ارزیابی روش‌های PLSR و bagging-PLSR در برآورد اجزای بافت خاک، کربنات کلسیم و pH خاک با استفاده از داده‌های طیفی

نوع مقاله : مقاله پژوهشی

نویسندگان

1 گروه علوم و مهندسی خاک، دانشکده کشاورزی، دانشگاه کردستان، سنندج، ایران

2 دانشیار گروه علوم و مهندسی خاک دانشگاه کردستان

3 گروه مدیریت مناطق خشک و بیابانی، دانشکده کشاورزی و منابع طبیعی، دانشگاه اردکان، اردکان، ایران.

چکیده

تعیین سریع، دقیق و کم‌هزینه ویژگی‌های خاکی در برنامه‌ریزی و مدیریت اراضی از اهمیت قابل‌توجهی برخوردار است. هدف از این پژوهش ارزیابی بازتاب طیفی نزدیک، به‌عنوان تکنیکی سریع، مقرون به صرفه و غیرمخرب در تخمین برخی ویژگی‌های خاک (شن، سیلت، رس، pH و کربنات کلسیم معادل (CCE)) با استفاده از روش‌های رگرسیون حداقل مربعات جزئی (PLSR) و رگرسیون حداقل مربعات جزئی توأم با بازنمونه‌گیری (bagging-PLSR) بود. بدین منظور، 220 نمونه مرکب خاک از عمق 0 تا 20 سانتی‌متری در شهریور 1398 از دشت قروه استان کردستان جمع آوری شد. سپس ویژگی‌های انتخابی خاک با روش‌های آزمایشگاهی استاندارد اندازه‌گیری شد. بازتاب طیفی نزدیک نمونه‌ی خاک‌ها در محدوده 350 تا 2500 نانومتر (Vis-NIR) با بهره‌گیری از دستگاه اسپکترورادیومتر آزمایشگاهی اندازه گیری شد. پس از ثبت طیف‌ها، انواع مختلف روش‌های پیش‌پردازش مورد ارزیابی قرار گرفت. نتایج نشان داد که روش PLSR  مقادیر R2 بین 58/0 تا 76/0 به‌دست می‌دهد، درحالی‌که این مقادیر در bagging-PLSR  بین 59/0 تا 74/0 متغیر می‌باشد. مقادیر RMSE معادل با 43/17، 65/7، 83/7، 94/7 و 66/0 در روش PLSR و همچنین مقادیر 66/16، 63/7، 13/8، 71/7 و 45/0 در روش bagging-PLSR به‌ترتیب برای شن، سیلت، رس، CCE و pH خاک به‌دست آمد. بر اساس مقادیر RPD (نسبت برآورد به انحراف)، در برآورد مقدار شن و CCE بهترین عملکرد توسط مدل bagging-PLSR به‌دست آمد؛ این در‌حالی است که در برآورد رس و pH خاک مدل ‌PLSR دقیق‌ترین بود. نتایج پیش‌بینی هر دو مدل برای سیلت یکسان بود (53/1 = RPD). در کل نتایج نشان داد که مدل‌های PLSR و bagging-PLSR در برآورد ویژگی‌های خاکی مورد‌مطالعه از دقت قابل قبولی برخوردار می‌باشند.

کلیدواژه‌ها

موضوعات


عنوان مقاله [English]

Evaluation of PLSR and bagging-PLSR methods in estimating soil texture, calcium carbonate, and pH using spectral data

نویسندگان [English]

  • Aram Shahabi 1
  • Masoud Davari 1
  • Kamal Nabiollahi 2
  • Rohullah Taghizadeh Mehrjardi 3
1 Department of Soil Science and Engineering, Faculty of Agriculture, University of Kurdistan, Sanandaj, Iran.
2 Associate Professor of Soil Science, Department of Soil Science, University of Kurdistan
3 Department of Arid and Desert Regions Management, Faculty of Agriculture and Natural Resources, Ardakan University, Ardakan, Iran
چکیده [English]

The rapid, accurate, and low-cost determination of soil properties has particularly important for land planning and management. The objective of this study was to evaluate the Vis-NIR spectral reflectance of soils, as a rapid, cost-effective, and non-destructive technique, for estimating some soil properties [sand, silt, clay, pH, and calcium carbonate equivalent (CCE)] by partial least-square regression (PLSR) and bagging-PLSR methods. For this purpose, a total of 220 composite soil samples were collected from 0-20 cm depth in Ghorveh Plain, Kurdistan province, in September 2019. The selected soil properties were measured by standard laboratory methods. The proximal spectral reflectance of soil samples was also measured within the 350-2500 nm range (Vis-NIR) using a handheld spectroradiometer. Different pre-processing methods were assessed after recording the spectra. The results indicated that the R2 values for the PLSR method ranged from 0.58 to 0.76, while the bagging-PLSR produced R2 values between 0.59 and 0.74. The RMSE values obtained for sand, silt, clay, CCE, and pH were 17.43, 7.65, 7.83, 7.94, and 0.66, respectively for the PLSR, and 16.66, 7.63, 8.13, 7.71, and 0.45 for the bagging-PLSR. Based on the ratio of prediction to deviation (RPD) values, the bagging-PLSR model achieved the best performance in predicting sand and CCE. However, for clay and pH prediction, the PLSR model was the most accurate. Both the PLSR and bagging-PLSR models yielded identical predictions for silt content, with an RPD value of 1.53. Overall, the results showed that PLSR and bagging-PLSR models have acceptable accuracy for estimating the proposed properties of the soils.

کلیدواژه‌ها [English]

  • CCE
  • Soil properties
  • Partial least-squares regression
  • Visible and near-infrared spectroscopy

Evaluation of PLSR and bagging-PLSR methods in estimating soil texture, pH, and calcium carbonate using spectral data

EXTENDED ABSTRACT

Introduction

The rapid, accurate, and low-cost determination of soil properties has particularly important for land planning and management. However, conventional soil sampling and reliable measurement of soil properties, especially on a large geographic scale, can be a laborious task, time-consuming, expensive, and require quantities of harmful chemicals substance for performing experiments. It is, perhaps, for these reasons that the proximal and remote sensing techniques are being considered as possible alternatives to enhance, complement or substitute traditional soil analysis methods. During the last few decades, the use of visible (Vis) and near-infrared (NIR) diffuse reflectance spectroscopy, as a proximal sensing technique, has attracted tremendous attention for assessing soil properties. Therefore, the objective of this study was to evaluate the Vis-NIR spectral reflectance of soils, as a rapid, cost-effective and non-destructive technique, for estimating some soil properties by PLSR and bagging-PLSR methods

Materials and Methods

A total of 220 composite soil samples were collected from the 0–20 cm depth in Ghorveh Plain, Kurdistan province, in September 2019. These soil samples were transported to the laboratory, air dried, grounded, and then sieved to a size fraction of smaller than 2-mm. The selected soil properties including sand, silt, clay, pH, and calcium carbonate equivalent (CCE) were measured by standard laboratory methods. In addition, the proximal spectral reflectance of soil samples was also measured within the 350-2500 nm range (Vis-NIR) using a handheld spectroradiometer. To minimize the impact of random noise and improve calibration models, different pre-processing methods were assessed after recording the spectra.

Results

The study found that the first derivative of the Savitzky-Golay smoothing filter was the most effective pre-processing technique for calibrating the PLS regression. The optimal number of factors for predicting sand, silt, clay, pH, and CCE were 14, 7, 23, 10, and 18, respectively, using the PLSR method. The R2 values for the PLSR method ranged from 0.58 to 0.76, while the bagging-PLSR produced R2 values between 0.59 and 0.74. The RMSE values obtained for sand, silt, clay, CCE, and pH were 17.43, 7.65, 7.83, 7.94, and 0.66, respectively for the PLSR, and 16.66, 7.63, 8.13, 7.71, and 0.45 for the bagging-PLSR. Based on the ratio of prediction to deviation (RPD) values, the bagging-PLSR model achieved the best performance in predicting sand and CCE. However, for clay and pH prediction, the PLSR model was the most accurate. Both the PLSR and bagging-PLSR models yielded identical predictions for silt content, with an RPD value of 1.53.

Conclusion

It is concluded that the bagging-PLSR outperformed PLSR in predicting sand content and CCE. However, PLSR was more effective for predicting clay content and pH. Both models produced similar results for silt content. In all, the accuracy levels of both PLSR and bagging-PLSR were high for clay content and moderate to good for sand and silt contents, pH, and CCE. These findings suggest that the Vis-NIR spectroscopy, as a complement or replacement approach to laboratory conventional methods, can be used for rapid and cost-efficient assessment of soil properties.

Adeline, K.R.M., Gomez, C., Gorretta, N., & Roger, J.M. (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.
Akpa, S. I. C., Odeh, I. O. A., & Bishop, T. F. A. (20140. Digital mapping of soil particle-size fractions for Nigeria. Soil Science Society of America Journal, 78: 1953-1966. http://doi.org/10.2136/sssaj 2014. 05.0202.
Akumu, C. E., Johnson, J. A., Etheridge, D., Uhlig, P., Woods, M., Pitt, D. G., & McMurray, S. (2015). GIS-fuzzy logic based approach in modeling soil texture: Using parts of the Clay Belt and Hornepayne region in Ontario Canada as a case study. Geoderma, 239-240: 13-24. https://doi.org/10.1016/j.geoderma.2014.09.021.
Azizi, K., Nabiollahi, K. & Davari, M. 2017. Evaluation of spectroscopic capability in estimating some properties of salt-affected soils. Agricultural Engineering (Agricultural Scientific Journal), 41(3): 1-16. https://doi.org/10.22055/AGEN.2019.25763.1427. (In Persian).
Breiman, L. (1996). Bagging predictors. Machine Learning. 24: 2. 123-140. https://doi.org/10.1007/BF00058655
Bellon-Maurel, V., & McBratney, A. (2011). Near-infrared (NIR) and mid-infrared (MIR) spectroscopic techniques for assessing the amount of carbon stock in soils critical review and research perspectives. Soil Biology and Biochemical, 43: 1398-1410. https://doi.org/10.1016/j.soilbio.2011.02.019.
Babaeian, A., & Jalali, V. (2015). Estimation of soil organic carbon using hyperspectral data in VIS-NIR-SWIR range. Journal of Soil Management and Sustainable Production, 6(2), 65-82. https://doi.org/10.22069/EJSMS.2016.3143. (In Persian).
Chatrenour, M., Landi, A., Bahrami, H.A. & Mirzaei, S. (2023). Dust source clay content and salinity estimation using VNIR spectrometry. Arid Land Research and Management, 37(3), 369–388. https://doi.org/10.1080/15324982.2023.2170837.
Chang, C.W., Laird, D.A., Mausbach, M.J., & Hurburgh, C.R. (2001). Near-infrared reflectance spectroscopy-principal components regression analyses of soil properties. Soil Science Society of America Journal. 65, 480–490. https://doi.org/10.2136/sssaj2001.652480x.
Chen, L., Sheng-lu, Z., Shao-hua, W., Qing, Z., & Qi, D. (2014). Spectral Response of Different Eroded Soils in Subtropical China: A Case Study in Changting County, China. Journal of Materials Science, 11: 697-707. https://doi.org/10.1007/s11629-013-2780-8.
Curcio D., Ciraolo G., D’Asaro F., & Minacapillia M. (2013). Prediction of soil texture distributions using VNIR-SWIR reflectance spectroscopy. Procedia Environmental Sciences, 19: 494 – 503. https://doi.org/10.1016/j.proenv.2013.06.056.
Dotto, A.C., Dalmolin, R.S.D., Grunwald, S., ten Caten, A., & Pereira Filho, W. )2017(. Two preprocessing techniques to reduce model covariables in soil property predictions by Vis-NIR spectroscopy. Soil Tillage and Research, 172, 59–68. https://doi.org/10.1016/j.still.2017.05.008.
Esbensen, K.H. (2006). Multivariate Data Analysis -In practice. CAMO Software AS. 5th Edition, 589 pages.
Gee, G.W., & Or, D. (2002). 2.4 Particle-Size Analysis. In: Dane, J.H., Topp, C.G. (Eds.), Methods of Soil Analysis: Part 4 Physical Methods. Soil Science Society of America, Madison, WI, pp. 255–293.
Gomez, C., Lagacherie, P. & Coulouma, G. (2008). Continuum removal versus PLSR method for clay and calcium carbonate content estimation from laboratory and airborne hyperspectral measurements. Geoderma, 148: 141-148. https://doi.org/10.1016/j.geoderma.2008.09.016.
Hartemink, A. E., & McBratney, A. B. (2008). A soil science renaissance. Geoderma, 148: 123-129. https://doi.org/10.1016/j.geoderma.2008.10.006.
Islam, K., Singh, B., & McBratney, A. (2003). Simultaneous estimation of several soil properties by ultra-violet, visible, and near-infrared reflectance spectroscopy. Australian Journal of Soil Research. 41:1101-1114. https://doi.org/10.1071/SR02137.
IUSS Working GWRB. (2015). World reference base for soil resources 2014, update 2015: International soil classification system for naming soils and creating legends for soil maps. World Soil Resources Reports. 106: 1–192
Jaconi, A., Vos, C., & Don, A. (2019). Near infrared spectroscopy as an easy and precise method to estimate soil texture. Geoderma, 337: 906–913. https://doi.org/10.1016/j.geoderma.2018.10.038.
Kodaira, M., & Shibusawa, S. (2013). Using a mobile real-time soil visible-near infrared sensor for high resolution soil property mapping. Geoderma, 199: 64-79. https://doi.org/10.1016/j.geoderma.2012.09.007.
Khayamim, F., Wetterlind, J., Khademi, H., Robertson, A.J., Cano, A.F., & Stenberg, B. (2015). Using visible and near infrared spectroscopy to estimate carbonates and gypsum in soils in arid and subhumid regions of Isfahan, Iran. Journal of Near Infrared Spectroscopy, 23(3): 155-165. https://doi.org/10.1255/jnirs.1157.
Karimi, S.A., Davari, M., Bahrami, H., Babaian, A., & Hosseini, S.M.T. (2017). Derivation and evaluation of spectral transfer function and soil transfer function in order to estimate cation exchange capacity. Journal of Soil Research (Soil and Water Sciences), A, 31(4): 573-585. https://doi.org/10.22092/IJSR.2018.115957. (In Persian).
Lacerda M.P.C., Demattê J.A.M., Sato M.V., Fongaro C.T., Gallo B.C., & Souza A.B. (2016). Tropical TextureDetermination by Proximal Sensing Using a Regional Spectral Library and Its Relationship with Soil Classification. Remote Sensing, 8(701):1-20. https://doi.org/10.3390/rs8090701.
Mousavi, F., Abdi, E., Ghalandarzadeh, A., Bahrami, H.A., Majnounian, B.,  &Ziadi, N. )2020(. Diffuse reflectance spectroscopy for rapid estimation of soil Atterberg limits. Geoderma, 361, 114083. https://doi.org/10.1016/j.geoderma.2019.114083.
Nocita, M., Stevens, A., Noon, C., & van Wesemael, B. (2013). Prediction of soil organic carbon for different levels of soil moisture using Vis-NIR spectroscopy. Geoderma, 199: 37-42. https://doi.org/10.1016/j.geoderma.2012.07.020.
Qiu, H. (2010). Studies on the potential ecological risk and homology correlation of heavy metal in the surface soil. Journal of Agricultural Science, 2:1916-9760. https://doi.org/10.5539/jas.v2n2p194.
Ramírez, P. B., Calderón, F. J., Jastrow, J. D., Chien-Lu Ping, Ch. & Matamala, R. (2023). Applying NIR and MIR spectroscopy for C and soil property prediction in northern cold-region ecosystems. Which approach works better? Geoderma Regional, 32, e00617. https://doi.org/10.1016/j.geodrs.2023.e00617.
Rasouli, N., Farpour, M. H., Khayamim, F., & Ranjbar, H. (2017). Prediction of selected soil properties using visible and near infrared spectroscopy in Bardsir area, Kerman Province. Iranian Journal of Soil Research, 32(2): 243-231. https://doi.org/20.1001.1.22287124.1397.32.2.8.9. (In Persian).
Rasooli, N., Farpoor, M. J., Mahmoodabadi, M., & Esfandiarpour-Boroujeni, I. (2023). Vis-NIR spectroscopy as an eco-friendly method for monitoring pedoenvironmental variations and pedological assessments in Lut Watershed, Central Iran. Soil and Tillage Research, 233: 105808. https://doi.org/10.1016/j.still.2023.105808.
Sparks, D.L., Page, A.L., Helmke, P.A., & Loeppert, R.H. (1996). Methods of Soil Analysis Part 3—Chemical Methods. Soil Science Society of America, American Society of Agronomy, Madison, WI, SSSA Book Series.
Stenberg, B., Viscarra Rossel, R., Mouazen, A., & Wetterlind, J. (2010). Visible and nea infrared spectroscopy in soil science. Advances in Agronomy, 107: 163-215. https://doi.org/10.1016/S0065-2113(10)07005-7.
Savvides, A., Corstanje, R. Baxter, S. J., Rawlins, B. G., & Lark, R. M. (2010). The relationship between diffuse spectral reflectance of the soil and its cation exchange capacity is scale-dependent. Geoderma 154: 353-358. https://doi.org/10.1016/j.geoderma.2009.11.007.
Shiferaw A., & Hergarten Ch. (2014). Visible near infra-red (VisNIR) spectroscopy for predicting soil organic carbon in Ethiopia. Journal of Ecology and the Natural Environment, 6:126-139. https://doi.org/10.5897/JENE2013.0374.
Silva E. B., ten Caten, Dalmolin R.S.D., Dotto A.C., Silva W.C., & Giasson E. (2016). Estimating Soil Texture from a Limited Region of the Visible/Near-Infrared Spectrum..p. 73–87. In A.E. Hartemink and B. Minasny (eds). Digital Soil Morphometr. Springer International Publishing, Switzerland.
Tayibi, M., Naderi, M., Mohammadi, J., & Hosseinjanizadeh, M. (2017). Comparison of different statistical methods in estimating soil texture components using spectral data in the visible-near and short-infrared range. Water and Soil Journal (Agricultural Sciences and Technology), 32(1): 73-85. https:// 10.22067/jsw.v32i1.63618. (In Persian).
Viscarra Rossel, R.A.V. (2008). ParLeS: Software for chemometric analysis of spectroscopic data. Chemometrics and Intelligent Laboratory Systems, 90, 72–83. https://doi.org/10.1016/j.chemolab.2007.06.006.
Viscara Rossel, R., & Behrens, T. (2010). Using data mining to model and interpret soil diffuse reflectance spectra. Geoderma, 158: 46-54. https://doi.org/10.1016/j.geoderma.2009.12.025.
Wold S., Sjostrom M., & Eriksson, L. (2001). PLS-regression: a basic tool of chemometrics. Chemometrics and Intelligent Laboratory Systems, 58: 109–130. https://doi.org/10.1016/S0169-7439(01)00155-1.
Zhu, A.X., Hudson, B., Burt, J., Lubich, K., & Simonson, D. (2001). Soil mapping using GIS, expert Knowledge, and fuzzy logic. Soil Science Society of America Journal, 65: 1463-1472. https://doi.org/10.2136/sssaj2001.6551463x.
Zhao, D., Arshad, M., Li, N., & Triantafilis, J. (2021). Predicting soil physical and chemical properties using vis-NIR in Australian cotton areas. Catena, 196: 104938. https://doi.org/10.1016/j.catena.2020.104938.
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 Regional, 30, e00566. https://doi.org/10.1016/j.geodrs.2022.e00566.