The adjoint method applied to time-distance helioseismology

Shravan Hanasoge, Aaron Birch, Laurent Gizon, Jeroen Tromp

Research output: Contribution to journalArticle

Abstract

For a given misfit function, a specified optimality measure of a model, its gradient describes the manner in which one may alter properties of the system to march toward a stationary point. The adjoint method, arising from partial-differential-equation-constrained optimization, describes a means of extracting derivatives of a misfit function with respect to model parameters through finite computation. It relies on the accurate calculation of wavefields that are driven by two types of sources, namely, the average wave-excitation spectrum, resulting in the forward wavefield, and differences between predictions and observations, resulting in an adjoint wavefield. All sensitivity kernels relevant to a given measurement emerge directly from the evaluation of an interaction integral involving these wavefields. The technique facilitates computation of sensitivity kernels (Fréchet derivatives) relative to three-dimensional heterogeneous background models, thereby paving the way for nonlinear iterative inversions. An algorithm to perform such inversions using as many observations as desired is discussed.

Original languageEnglish (US)
Article number100
JournalAstrophysical Journal
Volume738
Issue number1
DOIs
StatePublished - Sep 1 2011

Fingerprint

adjoint method
helioseismology
inversions
sensitivity
wave spectrum
wave excitation
partial differential equations
gradients
optimization
evaluation
prediction
predictions
interactions
inversion

Keywords

  • magnetohydrodynamics (MHD)
  • Sun: dynamo
  • Sun: helioseismology
  • Sun: interior
  • Sun: oscillations
  • waves

ASJC Scopus subject areas

  • Space and Planetary Science
  • Astronomy and Astrophysics
  • Nuclear and High Energy Physics

Cite this

The adjoint method applied to time-distance helioseismology. / Hanasoge, Shravan; Birch, Aaron; Gizon, Laurent; Tromp, Jeroen.

In: Astrophysical Journal, Vol. 738, No. 1, 100, 01.09.2011.

Research output: Contribution to journalArticle

@article{abaf067c0df843cc971cedc44de1f84c,
title = "The adjoint method applied to time-distance helioseismology",
abstract = "For a given misfit function, a specified optimality measure of a model, its gradient describes the manner in which one may alter properties of the system to march toward a stationary point. The adjoint method, arising from partial-differential-equation-constrained optimization, describes a means of extracting derivatives of a misfit function with respect to model parameters through finite computation. It relies on the accurate calculation of wavefields that are driven by two types of sources, namely, the average wave-excitation spectrum, resulting in the forward wavefield, and differences between predictions and observations, resulting in an adjoint wavefield. All sensitivity kernels relevant to a given measurement emerge directly from the evaluation of an interaction integral involving these wavefields. The technique facilitates computation of sensitivity kernels (Fr{\'e}chet derivatives) relative to three-dimensional heterogeneous background models, thereby paving the way for nonlinear iterative inversions. An algorithm to perform such inversions using as many observations as desired is discussed.",
keywords = "magnetohydrodynamics (MHD), Sun: dynamo, Sun: helioseismology, Sun: interior, Sun: oscillations, waves",
author = "Shravan Hanasoge and Aaron Birch and Laurent Gizon and Jeroen Tromp",
year = "2011",
month = "9",
day = "1",
doi = "10.1088/0004-637X/738/1/100",
language = "English (US)",
volume = "738",
journal = "Astrophysical Journal",
issn = "0004-637X",
publisher = "IOP Publishing Ltd.",
number = "1",

}

TY - JOUR

T1 - The adjoint method applied to time-distance helioseismology

AU - Hanasoge, Shravan

AU - Birch, Aaron

AU - Gizon, Laurent

AU - Tromp, Jeroen

PY - 2011/9/1

Y1 - 2011/9/1

N2 - For a given misfit function, a specified optimality measure of a model, its gradient describes the manner in which one may alter properties of the system to march toward a stationary point. The adjoint method, arising from partial-differential-equation-constrained optimization, describes a means of extracting derivatives of a misfit function with respect to model parameters through finite computation. It relies on the accurate calculation of wavefields that are driven by two types of sources, namely, the average wave-excitation spectrum, resulting in the forward wavefield, and differences between predictions and observations, resulting in an adjoint wavefield. All sensitivity kernels relevant to a given measurement emerge directly from the evaluation of an interaction integral involving these wavefields. The technique facilitates computation of sensitivity kernels (Fréchet derivatives) relative to three-dimensional heterogeneous background models, thereby paving the way for nonlinear iterative inversions. An algorithm to perform such inversions using as many observations as desired is discussed.

AB - For a given misfit function, a specified optimality measure of a model, its gradient describes the manner in which one may alter properties of the system to march toward a stationary point. The adjoint method, arising from partial-differential-equation-constrained optimization, describes a means of extracting derivatives of a misfit function with respect to model parameters through finite computation. It relies on the accurate calculation of wavefields that are driven by two types of sources, namely, the average wave-excitation spectrum, resulting in the forward wavefield, and differences between predictions and observations, resulting in an adjoint wavefield. All sensitivity kernels relevant to a given measurement emerge directly from the evaluation of an interaction integral involving these wavefields. The technique facilitates computation of sensitivity kernels (Fréchet derivatives) relative to three-dimensional heterogeneous background models, thereby paving the way for nonlinear iterative inversions. An algorithm to perform such inversions using as many observations as desired is discussed.

KW - magnetohydrodynamics (MHD)

KW - Sun: dynamo

KW - Sun: helioseismology

KW - Sun: interior

KW - Sun: oscillations

KW - waves

UR - http://www.scopus.com/inward/record.url?scp=80052783059&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=80052783059&partnerID=8YFLogxK

U2 - 10.1088/0004-637X/738/1/100

DO - 10.1088/0004-637X/738/1/100

M3 - Article

AN - SCOPUS:80052783059

VL - 738

JO - Astrophysical Journal

JF - Astrophysical Journal

SN - 0004-637X

IS - 1

M1 - 100

ER -