Improving Nowcasts of Road Surface Temperature by a Backpropagation Neural Network

J. Shao Vaisala TMI, Birmingham, United Kingdom

Search for other papers by J. Shao in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

Accurate numerical prediction of road ice is important in cutting winter road maintenance costs, reducing environmental damage from oversalting, and providing safer roads for road users. In this paper, the error of road surface temperature nowcasts (3 and 6 h ahead) by an automated numerical (Icebreak) model is regarded as a time series and is further treated by a three-layer neural network (NN) to increase accuracy of the nowcasts. The network is trained by an error-backpropagation algorithm with updated 7-day memory and progressively reduced learning rate. Its learning is based on historical and preliminary meteorological parameters measured at an automatic roadside weather station. The effectiveness of the network in improving the accuracy of numerical model forecasts was tested at both normal and problematic forecast sites in Austria, Italy, Japan, Norway, Switzerland, and England. Results of the tests show that the NN technique is able to reduce root-mean-square error of temperature forecasts and increase the accuracy of frost–ice prediction. The improvements were minor at normal sites but significant at problematic sites where complex environmental conditions and underlying nonlinear mechanisms are currently unresolvable by operational numerical models.

Corresponding author address: Dr. Jianmin Shao, Vaisala Thermal Mapping International (TMI) Limited, Vaisala House, 349 Bristol Road, Birmingham B5 7SW, United Kingdom.

Email: jianmin.shao@vaisala.com

Abstract

Accurate numerical prediction of road ice is important in cutting winter road maintenance costs, reducing environmental damage from oversalting, and providing safer roads for road users. In this paper, the error of road surface temperature nowcasts (3 and 6 h ahead) by an automated numerical (Icebreak) model is regarded as a time series and is further treated by a three-layer neural network (NN) to increase accuracy of the nowcasts. The network is trained by an error-backpropagation algorithm with updated 7-day memory and progressively reduced learning rate. Its learning is based on historical and preliminary meteorological parameters measured at an automatic roadside weather station. The effectiveness of the network in improving the accuracy of numerical model forecasts was tested at both normal and problematic forecast sites in Austria, Italy, Japan, Norway, Switzerland, and England. Results of the tests show that the NN technique is able to reduce root-mean-square error of temperature forecasts and increase the accuracy of frost–ice prediction. The improvements were minor at normal sites but significant at problematic sites where complex environmental conditions and underlying nonlinear mechanisms are currently unresolvable by operational numerical models.

Corresponding author address: Dr. Jianmin Shao, Vaisala Thermal Mapping International (TMI) Limited, Vaisala House, 349 Bristol Road, Birmingham B5 7SW, United Kingdom.

Email: jianmin.shao@vaisala.com

Save
  • Aussem, A., F. Murtagh, and M. Sarazin, 1995: Dynamical recurrent neural networks—Towards environmental time series prediction. Int. J. Neural Sys.,6, 145–170.

    • Crossref
    • Export Citation
  • Bankert, R. L., 1994: Cloud classification of AVHRR imagery in maritime regions using a probabilistic neural network. J. Appl. Meteor.,33, 909–918.

    • Crossref
    • Export Citation
  • ——, and D. W. Aha, 1996: Improvement to a neural network cloud classifier. J. Appl. Meteor.,35, 2036–2039.

    • Crossref
    • Export Citation
  • Beale, R., and T. Jackson, 1994: Neural Computing: An Introduction. Institute of Physics Publishing, 240 pp.

  • Boznar, M., M. Lesjak, and P. Mlakar, 1993: A neural network-based method for short-term predictions of ambient SO2 concentrations in highly polluted industrial areas of complex terrain. Atmos. Environ.,27B, 221–230.

    • Crossref
    • Export Citation
  • Eckert, P., D. Cattani, and J. Ambühl, 1996: Classification of ensemble forecasts by means of an artificial neural network. Meteor. Appl.,3, 169–178.

    • Crossref
    • Export Citation
  • Fausett, L., 1994: Fundamentals of Neural Networks. Prentice-Hall, 461 pp.

  • Gallant, S. I., 1993: Neural Network Learning and Expert Systems. The MIT Press, 365 pp.

    • Crossref
    • Export Citation
  • Jin, L., M. M. Gupta, and P. N. Nikiforuk, 1994: Computational neural architectures for control applications. Soft Computing: Fuzzy Logic, Neural Networks, and Distributed Artificial Intelligence, F. Aminzadeh and M. Jamshidi, Ed., Prentice-Hall, 121–152.

  • Marzban, C., and G. J. Stumpf, 1996: A neural network for tornado prediction based on Doppler radar-derived attributes. J. Appl. Meteor.,35, 617–626.

    • Crossref
    • Export Citation
  • McCann, D. W., 1992: A neural network short-term forecast of significant thunderstorms. Wea. Forecasting,7, 525–534.

    • Crossref
    • Export Citation
  • Shao, J., and P. J. Lister, 1996: An automated nowcasting model of road surface temperature and state for winter road maintenance. J. Appl. Meteor.,35, 1352–1361.

    • Crossref
    • Export Citation
  • Stephenson, T. E., 1988: Wisconsin’s winter weather system. Proc. Fourth Int. Conf. on Weather and Road Safety, Florence, Italy, Accademia Dei Georgofili, 61–101.

  • Thornes, J. E., 1994: Salt of the Earth. Surveyor, 8 December, 16–18.

  • ——, and J. Shao, 1991: Spectral analysis and sensitivity tests for a numerical road surface temperature prediction model. Meteor. Mag.,120, 117–124.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 510 183 8
PDF Downloads 264 92 12