Paper No. 96013 of the Journal of the American Water Resources Association (formerly Water Resources Bulletin). Discussions are open until December 1, 1997.
PREDICTION OF TWO-YEAR PEAK STREAM-DISCHARGES USING NEURAL NETWORKS1
Article first published online: 8 JUN 2007
JAWRA Journal of the American Water Resources Association
Volume 33, Issue 3, pages 625–630, June 1997
How to Cite
Muttiah, R. S., Srinivasan, R. and Allen, P. M. (1997), PREDICTION OF TWO-YEAR PEAK STREAM-DISCHARGES USING NEURAL NETWORKS. JAWRA Journal of the American Water Resources Association, 33: 625–630. doi: 10.1111/j.1752-1688.1997.tb03537.x
- Issue published online: 8 JUN 2007
- Article first published online: 8 JUN 2007
- neural networks;
- two year discharge;
- machine learning;
- regional scale;
- local scale predictions
ABSTRACT: The cascade correlation neural network was used to predict the two-year peak discharge (Q2) for major regional river basins of the continental United States (US). Watersheds ranged in size by four orders of magnitude. Results of the neural network predictions ranged from correlations of 0.73 for 104 test data in the Souris-Red Rainy river basin to 0.95 for 141 test data in California. These results are improvements over previous multilinear regressions involving more variables that showed correlations ranging from 0.26 to 0.94. Results are presented for neural networks trained and tested on drainage area, average annual precipitation, and mean basin elevation. A neural network trained on regional scale data in the Texas Gulf was comparable to previous estimates of Q2 by regression. Our research shows Q2 was difficult to predict for the Souris-Red Rainy, Missouri, and Rio Grande river basins compared to the rest of the US, and acceptable predictions could be made using only mean basin elevation and drainage areas of watersheds.