اشتقاق و اعتبارسنجی توابع انتقالی طیفی برای پیش‌بینی غلظت برخی فلزات سنگین در محدوده طیف مرئی تا مادون قرمز

نوع مقاله: مقاله پژوهشی

نویسندگان

1 کرمان

2 عضو هیات علمی دانشگاه شهید باهنر

3 دانشجوی دانشگاه شهید باهنر

چکیده

سابقه و هدف
ارتباط مستقیم بین افزایش غلظت فلزات سنگین خاک و ابتلا به سرطان‌های مختلف برای افرادی که در معرض آلودگی این فلزات هستند، توسط محققان مختلفی گزارش شده است. بنابراین پایش سریع و دوره‌ای گسترش مکانی این فلزات، بسیار با اهمیت است. اگرچه روش-های معمول اندازه‌گیری غلظت فلزات سنگین خاک که مبتنی بر روش هضم در اسیدهای غلیظ و قرائت توسط دستگاه ICP-OES و یا AAS انجام می‌گیرد از دقت کافی برخوردار است، این روش‌ها عمدتاً وقت‌گیر و پرهزینه بوده و نیاز به مواد شیمیایی و کارشناسان آموزش دیده دارند. توسعه روش‌های اسپکتروسکوپی در دامنه طیف‌های مرئی تا مادون قرمز نزدیک می‌تواند روش جایگزین مناسبی برای انجام تخمین محتوی فلزات سنگین خاک باشد. این روش جز روش‌های غیرتخریبی تقسیم بندی شده، احتیاج به حداقل آماده‌سازی نمونه پیش از انجام آزمایش داشته و نیازمند به استفاده از هیچ گونه مواد شیمیایی( خطرناک ) نیست. همچنین قرائت‌های این روش حداکثر چند ثانیه طول کشیده و همزمان می‌توان چندین ویژگی خاک را از یک قرائت تخمین زد. اطلاعات چندانی در زمینه استفاده از بازتاب‌های طیفی در تخمین فلزات سنگین آرسنیک و مولیبدن با استفاده از بازتاب‌های طیفی در محدوده مادون قرمز نزدیک و میانی در کشور وجود ندارد. بنابراین هدف این پژوهش بررسی قابلیت شبکه های عصبی مصنوعی در تخمین غلظت این عناصر بر اساس مطالعه بازتاب‌های طیفی در محدوده مادون قرمز نزدیک و میانی است.
مواد و روش‌ها
تعداد 58 نمونه سطحی از جزیره هرمز جمع‌آوری و غلظت فلزات سنگین مولیبدن و آرسنیک با استفاده از روش هضم چهار اسید (16) و توسط دستگاه ICP-OES تعیین شد. به منظور ‌اندازه‌گیری داده‌های طیفی نمونه‌های خاک، از دستگاه اسپکترورادیومتر زمینی(Field Spec 3, Analytical Spectral Device, ASD Inc) استفاده و بازتاب طیفی نمونه‌های سطحی در محدوده مادون قرمز نزدیک و میانی به دست آمد. سپس با استفاده از روش شبکه عصبی مصنوعی اقدام به استخراج توابع انتقالی طیفی و تخمین غلظت فلزات آرسنیک و مولیبدن گردید.
یافته‌ها
نتایج نشان داد که روش شبکه عصبی مصنوعی دارای قابلیت بالا در تخمین غلظت فلزات سنگین مورد مطالعه با استفاده از داده‌های طیفی می‌باشد. مقادیر ضریب همبستگی(R2) برای هر دو عنصر، مطلوب و بیشتر از 9/0 بوده است که نشان‌دهنده همراستایی بالای داده های واقعی و پیش‌بینی شده توسط مدل شبکه عصبی برای پیش‌بینی فلزات سنگین مورد مطالعه بوده است، در عین‌حال نتایج حاصل از سایر شاخص‌ها نشان داد که توانایی شبکه عصبی مصنوعی برای پیش‌بینی غلظت مولیبدن بهتر از آرسنیک بوده است، به طوری که نتایج نشان داد که مقدار خطای باقی مانده برای این عنصر کم (CRM=0.11)، ضریب آکائیک منفی(AIC=-345.8) و کارایی مدل‌سازی برای این عنصر نزدیک به یک بوده است (EF=0.97).

نتیجه گیری
در این تحقیق از بازتابش‌های طیفی در محدوده مادون قرمز در تخمین محتوای مولیبدن و آرسنیک خاک استفاده شد. همچنین شبکه های عصبی مصنوعی به عنوان ابزار برقراری ارتباط بین بازتابش‌های طیفی و میزان فلزات سنگین به کار گرفته شد. بطور کلی نتایج نشان داد که روش شبکه عصبی مصنوعی می‌تواند به عنوان روشی کارا در اشتقاق توابع انتقالی طیفی و تخمین قابل اعتماد غلظت مولیبدن و آرسنیک در غلظت‌های بالا به کار گرفته شود.

کلیدواژه‌ها


عنوان مقاله [English]

Deriving and validating spectral pedotransfer functions for estimating some soil heavy metal in Vis-NIR range

چکیده [English]

Background and objectives
Direct relationships between the incidences of cancer in people who are exposed to heavy metals, have been investigated and proved in various studies (28). So rapid and periodic monitoring of heavy metals in the areas vulnerable to pollution is important (9). Although conventional methods of soil metals content determination are sufficiently accurate, they are mostly based on wet digestion of soil samples in hot concentrated acids followed by atomic absorption spectrometry (AAS) or inductively coupled plasma (ICP) spectrometry, these methods are time consuming, expensive and require chemical agents and qualified staff (1). Development of visible-near infrared (Vis-NIR) diffuse reflectance spectroscopy provides an alternative to these conventional monitoring methods of the soil heavy metal contamination. Because there are many advantages with using the technique. It is non- destructive, requires a minimum of sample preparation and does not involve any (hazardous) chemicals. The measurements only take a few seconds and several soil properties can be estimated from a single scan. Moreover, the technique allows for flexible measurement configurations and in situ as well as laboratory-based measurements (35). Limited work has been done to predict soil heavy metal content with Vis-NIR through different models or data mining methods in Iran. The aim of this study was to explore the feasibility of ANN in estimating the heavy metal concentration using diffuse spectral reflectance data in the Vis-NIR range,
Materials and methods
A total of 57 soil samples were collected from the topsoil of Hormuz Island. The total concentrations of Mo and As elements were measured using inductively coupled plasma (ICP-OES) apparatus. Then reflectance spectra of the collected soil samples were measured using a portable spectroradiometer apparatus (Field Spec 3, Analytical Spectral Device, ASD Inc.) in the Vis-NIR (350-2500 nm) range. Artificial Neural Networks (ANN) method WAS used to predict heavy metal concentration from soil samples reflectance spectra.
Results
The results showed that ANN has high capability in estimating the concentration of studied heavy metals using spectral data. Coefficient of determination (R2) for both elements, were desirable and more than 0.9 that represents the correspondence of the observed and predicted data by the neural network model in predicting the Mo and As heavy metals. However, results from other index also indicated that the ability of artificial neural network to predict the concentration of molybdenum was better than arsenic heavy metal, So that the results showed that the coefficient of residual mass was low for this element (CRM = 0.11), the coefficient of Akaike was negative (AIC = -345.8) and modeling efficiency for this element has been close to a 1 (EF = 0.97).
Conclusions
In this paper we used hyperspectral reflectance data in visible and near infrared regions (350-2500 nm) to predict concentration of Mo and As heavy Using ANN as calibration model. Overall, results showed that artificial neural networks can be effectively used in deriving spectral-pedotransfer functions and bridging soil spectral reflectance to accurate estimates of molybdenum and arsenic heavy metals in high concentrations.

کلیدواژه‌ها [English]

  • ANN
  • Heavy metals
  • Hormuz island
  • spectral pedotransfer functions(STFs)
1.Babaeean, E., and Jalali, V.R. 2016. Estimating Soil Organic Carbon Using Hyperspectral
Data in Visible, Near-infrared and Shortwave-infrared (VIS-NIR-SWIR) Range. J. Soil
Manage. Sust. Prod. 6: 2. 65-82. (In Persian)
2.Babaeian, E., Homaee, M., Montzka, C., Vereecken, H., Norouzi, A.A., and van Genuchten,
M.Th. 2016. Soil moisture prediction of bare soil profiles using diffuse spectral reflectance
information and vadose zone flow modeling. Remote Sensing of Environment. 187: 218-229.
3.Babaeian, E., Homaee, M., Montzka, C., Vereecken, H., and Norouzi, A.A. 2015. Towards
retrieving soil hydraulic properties by hyperspectral remote sensing. Vadose zone J. 14: 3. 1-17.
4.Babaeian, E., Homaee, M., and Norouzi, A.A. 2014. Deriving and validating parametric
spectrotransfer functions for estimating soil hydraulic properties in VIS-NIR-SWIR range.
J. Water Soil Resour. Cons. 3: 3. 21-36. (In Persian)
5.Bouma, J. 1989. Using soil survey data for quantitative land evaluation. Advanced Soil
Science. 9: 177-213.
6.Bray, J.G.P., Viscarra Rossel, R.A., and McBratney, A.B. 2009. Diagnostic scree using
diffuse reflectance spectroscopy. Aust. J. Soil Res. 47: 433-442.
7.Cambou, A., Cardinael, R., Kouakoua, E., Villeneuve, M., Durand, C., and Barthès, B.G.
2016. Prediction of soil organic carbon stock using visible and near infrared reflectance
spectroscopy (VNIRS) in the field. Geoderma. 261: 151-159.
8.Cécile, C., Viscarra Rossel, R.A., and McBratney, A.B. 2008. Soil organic carbon prediction
by hyperspectral remote sensing and field Vis-NIR spectroscopy: An Australian case study.
Geoderma. 146: 3-4. 40-41.
9.Choe, E., Kim, K.W., Bang, S., Yoon, I.H., and Lee, K.Y. 2008. Qualitative analysis and
mapping of heavy metals in an abandoned Au-Ag mine area using NIR spectroscopy.
Environ. Geol. 58: 477-482.
10.Clark, R.N., King, T.V.V., Klejwa, M., Swayze, G.A., and Vergo, N. 1990. High spectral
resolution reflectance spectroscopy of minerals. Geophysic. Res. J. 95: 12653-12680.
11.DuBose, P., and Klimasauskas, C. 1989. Introduction to Neural Networks with Examples
and Applications. NeuralWare Inc., Pittsburgh, 317p.
12.Esmaeelnejad, L., Ramezanpour, H., Seyedmohammadi, J., and Shabanpour, M. 2015.
Selection of a suitable model for the prediction of soil water content in north of Iran. Spanish
J. Agric. Res. 13: 1. 12-20.
13.Garavand, M., Ghasemi, H., and Hafezi Moghddas, N. 2013. Geochemical and
Environmental Assessment of the Heavy Metals in the Soils Derived from the Gorgan
Schists. Sci. Quar. J. Geosci. 22: 86. 35-46. (In Persian)
14.Gholizadeh, A., Boruvka, L., Saberioon, M.M., and Vasat, R. 2013. Visible, near-infrared,
and mid-infrared spectroscopy applications for soil assessment with emphasis on soil organic
matter content and quality: State-of-the-art and key issues. Appl Spectrosc. 67: 1349-1362.
15.Gomez, C., Lagacherie, P., and Coulouma, G. 2008. Continuum removal versus PLSR
method for clay and calcium carbonate content estimation from laboratory and airborne
hyperspectral measurements. Geoderma. 148: 2. 141-148.
16.Hseu, Z.Y. 2004. Evaluating heavy metal contents in nine composts using four digestion
methods. Bioresource Technology. 95: 53-59.
17.Jalali, V., Asadi Kapourchal, S., and Homaee, M. 2017. Evaluating performance of
macroscopic water uptake models at productive growth stages of durum wheat under saline
conditions. Agric. Water Manage. J. 180: 13-21.
18.Kemper, T., and Sommer, S. 2002. Estimate of heavy metal contamination in soils after a
mining accident using reflectance spectroscopy. Environ. Sci. Technol. 36: 2742-2747.
19.Khashei Siuki, A., Jalali Moakhar, V.R., Noferesti, A.M., and Ramazani, Y. 2015.
Comparing nonparametric k-nearest neighbor technique with ANN model for predicting soil
saturated hydraulic conductivity. Soil management and sustainable production. 5: 3. 81-95.
(In Persian)
20.Khayamim, F., Khademi, H., Stenberg, B., and Wetterlind, J. 2015. Capability of Vis-NIR
Spectroscopy to Predict Selected Chemical Soil Properties in Isfahan Province. J. Water Soil
Sci. 19: 72. 81-92. (In Persian)
21.Kodaira, M., and Shibusawa, S. 2013. Using a mobile real-time soil visible-near infrared
sensor for high resolution soil property mapping. Geoderma. 199: 64-79.
22.Lagacherie, P., Baret, F., Feret, J.B., Netto, J.M., and Robbez-Masson, J.M. 2008. Estimation
of soil clay and calcium carbonate using laboratory, field and airborne hyperspectral
measurements. Rem. Sens. Environ. J. 112: 3. 825-835.
23.Leonard, T., and Hsu, J.S.J. 2001. Bayesian Methods: an analysis for statisticians and
interdisciplinary. Cambridge University Press, Cambridge, 333p.
24.Meysami, A. 2011. Look at the position of geology and mineral deposits of the island of
Hormuz in the Persian Gulf. The second congress. Islamic Azad University, Ashtiyan Unit.
http://www.civilica.com/Paper-GEOSYMAIAU02-GEOSYMAIAU02_262.html. (In Persian)
25.Miranda Salazar, D., Martınez Reyesa, H.L., Martınez-Rosasa, M.E., Miranda Velascoa,
M.M., and Arroyo Ortegaa, E. 2012. Visible-near infrared spectroscopy to assess soil
contaminated with cobalt. Proc. Engin. J. 35: 245-253.
26.Mohajer, R., Salehi, M.H., and Beigi Herchegani, H. 2009. Estimating Soil Cation Exchange
Capacity (in View of Pedotransfer Functions) Using Regression and Artificial Neural
Networks and the Effect of Data Partitioning on Accuracy and Precision of Functions.
Water and Soil Science. 13: 49. 99-110. (In Persian)
27.Mohammadi Moghaddam, T., Razavi, S.M.A., Taghizadeh, M., Sazgarnia, A., and Pradhan,
B. 2015. Vis-NIR hyperspectral imaging and multivariate analysis for prediction of the
moisture content and hardness of Pistachio kernels roasted in different conditions. J. Agric.
Machin. 5: 2. 281-291. (In Persian)
28.Moros, J., de Vallejuelo, S.F.O., Gredilla, A., de Diego, A., and Madariaga, J.M. 2009. Use
of reflectance infrared spectroscopy for monitoring the metal content of the estuarine
sediments of the Nerbioi-Ibaizabal River (Metropolitan Bilbao, Bay of Biscay, Basque
Country). Environ Sci. Technol. J. 43: 93. 14-9320.
29.Patil, N.G., and Singh, S.K. 2016. Pedotransfer Functions for Estimating Soil Hydraulic
Properties: A Review. Pedosphere. 26: 4. 417-430.
30.Pirie, A., Singh, B., and Islam, K. 2005. Ultra-violet, visible, near-infrared, and mid infrared
diffuse reflectance spectroscopic techniques to predict several soil properties. Aust. J. Soil
Res. 43: 713-721.
31.Sá, I., Semedo, M., and Cunha, M.E. 2016. Kidney cancer. Heavy metals as a risk factor.
Porto Biomed. J. 1: 1. 25-28.
32.Santra, P., Sahoo, R.N., Das, B.S., Samal, R.N., Pattanaik, A.K., and Gupta, V.K. 2009.
Estimation of soil hydraulic properties using proximal spectral reflectance in visible,
near-infrared, and short wave-infrared (VIS-NIR-SWIR) region. Geoderma. 152: 338-349.
33.Seyedmohammadi, J., Esmaeelnejad, L., and Shabanpour, M. 2016. Derivation the suitable
pedotransfer functions for prediction of some difficult available soil properties. J. Water Soil
Cons. 23: 4. 204-217. (In Persian)
34.Shamsadin, H., Jalali, V., and Jafari, A. 2015. Application of multivariate statistical methods
and environmental pollution indices in evaluation of distribution of heavy metals. J. Water
Soil Resour. Cons. 4: 3. 65-76. (In Persian)
35.Soil resources quality standards and its guidelines. 2012. Deputy of the human environment,
soil and water office Press, 166p. (In Persian)
36.Song, Y., Li, F., Yang, Z., Ayoko, G.A., and Frost, R.L. 2012. Diffuse reflectance
spectroscopy for monitoring potentially toxic elements in the agricultural soils of Changjiang
River Delta, China. Appl. Clay Sci. J. 64: 75-83.
37.Statistical Yearbook of Hormozgan Province. 2013. http://www.mpohr.gov.ir/index.aspx?
fkeyid=&siteid=1&pageid=130&newsview=1.
38.Vapnik, V. 1995. The Nature of Statistical Learning Theory. Springer-Verlag, New York,
330p.
39.Viscarra Rossel, R.A., Walvoort, D.J.J., McBratney, A.B., Janik, L.J., and Skjemstad, J.O.
2006. Visible, near infrared, mid infrared or combined diffuse reflectance spectroscopy for
simultaneous assessment of various soil properties. Geoderma. 131: 59-75.
40.Viscarra Rossel, R.A.V. 2008. ParLeS: Software for chemometric analysis of spectroscopic
data. Chemometrics and Intelligent Laboratory Systems. 90: 72-83.
41.Wang, J., Cui, L., Gao, W., Shi, T., Chen, Y., and Gao, Y. 2014. Prediction of low heavy
metal concentration in agriculture soils using visible and near-infrared reflectance
spectroscopy. Geoderma. 216: 1-9.
42.Williams, P.C., and Malley, D.F. 1997. Use of Near-Infrared reflectance spectroscopy in
prediction of heavy metal in freshwater sediments by their association with organic matter.
Environ. Sci. Technol. 31: 3461-3467.
43.Xie, X., Pan, X.Z., and Sun, B. 2012. Visible and near-infrared diffuse reflectance
spectroscopy for prediction of soil properties near a Copper smelter. Pedosphere. 22: 351-366.