Determination of in-situ salinized soil moisture content from visible-near infrared (VIS–NIR) spectroscopy by fractional order derivative and spectral variable selection algorithms

Congcong Lao, Zhitao Zhang, Junying Chen, Haorui Chen, Zhihua Yao, Zheng Xing, Xiang Tai, Jifeng Ning, Yinwen Chen

Abstract


Abstract: The measurement of salinized soil moisture content (SSMC) is essential to precise irrigation and avoidance of secondary salinization.  Visible and near infrared (VIS–NIR) spectroscopy has been effectively used to estimate soil moisture content (SMC) but not for SSMC.  The direct application of in-situ VIS–NIR spectroscopy to the estimation of SSMC can help save a large amount of time and labor, but the in-situ VIS–NIR was interfered by many factors, such as soil texture, soil surface debris and environmental temperature.  Spectral derivatives can be used to eliminate unnecessary interference for optimal spectral information, but traditional integer derivatives (i.e. first and second derivatives) often ignored some spectral information due to different integer order differential curves were obviously different.  In addition, the full spectrum usually contains redundant spectral variables.  These variables would affect the accuracy and estimation velocity of the model.  Different combinations of fractional order derivative (FOD) and spectral variable selection algorithms (i.e. variable importance projection (VIP), competitive adaptive weighted sampling (CARS) and random frog algorithm (RFA)) may offer some alternative solutions to these problems.  In order to test the effects of these combinations on VIS–NIR spectral model optimization, we measured the in-situ soil spectra of 163 sites in Shahaoqu Irrigation Area, Inner Mongolia, China.  Meanwhile, we collected soil samples and measured their SSMC and soil salt content (SSC).  Then the Extreme Learning Machine (ELM) model was applied to the SSMC estimation.  The results showed that SSC and SSMC had obvious effects on in-situ spectra.  With the increase of differential order, the spectral resolution increased gradually, but the spectral intensity decreased at the same time.  So, the spectral information may not increase.  However, FOD can balance the contradiction between spectral resolution and spectral intensity.  The estimation of ELM models based on 0.75 order derivatives that is the most accurate among the full spectrum ELM models.  The coefficient of determination (R2) was 0.83 and ratio of the performance to deviation (RPD) was 2.44.  In all the models (twenty-seven different combinations of FOD and variable selection algorithms), the best model was based on the combination of 0.75 derivative spectrum and random frog algorithm (R2 = 0.94, RPD = 3.80).  The results of this study also confirmed that the combination of RFA and FOD could effectively improve the accuracy of the in-situ spectral estimation of SSMC.  However, VIP was chosen as an alternative due to computational efficiency.

Keywords: salinized soil moisture content, in-situ visible and near-infrared spectroscopy, fractional order derivative, random frog algorithm, extreme learning machine

DOI: 10.33440/j.ijpaa.20200303.98

 

Citation: Lao C C, Zhang Z T, Chen J Y, Chen H R, Yao Z H, Xing Z, Tai X, Ning J F, Chen Y W.  Determination of in-situ salinized soil moisture content from visible-near infrared (VIS–NIR) spectroscopy by fractional order derivative and spectral variable selection algorithms.  Int J Precis Agric Aviat, 2020; 3(3): 21 –34.


Full Text:

PDF

References


Kevin Z M, Nashon K R M, Moses M, et al. The Role of Moisture in the Successful Rehabilitation of Denuded Patches of a Semi-Arid Environment in Kenya. Journal of Environmental Science and Technology, 2010; 3(4): 195–207. doi: 10.3923/jest.2010.195.207.

Chen L, Feng Q, Li F, et al. A bidirectional model for simulating soil water flow and salt transport under mulched drip irrigation with saline water. Agricultural Water Management, 2014; 146: 24–33. doi: 10.1016/j.agwat.2014.07.021.

Susha Lekshmi S U, Singh D N, Shojaei Baghini M. A critical review of soil moisture measurement. Measurement, 2014; 54: 92–105. doi: 10.1016/j.measurement.2014.04.007.

Robinet J, von Hebel C, Govers G, et al. Spatial variability of soil water content and soil electrical conductivity across scales derived from Electromagnetic Induction and Time Domain Reflectometry. Geoderma, 2018; 314(1): 60–74. doi: 10.1016/j.geoderma.2017.10.045.

Eller H, Denoth H. A capacitive soil moisture sensor. Journal of Hydrology, 1996; 185(1): 37–46. doi: 10.1016/0022-1694(95)03003-4.

Xu D, Sun R, Yeh T J, et al. Mapping soil layers using electrical resistivity tomography and validation: Sandbox experiments. Journal of Hydrology, 2019; 575(5): 23–36. doi: 10.1016/j.jhydrol.2019.05.036.

Alamry A S, Meijde M, Noomen M, et al. Spatial and temporal monitoring of soil moisture using surface electrical resistivity tomography in Mediterranean soils. Catena, 2017; 157(3): 88–96. doi: 10.1016/ j.catena.2017.06.001.

Wang Q, Li P, Pu Z, et al. Calibration and validation of salt-resistant hyperspectral indices for estimating soil moisture in arid land. Journal of Hydrology, 2011; 408(2): 76–85. doi: 10.1016/j.jhydrol.2011.08.012.

Hong Y, Chen Y, Yu L, et al. Combining Fractional Order Derivative and Spectral Variable Selection for Organic Matter Estimation of Homogeneous Soil Samples by VIS–NIR Spectroscopy. Remote Sens.-Basel, 2018; 10: 479. doi: 10.3390/rs10030479.

Nawar S, Buddenbaum H, Hill J, et al. Estimating the soil clay content and organic matter by means of different calibration methods of vis-NIR diffuse reflectance spectroscopy. Soil and Tillage Research, 2016; 155(5): 10-22. doi: 10.1016/j.still.2015.07.021.

Yue J, Tian Q, Tang S, et al. A dynamic soil endmember spectrum selection approach for soil and crop residue linear spectral unmixing analysis. International Journal of Applied Earth Observation and Geoinformation, 2019; 78(3): 06–17. doi:10.1016/j.jag.2019.02.001.

Wang X, Zhang F, Kung H, et al. Spectral response characteristics and identification of typical plant species in Ebinur lake wetland national nature reserve (ELWNNR) under a water and salinity gradient. Ecological Indicators, 2017; 81(2): 22–34. doi: 10.1016/j.ecolind.2017.05.071.

Ji R, Zhao Z, Yu X, et al. Determination of rhodamine B in capsicol using the first derivative absorption spectrum. Optik, 2019; 181: 796–801. doi: 10.1016/j.ijleo.2018.12.141.

Zhang D, Tiyip T, Ding J, et al. Quantitative Estimating Salt Content of Saline Soil Using Laboratory Hyperspectral Data Treated by Fractional Derivative. Journal of Spectroscopy, 2016; 2016: 1–11. doi: 10.1155/ 2016/1081674.

Wang J, Ding J, Abulimiti A, et al. Quantitative estimation of soil salinity by means of different modeling methods and visible-near infrared (VIS–NIR) spectroscopy, Ebinur Lake Wetland, Northwest China. PeerJ, 2018; 6: e4703. doi: 10.7717/peerj.4703.

Wang X, Zhang F, Kung H, et al. New methods for improving the remote sensing estimation of soil organic matter content (SOMC) in the Ebinur Lake Wetland National Nature Reserve (ELWNNR) in northwest China. Remote Sensing of Environment, 2018; 218(1): 04–18. doi: 10.1016/ j.rse.2018.09.020.

Hong Y, Liu Y, Chen Y, et al. Application of fractional-order derivative in the quantitative estimation of soil organic matter content through visible and near-infrared spectroscopy. Geoderma, 2019; 337(7): 58–69. doi: 10.1016/j.geoderma.2018.10.025.

Vohland M, Ludwig M, Thiele-Bruhn S, et al. Quantification of Soil Properties with Hyperspectral Data: Selecting Spectral Variables with Different Methods to Improve Accuracies and Analyze Prediction Mechanisms. Remote Sens.-Basel, 2017; 9: 1103. doi: 10.3390/rs9111103.

Rossel R A V, Behrens T. Using data mining to model and interpret soil diffuse reflectance spectra. Geoderma, 2010; 158: 46–54. doi: 10.1016/j.geoderma.2009.12.025.

Liu X, Xu L. The universal consistency of extreme learning machine. Neurocomputing, 2018; 311: 176–82. doi: 10.1016/j.neucom.2018.05. 066.

Khosravi V, Doulati Ardejani F, Yousefi S, et al. Monitoring soil lead and zinc contents via combination of spectroscopy with extreme learning machine and other data mining methods. Geoderma, 2018; 318: 29–41. doi: 10.1016/j.geoderma.2017.12.025.

Cai L, Ding J. Prediction for Soil Water Content Based on Variable Preferred and Extreme Learning. Spectroscopy and Spectral Analysis, 2018; 38(7): 2209–2214. doi: 10.3964/j.issn.1000-0593(2018)07-2209-06. (in Chinese)

Vohland M, Ludwig M, Thiele-Bruhn S, et al. Determination of soil properties with visible to near- and mid-infrared spectroscopy: Effects of spectral variable selection. Geoderma, 2014; 223(225): 88–96. doi: 10.1016/j.geoderma.2014.01.013.

Sarathjith M C, Das B S, Wani S P, et al. Variable indicators for optimum wavelength selection in diffuse reflectance spectroscopy of soils. Geoderma, 2016; 267: 1–9. doi: 10.1016/j.geoderma.2015.12.031.

Zou X, Zhao J, Malcolm J W, et al. Variables selection methods in near-infrared spectroscopy. Analytica Chimica Acta, 2010; 667(1-2): 14–32. doi: 10.1016/j.aca.2010.03.048.

Bin J, Ai F, Fan W, et al. An efficient variable selection method based on variable permutation and model population analysis for multivariate calibration of NIR spectra. Chemometrics & Intelligent Laboratory Systems, 2016; 158: 1–13. doi: 10.1016/j.chemolab.2016.08.006.

Chong I, Jun, C. Performance of some variable selection methods when multicollinearity is present. Chemometrics & Intelligent Laboratory Systems, 2005; 78(1): 03–12. doi: 10.1016/j.chemolab.2004.12.011.

Abrahamsson C, Johansson J, Sparén A, et al. Comparison of different variable selection methods conducted on NIR transmission measurements on intact tablets. Chemometrics & Intelligent Laboratory Systems, 2003; 69: 3–12. doi: 10.1016/S0169-7439(03)00064-9.

Araújo M C U, Saldanha T C B, Galvão R K H, et al. The successive projections algorithm for variable selection in spectroscopic multicomponent analysis. Chemometr. Intell. Lab. 2001; 57(2): 65–73. doi: 10.1016/S0169-7439(01)00119-8.

Wang H, Chen Y, Zhang Z, et al. Quantitatively estimating main soil water-soluble salt ions content based on Visible-near infrared wavelength selected using GC, SR and VIP. Peer J, 2019, 7: e6310. doi: 10.7717/ peerj.6310.

Xu S, Zhao Y, Wang M, et al. Determination of rice root density from Vis–NIR spectroscopy by support vector machine regression and spectral variable selection techniques. Catena, 2017; 157: 12–23. doi: 10.1016/ j.catena.2017.05.008.

Oussama A, Elabadi F, Platikanov S, et al. Detection of Olive Oil Adulteration Using FT-IR Spectroscopy and PLS with Variable Importance of Projection (VIP) Scores. Journal of the American Oil Chemists' Society, 2012; 89(18): 07–12. doi: 10.1007/s11746-012-2091-1.

Rahman A, Faqeerzada M A, Joshi R, et al. Quality Analysis of Stored Bell Peppers Using Near-Infrared Hyperspectral Imaging. Transactions of the Asabe, 2018; 61(1): 199–207. doi: 10.13031/trans.12482.

GREEN B P J. Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika, 1995; 4(7): 11–32. doi: 10.1093/biomet/82.4.711.

Li H, Xu Q, Liang Y. Random frog: An efficient reversible jump Markov Chain Monte Carlo-like approach for variable selection with applications to gene selection and disease classification. Analytica Chimica Acta, 2012; 740(31): 20–6. doi: 10.1016/j.aca.2012.06.031.

Hu M, Dong Q, Liu B, et al. Estimating blueberry mechanical properties based on random frog selected hyperspectral data. Postharvest Biology and Technology, 2015; 106(3): 1–10. doi: 10.1016/j.postharvbio.2015.03. 014.

Yu R, Liu T, Xu Y, et al. Analysis of salinization dynamics by remote sensing in Hetao Irrigation District of North China. Agricultural Water Management, 2010; 97(19): 52–60. doi: 10.1016/j.agwat.2010.03.009.

Gao X, Huo Z, Bai Y, et al. Soil salt and groundwater change in flood irrigation field and uncultivated land: a case study based on 4-year field observations. Environmental Earth Sciences, 2015; 73(21): 27–39. doi: 10.1007/s12665-014-3563-4.

Xu C, Zeng W, Huang J, et al. Prediction of Soil Moisture Content and Soil Salt Concentration from Hyperspectral Laboratory and Field Data. Remote Sensing, 2016; 8(42): 1–20. doi: 10.3390/rs8010042.

Huang Q, Xu X, Lü L, et al. Soil salinity distribution based on remote sensing and its effect on crop growth in Hetao Irrigation District. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018; 34(1): 102–109. doi: 10.11975/ j.issn.1002-6819.2018.01.014. (in Chinese)

Li H, Liang Y, Xu Q, et al. Key wavelengths screening using competitive adaptive reweighted sampling method for multivariate calibration. Anal Chim Acta, 2009; 648(1): 77–84. doi: 10.1016/j.aca.2009.06.046.

Li H, Xu Q, Liang Y. libPLS: An integrated library for partial least squares regression and linear discriminant analysis. Chemometr Intell Lab, 2018; 176(15): 34–43. doi: 10.1016/j.chemolab.2018.03.003.

Huang G, Zhu Q, Siew C. Extreme learning machine: Theory and applications. Neurocomputing, 2006; (70): 489–501. doi: 10.1016/ j.neucom.2005.12.126.

Hong Y, Chen S, Zhang Y, et al. 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(12): 32–43. doi: 10.1016/j.scitotenv.2018. 06.319.

Viscarra Rossel R A, Taylor H J, McBratney A B. Multivariate calibration of hyperspectral ray energy spectra for proximal soil sensing. Eur J Soil Sci, 2007; 58(3): 43–53. doi: 10.1111/j.1365-2389.2006. 00859.x.

Guo Y, Ni Y, Kokot S. Evaluation of chemical components and properties of the jujube fruit using near infrared spectroscopy and chemometrics. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2016; (153): 79–86. doi: 10.1016/j.saa.2015.08.006.

Morellos A, Pantazi X, Moshou D, Alexandridis T, Whetton R, Tziotzios G, Wiebensohn J, Bill R, Mouazen A M. Machine learning based prediction of soil total nitrogen, organic carbon and moisture content by using VIS-NIR spectroscopy. Biosyst. Eng, 2016; 152(1): 04–16. doi: 10.1016/j.biosystemseng.2016.04.018.

Oltra-Carrió R, Baup F, Fabre S, et al. Improvement of Soil Moisture Retrieval from Hyperspectral VNIR-SWIR Data Using Clay Content Information: From Laboratory to Field Experiments. Remote Sens.-Basel, 2015; 7(3): 184–205. doi: 10.3390/rs70303184.

Tian G, Yuan H, Chu X, et al. Near Infrared Spectra (NIR) Analysis of Octane Number by Wavelet Denoising-Derivative Method. Spectroscopy and Spectral Analysis, 2005; 25(04): 516–520. doi: 10.1016/j.saa.2004. 06.052.

Hong Y, Chen S, Liu Y, et al. Combination of fractional order derivative and memory-based learning algorithm to improve the estimation accuracy of soil organic matter by visible and near-infrared spectroscopy. Catena, 2019; 174(1): 04–16. doi: 10.1016/j.catena.2018.10.051.

Raj A, Chakraborty S, Duda B M, et al. Soil mapping via diffuse reflectance spectroscopy based on variable indicators: An ordered predictor selection approach. Geoderma, 2018; (314): 146–59. doi: 10.1016/ j.geoderma.2017.10.043.

Guezenoc J, Bassel L, Gallet-Budynek A, et al. Variables selection: A critical issue for quantitative laser-induced breakdown spectroscopy. Spectrochimica Acta Part B: Atomic Spectroscopy, 2017; (134): 6–10. doi: 10.1016/j.sab.2017.05.009.

Jia S, Li H, Wang Y, et al. Recursive variable selection to update near-infrared spectroscopy model for the determination of soil nitrogen and organic carbon. Geoderma, 2016; 268(9): 2–9. doi: 10.1016/ j.geoderma.2016.01.018.

Vohland M, Ludwig M, Thiele-Bruhn S, et al. Determination of soil properties with visible to near- and mid-infrared spectroscopy: Effects of spectral variable selection. Geoderma, 2014; 223(225): 88–96. doi: 10.1016/j.geoderma.2014.01.013.

Yao X, Yang W, Li M, et al. Prediction of Total Nitrogen in Soil Based on Random Frog Leaping Wavelet Neural Network. IFAC-PapersOnLine, 2018; 51(6): 1–5. doi: 10.1016/j.ifacol.2018.08.121.

Wijewardane N K, Ge Y, Morgan, et al. Moisture insensitive prediction of soil properties from VNIR reflectance spectra based on external parameter orthogonalization. Geoderma, 2016; 26(7): 92–101. doi: 10.1016/j.geoderma.2015.12.014.

Yun Y, Li H, Wood L R, et al. An efficient method of wavelength interval selection based on random frog for multivariate spectral calibration. Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy, 2013; 111(3): 1–6. doi: 10.1016/j.saa.2013.03.083.

Yu L, Zhu Y, Hong Y, et al. Determination of soil moisture content by hyperspectral technology with CARS algorithm. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2016; 32(22): 138–145. doi: 10.11975/j.issn.1002-6819.2016.22.019. (in Chinese)

Zhang K, Luo M. Outlier-robust extreme learning machine for regression problems. Neurocomputing, 2015; 151(15): 19–27. doi: 10.1016/ j.neucom.2014.09.022.


Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.