Efficient Reconstructions of Common Era Climate via Integrated Nested Laplace Approximations

Luis A. Barboza, Julien Emile-geay, Bo Li, Wan He

Research output: Contribution to journalArticle

Abstract

Paleoclimate reconstruction on the Common Era (1–2000 AD) provides critical context for recent warming trends. This work leverages integrated nested Laplace approximations (INLA) to conduct inference under a Bayesian hierarchical model using data from three sources: a state-of-the-art proxy database (PAGES 2k), surface temperature observations (HadCRUT4), and latest estimates of external forcings. INLA’s computational efficiency allows to explore several model formulations (with or without forcings, explicitly modeling internal variability or not), as well as five data reduction techniques. Two different validation exercises find a small impact of data reduction choices, but a large impact for model choice, with best results for the two models that incorporate external forcings. These models confirm that man-made greenhouse gas emissions are the largest contributor to temperature variability over the Common Era, followed by volcanic forcing. Solar effects are indistinguishable from zero. INLA provide an efficient way to estimate the posterior mean, comparable with the much costlier Monte Carlo Markov Chain procedure, but with wider uncertainty bounds. We recommend using it for exploration of model designs, but full MCMC solutions should be used for proper uncertainty quantification.
Original languageEnglish (US)
JournalJournal of Agricultural, Biological, and Environmental Statistics
DOIs
StateAccepted/In press - Jul 22 2019

Fingerprint

Laplace Approximation
Climate
Forcing
Uncertainty
climate
Markov Chains
Temperature
Data Reduction
Information Storage and Retrieval
Proxy
Gases
Monte Carlo Markov Chain
Databases
Exercise
Model Choice
Posterior Mean
Bayesian Hierarchical Model
Uncertainty Quantification
Greenhouse Gases
Data reduction

Keywords

  • Hierarchical Bayesian model
  • INLA
  • Paleoclimate reconstruction

Cite this

Efficient Reconstructions of Common Era Climate via Integrated Nested Laplace Approximations. / Barboza, Luis A.; Emile-geay, Julien; Li, Bo; He, Wan.

In: Journal of Agricultural, Biological, and Environmental Statistics, 22.07.2019.

Research output: Contribution to journalArticle

@article{e7eb6a9cff6a42c09675a1a16e755a20,
title = "Efficient Reconstructions of Common Era Climate via Integrated Nested Laplace Approximations",
abstract = "Paleoclimate reconstruction on the Common Era (1–2000 AD) provides critical context for recent warming trends. This work leverages integrated nested Laplace approximations (INLA) to conduct inference under a Bayesian hierarchical model using data from three sources: a state-of-the-art proxy database (PAGES 2k), surface temperature observations (HadCRUT4), and latest estimates of external forcings. INLA’s computational efficiency allows to explore several model formulations (with or without forcings, explicitly modeling internal variability or not), as well as five data reduction techniques. Two different validation exercises find a small impact of data reduction choices, but a large impact for model choice, with best results for the two models that incorporate external forcings. These models confirm that man-made greenhouse gas emissions are the largest contributor to temperature variability over the Common Era, followed by volcanic forcing. Solar effects are indistinguishable from zero. INLA provide an efficient way to estimate the posterior mean, comparable with the much costlier Monte Carlo Markov Chain procedure, but with wider uncertainty bounds. We recommend using it for exploration of model designs, but full MCMC solutions should be used for proper uncertainty quantification.",
keywords = "Hierarchical Bayesian model, INLA, Paleoclimate reconstruction",
author = "Barboza, {Luis A.} and Julien Emile-geay and Bo Li and Wan He",
year = "2019",
month = "7",
day = "22",
doi = "10.1007/s13253-019-00372-4",
language = "English (US)",
journal = "Journal of Agricultural, Biological, and Environmental Statistics",
issn = "1085-7117",
publisher = "Springer New York",

}

TY - JOUR

T1 - Efficient Reconstructions of Common Era Climate via Integrated Nested Laplace Approximations

AU - Barboza, Luis A.

AU - Emile-geay, Julien

AU - Li, Bo

AU - He, Wan

PY - 2019/7/22

Y1 - 2019/7/22

N2 - Paleoclimate reconstruction on the Common Era (1–2000 AD) provides critical context for recent warming trends. This work leverages integrated nested Laplace approximations (INLA) to conduct inference under a Bayesian hierarchical model using data from three sources: a state-of-the-art proxy database (PAGES 2k), surface temperature observations (HadCRUT4), and latest estimates of external forcings. INLA’s computational efficiency allows to explore several model formulations (with or without forcings, explicitly modeling internal variability or not), as well as five data reduction techniques. Two different validation exercises find a small impact of data reduction choices, but a large impact for model choice, with best results for the two models that incorporate external forcings. These models confirm that man-made greenhouse gas emissions are the largest contributor to temperature variability over the Common Era, followed by volcanic forcing. Solar effects are indistinguishable from zero. INLA provide an efficient way to estimate the posterior mean, comparable with the much costlier Monte Carlo Markov Chain procedure, but with wider uncertainty bounds. We recommend using it for exploration of model designs, but full MCMC solutions should be used for proper uncertainty quantification.

AB - Paleoclimate reconstruction on the Common Era (1–2000 AD) provides critical context for recent warming trends. This work leverages integrated nested Laplace approximations (INLA) to conduct inference under a Bayesian hierarchical model using data from three sources: a state-of-the-art proxy database (PAGES 2k), surface temperature observations (HadCRUT4), and latest estimates of external forcings. INLA’s computational efficiency allows to explore several model formulations (with or without forcings, explicitly modeling internal variability or not), as well as five data reduction techniques. Two different validation exercises find a small impact of data reduction choices, but a large impact for model choice, with best results for the two models that incorporate external forcings. These models confirm that man-made greenhouse gas emissions are the largest contributor to temperature variability over the Common Era, followed by volcanic forcing. Solar effects are indistinguishable from zero. INLA provide an efficient way to estimate the posterior mean, comparable with the much costlier Monte Carlo Markov Chain procedure, but with wider uncertainty bounds. We recommend using it for exploration of model designs, but full MCMC solutions should be used for proper uncertainty quantification.

KW - Hierarchical Bayesian model

KW - INLA

KW - Paleoclimate reconstruction

U2 - 10.1007/s13253-019-00372-4

DO - 10.1007/s13253-019-00372-4

M3 - Article

JO - Journal of Agricultural, Biological, and Environmental Statistics

JF - Journal of Agricultural, Biological, and Environmental Statistics

SN - 1085-7117

ER -