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

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

Biomedical Engineering

Research output: Contribution to journal › Article › peer-review

55 Scopus citations

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 (10⁷) 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.

Original language	English (US)
Pages (from-to)	462-478
Number of pages	17
Journal	Neural Networks
Volume	20
Issue number	4
DOIs	https://doi.org/10.1016/j.neunet.2007.04.023
State	Published - Jan 1 2007

Funding

This work was supported by NSF grant OCI-0121475. The authors thank Joseph Zhang and Eric Wan for helpful discussions.

Funders	Funder number
Author National Science Foundation National Science Foundation National Institutes of Health National Institutes of Health National Institutes of Health National Institutes of Health National Science Foundation National Science Foundation	OCI-0121475

Keywords

Computational oceanography
Data assimilation
Fast neural network dynamic surrogates
High-dimensional time series prediction
Physics-based models
River estuary modelling

ASJC Scopus subject areas

Cognitive Neuroscience
Artificial Intelligence

Access to Document

10.1016/j.neunet.2007.04.023

Cite this

@article{54fa371495cb47c7a89e0df524f1ddad,

title = "Fast neural network surrogates for very high dimensional physics-based models in computational oceanography",

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 (107) 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.",

keywords = "Computational oceanography, Data assimilation, Fast neural network dynamic surrogates, High-dimensional time series prediction, Physics-based models, River estuary modelling",

author = "\{van der Merwe\}, Rudolph and Leen, \{Todd K.\} and Zhengdong Lu and Sergey Frolov and Baptista, \{Antonio M.\}",

note = "Funding Information: This work was supported by NSF grant OCI-0121475. The authors thank Joseph Zhang and Eric Wan for helpful discussions. ",

year = "2007",

month = jan,

day = "1",

doi = "10.1016/j.neunet.2007.04.023",

language = "English (US)",

volume = "20",

pages = "462--478",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Ltd",

number = "4",

}

TY - JOUR

T1 - Fast neural network surrogates for very high dimensional physics-based models in computational oceanography

AU - van der Merwe, Rudolph

AU - Leen, Todd K.

AU - Lu, Zhengdong

AU - Frolov, Sergey

AU - Baptista, Antonio M.

N1 - Funding Information: This work was supported by NSF grant OCI-0121475. The authors thank Joseph Zhang and Eric Wan for helpful discussions.

PY - 2007/1/1

Y1 - 2007/1/1

N2 - 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 (107) 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.

AB - 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 (107) 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.

KW - Computational oceanography

KW - Data assimilation

KW - Fast neural network dynamic surrogates

KW - High-dimensional time series prediction

KW - Physics-based models

KW - River estuary modelling

UR - https://www.scopus.com/pages/publications/34249876623

UR - https://www.scopus.com/pages/publications/34249876623#tab=citedBy

U2 - 10.1016/j.neunet.2007.04.023

DO - 10.1016/j.neunet.2007.04.023

M3 - Article

C2 - 17517493

AN - SCOPUS:34249876623

SN - 0893-6080

VL - 20

SP - 462

EP - 478

JO - Neural Networks

JF - Neural Networks

IS - 4

ER -

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

Abstract

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this