Stochastic superparameterization and multiscale filtering of turbulent tracers

Yoonsang Lee, Andrew J. Majda, Di Qi

Research output: Contribution to journalArticle

Abstract

Data assimilation or filtering combines a numerical forecast model and observations to provide accurate statistical estimation of the state of interest. In this paper we are concerned with accurate data assimilation of a sparsely observed passive tracer advected in turbulent flows using a reduced-order forecast model. The turbulent flows which contain anisotropic and inhomogeneous structures such as jets are typical in geophysical turbulent flows in atmosphere and ocean science, and passive tracers with a mean gradient can exhibit anisotropic transport with intermittent extreme events, as shown below. Stochastic superparameterization, which is a seamless multiscale method developed for large-scale models of atmosphere and ocean models without scale-gap between the resolved and unresolved scales, generates large-scale turbulent velocity fields using a significantly smaller degree of freedom compared to a direct fine resolution numerical simulation. In a large-scale model of the tracer transport, the tracer is advected by the large-scale velocity field generated by superparameterization with a parameterization of eddies, an additional eddy diffusion given by an anisotropic biharmonic diffusion. The turbulent tracer is sparsely observed in space in only the upper surface layer. These observations naturally mix the resolved and unresolved scales, and so we develop an ensemble multiscale data assimilation algorithm which provides estimates of the resolved scales using the mixed observations. The reduced-order model is 200 times cheaper than the fine resolution solution and thus enables us to increase the number of ensemble members for accurate predictions of state distributions. Numerical experiments show positive results in the estimation of the resolved scales of the tracer as well as in capturing anisotropic intermittent extreme events for the unresolved portions of the tracer field.

Original languageEnglish (US)
Pages (from-to)215-234
Number of pages20
JournalMultiscale Modeling and Simulation
Volume15
Issue number1
DOIs
StatePublished - 2017

Fingerprint

tracers
Filtering
tracer
Data Assimilation
Turbulent Flow
Turbulent flow
Extreme Events
assimilation
turbulent flow
data assimilation
Ocean
Velocity Field
Atmosphere
Forecast
Ensemble
scale models
extreme event
Geophysical Flows
forecasting
Statistical Estimation

Keywords

  • Data assimilation
  • Multiscale
  • Stochastic superparameterization
  • Turbulent tracer

ASJC Scopus subject areas

  • Chemistry(all)
  • Modeling and Simulation
  • Ecological Modeling
  • Physics and Astronomy(all)
  • Computer Science Applications

Cite this

Stochastic superparameterization and multiscale filtering of turbulent tracers. / Lee, Yoonsang; Majda, Andrew J.; Qi, Di.

In: Multiscale Modeling and Simulation, Vol. 15, No. 1, 2017, p. 215-234.

Research output: Contribution to journalArticle

@article{04cb9f359cbb43d1b4eada4c44fcabdd,
title = "Stochastic superparameterization and multiscale filtering of turbulent tracers",
abstract = "Data assimilation or filtering combines a numerical forecast model and observations to provide accurate statistical estimation of the state of interest. In this paper we are concerned with accurate data assimilation of a sparsely observed passive tracer advected in turbulent flows using a reduced-order forecast model. The turbulent flows which contain anisotropic and inhomogeneous structures such as jets are typical in geophysical turbulent flows in atmosphere and ocean science, and passive tracers with a mean gradient can exhibit anisotropic transport with intermittent extreme events, as shown below. Stochastic superparameterization, which is a seamless multiscale method developed for large-scale models of atmosphere and ocean models without scale-gap between the resolved and unresolved scales, generates large-scale turbulent velocity fields using a significantly smaller degree of freedom compared to a direct fine resolution numerical simulation. In a large-scale model of the tracer transport, the tracer is advected by the large-scale velocity field generated by superparameterization with a parameterization of eddies, an additional eddy diffusion given by an anisotropic biharmonic diffusion. The turbulent tracer is sparsely observed in space in only the upper surface layer. These observations naturally mix the resolved and unresolved scales, and so we develop an ensemble multiscale data assimilation algorithm which provides estimates of the resolved scales using the mixed observations. The reduced-order model is 200 times cheaper than the fine resolution solution and thus enables us to increase the number of ensemble members for accurate predictions of state distributions. Numerical experiments show positive results in the estimation of the resolved scales of the tracer as well as in capturing anisotropic intermittent extreme events for the unresolved portions of the tracer field.",
keywords = "Data assimilation, Multiscale, Stochastic superparameterization, Turbulent tracer",
author = "Yoonsang Lee and Majda, {Andrew J.} and Di Qi",
year = "2017",
doi = "10.1137/16M1080239",
language = "English (US)",
volume = "15",
pages = "215--234",
journal = "Multiscale Modeling and Simulation",
issn = "1540-3459",
publisher = "Society for Industrial and Applied Mathematics Publications",
number = "1",

}

TY - JOUR

T1 - Stochastic superparameterization and multiscale filtering of turbulent tracers

AU - Lee, Yoonsang

AU - Majda, Andrew J.

AU - Qi, Di

PY - 2017

Y1 - 2017

N2 - Data assimilation or filtering combines a numerical forecast model and observations to provide accurate statistical estimation of the state of interest. In this paper we are concerned with accurate data assimilation of a sparsely observed passive tracer advected in turbulent flows using a reduced-order forecast model. The turbulent flows which contain anisotropic and inhomogeneous structures such as jets are typical in geophysical turbulent flows in atmosphere and ocean science, and passive tracers with a mean gradient can exhibit anisotropic transport with intermittent extreme events, as shown below. Stochastic superparameterization, which is a seamless multiscale method developed for large-scale models of atmosphere and ocean models without scale-gap between the resolved and unresolved scales, generates large-scale turbulent velocity fields using a significantly smaller degree of freedom compared to a direct fine resolution numerical simulation. In a large-scale model of the tracer transport, the tracer is advected by the large-scale velocity field generated by superparameterization with a parameterization of eddies, an additional eddy diffusion given by an anisotropic biharmonic diffusion. The turbulent tracer is sparsely observed in space in only the upper surface layer. These observations naturally mix the resolved and unresolved scales, and so we develop an ensemble multiscale data assimilation algorithm which provides estimates of the resolved scales using the mixed observations. The reduced-order model is 200 times cheaper than the fine resolution solution and thus enables us to increase the number of ensemble members for accurate predictions of state distributions. Numerical experiments show positive results in the estimation of the resolved scales of the tracer as well as in capturing anisotropic intermittent extreme events for the unresolved portions of the tracer field.

AB - Data assimilation or filtering combines a numerical forecast model and observations to provide accurate statistical estimation of the state of interest. In this paper we are concerned with accurate data assimilation of a sparsely observed passive tracer advected in turbulent flows using a reduced-order forecast model. The turbulent flows which contain anisotropic and inhomogeneous structures such as jets are typical in geophysical turbulent flows in atmosphere and ocean science, and passive tracers with a mean gradient can exhibit anisotropic transport with intermittent extreme events, as shown below. Stochastic superparameterization, which is a seamless multiscale method developed for large-scale models of atmosphere and ocean models without scale-gap between the resolved and unresolved scales, generates large-scale turbulent velocity fields using a significantly smaller degree of freedom compared to a direct fine resolution numerical simulation. In a large-scale model of the tracer transport, the tracer is advected by the large-scale velocity field generated by superparameterization with a parameterization of eddies, an additional eddy diffusion given by an anisotropic biharmonic diffusion. The turbulent tracer is sparsely observed in space in only the upper surface layer. These observations naturally mix the resolved and unresolved scales, and so we develop an ensemble multiscale data assimilation algorithm which provides estimates of the resolved scales using the mixed observations. The reduced-order model is 200 times cheaper than the fine resolution solution and thus enables us to increase the number of ensemble members for accurate predictions of state distributions. Numerical experiments show positive results in the estimation of the resolved scales of the tracer as well as in capturing anisotropic intermittent extreme events for the unresolved portions of the tracer field.

KW - Data assimilation

KW - Multiscale

KW - Stochastic superparameterization

KW - Turbulent tracer

UR - http://www.scopus.com/inward/record.url?scp=85017026957&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017026957&partnerID=8YFLogxK

U2 - 10.1137/16M1080239

DO - 10.1137/16M1080239

M3 - Article

AN - SCOPUS:85017026957

VL - 15

SP - 215

EP - 234

JO - Multiscale Modeling and Simulation

JF - Multiscale Modeling and Simulation

SN - 1540-3459

IS - 1

ER -