Resolution and error propagation analysis for tomographic data with correlated errors

Lars Nielsen; Bo Holm Jacobsen

doi:10.1007/BFb0011770

Resolution and error propagation analysis for tomographic data with correlated errors

Lars Nielsen
, Bo Holm Jacobsen

Research output: Chapter in Book/Report/Conference proceeding › Chapter in book › Research › peer-review

Abstract

Tomographic data are likely to be contaminated with correlated errors. We use an approach based on data error covariance specification in order to analyse the effect of such errors. This approach is valid for arbitrary irregular measurement geometries and therefore offers great flexibility.

We have shown that the estimates in (1) and (2) behave differently in the presence of correlated errors. Taking the correct error correlations into account we obtain a more realistic description of error propagation, and we produce better model resolution with less smearing.

Original language	English
Title of host publication	Inverse methods
Subtitle of host publication	Interdisciplinary elements of methodology, computation, and applications
Editors	Bo Holm Jacobsen, Klaus Mosegaard, Paolo Sibani
Publisher	Springer
Pages	131–138
Number of pages	8
ISBN (Print)	978-3-540-61693-1
DOIs	https://doi.org/10.1007/BFb0011770
Publication status	Published - 1996
Externally published	Yes

Publication series

Series	Lecture Notes in Earth Sciences
Volume	63
ISSN	0930-0317

Programme Area

Programme Area 3: Energy Resources

Access to Document

10.1007/BFb0011770

Cite this

@inbook{fe30132888614d0c9aaf9cd15bbe5b8e,

title = "Resolution and error propagation analysis for tomographic data with correlated errors",

abstract = "Tomographic data are likely to be contaminated with correlated errors. We use an approach based on data error covariance specification in order to analyse the effect of such errors. This approach is valid for arbitrary irregular measurement geometries and therefore offers great flexibility. We have shown that the estimates in (1) and (2) behave differently in the presence of correlated errors. Taking the correct error correlations into account we obtain a more realistic description of error propagation, and we produce better model resolution with less smearing.",

author = "Lars Nielsen and Jacobsen, \{Bo Holm\}",

year = "1996",

doi = "10.1007/BFb0011770",

language = "English",

isbn = "978-3-540-61693-1",

series = "Lecture Notes in Earth Sciences",

publisher = "Springer",

pages = "131–138",

editor = "Jacobsen, \{Bo Holm\} and Klaus Mosegaard and Paolo Sibani",

booktitle = "Inverse methods",

address = "China",

}

Resolution and error propagation analysis for tomographic data with correlated errors. / Nielsen, Lars; Jacobsen, Bo Holm.
Inverse methods: Interdisciplinary elements of methodology, computation, and applications. ed. / Bo Holm Jacobsen; Klaus Mosegaard; Paolo Sibani. Springer, 1996. p. 131–138 (Lecture Notes in Earth Sciences, Vol. 63).

Research output: Chapter in Book/Report/Conference proceeding › Chapter in book › Research › peer-review

TY - CHAP

T1 - Resolution and error propagation analysis for tomographic data with correlated errors

AU - Nielsen, Lars

AU - Jacobsen, Bo Holm

PY - 1996

Y1 - 1996

N2 - Tomographic data are likely to be contaminated with correlated errors. We use an approach based on data error covariance specification in order to analyse the effect of such errors. This approach is valid for arbitrary irregular measurement geometries and therefore offers great flexibility. We have shown that the estimates in (1) and (2) behave differently in the presence of correlated errors. Taking the correct error correlations into account we obtain a more realistic description of error propagation, and we produce better model resolution with less smearing.

AB - Tomographic data are likely to be contaminated with correlated errors. We use an approach based on data error covariance specification in order to analyse the effect of such errors. This approach is valid for arbitrary irregular measurement geometries and therefore offers great flexibility. We have shown that the estimates in (1) and (2) behave differently in the presence of correlated errors. Taking the correct error correlations into account we obtain a more realistic description of error propagation, and we produce better model resolution with less smearing.

U2 - 10.1007/BFb0011770

DO - 10.1007/BFb0011770

M3 - Chapter in book

SN - 978-3-540-61693-1

T3 - Lecture Notes in Earth Sciences

SP - 131

EP - 138

BT - Inverse methods

A2 - Jacobsen, Bo Holm

A2 - Mosegaard, Klaus

A2 - Sibani, Paolo

PB - Springer

ER -

Resolution and error propagation analysis for tomographic data with correlated errors

Abstract

Publication series

Programme Area

Access to Document

Fingerprint

Cite this