Application of Bayesian Regularized BP Neural Network Model for Trend Analysis, Acidity and Chemical Composition of Precipitation in North Carolina
Rent the article at a discountRent now
* Final gross prices may vary according to local VAT.Get Access
Bayesian regularized back-propagation neural network (BRBPNN) was developed for trend analysis, acidity and chemical composition of precipitation in North Carolina using precipitation chemistry data in NADP. This study included two BRBPNN application problems: (i) the relationship between precipitation acidity (pH) and other ions (NH4 +, NO3 −, SO4 2−, Ca2+, Mg2+, K+, Cl− and Na+) was performed by BRBPNN and the achieved optimal network structure was 8-15-1. Then the relative importance index, obtained through the sum of square weights between each input neuron and the hidden layer of BRBPNN(8-15-1), indicated that the ions' contribution to the acidity declined in the order of NH4 + > SO4 2− > NO3 −; and (ii) investigations were also carried out using BRBPNN with respect to temporal variation of monthly mean NH4 +, SO4 2− and NO3 − concentrations and their optimal architectures for the 1990–2003 data were 4-6-1, 4-6-1 and 4-4-1, respectively. All the estimated results of the optimal BRBPNNs showed that the relationship between the acidity and other ions or that between NH4 +, SO4 2−, NO3 − concentrations with regard to precipitation amount and time variable was obviously nonlinear, since in contrast to multiple linear regression (MLR), BRBPNN was clearly better with less error in prediction and of higher correlation coefficients. Meanwhile, results also exhibited that BRBPNN was of automated regularization parameter selection capability and may ensure the excellent fitting and robustness. Thus, this study laid the foundation for the application of BRBPNN in the analysis of acid precipitation.
- Aherne, J. and Farrell, E. P.: 2002, ‘Deposition of sulphur, nitrogen and acidity in precipitation over Ireland: chemistry, spatial distribution and long-term trends’, Atmos. Environ. 36, 1379–1389. CrossRef
- Archontoula, C., Michaela, S. and Nikolas, S.: 2003, ‘omparative assessment of neural networks and regression models for forecasting summertime ozone in Athens’, The Science of the Total Environment 313, 1–13. CrossRef
- Baez, A. P., Belmont, R. D. and Padilla, H. G.: 1996, ‘hemical composition of precipitation at two sampling sites in Mexico: a 7-year study’, Atmos. Environ. 31, 915–925. CrossRef
- Buishand, T. A., Kempen, G. T., Frantzen, A. J., Reijnders, H. F. R. and Eshof, A. J.: 1988, ‘Trend and seasonal variation of precipitation chemistry data in the Netherlands’, Atmos. Environ. 22, 339–348. CrossRef
- Burden, F. R. and Winkler, D. Q.: 1999, ‘Robust QSAR models using Bayesian regularized neural networks’, J. Med. Chem. 42, 3183–3187. CrossRef
- Burden, F. R. and Winkler, D. A.: 2000, ‘A quantitative structure-activity relationships model for the acute toxicity of substituted benzenes to tetrahymena pyriformis Using Bayesian-regularized neural networks’, Chem. Res. Toxicol. 13, 436–440. CrossRef
- Christopher, M. B. L., Van, C. B. and Susan, M.: 2005, ‘Larson spatial and temporal trends of precipitation chemistry in the United States, 1985–2002’, Environ. Pollut. 135, 347–361. CrossRef
- Dana, M. T. and East, R. C.: 1987, ‘Statistical summary and analyses of event precipitation chemistry from the MAP3S network, 1976–1983’, Atmos. Environ. 21, 113–128. CrossRef
- Foresee, F. D. and Hagan, M. T.: 1997, ‘Gauss-Newton approximation to Bayesian regularization’, Proceedings of the 1997 International Joint Conference on Neural Networks, Houston, 1930–1935.
- George, M. H.: 2003, ‘Snowpack and precipitation chemistry at high altitudes’, Atmos. Environ. 37,1231–1242. CrossRef
- Grimma, J. W. and Lynch, J. A.: 2005, ‘Improved daily precipitation nitrate and ammonium concentration models for the Chesapeake Bay Watershed’, Environ. Pollut. 135, 445–455. CrossRef
- Husmeier, D., Penny, W. D. and Roberts, S. J.: 1999, ‘An empirical evaluation of Bayesian sampling with Monte Carlo for training neural net classifiers’, Neural Networks 12, 677–705. CrossRef
- Joseph, H. W. L., Huang, Y. and Dickman, M.: 2003, ‘Neural network modelling of coastal algal blooms’, Ecol. Model. 159, 179–201. CrossRef
- Jouko, L. and Aki, V.: 2001, ‘Bayesian approach for neural networks–review and case studies’, Neural Networks 14, 257–274. CrossRef
- John, T. W., Viney, P. A. and David, A. D.: 2000, ‘Atmospheric transport and wet deposition of ammonium in North Carolina’, Atmos. Environ. 34, 3407–3418. CrossRef
- Khawaja, H. A. and Husain, L.: 1990, ‘hemical characterization of acid precipitation in Albany, New York’, Atmos. Environ. 24A, 1869–1882.
- Li, Z. G.: 1999, ‘Analysis of the acidity and chemical characteristics of precipitation in Zhejiang Province China’, Environ. Science 19, 436–440 (in Chinese).
- Li, Z. Y.: 1999, ‘Relation between ion concentration and pH in precipitation in number of Chinese cities’, Acta Scientiae Circumstantize 19, 303–306.
- Lim, B., Jickells, T. D. and Davies, T. D.: 1991, ‘Sequential sampling of particles major ions and total trace metals in wet deposition’, Atmos. Environ. 25A, 745–762.
- MacKay, D. J. C.: 1992, ‘A practical Bayesian framework for back-propagation networks’, Neural Comput. 4, 415–447. CrossRef
- Mackay, D. J. C.: 1995, ‘Probable networks and plausible predictions: a review of practical Bayesian methods for supervised neural networks’, Comput. Neural Syst. 6, 469–505. CrossRef
- MAP3S/RAINE Research Community: 1982, ‘The MAP3S/RAINE precipitation chemistry network: Statistical overview for the period 1977–1980’, Atmos. Environ. 16, 1603–1631. CrossRef
- Migliavacca, D., Teixeira, E. C. and Pires, M.: 2004, ‘Study of chemical elements in atmospheric precipitation in South Brazil’, Atmos. Environ. 38, 1641–1656. CrossRef
- Page, T., Whyatt, J. D., Beven, K. J. and Metcalfe, S. E.: 2004, ‘Uncertainty in modeled estimates of acid deposition across Wales: a GLUE approach’, Atmos. Environ. 38, 2079–2090. CrossRef
- Yu, S. C., Gao, C. T. and Cheng, Z. M.: 1998, ‘An analysis of chemical composition of different rain types in ‘Minnan Golden Triangle’ region in the southeastern coast of China’, Atmos. Res. 47, 245–269. CrossRef
- Sinya, S., Akira, N. and Izumi, N.: 2002, ‘Annual and seasonal trends in chemical composition of precipitation in Japan during 1989–1998’, Atmos. Environ. 36, 3505–3517. CrossRef
- Sovan, L., Marc, D., Philippe, B., Ioannis, D., Jacques, L. and Stephane, A.: 1996, ‘Application of neural networks to modelling nonlinear relationships in ecology’, Ecol. Model. 90, 39–52. CrossRef
- Sun, W., Zeng, G.M., Wei, W.Z. and Huang G.H.: 2005, ‘Bayesian regularized BP neural network model for quantitative relationship between the electrochemical reduction potential and molecular structures of chlorinated aromatic compounds’, Environ. Science 26, 21–27 (in Chinese).
- The MathWorks Inc. http://www.mathworks.com/access/helpdesk/help/pdf_doc/nnet/nnet.pdf, Natick, MA, USA.
- Yasushi, N., Kei S., Keiichi, H., Tamami, I., Shigeru, T., Yukiko, D., Morikazu, H. and Kazuhiko, H.: 2001, ‘Long term trend of chemical constituents in Tokyo metropolitan area in Japan’, Water Air Soil Pollut. 130, 1649–1554. CrossRef
- Zeng, Y. and Flopke, P. K.: 1989, ‘A study of the sources of acid precipitation in Ontario, Canada’, Atmos. Environ. 23, 1499–1509. CrossRef
- Application of Bayesian Regularized BP Neural Network Model for Trend Analysis, Acidity and Chemical Composition of Precipitation in North Carolina
Water, Air, and Soil Pollution
Volume 172, Issue 1-4 , pp 167-184
- Cover Date
- Print ISSN
- Online ISSN
- Springer Netherlands
- Additional Links
- Bayesian regularized back-propagation neural network (BRBPNN)
- chemical composition
- temporal trend
- the sum of square weights
- Industry Sectors
- Author Affiliations
- 1. College of Environmental Science and Engineering, Hunan University, Changsha, 410082, China
- 2. Sino-Canadian Center of Energy and Environment Research, University of Regina, Regina, SK, S4S 0A2, Canada