Estimation of Soil Aggregate Stability in Forest`s Soils of Guilan Province by Artificial Neural Networks and Regression Pedotransfer Functions

Document Type : Research Paper

Authors

Abstract

Using artificial neural networks (ANNs) and regression pedotransfer functions to predict the surrogate soil properties such as aggregate stability reduces time and cost needed for their direct measurements. In this research, 100 soil samples were collected from the forest soils of Guilan province. Organic matter, bulk density, equivalent carbonate calcium, particle density, porosity, soil mechanical resistance, clay, sand, silt, pH and electrical conductivity all were measured as independent variables. Geometric mean diameter (GMD) was computed as dependent variable by appropriate method. The samples were divided into two data subsets randomly: 80 for model calibration and 20 for model test. Regression pedotransfer functions were generated by stepwise method. For establishing ANNs we used Marquardt-Levenburg training algorithm and a 3-layer perceptron structure with 6 neurons in one hidden layer. According to the correlation matrix between GMD as dependent variable and independent variables, 10 groups input variables were selected. The were employed once by multi-variate regression pedotransfer functions and once by artificial neural networks. According to the adjusted coefficient of determination (R2ady), root mean square error (RMSE) and relative improvement (RI) a model resulted from applying ANNs and using input variables of pH, particle density, silt and soil mechanical resistance turned to be the best model for predicting GMD of the examined soils.

Keywords

References

بای بوردی  م، 1372.  فیزیک خاک. انتشارات دانشگاه تهران.
رضایی ع و سلطانی ا، 1377. مقدمه­ای بر تحلیل رگرسیون کاربردی. انتشارات دانشگاه صنعتی اصفهان.
منهاج م ب، 1381. مبانی شبکه­های عصبی مصنوعی (جلد اول). انتشارات دانشگاه صنعتی امیرکبیر.
نوابیان م ، 1382. تخمین هدایت آبی اشباع با استفاده از توابع انتقالی. پایان نامه­ی کارشناسی ارشد، دانشکده کشاورزی دانشگاه تهران.
Barral MT, Arias M and Guerif J, 1998. Effect of iron and organic matter on the porosity and structural stability of soil aggregates. Soil & Tillage Research 46: 261-272.
Chenu C, Le Bissonnias Yand Arrouays  D ,2000. Organic matter influence on clay wettability and soil aggregate stability. Soil Sci Soc Am J  64: 1479-1486.
Doai M, Shabanpaour Shahrestani  M, Bagheri F and Navabiyan  M, 2006. Comparison of regression pedotransfer functions and artificial neural networks to simulation of soil hydraulic properties. Accepted in. 18th WCSS. USA.
Emerson WW, 1991. Structural decline of soils, assessment and  prevention. Aust  J Soil Res   24: 905-921.
Ghielmi L  and  Eccel  E, 2006. Descriptive models and artificial neural networks for spring frost prediction in an agricultural mountain area. Computers and Electronics in Agriculture   54: 101-114.
Heuvelmans G , Muys  B and Feyen  J, 2006. Regionalisation of the parameters of a hydrological model: Comparison of linear regression models with artificial neural nets. Journal of Hydrology   319: 245-265.
Klut A (ed.), 1986. Method of Soil Analysis. Part 1. Physical and Mineralogical Properties. ASA and  SSSA. Madison, WI.
Lentzsch P, Wieland  R and Wirth  S, 2005. Application of multiple regression and neural network approaches for landscape-scale assessment of soil microbial biomass. Soil Biology and Biochemistry 37: 1577-1580.
Marquardt  DW, 1963. An algorithm for least-squares estimation of nonlinear parameter. J SocInd Appl Math 11: 431–441.
Merdun H, Cinar O, Meral R and Apan M, 2006. Comparioson of artificial neural network and regression pedotramsfer functions for prediction water retention and saturated hydraulic counductivity. Soil & Tillage Res 90: 108-116.
Minasny B and  McbartneyAB, 2002.  The neuro method for fitting neural network parametric pedotransfer functions. Soil Sci  Soc  Am  J 66: 352-361.
Minasny  B,  Hopman  JW, Harter TX, Eching T, Toli A and DentonMA , 2004. Neural networks prediction of soil hydraulic functions for alluvial soils using multi step outflow data. Soil Sci  Soc  Am  J 68: 417- 429.
Mohammadi J, 2002.Testing an artificial neural network for predicting soil water retention characteristics from soil physical and chemical properties. Paper No. 378 and 943. 17th WCSS. Thailand. Paper no:378. Paper no: 943
Neufeldet H, Ayarza MA,. Resck DVS and Zech W, 1999. Distribution of water-stable aggregate in Cerrado Oxisols. Soil  & Tillage Res 93: 85-99.
Page AL, Miller RH and Keeney DR, 1982. Method of Soil Analysis. Part 2. Chemical and Microbiological Properties. ASA and SSSA. Madison, WI.
Ryan M, Müller C, Di HJ and Cameron KC, 2005. The use of artificial neural networks (ANNs) to simulate N2O emissions from a temperate grassland ecosystem Ecological Modeling 175: 189-194.
Schaap MG, Leij FJ and Van Genuchten MTh, 2001. A computer program for estimating soil hydraulic parameters with hierachical pedotransfer functions. Journal of Hydrology 251: 202-220.
Shrestha BM, Singh BR, Sitaula BK, Lai R and Barjacharya RM, 2007. Soil aggregate and particle-associated organic carbon under different land use in Nepal. Soil Sci Soc Am J 71: 1194-1203.
Tamari S, Wosten JHM and Ruiz-Suarez JC, 1996. Testing an artificial neural network for predicting soil hydraulic conductivity. Soil Sci Soc Am J 60: 1732-1741.
Water and Soil Science
Volume 21, Issue 3 - Serial Number 3
November 2011
Pages 153-162

Files

Share

How to cite

Statistics