ارزیابی مدل‌های رگرسیونی و هوشمند برای برآورد میانگین وزنی قطر خاکدانه‌های تر

نوع مقاله : مقالات پژوهشی

نویسندگان

• شکراله اصغری ¹
• کیمیا حیدری ¹
• مهسا حسنپور کاشانی ²
• حسین شهاب آرخازلو ¹

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

² گروه مهندسی آب، دانشکده کشاورزی و منابع طبیعی، دانشگاه محقق اردبیلی، اردبیل، ایران

10.22067/jsw.2025.91071.1454

چکیده

اندازه­گیری مستقیم میانگین وزنی قطر (MWD) خاکدانه­های تر به­عنوان یکی از شاخص­های مهم برای ارزیابی پایداری ساختمان خاک، کاری وقت­گیر و پرهزینه است. هدف پژوهش حاضر مقایسه دقت مدل رگرسیون خطی چندگانه (MLR) و دو مدل هوشمند شامل شبکه عصبی مصنوعی (ANN) و برنامه­ریزی بیان ژن (GEP) در برآورد MWD از روی متغیرهای زودیافت و سهل­الوصول خاک بود. برای این منظور، 80 نمونه خاک سطحی دست­خورده و دست­نخورده از عمق صفر تا cm 10 اراضی جنگلی، مرتعی و زراعی منطقه فندقلوی استان اردبیل جمع­آوری شد. سپس برخی ویژگی­های فیزیکی و شیمیایی زودیافت آن­ها و میانگین وزنی قطر (MWD) خاکدانه­های تر مطابق روش­های معمول و استاندارد تعیین گردید. داده­ها به­طور تصادفی به دو مجموعه آموزشی (60 داده) و آزمونی (20 داده) تقسیم گردید. مدل­های MLR، ANN و GEP به­ترتیب با به­کارگیری نرم­افزارهای SPSS، MATLAB و Gene Xpro Tools ایجاد شدند. نتایج نشان­دهنده همبستگی مثبت و معنی­دار (^**59/0 r=) بین شن و کربن آلی خاک بود. شن، کربن آلی و میانگین هندسی قطر (GMD) خاکدانه­های خشک از مهمترین متغیرهای ورودی مدل­ها در برآورد MWD خاکدانه­های تر بودند. براساس داده­های آزمونی، مقادیر ضریب تبیین (R²)، ریشه میانگین مربعات خطا (RMSE)، میانگین خطا (ME) و ضریب نش-ساتکلیف (NS) برابر 52/0، mm 48/0، mm 13/0 و 48/0، 85/0، mm 30/0 ، mm 03/0 و 78/0، 79/0، mm 35/0، mm 10/0 - و 95/0 به­ترتیب در بهترین مدل MLR، ANN و GEP به­دست آمد. بنابراین مدل MLR در مقایسه با مدل­های هوشمند از دقت کمتری و خطای بیشتری در برآورد MWD برخوردار بود؛ مدل ANN به علت داشتن R² و NS بالا و RMSE پایین توانست در مقایسه با دو مدل دیگر، MWD خاکدانه­های تر را با دقت زیاد و خطای کم در منطقه مورد مطالعه تخمین بزند.

کلیدواژه‌ها

• ارضی شیبدار
• توابع انتقالی خاک
• برنامه‌ریزی بیان ژن
• پایداری خاکدانه
• شبکه عصبی مصنوعی

موضوعات

• علوم خاک

عنوان مقاله [English]

Evaluation of Regression and Intelligent Models for Estimating Mean Weight Diameter of Wet Aggregates

نویسندگان [English]

• Sh. Asghari ¹
• K. Heidari ¹
• M. Hasanpour Kashani ²
• H. Shahab Arkhazloo ¹

¹ Department of Soil Sciences and Engineering, Faculty of Agriculture and Natural Resources, University of Mohaghegh Ardabili, Ardabil, Iran

² Water Engineering Department, Faculty of Agriculture and Natural Resources, University of Mohaghegh Ardabili, Ardabil, Iran

چکیده [English]

Introduction
 The study of soil mean weight diameter (MWD) of wet aggregates that is important for sustainable soil management, has recently received much attention. As the prediction of MWD is challenging, laborious, and time-consuming, there is a crucial need to develop a predictive estimation method to generate helpful information required for the soil health assessment to save time and cost involved in soil analysis. Therefore, it is useful to use different models such as multiple linear regression (MLR) and intelligent models including artificial neural network (ANN) and gene expression programming (GEP) to estimate MWD of wet aggregates through easily accessible and low-cost soil properties. The objectives of this study were (1) to creating MLR, ANN and GEP models for predicting MWD from the easily measurable soil variables in forest, range and cultivated lands of the Fandoghloo region of Ardabil province, (2) to compare the precision of the mentioned models in the prediction of MWD of wet aggregates using the coefficient of determination (R²), root mean square error (RMSE), mean error (ME) and Nash-Sutcliffe coefficient (NS) criteria.
 
Materials and Methods
 Disturbed and undisturbed soil samples (n= 80) were nearly systematically taken from 0-10 cm depth with nearly 50 m distance in forest (n= 20), range (n= 23) and cultivated (n= 37) lands of the Fandoghloo region of Ardabil province, Iran (lat. 38° 24' 10" to 38° 24' 25" N, long. 48° 32' 45" to 48° 33' 5" E) in summer 2023. The contents of sand, silt, clay, CaCO₃, pH, EC, bulk (BD) and particle (PD) density, organic carbon (OC), geometric mean diameter (GMD) of dry aggregates were determined in the laboratory using standard methods. Total porosity (n) was calculated using BD and PD data (n= 1-BD/PD). The mean geometric diameter (d_g) and geometric standard deviation (σ_g) of soil particles were computed by sand, silt and clay percentages. The mean weight diameter (MWD) of wet aggregates was measured in the aggregates smaller than 4.75 mm by wet sieving equipment using sieves with 2, 1, 0.5, 0.25 and 0.106 mm pore diameter. All data were randomly divided into two series as 60 data for training and 20 data for testing of models. The SPSS 22 software with the stepwise method, MATLAB and Gene Xpro Tools 4.0 software were used to derive multiple linear regression (MLR), artificial neural network (ANN) and gene expression programming (GEP) models, respectively. A feed forward three-layer (9, 8, 6 and 6 neurons in the hidden layer) perceptron network and the tangent sigmoid transfer function were used for the ANN modeling. A set of optimal parameters were chosen before developing the best GEP model. The number of chromosomes and genes, head size and linking function were selected by the trial and error method, and they are 30, 3, 8, and +, respectively. The rates of genetic operators were chosen according to literature studies. The precision of MLR, ANN and GEP models in predicting MWD of wet aggregates were evaluated by the coefficient of determination (R²), root mean square error (RMSE), mean error (ME) and Nash-Sutcliffe coefficient (NS) statistics.
 
Results and Discussion
The values of sand (13.14 to 64.79 %), silt (21.11 to 74.96 %), clay (3 to 42.18 %), OC (1.01 to 7.17 %), PD (2.00 to 2.67 g cm^-3), n (0.39 to 0.87 cm³ cm^-3), GMD of dry aggregates (0.8 to 1.33 mm) and MWD of wet aggregates (0.35 to 2.65 mm) showed good variations in the soils of the studied region. The studied soils had clay loam (n= 11), sandy clay loam (n= 6), sandy loam (n= 12), loam (n= 13), silty clay loam (n= 14), silty clay (n= 1) and silt loam (n= 23) textural classes. There were found significant correlations between MWD with OC (r= 0.67^**), sand (r= 0.70^**), GMD (r= 0.30^**) and PD (r= -0.46^**). Also, significant and positive correlation was found between OC and sand (r= 0.59^**). Due to the multicollinearity of sand with d_g (r= 0.87^**), we did not use the d_g as an input variable to estimate MWD of wet aggregates. Generally, four MLR, ANN and GEP models were constructed to predict MWD of wet aggregates from measured readily available soil variables. The results of MLR, ANN and GEP models indicated that the most suitable variables to estimate MWD of wet aggregates were sand, OC and GMD of dry aggregates. The values of R², RMSE, ME and NS criteria were obtained equal 0.52, 0.48 mm, 0.13 mm and 0.48, and 0.85, 0.30 mm, 0.03 mm and 0.78, 0.79, 0.35 mm, -0.10 mm, 0.95 for the best MLR, ANN and GEP models in the testing data set, respectively. Many researchers also reported that there is a positive and significant correlation between MWD of wet aggregates and OC.
 
Conclusion
 The results showed that sand, OC and GMD of dry aggregates were the most important and readily available soil variables to predict the mean weight diameter (MWD) of wet aggregates in the Fandoghloo region of Ardabil province. According to the lowest values of RMSE and the highest values of R² and NS, the precision of ANN models to predict MWD of wet aggregates was more than MLR and GEP models in this study. Because ANN is more flexible and effectively captures non-linear relationships, it performed better than the other models in predicting MWD.    

کلیدواژه‌ها [English]

• Aggregate stability
• Artificial neural network
• Gene expression programming
• Sloped lands
• Soil pedotransfer functions

©2024 The author(s). This is an open access article distributed under Creative Commons Attribution 4.0 International License (CC BY 4.0).

مراجع

1. Alijanpour Shalmani, A., Shabanpour, M., Asadi, H., & Bagheri, F. (2011). Estimation of soil aggregate stability in forest`s soils of Guilan province by artificial neural networks and regression pedotransfer functions. Water and Soil Science, 21(3), 153-162. (In Persian with English abstract)
2. Amirabedi,, Asghari, Sh., Mesri, T., & Balandeh, N. (2016). Prediction of mean weight diameter of aggregates using artificial neural network and regression models. Applied Soil Research, 4(1), 39-53. (In Persian with English abstract)
3. Ahmadzadeh Kaleibar, F., & Fuladipanah, M. (2023). Assessment of regression, support vector machine, and gene expression programming transfer functions to predict soil humidity parameters in Arasbaran plain. Journal of Water and Soil Science, 27(2), 135-149. https://doi.org/10.47176/jwss.27.2.42532
4. Asghari, Sh., Dizajghoorbani Aghdam, S., & Esmali, A. (2016). Investigation the spatial variability of some soil physical quality indices in Fandoghlou region of Ardabil using geostatistics. Water and Soil, 28(6), 1271-1283. https://doi.org/10.22067/jsw.v0i0.33460
5. Asghari, Sh., Alimohammadi, M., Ahmadi, A., & Davatgar, N. (2017). Derivation of pedotransfer functions for estimating wet aggregate stability using fractal dimensions of particles and aggregates. Water and Soil Science, 27(1), 107-119.
6. Asghari, Sh., Hatamvand, M., & Hasanpour Kashani, M. (2021). Estimating wet aggregates stability from easily available soil properties in north west of Urmia Lake. Applied Soil Research, 9(2), 102-115. (In Persian with English abstract)
7. Bhattacharya, P., Maity, P.P., Ray, M., & Mridha, N. (2021). Prediction of mean weight diameter of soil using machine learning approaches. Agronomy Journal, 113(2), 1303–1316. https://doi.org/10.1002/agj2.20469
8. Blake, G.R., & Hartge, K.H. (1986a). Bulk density. p. 363-375. In: Klute A. (ed). Methods of Soil Analysis Part 1, Physical and Mineralogical Methods. 2nd ed. American Society of Agronomy, Madison, WI.
9. Blake, G.R., & Hartge, K.H. (1986b). Particle density. p. 377-381. In: Klute A. (ed). Methods of Soil Analysis Part 1, Physical and Mineralogical Methods. 2nd ed. ASA and SSSA, Madison, WI.
10. Ferreira, C. (2001). Gene expression programming: a new adaptive algorithm for solving problems. Complex Systematic, 13, 87–129. https://doi.org/10.48550/arXiv.cs/0102027
11. Ghorbani, M.A., Deo, R.C., Kashani, M.H., Shahabi M., & Ghorbani, S. (2019). Artificial intelligence-based fast and efficient hybrid approach for spatial modeling of soil electrical conductivity. Soil and Tillage Research, 186, 152–164. https://doi.org/10.1016/j.still.2018.09.012
12. Gee, G.W., & Or, D. (2002). Particle-size analysis. p. 255–293. In: Dane J.H., & Topp G.C. (eds.). Methods of Soil Analysis. Part 4. SSSA Book Series No. 5. Soil Science Society of America, Madison, WI.
13. Kemper, W.D., & Rosenau, K. (1986). Size distribution of aggregates. 425-442. In: Klute A. (ed). Methods of Soil Analysis Part 1, Physical and Mineralogical Methods. 2nd ed. ASA and SSSA, Madison, WI.
14. Kozak, E., Pachepsky, Y.A., Sokolowski, S., Sokolowska, Z., & Stepniewski, W. (1996). A modified number-based method for estimating fragmentation fractal dimensions of soils. Soil Science Society of America Journal. 60, 1291- 1297.
15. Marashi, M., Mohammadi Torkashvand, A., Ahmadi, A., & Esfandyari, M. (2019). Adaptive neurofuzzy inference system: estimation of soil aggregates stability. Acta Ecologica Sinica, 39, 95–101. https://doi.org/10.1016/j.chnaes. 2018.05.002
16. Nelson, D.W., & Sommers, L.E. (1982). Total carbon, organic carbon, and organic matter. p. 539–579. In A.L. Page et al. (ed.) Methods of Soil Analysis. Part 2. 2^nd Agron. Monogr. 9. ASA and SSSA, Madison, WI.
17. Page, A.L. (ed.). (1985). Methods of Soil Analysis. Part 2. Chemical and Microbiological Methods. Agronomy No. 9. American Society of Agronomy, Madison, WI.
18. Sarkar, A., Maity, P.P., Ray, M., Chakraborty, D., Das, B., & Bhatia, A. (2023). Inclusion of fractal dimension in four machine learning algorithms improves the prediction accuracy of mean weight diameter of soil. Ecological Informatics. 74, 1-15. https://doi.org/10.1016/j.ecoinf.2022.101959
19. Shirazi, M.A., & Boersma, L. (1984). A unifying quantitative analysis of soil texture. Soil Science Society American Journal, 48(1), 142–147.
20. Yazdani, A., Mosaddeghi, M.R., Khademi, H., Ayoubi, S., & Khayamim, F. (2014). Relationship between surface aggregate stability and some soil and climate properties in Isfahan province. Soil Management, 3(2), 23-31. (In Persian with English abstract)
21. Yoder, R.E. (1936). A direct method of aggregate analysis of soils and a study of the physical nature of erosion losses. Journal of American Society Agronomy, 28, 337-35.
22. Zhang, R., & Zhang, S., (2024). Coefficient of permeability prediction of soils using gene expression programming. Engineering Applications of Artificial Intelligence, 128(107504). https://doi.org/10.1016/j.engappai.2023.107504