Fast neural network surrogates for very high dimensional physics-based models in computational oceanography

Rudolph van der Merwe; Todd K Leen; Zhengdong Lu; Sergey Frolov; Antonio M Baptista

doi:10.1016/j.neunet.2007.04.023

Fast neural network surrogates for very high dimensional physics-based models in computational oceanography

Neural Netw. 2007 May;20(4):462-78. doi: 10.1016/j.neunet.2007.04.023. Epub 2007 May 3.

Authors

Rudolph van der Merwe¹, Todd K Leen, Zhengdong Lu, Sergey Frolov, Antonio M Baptista

Affiliation

¹ Department of Computer Science and Electrical Engineering, OGI School of Science and Engineering, Oregon Health and Science University, Portland, OR 97006, USA.

PMID: 17517493
DOI: 10.1016/j.neunet.2007.04.023

Abstract

We present neural network surrogates that provide extremely fast and accurate emulation of a large-scale circulation model for the coupled Columbia River, its estuary and near ocean regions. The circulation model has O(10(7)) degrees of freedom, is highly nonlinear and is driven by ocean, atmospheric and river influences at its boundaries. The surrogates provide accurate emulation of the full circulation code and run over 1000 times faster. Such fast dynamic surrogates will enable significant advances in ensemble forecasts in oceanography and weather.

Publication types

Research Support, U.S. Gov't, Non-P.H.S.

MeSH terms

Artificial Intelligence*
Computer Simulation*
Feedback
Neural Networks, Computer*
Nonlinear Dynamics
Oceanography*
Physical Phenomena
Physics*
Rivers
Spectrum Analysis
Time Factors
Weather