مدل‌سازی آبگریزی خاک با استفاده از رگرسیون، شبکه عصبی مصنوعی و برنامه‌ریزی بیان ژن

نوع مقاله : مقاله پژوهشی

نویسندگان

1 دانشجوی کارشناسی ارشد، گروه علوم و مهندسی خاک، دانشکده کشاورزی و منابع طبیعی، دانشگاه محقق اردبیلی، اردبیل، ایران

2 فیزیک و فرسایش دانشگاه محقق اردبیلی

3 فرسایش خاک، گروه علوم و مهندسی خاک، دانشکده کشاورزی و منابع طبیعی، دانشگاه محقق اردبیلی، اردبیل، ایران

4 گروه مهندسی آب، دانشکده کشاورزی و منابع طبیعی، دانشگاه محقق اردبیلی، اردبیل، ایران

10.30466/asr.2025.55310.1846

چکیده

آبگریزی ­خاک (SWR) از ویژگی­های پویای خاک بوده که نفوذ آب به خاک را کاهش و بر روابط خاک و آب تأثیر دارد. اندازه­گیری مستقیم SWR کاری پرزحمت و وقت­گیر می­باشد. هدف از پژوهش حاضر ارائه توابع رگرسیونی خطی چندگانه (MLR)، شبکه عصبی مصنوعی (ANN) و برنامه­ریزی بیان ژن (GEP) برای برآورد SWR در منطقه فندقلوی اردبیل بود. هشتاد نمونه خاک دست­خورده و دست­نخورده از عمق 0 تا 10 سانتی­متری سه کاربری به­هم چسبیده جنگلی، مرتعی و زراعی برای تعیین برخی ویژگی­های فیزیکی و شیمیایی زودیافت خاک برداشته شد. متغیر SWR به روش زمان نفوذ قطره آب در آزمایشگاه اندازه­گیری شد. از 60 نمونه برای آموزش توابع و 20 نمونه برای آزمون توابع استفاده گردید. همبستگی مثبت و معنی­دار بین میانگین هندسی قطر ذرات خاک (dg) با کربن آلی (**61/0) یافت شد. همبستگی مثبت و معنی­دار بین SWR با کربن آلی (**37/0) و میانگین هندسی قطر ذرات خاک (**62/0) و همبستگی منفی و معنی­دار بین SWR با سیلت (**57/0-) و جرم مخصوص ظاهری (**37/0-) به­دست آمد. نتایج توابع انتقالی نشان داد dg، سیلت و جرم مخصوص ظاهری از مهمترین متغیر­های زودیافت خاک در برآورد SWR بودند. مقادیر آماره­های ضریب تبیین (R2) ، مجذور میانگین مربعات خطا( RMSE) ، میانگین خطا (ME) و نش ساتکلیف(NS)  به ترتیب 18/0،  sec89/16،sec  34/10-، 99/20- و 46/0، sec 85/2، sec 58/0، 37/0 و 19/0، sec 39/13، sec 38/6-، 82/12- به­ترتیب برای بهترین تابع MLR، ANN و GEP در داده­های آزمونی به­دست آمد. بنابراین توابع ANN به دلیل داشتن R2 بالا، RMSE پایین، ME نزدیک به صفر و NS نزدیک به یک در مقایسه با توابع MLR و GEP از دقت بالایی در برآورد SWR در خاک­های منطقه مورد مطالعه برخوردار بودند.

کلیدواژه‌ها

موضوعات

عنوان مقاله [English]

Modeling of Soil Water Repellency Using Regression, Artificial Neural Network and Gene Expression Programming

نویسندگان [English]

  • Kimia Heydari 1
  • shokrollah asghari 2
  • Hossain Shahab Arkhazloo 3
  • Mahsa Hasanpour Kashani 4
1 MSc student, Department of Soil Sciences and engineering, Faculty of Agriculture and Natural Resources, University of Mohaghegh Ardabili, Ardabil, Iran
2 Department of soil science and engineering, faculty of agriculture and natural resources, University of Mohaghegh Ardabili
3 Soil Erosion, Department of Soil Sciences and engineering, Faculty of Agriculture and Natural Resources, University of Mohaghegh Ardabili, Ardabil
4 Water Engineering Department, Faculty of Agriculture and Natural Resources, University of Mohaghegh Ardabili, Ardabil, Iran.
چکیده [English]

Increasing the concentration of heavy metals in soil has adverse effects on ecosystem and causes serious damage to humans. Quantifying pollution can be helpful for soil management. In this study, the pollution of heavy metals has been studied and quantified in agricultural lands around some industrial units at Ardabil plain. For this, we selected 9 industrial units and prepared 46 soil samples (0 to 30 cm). Clay, sand and silt percentages, soil organic carbon content, pH and EC were measured. The heavy metals were extracted by digestion using HNO3 and HCl and the concentration of Pb, Zn, Cu and Cd were measured by AAS. Pollution index (Pi), comprehensive pollution index (Pj), ecological risk (Er) and potential ecological risk (RI) were calculated. The average concentration of heavy metals varied from 0.724 mgkg-1 of Cd to 120.58 mg/kg-1 of Cu. All regions had Pi greater than 2 and showed mild pollution except region 4 which had slight pollution. The pollution index of Cd had the highest value among all heavy metals. Pj had the lowest (1.268) and highest (3.636) mean values in regions 2 and 5, respectively. Region 2 had slight pollution class, regions 1, 6 and 7 had a mild class and regions 3, 4, 5, 8 and 9 had a moderate class. The ecological risk of Pb, Zn and Cu was lower than 3.0 in all regions and ErCd was between 120 and 240 for regions 3, 5, 8 and 9 increasing a serious pollution class and greater than 240 indicating a severe pollution class for other areas. All regions were in the serious pollution class according to the RI index. There is a significant difference between the concentration of Pb (sig. 5%) and Zn and Cu (sig.1%) in all regions. It shows differences between regions cannot be related to parent material and human activities have resulted in an increase in metal concentration.

کلیدواژه‌ها [English]

  • Intelligent Models
  • Land use
  • Pedotransfer functions
  • Readily available soil variables
  • Water repellency

مراجع

Aelamanesh P., Mosaddeghi M.R., Mahboubi A.A., Ahrens B., and Safari Sinegani A.A. 2014. Water repellency in calcareous soils under different land uses in western Iran. Pedosphere, 24(3): 378-390.
Amirabedi H., Asghari Sh., Mesri T., Balandeh N., and Johari E. 2018. Estimating the soil saturated hydraulic conductivity in Ardabil plain soils using artificial neural networks and regression models. Applied Soil Research, 7(4): 88-109. (In Persian)
Asghari Sh., Hatamvand M., and Hasanpour Kashani M. 2019. Deriving pedotransfer functions for estimating soil saturated hydraulic conductivity in northwest of Urmia Lake. Quarterly Journal of Environmental Erosion Research, 9(3):102-118. (In Persian)
Ahmadi A., Alimohammadi M., and Asghari Sh. 2019. Pedotransfer functions for estimating soil moisture content using fractal parameters in Ardabil province. Quarterly Journal of Environmental Erosion Research, 34(9:2):37-52. (In Persian)
Beyrami H., Neyshabouri M., Nazemi A., and Abbasi, F. 2015. Effects of soil water repellency on infiltration characteristics of two sandy loam and clay loam soils. Water and Soil Science, 25(2): 181-192. (In Persian)
Beyrami H., Neyshabouri M., Abbasi F., and Nazemi, A. 2016. Effects of soil water repellency on soil moisture retention curve and S-index in two soils with different textures. Water and Soil Science, 25(1-4): 17-26. (In Persian)
Blake G.R., and Hartge K.H. 1986a. Bulk density, In: Klute, A. (Ed.). Methods of Soil Analysis. Part 1. 2nd ed. Agronomy. Monograph. 9. Madison, WI: Soil Science Society of America; pp. 363-375.
Blake G.R., and Hartge K.H. 1986b. Particle Density. In: Klute, A. (Ed.). Methods of Soil Analysis. Part 1. 2nd ed. Agronomy. Monograph. 9. Madison, WI: Soil Science Society of America; pp. 377-382.
Campbell G.S. 1985. Soil Physics with Basic: Transport Models for Soil–Plant System. Elsevier, New York. 150 pp.
Danielson R.E., and Sutherland, P.L. 1986. Porosity. In: Klute A. 2nd Ed. Methods of Soil Analysis. Part 1. Physical and mineralogical methods. Soil Science Society of America and American Society of Agronomy, Madison, WI, P 443-461.
Dekker L.W., Ritsema C.J. 1994. How water moves in a water repellent sandy soil—Part 1. Potential and actual water repellency. Water Resources Research, 30: 2507–2517.
Dekker L.W., and Jungerius P.D. 1990. Water repellency in the dunes with special reference to The Netherlands. Catena, 18: 173-183.
Doerr S.H., and Thomas A.D. 2000. The role of soil moisture in controlling water repellency: new evidence from forest soils in Portugal. Journal of Hydrology, 231-232: 134–147.
Ferreira C. 2005. Gene expression programming and the evolution of computer programs. In recent developments in biologically inspired computing, pp. 82-103.
Farzadmehr M., Dastourani M., Khashei Siuki A., and Jalali Moakhar V.R. 2020. Estimating the saturated hydraulic conductivity of soil using Gene expression programming method and comparing it whit the pedotransfer functions. Journal of watershed management research, 11(22):155-164. (In Persian)
Goodarzinejad A. 2001. Artificial Intelligence and Modeling. Shahid Chamran University of Ahvaz Publications. 325 p. (In Persian)
Gee G.W., and Or D. 2002. Particle-size analysis. In: Dane J. H., and Topp G. C. (Eds.), Methods of soil analysis—Part 4. Physical Methods—SSSA Book Series No. 5. Soil Science Society of America, Madison, WI, pp. 255-293.
Kozak E., Pachepsky Y.A., Sokolowski S., Sokolowska Z., and Stepniewski W. 1996. A modified number-based method for estimating fragmentation fractal dimensions of soils. Soil Science Society of America Journal, 60: 1291-1297.
Moghimi S., Parvizi Y., Mahdian M.H., and Masih Abadi M.H.  2015. Comparison of applying multi linear regression analysis and artificial neural network methods for simulating topographic factors effect on soil organic carbon. Watershed Engineering and Management, 6(4): 312-322. (In Persian)
Mirbabaei S.M., Shabanpour M., Zolfaghari A., and Abkenar K.T. 2013. Relationship between soil water repellency and some of soil properties in northern Iran. Catena, 108: 26-34.
Martínez-Zavala L., and Jordán-López A. 2009. Influence of different plant species on water repellency in Mediterranean heathland soils. Catena, 76(3): 215–223.
Mousavi zadeh mojarad R. A., Tabatabaei S., Ghorbani B., and Nourmahnad N. 2020. Effect of soil water repellency on runoff and erosion under different rainfall intensities. Iranian Water Researches Journal, 14(2): 51-60. (In Persian)
Moriasi D.N., J.G. Arnold M.W. Van Liew R.L. Bingner, R.D. Harmel, and T.L Veith. 2007. Model evaluation guidelines for systematic quantification of accuracy in watershed simulations. Transactions of the ASABE, 50(3): 885-900.
Pishnamaz Amoli F., Mosaddeghi M.R., Davatgar N., Chavoshi E., and Torabi Golsefidi H. 2022. The impact of land use on soil water repellency in the central part of Mazandaran province. Journal of soil Management and Sustainable production, 12(1): 59- 79. (In Persian)
Page A.L. (ed.).1985. Methods of Soil Analysis. Part 2. Chemical and Microbiological Methods. Agronomy No. 9. American Society of Agronomy, Madison, WI.
Raheli B., Aalami M.T., El-Shafie A., Ghorbani M.A., and Deo R.C. 2017. Uncertainty assessment of the multilayer perceptron (MLP) neural network model with implementation of the novel hybrid MLP-FFA method for prediction of biochemical oxygen demand and dissolved oxygen: A case study of Langat River. Environmental Earth Sciences, 76(14): 503.
Shahmohammadi-Kalalagh Sh, and Beyrami H. 2018. Effect of soil surface water repellency on the evaporation rate from three soils with different textures. Journal of Water and Soil Resources Conservation, 7(2): 123-136. (In Persian)
Tillman R. W., Scotter D. R., Wallis M. G., and Clothier B. E. 1989. Water repellency and its measurement by using intrinsic sorptivity. Australian Journal of Soil Research, 27(4): 637–644.
Yazdani A., Mosaddeghi M.R., Khademi H., Ayoubi S., and Khayamim F. 2016. Relationship of structural stability of surface soils with soil and climate properties in Isfahan province. Soil Management, 3(2): 23-31. (In Persian)
Zulfiqari A.A., and Haj Abbasi M.A. 2008. The effects of land use change on physical properties and water repellency of soils in Lordegan forest and Freidunshar pasture. Journal of Water and Soil, 22(2): 251-262. (In Persian)
Van Bavel C.H.M. 1949. Mean weight diameter of soil aggregates as a statistical index of aggregation. Soil Science Society of America proceeding, 14: 20-23.

فایل ها

سابقه مقاله

  • تاریخ دریافت: 02 اردیبهشت 1403
  • تاریخ بازنگری: 09 تیر 1403
  • تاریخ پذیرش: 11 تیر 1403

هم رسانی

ارجاع به این مقاله

آمار