Eintrag weiter verarbeiten
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success
Gespeichert in:
Zeitschriftentitel: | Geoscientific Model Development |
---|---|
Personen und Körperschaften: | , , , , , , |
In: | Geoscientific Model Development, 12, 2019, 7, S. 2941-2960 |
Format: | E-Article |
Sprache: | Englisch |
veröffentlicht: |
Copernicus GmbH
|
Schlagwörter: |
author_facet |
Scalzo, Richard Kohn, David Olierook, Hugo Houseman, Gregory Chandra, Rohitash Girolami, Mark Cripps, Sally Scalzo, Richard Kohn, David Olierook, Hugo Houseman, Gregory Chandra, Rohitash Girolami, Mark Cripps, Sally |
---|---|
author |
Scalzo, Richard Kohn, David Olierook, Hugo Houseman, Gregory Chandra, Rohitash Girolami, Mark Cripps, Sally |
spellingShingle |
Scalzo, Richard Kohn, David Olierook, Hugo Houseman, Gregory Chandra, Rohitash Girolami, Mark Cripps, Sally Geoscientific Model Development Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success Polymers and Plastics General Environmental Science |
author_sort |
scalzo, richard |
spelling |
Scalzo, Richard Kohn, David Olierook, Hugo Houseman, Gregory Chandra, Rohitash Girolami, Mark Cripps, Sally 1991-9603 Copernicus GmbH Polymers and Plastics General Environmental Science http://dx.doi.org/10.5194/gmd-12-2941-2019 <jats:p>Abstract. The rigorous quantification of uncertainty in geophysical inversions is a challenging problem. Inversions are often ill-posed and the likelihood surface may be multi-modal; properties of any single mode become inadequate uncertainty measures, and sampling methods become inefficient for irregular posteriors or high-dimensional parameter spaces. We explore the influences of different choices made by the practitioner on the efficiency and accuracy of Bayesian geophysical inversion methods that rely on Markov chain Monte Carlo sampling to assess uncertainty using a multi-sensor inversion of the three-dimensional structure and composition of a region in the Cooper Basin of South Australia as a case study. The inversion is performed using an updated version of the Obsidian distributed inversion software. We find that the posterior for this inversion has a complex local covariance structure, hindering the efficiency of adaptive sampling methods that adjust the proposal based on the chain history. Within the context of a parallel-tempered Markov chain Monte Carlo scheme for exploring high-dimensional multi-modal posteriors, a preconditioned Crank–Nicolson proposal outperforms more conventional forms of random walk. Aspects of the problem setup, such as priors on petrophysics and on 3-D geological structure, affect the shape and separation of posterior modes, influencing sampling performance as well as the inversion results. The use of uninformative priors on sensor noise enables optimal weighting among multiple sensors even if noise levels are uncertain. </jats:p> Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success Geoscientific Model Development |
doi_str_mv |
10.5194/gmd-12-2941-2019 |
facet_avail |
Online Free |
finc_class_facet |
Chemie und Pharmazie |
format |
ElectronicArticle |
fullrecord |
blob:ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuNTE5NC9nbWQtMTItMjk0MS0yMDE5 |
id |
ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuNTE5NC9nbWQtMTItMjk0MS0yMDE5 |
institution |
DE-D275 DE-Bn3 DE-Brt1 DE-Zwi2 DE-D161 DE-Gla1 DE-Zi4 DE-15 DE-Pl11 DE-Rs1 DE-105 DE-14 DE-Ch1 DE-L229 |
imprint |
Copernicus GmbH, 2019 |
imprint_str_mv |
Copernicus GmbH, 2019 |
issn |
1991-9603 |
issn_str_mv |
1991-9603 |
language |
English |
mega_collection |
Copernicus GmbH (CrossRef) |
match_str |
scalzo2019efficiencyandrobustnessinmontecarlosamplingfor3dgeophysicalinversionswithobsidianv012settingupforsuccess |
publishDateSort |
2019 |
publisher |
Copernicus GmbH |
recordtype |
ai |
record_format |
ai |
series |
Geoscientific Model Development |
source_id |
49 |
title |
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_unstemmed |
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_full |
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_fullStr |
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_full_unstemmed |
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_short |
Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_sort |
efficiency and robustness in monte carlo sampling for 3-d geophysical inversions with obsidian v0.1.2: setting up for success |
topic |
Polymers and Plastics General Environmental Science |
url |
http://dx.doi.org/10.5194/gmd-12-2941-2019 |
publishDate |
2019 |
physical |
2941-2960 |
description |
<jats:p>Abstract. The rigorous quantification of uncertainty in geophysical inversions is a
challenging problem. Inversions are often ill-posed and the likelihood surface
may be multi-modal; properties of any single mode become inadequate uncertainty
measures, and sampling methods become inefficient for irregular posteriors
or high-dimensional parameter spaces. We explore the
influences of different choices made by the practitioner on the efficiency and accuracy of Bayesian geophysical inversion methods that rely on Markov chain Monte Carlo sampling to assess uncertainty using a multi-sensor inversion
of the three-dimensional structure and composition of a region in the
Cooper Basin of South Australia as a case study. The inversion is performed
using an updated version of the Obsidian distributed inversion software.
We find that the posterior for this
inversion has a complex local covariance structure, hindering the efficiency of
adaptive sampling methods that adjust the proposal based on the chain history.
Within the context of a parallel-tempered Markov chain Monte Carlo scheme for
exploring high-dimensional multi-modal posteriors, a preconditioned
Crank–Nicolson proposal outperforms more conventional forms of random walk.
Aspects of the problem setup, such as priors on petrophysics and on 3-D
geological structure, affect the shape and separation of posterior modes,
influencing sampling performance as well as the inversion results. The use of
uninformative priors on sensor noise enables
optimal weighting among multiple sensors even if noise levels are uncertain.
</jats:p> |
container_issue |
7 |
container_start_page |
2941 |
container_title |
Geoscientific Model Development |
container_volume |
12 |
format_de105 |
Article, E-Article |
format_de14 |
Article, E-Article |
format_de15 |
Article, E-Article |
format_de520 |
Article, E-Article |
format_de540 |
Article, E-Article |
format_dech1 |
Article, E-Article |
format_ded117 |
Article, E-Article |
format_degla1 |
E-Article |
format_del152 |
Buch |
format_del189 |
Article, E-Article |
format_dezi4 |
Article |
format_dezwi2 |
Article, E-Article |
format_finc |
Article, E-Article |
format_nrw |
Article, E-Article |
_version_ |
1792347755333550081 |
geogr_code |
not assigned |
last_indexed |
2024-03-01T18:00:19.382Z |
geogr_code_person |
not assigned |
openURL |
url_ver=Z39.88-2004&ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fvufind.svn.sourceforge.net%3Agenerator&rft.title=Efficiency+and+robustness+in+Monte+Carlo+sampling+for+3-D+geophysical+inversions+with+Obsidian+v0.1.2%3A+setting+up+for+success&rft.date=2019-07-15&genre=article&issn=1991-9603&volume=12&issue=7&spage=2941&epage=2960&pages=2941-2960&jtitle=Geoscientific+Model+Development&atitle=Efficiency+and+robustness+in+Monte+Carlo+sampling+for+3-D+geophysical+inversions+with+Obsidian+v0.1.2%3A+setting+up+for+success&aulast=Cripps&aufirst=Sally&rft_id=info%3Adoi%2F10.5194%2Fgmd-12-2941-2019&rft.language%5B0%5D=eng |
SOLR | |
_version_ | 1792347755333550081 |
author | Scalzo, Richard, Kohn, David, Olierook, Hugo, Houseman, Gregory, Chandra, Rohitash, Girolami, Mark, Cripps, Sally |
author_facet | Scalzo, Richard, Kohn, David, Olierook, Hugo, Houseman, Gregory, Chandra, Rohitash, Girolami, Mark, Cripps, Sally, Scalzo, Richard, Kohn, David, Olierook, Hugo, Houseman, Gregory, Chandra, Rohitash, Girolami, Mark, Cripps, Sally |
author_sort | scalzo, richard |
container_issue | 7 |
container_start_page | 2941 |
container_title | Geoscientific Model Development |
container_volume | 12 |
description | <jats:p>Abstract. The rigorous quantification of uncertainty in geophysical inversions is a challenging problem. Inversions are often ill-posed and the likelihood surface may be multi-modal; properties of any single mode become inadequate uncertainty measures, and sampling methods become inefficient for irregular posteriors or high-dimensional parameter spaces. We explore the influences of different choices made by the practitioner on the efficiency and accuracy of Bayesian geophysical inversion methods that rely on Markov chain Monte Carlo sampling to assess uncertainty using a multi-sensor inversion of the three-dimensional structure and composition of a region in the Cooper Basin of South Australia as a case study. The inversion is performed using an updated version of the Obsidian distributed inversion software. We find that the posterior for this inversion has a complex local covariance structure, hindering the efficiency of adaptive sampling methods that adjust the proposal based on the chain history. Within the context of a parallel-tempered Markov chain Monte Carlo scheme for exploring high-dimensional multi-modal posteriors, a preconditioned Crank–Nicolson proposal outperforms more conventional forms of random walk. Aspects of the problem setup, such as priors on petrophysics and on 3-D geological structure, affect the shape and separation of posterior modes, influencing sampling performance as well as the inversion results. The use of uninformative priors on sensor noise enables optimal weighting among multiple sensors even if noise levels are uncertain. </jats:p> |
doi_str_mv | 10.5194/gmd-12-2941-2019 |
facet_avail | Online, Free |
finc_class_facet | Chemie und Pharmazie |
format | ElectronicArticle |
format_de105 | Article, E-Article |
format_de14 | Article, E-Article |
format_de15 | Article, E-Article |
format_de520 | Article, E-Article |
format_de540 | Article, E-Article |
format_dech1 | Article, E-Article |
format_ded117 | Article, E-Article |
format_degla1 | E-Article |
format_del152 | Buch |
format_del189 | Article, E-Article |
format_dezi4 | Article |
format_dezwi2 | Article, E-Article |
format_finc | Article, E-Article |
format_nrw | Article, E-Article |
geogr_code | not assigned |
geogr_code_person | not assigned |
id | ai-49-aHR0cDovL2R4LmRvaS5vcmcvMTAuNTE5NC9nbWQtMTItMjk0MS0yMDE5 |
imprint | Copernicus GmbH, 2019 |
imprint_str_mv | Copernicus GmbH, 2019 |
institution | DE-D275, DE-Bn3, DE-Brt1, DE-Zwi2, DE-D161, DE-Gla1, DE-Zi4, DE-15, DE-Pl11, DE-Rs1, DE-105, DE-14, DE-Ch1, DE-L229 |
issn | 1991-9603 |
issn_str_mv | 1991-9603 |
language | English |
last_indexed | 2024-03-01T18:00:19.382Z |
match_str | scalzo2019efficiencyandrobustnessinmontecarlosamplingfor3dgeophysicalinversionswithobsidianv012settingupforsuccess |
mega_collection | Copernicus GmbH (CrossRef) |
physical | 2941-2960 |
publishDate | 2019 |
publishDateSort | 2019 |
publisher | Copernicus GmbH |
record_format | ai |
recordtype | ai |
series | Geoscientific Model Development |
source_id | 49 |
spelling | Scalzo, Richard Kohn, David Olierook, Hugo Houseman, Gregory Chandra, Rohitash Girolami, Mark Cripps, Sally 1991-9603 Copernicus GmbH Polymers and Plastics General Environmental Science http://dx.doi.org/10.5194/gmd-12-2941-2019 <jats:p>Abstract. The rigorous quantification of uncertainty in geophysical inversions is a challenging problem. Inversions are often ill-posed and the likelihood surface may be multi-modal; properties of any single mode become inadequate uncertainty measures, and sampling methods become inefficient for irregular posteriors or high-dimensional parameter spaces. We explore the influences of different choices made by the practitioner on the efficiency and accuracy of Bayesian geophysical inversion methods that rely on Markov chain Monte Carlo sampling to assess uncertainty using a multi-sensor inversion of the three-dimensional structure and composition of a region in the Cooper Basin of South Australia as a case study. The inversion is performed using an updated version of the Obsidian distributed inversion software. We find that the posterior for this inversion has a complex local covariance structure, hindering the efficiency of adaptive sampling methods that adjust the proposal based on the chain history. Within the context of a parallel-tempered Markov chain Monte Carlo scheme for exploring high-dimensional multi-modal posteriors, a preconditioned Crank–Nicolson proposal outperforms more conventional forms of random walk. Aspects of the problem setup, such as priors on petrophysics and on 3-D geological structure, affect the shape and separation of posterior modes, influencing sampling performance as well as the inversion results. The use of uninformative priors on sensor noise enables optimal weighting among multiple sensors even if noise levels are uncertain. </jats:p> Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success Geoscientific Model Development |
spellingShingle | Scalzo, Richard, Kohn, David, Olierook, Hugo, Houseman, Gregory, Chandra, Rohitash, Girolami, Mark, Cripps, Sally, Geoscientific Model Development, Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success, Polymers and Plastics, General Environmental Science |
title | Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_full | Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_fullStr | Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_full_unstemmed | Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_short | Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
title_sort | efficiency and robustness in monte carlo sampling for 3-d geophysical inversions with obsidian v0.1.2: setting up for success |
title_unstemmed | Efficiency and robustness in Monte Carlo sampling for 3-D geophysical inversions with Obsidian v0.1.2: setting up for success |
topic | Polymers and Plastics, General Environmental Science |
url | http://dx.doi.org/10.5194/gmd-12-2941-2019 |