analysis of agricultural soils. J. Near Infrared Spectrosc. 1999, 7, 179–193. [Google Scholar] [CrossRef]
- Rossel, R.V.; Adamchuk, V.; Sudduth, K.; McKenzie, N.; Lobsey, C. Proximal soil sensing: An effective approach for soil measurements in space and time. Adv. Agron. 2011, 113, 243–291. [Google Scholar]
- Hobley, E.; Prater, I. Estimating soil texture from vis–NIR spectra. Eur. J. Soil Sci. 2019, 70, 83–95. [Google Scholar] [CrossRef] [Green Version]
- Chen, C.H.; Tanaka, K.; Funatsu, K. Random Forest Model with Combined Features: A Practical Approach to Predict Liquid-crystalline Property. Mol. Inform. 2019, 38, 1800095. [Google Scholar] [CrossRef] [PubMed]
References | Region | n | Model | R2Validation | |||||
---|---|---|---|---|---|---|---|---|---|
SOC/SOM | pH | EC | Sand | Clay | Silt | ||||
Johnson, et al. [7] | SSA | 2845 | PLSR | - | 0.59 | 0.37 | 0.54 | 0.70 | 0.47 |
Gupta, et al. [8] | India | 954 | PLSRLW | 0.70 | - | - | 0.72 | 0.61 | - |
Zhang et al. [2] | Canada | 257 | Cubist | 0.66 | 0.67 | 0.12 | 0.50 | 0.70 | 0.00 |
Conforti, et al. [9] | Italy | 267 | PLSR | 0.88 | 0.70 | - | 0.81 | 0.80 | 0.70 |
Terra, et al. [10] | Brazil | 1259 | SVM | 0.65 | 0.24 | - | 0.89 | 0.86 | - |
Gholizade, et al. [11] | Malaysia | 118 | SMLR | 0.81 | 0.59 | 0.51 | - | - | - |
P Leone, et al. [12] | Italy | 374 | PLSR | 0.91 | - | - | 0.58 | 0.83 | 0.51 |
Lee, et al. [13] | USA | 165 | SMLR | - | - | - | 0.76 | 0.80 | 0.80 |
Viscarra Rossel, et al. [14] | Australia | 116 | PLSR | 0.72 | 0.73 | 0.29 | 0.75 | 0.67 | 0.52 |
Islam, et al. [15] | Australia | 161 | PCR | 0.76 | 0.71 | 0.10 | 0.53 | 0.72 | 0.05 |
Preprocessing Algorithm | Impact | Equation |
---|---|---|
1st Derivative | Reduce the drift of the baseline and highlight some parts of the spectral information [38]. | |
2nd Derivative | Reduce the drift of the baseline and liner trend. Also highlight some parts of the spectral information [38]. | |
Gap Derivative | Remove both additive and multiplicative effects. These methods enhance spectral resolution and eliminate background effects. | |
Savitzky-Golay | Remove the high frequency noise from samples | |
Standard Normal Variate (SNV) | It performs both the centeringand scaling together by subtracting the mean and normalizing with the standard deviation for each reflectance spectrum [38]. | |
Detrend | It involves fitting a 2nd order polynomial to the SNV transformed spectrum and subtracted from it to correct for wavelength dependent scattering effects |
Preprocessing | 1st Derivative, 2nd Derivative, Gap Derivative, Savitzky-Golay, SNV, 1st Derivative + Gap, 2nd Derivative + Gap, Savitzky-Golay + Gap, Savitzky-Golay + 1st Derivative, Savitzky-Golay + 2nd Derivative, Savitzky-Golay + SNV, Savitzky-Golay + SNV + Detrend, SNV + Detrend |
Modeling | Partial Least Square Regression (PLSR), Random Forest (RF), Cubist, Extreme Learning Machine (ELM) |
Indicator | Meaning | Formula |
---|---|---|
R2 | Correlation coefficient of determination explains how well the variance of the spectral predicted values align with the lab measured values | ; is the sum of squared of residuals or predicted, is the total sum of squared |
R2adj | Adjusted R2 or modified version of R2 adjusts for the number variables in the prediction model. While more predictor variables tend to increase (called overfitting) and often return an unwarranted high R2, adjusted R2 can determine how reliable the correlation is and how it is determined by the addition of more predictor variables. It compensates for addition of variables and only increase if the new variable enhances the model above what that would be obtained by chance. | ; is the sum of squared of residuals or predicted, y-measured, x; is the total sum of squared, n is the number of data points and k is the number of variables in the model. |
CCC | Concordance correlation coefficient measuring the agreement between the measured and predicted values of soil properties or reproducibility or how close the predicted values are to the measured values (closeness to 1:1 line). | ; r is the correlation coefficient, is the mean of the measured, is the mean of the predicted, variance of measured and is the variance of the predicted values. |
MSE | Mean squared error measures the average squares of the error or the difference between predicted and measured values. | ; n is the number of data points, are the predicted values and are the measured values. |
RMSE | Root mean squared error measures the difference between values predicted by a model and is the square root of the MSE. | |
MSEc | Mean squared error of calibration dataset measuring how well the calibration worked | Same as MSE but for calibration dataset |
RMSEc | Root mean squared error of calibration measuring how well the calibration worked | Same as RMSE but for calibration dataset |
RPD | Ratio of performance of deviation or the ratio between the standard deviation of a variable and the standard error of prediction | ; SD is the standard deviation of the sample and SEP is the standard error of prediction (calculated as RMSE) |
RPIQ | Ratio of performance of interquartile distance is the interquartile range of the measured values divided by the RMSE | ; IQ is the interquartile range and SEP is the standard error of prediction (calculated as RMSE) |
Properties | Mean | Median | Min | Max | σ | n |
---|---|---|---|---|---|---|
EC, μs cm−1 | 309.30 | 265.90 | 26.25 | 2034.00 | 197.86 | 1038 |
SOM, % | 2.69 | 2.11 | 0.39 | 17.13 | 1.82 | 1025 |
pH | 7.71 | 7.71 | 5.08 | 9.10 | 0.55 | 1041 |
Sand, % | 45.11 | 41.85 | 0.49 | 93.91 | 20.20 | 238 |
Silt, % | 43.23 | 45.46 | 4.70 | 87.86 | 17.07 | 238 |
Clay, % | 11.67 | 10.68 | 1.38 | 31.73 | 6.23 | 238 |
VCS, % | 3.69 | 2.06 | 0.03 | 41.29 | 5.57 | 208 |
CS, % | 5.66 | 3.80 | 0.00 | 46.15 | 6.45 | 208 |
ms, % | 15.57 | 12.28 | 0.91 | 69.94 | 10.96 | 208 |
fs, % | 22.82 | 21.66 | 1.32 | 68.47 | 11.23 | 208 |
Preprocessing Algorithms | Calibration R2adj | Validation R2adj | ||||||
---|---|---|---|---|---|---|---|---|
PLSR | Cubist | RF | ELM | PLSR | Cubist | RF | ELM | |
1st Derivative | 0.81 | 0.84 | 0.97 | 0.45 | 0.75 | 0.79 | 0.79 | 0.62 |
1st Derivative + Gap | 0.77 | 0.91 | 0.97 | 0.63 | 0.83 | 0.89 | 0.87 | 0.77 |
2nd Derivative | 0.73 | 0.76 | 0.97 | 0.14 | 0.70 | 0.70 | 0.70 | 0.13 |
2nd Derivative + Gap | 0.76 | 0.88 | 0.97 | 0.49 | 0.84 | 0.88 | 0.87 | 0.81 |
Savitzky-Golay + Gap | 0.74 | 0.75 | 0.97 | 0.67 | 0.83 | 0.69 | 0.84 | 0.76 |
Gap Derivative | 0.77 | 0.80 | 0.97 | 0.75 | 0.71 | 0.77 | 0.77 | 0.70 |
Savitzky-Golay | 0.77 | 0.89 | 0.97 | 0.71 | 0.78 | 0.82 | 0.80 | 0.70 |
Savitzky-Golay + 1st Derivative | 0.79 | 0.70 | 0.97 | 0.62 | 0.74 | 0.61 | 0.78 | 0.40 |
Savitzky-Golay + 2nd Derivative | 0.78 | 0.61 | 0.97 | 0.40 | 0.68 | 0.37 | 0.71 | 0.29 |
Savitzky-Golay + SNV | 0.72 | 0.92 | 0.96 | 0.28 | 0.64 | 0.76 | 0.75 | 0.20 |
Savitzky-Golay + SNV + Detrend | 0.74 | 0.89 | 0.96 | 0.58 | 0.52 | 0.64 | 0.66 | 0.32 |
SNV | 0.77 | 0.90 | 0.96 | 0.71 | 0.59 | 0.70 | 0.66 | 0.56 |
SNV + Detrend | 0.78 | 0.90 | 0.96 | 0.57 | 0.59 | 0.65 | 0.71 | 0.26 |
R2 | CCC | MSE | RMSE | Bias | MSEc | RMSEc | RPD | RPIQ | ||
---|---|---|---|---|---|---|---|---|---|---|
PLSR | Calibration | 0.76 | 0.86 | 0.88 | 0.94 | 0.00 | 0.88 | 0.94 | 2.03 | 2.46 |
Validation | 0.73 | 0.85 | 0.76 | 0.87 | 0.04 | 0.76 | 0.87 | 1.86 | 2.11 | |
External Validation | 0.75 | 0.86 | 0.72 | 0.85 | 0.00 | 0.72 | 0.85 | 2.00 | 2.52 | |
Cubist | Calibration | 0.83 | 0.90 | 0.63 | 0.79 | −0.08 | 0.62 | 0.79 | 2.41 | 2.91 |
Validation | 0.82 | 0.89 | 0.48 | 0.69 | 0.03 | 0.48 | 0.69 | 2.35 | 2.65 | |
External Validation | 0.81 | 0.89 | 0.56 | 0.75 | −0.08 | 0.55 | 0.74 | 2.27 | 2.87 | |
RF | Calibration | 0.94 | 0.95 | 0.29 | 0.54 | 0.01 | 0.29 | 0.54 | 3.53 | 4.28 |
Validation | 0.65 | 0.76 | 0.95 | 0.97 | 0.09 | 0.94 | 0.97 | 1.67 | 1.89 | |
External Validation | 0.67 | 0.78 | 0.96 | 0.98 | 0.03 | 0.96 | 0.98 | 1.73 | 2.18 | |
ELM | Calibration | 0.57 | 0.72 | 1.57 | 1.25 | 0.00 | 1.57 | 1.25 | 1.52 | 1.84 |
Validation | 0.60 | 0.75 | 1.13 | 1.06 | 0.23 | 1.08 | 1.04 | 1.53 | 1.73 | |
External Validation | 0.60 | 0.76 | 1.17 | 1.08 | 0.11 | 1.16 | 1.08 | 1.57 | 1.98 |
Properties. | 1st Derivative + Gap | 2nd Derivative + Gap | SNV | |||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
A | B | C | D | A | B | C | D | A | B | C | D | |
SOM, % | 0.83 | 0.89 | 0.87 | 0.77 | 0.84 | 0.88 | 0.87 | 0.81 | 0.59 | 0.70 | 0.66 | 0.56 |
EC, μs cm−1 | −0.02 | 0.00 | −0.02 | −0.02 | −0.01 | −0.03 | −0.03 | 0.22 | −0.01 | −0.03 | −0.03 | 0.22 |
pH | 0.57 | 0.62 | 0.63 | 0.52 | 0.48 | 0.54 | 0.53 | 0.48 | 0.48 | 0.54 | 0.53 | 0.48 |
Sand, % | 0.48 | 0.47 | 0.70 | 0.53 | 0.29 | 0.40 | 0.46 | 0.45 | 0.29 | 0.40 | 0.46 | 0.45 |
Silt, % | 0.46 | 0.53 | 0.70 | 0.60 | 0.40 | 0.39 | 0.42 | 0.25 | 0.4 | 0.39 | 0.42 | 0.25 |
Clay, % | 0.13 | 0.26 | 0.20 | 0.19 | 0.23 | 0.20 | 0.25 | 0.25 | 0.23 | 0.20 | 0.25 | 0.25 |
VCS, % | 0.18 | −0.02 | 0.17 | 0.04 | 0.11 | 0.00 | 0.02 | −0.01 | 0.11 | 0.00 | 0.02 | −0.01 |
CS, % | 0.68 | 0.08 | 0.15 | 0.46 | 0.30 | 0.58 | 0.22 | 0.02 | 0.30 | 0.58 | 0.22 | 0.02 |
ms, % | 0.50 | 0.24 | 0.53 | 0.39 | 0.31 | 0.28 | 0.32 | 0.09 | 0.31 | 0.28 | 0.32 | 0.09 |
fs, % | −0.01 | 0.49 | −0.02 | −0.02 | 0.01 | 0.03 | 0.14 | −0.01 | 0.01 | 0.03 | 0.14 | −0.01 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Vestergaard, R.-J.; Vasava, H.B.; Aspinall, D.; Chen, S.; Gillespie, A.; Adamchuk, V.; Biswas, A. Evaluation of Optimized Preprocessing and Modeling Algorithms for Prediction of Soil Properties Using VIS-NIR Spectroscopy. Sensors 2021, 21, 6745. https://doi.org/10.3390/s21206745
Vestergaard R-J, Vasava HB, Aspinall D, Chen S, Gillespie A, Adamchuk V, Biswas A. Evaluation of Optimized Preprocessing and Modeling Algorithms for Prediction of Soil Properties Using VIS-NIR Spectroscopy. Sensors. 2021; 21(20):6745. https://doi.org/10.3390/s21206745
Chicago/Turabian StyleVestergaard, Rebecca-Jo, Hiteshkumar Bhogilal Vasava, Doug Aspinall, Songchao Chen, Adam Gillespie, Viacheslav Adamchuk, and Asim Biswas. 2021. "Evaluation of Optimized Preprocessing and Modeling Algorithms for Prediction of Soil Properties Using VIS-NIR Spectroscopy" Sensors 21, no. 20: 6745. https://doi.org/10.3390/s21206745
APA StyleVestergaard, R. -J., Vasava, H. B., Aspinall, D., Chen, S., Gillespie, A., Adamchuk, V., & Biswas, A. (2021). Evaluation of Optimized Preprocessing and Modeling Algorithms for Prediction of Soil Properties Using VIS-NIR Spectroscopy. Sensors, 21(20), 6745. https://doi.org/10.3390/s21206745