Evaluation of the effect of grouping based on different characteristics on the performance of functions in estimating soil cation exchange capacity

Document Type : Research Paper

Authors

1 Department of Soil Science, Faculty of Agriculture, Bu Ali Sina University, Hamadan, IRAN

2 Department of Soil Science, Faculty of Agriculture, Bu Ali Sina University, Hamadan,, IRAN

3 Ph. D. Student, Soil Science Department, Faculty of Agricultural Sciences, University of Guilan, RASHT, IRAN

Abstract

This study addresses the challenge of measuring soil cation exchange capacity (CEC), a vital factor influencing soil fertility, by exploring the impact of grouping soil samples based on different characteristics on the performance of estimation models. Recognizing the difficulties associated with traditional CEC measurement methods, the study employs a cost-effective and rapid approach using various models and equations. The research, conducted at Bu Ali Sina University in Hamedan, utilizes a substantial dataset of 45,948 soil samples from the standardized database of world soils. Soil samples are initially categorized into different groups, and nine estimator variables are examined across 11 models for the entire dataset and specific classes within each group. These variables include soil texture components, organic carbon, calcium sulfate, calcium carbonate, bulk density, base saturation percentage, total exchangeable base cations, and soil reaction. The results demonstrate that grouping soil samples, especially based on texture classes, significantly improves the performance of artificial neural network models, with a remarkable 87% relative improvement coefficient in the test section. The study reveals that data grouping enhances the model's estimation capabilities, as evidenced by reduced root mean square error (RMSE) values in the test sections for different texture classes. In conclusion, the findings suggest that utilizing functions derived from grouped data offers an effective and cost-efficient method for estimating soil cation exchange capacity. This approach provides valuable insights for soil fertility management, offering a simplified yet accurate means of assessing this critical soil parameter.

Keywords

Main Subjects


Evaluation of the Effect of Grouping Based on Different Characteristics on the Performance of Functions in Estimating Soil Cation Exchange Capacity

EXTENDED ABSTRACT

 

Introduction

The cation exchange capacity of the soil is one of the most important characteristics, as it greatly influences soil fertility. This capacity is influenced by various physical and chemical characteristics of the soil, resulting in significant variability. However, measuring soil cation exchange capacity can be challenging and expensive. The objective of this study is to assess the impact of grouping based on different characteristics on the accuracy of estimating soil cation exchange capacity. Additionally, we aim to introduce a grouping method that yields the best estimation results and compare the effectiveness of linear regression and artificial neural networks in estimating cation exchange capacity.

Material and Methods

For this research, we utilized 45,948 soil samples from a standardized global soils database. Initially, these samples were grouped into different categories. Subsequently, we evaluated nine predictor variables including soil texture components, organic carbon content, calcium sulfate and carbonate levels, apparent specific gravity, base saturation percentage, total exchangeable base cations, and soil reaction in 11 models for both the entire dataset and each group separately. To determine which grouping method yields more accurate estimations for soil cation exchange capacity and which characteristic contributes to better results, transfer functions were developed for each group using all data and then again for each class within the groups. In this study, we employed multilayer perceptron neural networks with one hidden layer and a hyperbolic tangent activation function. The number of cells in the hidden layer ranged from 3 to 8. The network with the optimal number of hidden cells was selected as the final network based on its performance. Additionally, linear regression modeling was conducted using Datafit 9 software to compare its effectiveness with artificial neural networks in estimating cation exchange capacity.

Results and Discussion

The correlation between the cation exchange capacity of soil and clay, apparent specific gravity, the ratio of silt to positive sand, calcium carbonate, total exchangeable base cations, and soil reaction was found to be significant at the 1% level. When comparing the RMSE and RI statistics in two sets of training and test data for the entire dataset and for classes separated based on soil texture groups, it was observed that data separation improved the mean in estimating the cation exchange capacity of the soil. Two methods, linear regression and artificial neural networks, were used to estimate the cation exchange capacity. The results showed that for all coarse, medium, and fine texture groups, the artificial neural network test section had a relative improvement coefficient of 87%, while linear regression had a value of 17%. The grouping of data generally increased the estimation ability of the models. Among the groups studied in this research, the texture group showed higher accuracy in predicting cation exchange capacity compared to other groups. Overall, using functions obtained through grouping proved to be an easy and cost-effective method for estimating cation exchange capacity.

Conclusion

In general, artificial neural networks outperformed linear regression in estimating cation exchange capacity in most classes within each group. This may be due to not requiring a specific type of equation in designing artificial neural networks. By establishing a suitable relationship between output and input data, accurate results can be achieved.

Amini, M., Abbaspour, K. C., Khademi, H., Fathianpour, N., Afyuni, M., & Schulin, R. (2005). Neural network models to predict cation exchange capacity in arid regions of Iran. European Journal of Soil Science, 56, 551-559.
Asadu, C. L. A., & Akamigbo, F. O. R. (1990). Relative contribution of organic matter and clay fractions to cation exchange capacity of soils in southern Nigeria. Samaru: Journal of Agriculture Research, 7, 17–23.
Asadu, C. L. A., Diels, J., & Vanlauwe, B. (1997). A comparison of the contributions of clay, silt and organic matter to the effective CEC of soils of sub-Saharan Africa. Soil Science, 162, 785-794.
Bayat, B. M., Neyshabouri, R., Hajabbasi, M. A., Mahboubi, A. A., & Mosaddeghi, M. R. (2008). Comparing neural networks, linear and nonlinear regression techniques to model penetration resistance. Turkish Journal of Agriculture and Forestry, 32, 425–433.
Bayat, H., Davatgar, N., & Jalali, M. (2013). Prediction of CEC using fractal parameters by artificial neural networks. International Agrophysics, 28, 143-152.
Bayat, H., Davatgar, N., & Moallemi, S. (2012). Using of specific surface to improve the prediction of Ssoil CEC by artificial neural networks. Journal of Water and Soil Science, 21(4), 105-119. (In Persian).
Cai, Z., Yang, C., Du, X., Zhang, L., Wen, S., & Yang, Y. (2023). Parent material and altitude influence red soil acidification after converted rice paddy to upland in a hilly region of southern China. Journal of Soils and Sediments, 1-13.
Costa, J. L., Aparicio, V., & Cerdà, A. (2015). Soil physical quality changes under different management systems after 10 years in the Argentine humid pampa. Solid Earth6(1), 361-371.
Drake, E. H., & Motto, H. L. (1982). An analysis of the effect of clayand organic matter content on the cation exchange capacity of New Jersey soils. Soil Science, 133, 281–288.
Edmeades, D. C. (1982). Effects of lime on effective cation exchange capacity and exchangeable cations on a range of New Zealand soils. New Zealand Journal of Agricultural Research, 25, 27-33.
Emamgholizadeh, S., Bazoobandi, A., Mohammadi, B., Ghorbani, H., & Sadeghi, M. A. (2023). Prediction of soil cation exchange capacity using enhanced machine learning approaches in the southern region of the Caspian Sea. Ain Shams Engineering Journal, 14(2), 101876.
FAO/IIASA/ISRIC/ISS-CAS/JRC. (2012). Harmonized World Soil Database (version 1.2). FAO, Rome, Italy and IIASA, Laxenburg, Austria.
Firat Pulat, H., Tayfur, G., & Yukselen-Aksoy, Y. (2014). Developing cation exchange capacity and soil index properties relationships using a neuro-fuzzy approach. Bulletin of Engineering Geology and the Environment, 73, 1141–1149.
Fouladmand, H. R. (2007). Estimation of soil cation exchange capacity using some soil physicochemical properties. Journal of Agricultural Sciences and Natural Resources, 1, 1-8. (In Persian).
Hezarjaribi, A., Nosrati Karizak, F., Abdollahnezhad, K., & Ghorbani, Kh. (2013). The prediction possibility of soil cation exchange capacity by using of easily accessible soil parameters. Journal of Water and Soil, 27(4), 712-719. (In Persian).
Ho, R. (2006). Handbook of univariate and multivariate data analysis and interpretation with SPSS. Chapman and Hall/CRC, 403p.
Horn, A. L., Düring, R. A., & Gäth, S. (2005). Comparison of the prediction efficiency of two pedotransfer functions for soil cation-exchange capacity. Journal of Plant Nutrition and Soil Science, 168, 372-374.
Ibrahim, O. M., El-Gamal, E. H., Darwish, K. M., & Kianfar, N. (2022). Modeling main and interactional effects of some physiochemical properties of Egyptian soils on cation exchange capacity via artificial neural networks. Eurasian Soil Science, 55(8), 1052-1063.
Kar, R., Bose, P. C., and Bajpai, A. K. (2008). Prediction of cation exchange capacity of soils of mulberry garden based on their clay and organic carbon content in Eastern India. Journal of Crop and Weed, 4(2), 47-49.
Keshavarzi, A., Sarmadian, F., Sadeghnejad, M., & Pezeshki, P. (2010). Developing pedotransfer functions for estimating some soil properties using artificial neural network and multivariate regression approaches. Pro Environment, 3, 322 – 330.
Khodaverdiloo, H., & Hosseini Arablu, N. (2014). Derivation, validation and comparison of class and continuous pedotransfer functions for predicting soil cation exchange capacity in several textural classes. Journal of Science and Technology of Agriculture and Natural Resources, 18(67), 311-320. (In Persian).
Khormali, F., Abtahi, A., Mahmoodi, S., & Stoops, G. (2003). Argillic horizondevelopment in calcareous soils of arid and semiarid regions of southern Iran. Catena, 53, 273-301.
Khormali, F., Ghorbani, R., and Amoozadeh Omrani, R. (2005). Variations in soil properties as affected by deforestation on loess-derived hillslopes of Golestan Province, northern Iran. Sociedade and Natureza, Uberlândia, Brazil, Pp: 440-445.
Kissel, D. E., & Sonon, L., (eds). )2008 (. Soil Test Handbook for Georgia. http://aesl.ces.uga.edu/publications/soil/STHandbook.pdf
Krogh, L., Breuning-madsen, H., & Greve, M. H. (2000). Cation exchange capacity pedotransfer function for Danish soils. Plant and Soil, 50, 1-12.
MacDonald, K.B. (1998). Development of Pedotransfer Functions of Southern Ontario Soils, pp: 1–23. Report from greenhouse and processing crops research center, Harrow, Ontario, No: 01686-8- 0436
Madeira, M., Auxtero, E., & Sousa, E. (2003). Cation and anion exchange properties of Andisols from the Azores, Portugal, as determined by the compulsive exchange and the ammonium acetate methods. Geoderma, 117-225.
Manrique, L. A., Jones, C. A., & Dyke, P. T. (1991). Predicting cation exchange capacity from soil physical and chemical properties. Soil Science Society of America Journal, 55:787–794.
Marschner, P., & Rengel, Z. (2023). Nutrient availability in soils. In Marschner's Mineral Nutrition of Plants (pp. 499-522). Academic press.
Matos, A. T., Fontes, M. P. F., Costa, L. M., & Martinez, M. A. (2001). Mobility of heavy metals as related to soil chemical and mineralogical characteristics of Brazilian soils. Environmental Pollution, 111, 429-435.
Merdun, H., Meral, O. C., & Apan, R. M. (2006). Comparison of artificial neural network and regression pedo transfer funection for predict of water retention and standard hydraulic conductivity. Soil and Tillage Research, 90, 108-116.
Mishra, G., Das, J., & Sulieman, M. (2019). Modelling soil cation exchange capacity in different land-use systems using artificial neural networks and multiple regression analysis. Current Science, 116(12), 2020-2027.
Moallemi, S., & Davatgar, N. (2011). Comparison of artificial neural network and regression pedotransfer functions for prediction of cation exchange capacity in Guilan Province soils. Journal of Science and Technology of Agriculture and Natural Resources, 15(55), 169-182. (In Persian).
Moghadam, M. R. (2001). Statistics and description of vegetation ecology. Tehran University Publications. 285 pages. (In Persian).
Mohajer, R., Salehi, M., & 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. Journal of Science and Technology of Agriculture and Natural Resources, 13(49), 83-97. (In Persian).
Mohammadi, M. (2006). Agricultural soil science. Tehran. Sepehr Publishing Center, 245 pages. (In Persian).
Morras, H. J. M. (1995). Mineralogy and cation exchange capacity of the fine silt fraction in teo soils from the Chaco region (Argentina). Geoderma, 64, 281-295.
Mukherjee, A., & Zimmerman. A.R. (2013). Organic carbon and nutrient release from a range of laboratory-produced biochars and biochar–soil mixtures. Geoderma.193–194(0):122- 30.
Nourbakhsh, F., Jalalian, A., & Shariatmadari, H. (2003). Estimation of cation exchange capacity from some soil physical and chemical properties. Journal of Science and Technology of Agriculture and Natural Resources, 7(3), 107-118. (In Persian).
Obalum, S. E., Watanabe, Y., Igwe, C. A., Obi, M. E., & Wakatsuki, T. (2012). Improving on the prediction of cation exchange capacity for highly weathered and structurally contrasting tropical soils from their fine-earth fractions. Taylor and Francis Group, 44,1831–1848.
Olorunfemi, I., Fasinmirin, J., & Ojo, A. (2016). Modeling cation exchange capacity and soil water holding capacity from basic soil properties. Eurasian Journal of Soil Science5(4), 266-274.
Pachepsky, Y. A., & Rawls, W. J. (1999). Accuracy and reliability of pedotransfer functions as affected by grouping soils. Soil Science Society of America Journal, 63,1748–1757.
Pulido, M., Schnabel, S., Lavado Contador, J. F., Lozano‐Parra, J., & Gonzalez, F. (2018). The impact of heavy grazing on soil quality and pasture production in rangelands of SW Spain. Land Degradation & Development29(2), 219-230.
Rahimi Lake, H., Akbarzadeh, A., & Taghizadeh Mehrjardi, R. (2009). Development of pedo transfer functions (PTFs) to predict soil physico-chemical and hydrological characteristics in southern coastal zones of the Caspian Sea. Journal of Ecology and The Natural Environment, 1(7), 160-172.
Salchow, E., Lal, R., Fausey, N. R., & Ward, A. (1996). Pedotransfer functions for variable alluvial soils in southern Ohio. Geoderma, 73, 165-181.
Sarmadian, F., Azimi, S., Keshavarzi Ahmadi, A. (2013). Neural computing model for prediction of Soil Cation Exchange Capacity: A Data Mining Approach. International Journal of Plant Production, 4, 1706-1712.
Sayegh, A. H., Khan, N. A., Khan, P., & Ryan, J. (1978). Factors affecting gypsum and CEC determinations in gypsiferous soils. Soil Science, 125(5), 294-300.
Seybold, C. A., Grossman, R. B., & Reinsch, T. G. (2005). Preicting Cation Exchange Capacity for Soil Survey Using Linear Models. Soil Science Society of America Journal, 69, 856-86.
Soares, M. R., Alleoni, L. R. F., Vidal-Torrado, P., & Cooper, M. (2005). Mineralogy and ion exchange properties of the particle-size fractions of some Brazilian soils in tropical humid areas. Geoderma, 125, 355–367.
Sparks, D. L. (1995). Enviromental Soil Chemistry. In Academic Press Inc. University of Delaware London.
Syers, J. K., Campbell, A. S., & Walker, T. W. (1970). Contribution of organic carbon and clay to cation exchange capacity in a chronosequence of sandy soils. Plant and Soil, 33, 104–112.
Tamari, S., Wosten, J. H. M., & Ruz-suarez, J. C. (1996). Testing an artificial neural network for predicting soil hydraulic conductivity. Soil Science Society of America Journal, 60, 1732-1741.
Tang, L., Zeng, G. M., Nourbakhsh, F., & Shen, G. L., (2008). Artificial neural network approach for predicting cation exchange capacity in soil based on physico-chemical properties. Environmental Engineering Science, 26(2), 1-10.
Tessler, M., David, F. J., Cunningham, S. W., & Herstoff, E. M. (2023). Rewilding in miniature: suburban meadows can improve soil microbial biodiversity and soil health. Microbial Ecology, 1-10.
Thompson, M. L., Zhang, H., Kazemi, M., & Sandor. J. A. (1989). Contribution of organic matter to cation exchange capacity and specific surface area of fractionated soil materials. Soil Science, 148: 250-257.
Turpault, M. P., Bonnaud, P., Fichter, J., Ranger, J., & Dambrine, E. (1996). Distribution of cation exchange capacity between organic matter and mineral fractions in acid forest soils (Vosges mountains, France). European Journal of Soil Science, 47, 545-556.
Wagner, B., Hennings, V., Muller, U., Wessolek, G., & Plagge, R. (2001). Evaluation of pedotransfer functions for unsaturated soil hydraulic conductivity using an independent data set. Geoderma, 102, 275-279.
Wilding, L. P., Smeck, N. E., & Hall, G. F. (1983). Pedogenesis and soil taxonomy. I. Concepts and interactions. Elsevier Publishing Company, 303p.
Yukselen, Y., & Kaya, A. (2006). Prediction of cation exchange capacity from soil index properties. Clay Minerals, 41, 827–837.
Zeraat Pishe, M., Khormali, F., Kiani, F., & Pahlavani, M. H. (2013). Studying clay minerals in soils formed on loess parent materials in a climatic gradient in Golestan Province. Soil Research, 26(3), 303-316. (In Persian).