OpenLB
Open Library of Bioscience
Advanced Search
Scientific reports
09 01, 2021;11 (1): 17503.

Wavelet geographically weighted regression for spectroscopic modelling of soil properties.

Yongze Song, Zefang Shen, Peng Wu, R A Viscarra Rossel
Author Information
  1. Yongze Song: School of Design and the Built Environment, Curtin University, GPO Box U1987, Perth, WA, 6845, Australia.
  2. Zefang Shen: Soil and Landscape Science, School of Molecular and Life Sciences, Curtin University, GPO Box U1987, Perth, WA, 6845, Australia.
  3. Peng Wu: School of Design and the Built Environment, Curtin University, GPO Box U1987, Perth, WA, 6845, Australia.
  4. R A Viscarra Rossel: Soil and Landscape Science, School of Molecular and Life Sciences, Curtin University, GPO Box U1987, Perth, WA, 6845, Australia. r.viscarra-rossel@curtin.edu.au.
PMID: 34471173 DOI: 10.1038/s41598-021-96772-z

Abstract

Soil properties, such as organic carbon, pH and clay content, are critical indicators of ecosystem function. Visible-near infrared (vis-NIR) reflectance spectroscopy has been widely used to cost-efficiently estimate such soil properties. Multivariate modelling, such as partial least squares regression (PLSR), and machine learning are the most common methods for modelling soil properties with spectra. Often, such models do not account for the multiresolution information presented in the vis-NIR signal, or the spatial variation in the data. To address these potential shortcomings, we used wavelets to decompose the vis-NIR spectra of 226 soils from agricultural and forested regions in south-western Western Australia and developed a wavelet geographically weighted regression (WGWR) for estimating soil organic carbon content, clay content and pH. To evaluate the WGWR models, we compared them to linear models derived with multiresolution data from a wavelet decomposition (WLR) and PLSR without multiresolution information. Overall, validation of the WGWR models produced more accurate estimates of the soil properties than WLR and PLSR. Around 3.5-49.1% of the improvement in the estimates was due to the multiresolution analysis and 1.0-5.2% due to the integration of spatial information in the modelling. The WGWR improves the modelling of soil properties with spectra.

References

  1. Schmidt, M. W. et al. Persistence of soil organic matter as an ecosystem property. Nature 478, 49–56 (2011). [PMID: 21979045]
  2. Drobnik, T., Greiner, L., Keller, A. & Grêt-Regamey, A. Soil quality indicators-from soil functions to ecosystem services. Ecol. Ind. 94, 151–169 (2018). [DOI: 10.1016/j.ecolind.2018.06.052]
  3. Bradford, M. A. et al. Managing uncertainty in soil carbon feedbacks to climate change. Nat. Clim. Change 6, 751–758 (2016). [DOI: 10.1038/nclimate3071]
  4. Viscarra Rossel, R. et al. A global spectral library to characterize the world’s soil. Earth Sci. Rev. 155, 198–230 (2016).
  5. Amundson, R. & Biardeau, L. Opinion: Soil carbon sequestration is an elusive climate mitigation tool. Proc. Natl. Acad. Sci. 115, 11652–11656 (2018). [PMID: 30425181]
  6. Smith, P. et al. How to measure, report and verify soil carbon change to realize the potential of soil carbon sequestration for atmospheric greenhouse gas removal. Glob. Change Biol. 26, 219–241 (2020). [DOI: 10.1111/gcb.14815]
  7. Zhang, S., Yu, Z., Lin, J. & Zhu, B. Responses of soil carbon decomposition to drying-rewetting cycles: A meta-analysis. Geoderma 361, 114069 (2020). [DOI: 10.1016/j.geoderma.2019.114069]
  8. Bot, A. & Benites, J. The Importance of Soil Organic Matter: Key to Drought-Resistant Soil and Sustained Food Production. 80 (Food & Agriculture Org., 2005).
  9. Rawles, W. J. & Brakensiek, D. Estimating soil water retention from soil properties. J. Irrig. Drain. Div. 108, 166–171 (1982). [DOI: 10.1061/JRCEA4.0001383]
  10. Zhao, D., Zhao, X., Khongnawang, T., Arshad, M. & Triantafilis, J. A Vis–NIR spectral library to predict clay in Australian cotton growing soil. Soil Sci. Soc. Am. J. 82, 1347–1357 (2018). [DOI: 10.2136/sssaj2018.03.0100]
  11. Demattê, J. A., Campos, R. C., Alves, M. C., Fiorio, P. R. & Nanni, M. R. Visible–NIR reflectance: A new approach on soil evaluation. Geoderma 121, 95–112 (2004). [DOI: 10.1016/j.geoderma.2003.09.012]
  12. Viscarra Rossel, R. Fine-resolution multiscale mapping of clay minerals in Australian soils measured with near infrared spectra. J. Geophys. Res. Earth Surf. 116 (2011).
  13. Viscarra Rossel, R., 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 131, 59–75 (2006).
  14. Viscarra Rossel, R. & Lark, R. Improved analysis and modelling of soil diffuse reflectance spectra using wavelets. Eur. J. Soil Sci. 60, 453–464 (2009).
  15. Abdi, H. & Williams, L. J. Principal component analysis. Wiley Interdiscip. Rev. Comput. Stat. 2, 433–459 (2010). [DOI: 10.1002/wics.101]
  16. Næs, T. & Martens, H. Principal component regression in NIR analysis: Viewpoints, background details and selection of components. J. Chemom. 2, 155–167 (1988). [DOI: 10.1002/cem.1180020207]
  17. Geladi, P. & Kowalski, B. R. Partial least-squares regression: A tutorial. Anal. Chim. Acta 185, 1–17 (1986). [DOI: 10.1016/0003-2670(86)80028-9]
  18. Rossel, R. V. Robust modelling of soil diffuse reflectance spectra by “bagging-partial least squares regression”. J. Near Infrared Spectrosc. 15, 39–47 (2007).
  19. Viscarra Rossel, R. & Behrens, T. Using data mining to model and interpret soil diffuse reflectance spectra. Geoderma 158, 46–54 (2010).
  20. Tsakiridis, N. L., Keramaris, K. D., Theocharis, J. B. & Zalidis, G. C. Simultaneous prediction of soil properties from VNIR–SWIR spectra using a localized multi-channel 1-d convolutional neural network. Geoderma 367, 114208 (2020). [DOI: 10.1016/j.geoderma.2020.114208]
  21. Yang, J., Wang, X., Wang, R. & Wang, H. Combination of convolutional neural networks and recurrent neural networks for predicting soil properties using Vis–NIR spectroscopy. Geoderma 380, 114616 (2020). [DOI: 10.1016/j.geoderma.2020.114616]
  22. Shen, Z. & Viscarra Rossel, R. A. Automated spectroscopic modelling with optimised convolutional neural networks. Sci. Rep. 11, 208. https://doi.org/10.1038/s41598-020-80486-9 (2021). [DOI: 10.1038/s41598-020-80486-9]
  23. Li, F., Wang, L., Liu, J., Wang, Y. & Chang, Q. Evaluation of leaf n concentration in winter wheat based on discrete wavelet transform analysis. Remote Sens. 11, 1331 (2019). [DOI: 10.3390/rs11111331]
  24. Meng, X. et al. Regional soil organic carbon prediction model based on a discrete wavelet analysis of hyperspectral satellite data. Int. J. Appl. Earth Obs. Geoinf. 89, 102111 (2020).
  25. Jiang, B. Geospatial analysis requires a different way of thinking: The problem of spatial heterogeneity. GeoJournal 80, 1–13 (2015). [DOI: 10.1007/s10708-014-9537-y]
  26. Song, Y., Wang, J., Ge, Y. & Xu, C. An optimal parameters-based geographical detector model enhances geographic characteristics of explanatory variables for spatial heterogeneity analysis: Cases with different types of spatial data. GISci. Remote Sens. 57, 593–610 (2020). [DOI: 10.1080/15481603.2020.1760434]
  27. Yang, Z. et al. The effect of environmental heterogeneity on species richness depends on community position along the environmental gradient. Sci. Rep. 5, 1–7 (2015).
  28. Jenny, H. Factors of Soil Formation (McGraw-Hill, 1941).
  29. Ye, H. et al. Effects of different sampling densities on geographically weighted regression kriging for predicting soil organic carbon. Spat. Stat. 20, 76–91 (2017). [DOI: 10.1016/j.spasta.2017.02.001]
  30. Viscarra Rossel, R. & Webster, R. Predicting soil properties from the Australian soil visible-near infrared spectroscopic database. Eur. J. Soil Sci. 63. https://doi.org/10.1111/j.1365-2389.2012.01495.x (2012).
  31. Sila, A., Pokhariyal, G. & Shepherd, K. Evaluating regression-kriging for mid-infrared spectroscopy prediction of soil properties in western Kenya. Geoderma Reg. 10, 39–47 (2017). [DOI: 10.1016/j.geodrs.2017.04.003]
  32. Fotheringham, A. S., Brunsdon, C. & Charlton, M. Geographically Weighted Regression: The Analysis of Spatially Varying Relationships (Wiley, 2003).
  33. Brunsdon, C., Fotheringham, A. S. & Charlton, M. E. Geographically weighted regression: A method for exploring spatial nonstationarity. Geogr. Anal. 28, 281–298 (1996). [DOI: 10.1111/j.1538-4632.1996.tb00936.x]
  34. Bidanset, P. E. & Lombard, J. R. The effect of kernel and bandwidth specification in geographically weighted regression models on the accuracy and uniformity of mass real estate appraisal. J. Prop. Tax Assess. Admin. 11, 5–14 (2014).
  35. Brunsdon, C., Fotheringham, A. & Charlton, M. Geographically weighted summary statistics? A framework for localised exploratory data analysis. Comput. Environ. Urban Syst. 26, 501–524 (2002). [DOI: 10.1016/S0198-9715(01)00009-6]
  36. Comber, A. et al. The GWR route map: A guide to the informed application of geographically weighted regression. arXiv preprint arXiv:2004.06070 (2020).
  37. Fotheringham, A. S., Yang, W. & Kang, W. Multiscale geographically weighted regression (MGWR). Ann. Am. Assoc. Geogr. 107, 1247–1265 (2017).
  38. Yu, H. et al. Inference in multiscale geographically weighted regression. Geogr. Anal. 52, 87–106 (2020). [DOI: 10.1111/gean.12189]
  39. Wheeler, D. & Tiefelsdorf, M. Multicollinearity and correlation among local regression coefficients in geographically weighted regression. J. Geogr. Syst. 7, 161–187 (2005). [DOI: 10.1007/s10109-005-0155-6]
  40. Harris, P., Fotheringham, A. S. & Juggins, S. Robust geographically weighted regression: A technique for quantifying spatial relationships between freshwater acidification critical loads and catchment attributes. Ann. Assoc. Am. Geogr. 100, 286–306 (2010). [DOI: 10.1080/00045600903550378]
  41. Harris, P., Brunsdon, C., Lu, B., Nakaya, T. & Charlton, M. Introducing bootstrap methods to investigate coefficient non-stationarity in spatial regression models. Spat. Stat. 21, 241–261 (2017). [DOI: 10.1016/j.spasta.2017.07.006]
  42. Cho, S.-H., Lambert, D. M. & Chen, Z. Geographically weighted regression bandwidth selection and spatial autocorrelation: An empirical example using Chinese agriculture data. Appl. Econ. Lett. 17, 767–772 (2010). [DOI: 10.1080/13504850802314452]
  43. Lu, B., Yang, W., Ge, Y. & Harris, P. Improvements to the calibration of a geographically weighted regression with parameter-specific distance metrics and bandwidths. Comput. Environ. Urban Syst. 71, 41–57 (2018). [DOI: 10.1016/j.compenvurbsys.2018.03.012]
  44. Arabameri, A., Pradhan, B. & Rezaei, K. Gully erosion zonation mapping using integrated geographically weighted regression with certainty factor and random forest models in gis. J. Environ. Manag. 232, 928–942 (2019). [DOI: 10.1016/j.jenvman.2018.11.110]
  45. Li, X. et al. Mapping soil organic carbon and total nitrogen in croplands of the corn belt of northeast china based on geographically weighted regression kriging model. Comput. Geosci. 135, 104392 (2020). [DOI: 10.1016/j.cageo.2019.104392]
  46. Cao, K., Diao, M. & Wu, B. A big data-based geographically weighted regression model for public housing prices: A case study in Singapore. Ann. Am. Assoc. Geogr. 109, 173–186 (2019).
  47. Ge, Y. et al. Geographically weighted regression-based determinants of malaria incidences in northern China. Trans. GIS 21, 934–953 (2017). [DOI: 10.1111/tgis.12259]
  48. Viscarra Rossel, R. A. & Hicks, W. S. Soil organic carbon and its fractions estimated by visible–near infrared transfer functions. Eur. J. Soil Sci. 66(3), 438–450 (2015). [DOI: 10.1111/ejss.12237]
  49. Wight, J. P., Ashworth, A. J. & Allen, F. L. Organic substrate, clay type, texture, and water influence on NIR carbon measurements. Geoderma 261, 36–43 (2016). [DOI: 10.1016/j.geoderma.2015.06.021]
  50. Costa, L. R., Tonoli, G. H. D., Milagres, F. R. & Hein, P. R. G. Artificial neural network and partial least square regressions for rapid estimation of cellulose pulp dryness based on near infrared spectroscopic data. Carbohyd. Polym. 224, 115186 (2019). [DOI: 10.1016/j.carbpol.2019.115186]
  51. Murphy, R. J., Schneider, S., Taylor, Z. & Nieto, J. Mapping clay minerals in an open-pit mine using hyperspectral imagery and automated feature extraction. In Vertical Geology, From Remote Sensing to 3D Geological Modelling. Proceedings of the first Vertical Geology Conference, Lausanne, Switzerland, 5–7 (2014).
  52. Todorova, M. H. & Atanassova, S. L. Near infrared spectra and soft independent modelling of class analogy for discrimination of chernozems, luvisols and vertisols. J. Near Infrared Spectrosc. 24, 271–280 (2016). [DOI: 10.1255/jnirs.1223]
  53. Stenberg, B., Viscarra Rossel, R., Mouazen, A. & Wetterlind, J. Visible and Near Infrared Spectroscopy in Soil Science, vol. 107 (Academic Press, 2010).
  54. Harris, P., Fotheringham, A., Crespo, R. & Charlton, M. The use of geographically weighted regression for spatial prediction: An evaluation of models using simulated data sets. Math. Geosci. 42, 657–680 (2010). [DOI: 10.1007/s11004-010-9284-7]
  55. Department of Primary Industries and Regional Development, Western Australia. South West Agricultural Region (dpird-008) (2020).
  56. Australian Bureau of Statistics. Value of Agricultural Commodities Produced, Australia (2020).
  57. Department of Primary Industries and Regional Development, Western Australia. Western Australian Wheat Industry (2019).
  58. Rayment, G. E. & Lyons, D. J. Soil Chemical Methods—Australasia (CSIRO Publishing, 2010).
  59. Dolui, S. et al. Structural correlation-based outlier rejection (score) algorithm for arterial spin labeling time series. J. Magn. Reson. Imaging 45, 1786–1797 (2017). [PMID: 27570967]
  60. Pollet, T. V. & van der Meij, L. To remove or not to remove: The impact of outlier handling on significance testing in testosterone data. Adapt. Hum. Behav. Physiol. 3, 43–60 (2017). [DOI: 10.1007/s40750-016-0050-z]
  61. Leys, C., Ley, C., Klein, O., Bernard, P. & Licata, L. Detecting outliers: Do not use standard deviation around the mean, use absolute deviation around the median. J. Exp. Soc. Psychol. 49, 764–766 (2013). [DOI: 10.1016/j.jesp.2013.03.013]
  62. Mallat, S. G. A theory for multiresolution signal decomposition: The wavelet representation. IEEE Trans. Pattern Anal. Mach. Intell. 11, 674–693 (1989). [DOI: 10.1109/34.192463]
  63. Whitcher, B. waveslim: Basic Wavelet Routines for One-, Two-, and Three-Dimensional Signal Processing (2020).
  64. O’brien, R. M. A caution regarding rules of thumb for variance inflation factors. Qual. Quant. 41, 673–690 (2007).
  65. Akinwande, M. O. et al. Variance inflation factor: As a condition for the inclusion of suppressor variable (s) in regression analysis. Open J. Stat. 5, 754 (2015). [DOI: 10.4236/ojs.2015.57075]
  66. Webster, R. & Oliver, M. A. Sample adequately to estimate variograms of soil properties. J. Soil Sci. 43, 177–192 (1992). [DOI: 10.1111/j.1365-2389.1992.tb00128.x]
  67. Atteia, O., Dubois, J.-P. & Webster, R. Geostatistical analysis of soil contamination in the Swiss Jura. Environ. Pollut. 86, 315–327 (1994). [PMID: 15091623]
  68. Brunsdon, C., Fotheringham, S. & Charlton, M. Geographically weighted regression-modelling spatial non-stationarity. J. R. Stat. Soc. Ser. D (The Statistician) 47, 431–443 (1998).
  69. Gollini, I., Lu, B., Charlton, M., Brunsdon, C. & Harris, P. Gwmodel: An r package for exploring spatial heterogeneity using geographically weighted models. arXiv preprint arXiv:1306.0413 (2013).
  70. Lu, B., Harris, P., Charlton, M. & Brunsdon, C. The GWmodel R package: Further topics for exploring spatial heterogeneity using geographically weighted models. Geo Spat. Inf. Sci. 17, 85–101 (2014). [DOI: 10.1080/10095020.2014.917453]
  71. Hastie, T., Tibshirani, R. & Friedman, J. The Elements of Statistical Learning: Data Mining, Inference, and Prediction (Springer Science & Business Media, 2009).
  72. Burnham, K. P. & Anderson, D. R. A practical information-theoretic approach. Model selection and multimodel inference 2 (2002).
  73. R Core Team. R: A Language and Environment for Statistical Computing. (R Foundation for Statistical Computing, Vienna, Austria 2020).
  74. Mevik, B.-H., Wehrens, R., Liland, K. H. & Hiemstra, P. pls: Partial Least Squares and Principal Component Regression (2020).
  75. Bivand, R., Yu, D., Nakaya, T. & Garcia-Lopez, M. spgwr: Geographically weighted regression. R Package Version 0.6-34. http://cran.r-project.org/web/packages/spgwr/ . Accessed August 30th 2020 (2020).
Journal Article Research Support, Non-U.S. Gov't
Article has an altmetric score of 6

Word Cloud

Created with Highcharts 10.0.0propertiessoilmodellingmodelsmultiresolutionWGWRcontentvis-NIRregressionPLSRspectrainformationorganiccarbonpHclayusedspatialdatawaveletgeographicallyweightedWLRestimatesdueSoilcriticalindicatorsecosystemfunctionVisible-nearinfraredreflectancespectroscopywidelycost-efficientlyestimateMultivariatepartialleastsquaresmachinelearningcommonmethodsOftenaccountpresentedsignalvariationaddresspotentialshortcomingswaveletsdecompose226soilsagriculturalforestedregionssouth-westernWesternAustraliadevelopedestimatingevaluatecomparedlinearderiveddecompositionwithoutOverallvalidationproducedaccurateAround35-491%improvementanalysis10-52%integrationimprovesWaveletspectroscopic

Similar Articles

Cited By