Pragmatic aspects of uncertainty propagation

A conceptual review

W. Carlisle Thacker, Mohamed Iskandarani, Rafael C. Gonçalves, Ashwanth Srinivasan, Omar M. Knio

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

When quantifying the uncertainty of the response of a computationally costly oceanographic or meteorological model stemming from the uncertainty of its inputs, practicality demands getting the most information using the fewest simulations. It is widely recognized that, by interpolating the results of a small number of simulations, results of additional simulations can be inexpensively approximated to provide a useful estimate of the variability of the response. Even so, as computing the simulations to be interpolated remains the biggest expense, the choice of these simulations deserves attention. When making this choice, two requirement should be considered: (i) the nature of the interpolation and (ii) the available information about input uncertainty. Examples comparing polynomial interpolation and Gaussian process interpolation are presented for three different views of input uncertainty.

Original languageEnglish (US)
Pages (from-to)25-36
Number of pages12
JournalOcean Modelling
Volume95
DOIs
StatePublished - Nov 1 2015

Fingerprint

Interpolation
interpolation
simulation
Polynomials
Uncertainty

Keywords

  • Error propagation
  • Gaussian process
  • Ocean modeling
  • Optimal interpolation
  • Polynomial chaos
  • Uncertainty quantification

ASJC Scopus subject areas

  • Atmospheric Science
  • Oceanography
  • Geotechnical Engineering and Engineering Geology
  • Computer Science (miscellaneous)

Cite this

Pragmatic aspects of uncertainty propagation : A conceptual review. / Thacker, W. Carlisle; Iskandarani, Mohamed; Gonçalves, Rafael C.; Srinivasan, Ashwanth; Knio, Omar M.

In: Ocean Modelling, Vol. 95, 01.11.2015, p. 25-36.

Research output: Contribution to journalArticle

Thacker, W. Carlisle ; Iskandarani, Mohamed ; Gonçalves, Rafael C. ; Srinivasan, Ashwanth ; Knio, Omar M. / Pragmatic aspects of uncertainty propagation : A conceptual review. In: Ocean Modelling. 2015 ; Vol. 95. pp. 25-36.
@article{3211bf3e174743b593f5dfb20afeb75e,
title = "Pragmatic aspects of uncertainty propagation: A conceptual review",
abstract = "When quantifying the uncertainty of the response of a computationally costly oceanographic or meteorological model stemming from the uncertainty of its inputs, practicality demands getting the most information using the fewest simulations. It is widely recognized that, by interpolating the results of a small number of simulations, results of additional simulations can be inexpensively approximated to provide a useful estimate of the variability of the response. Even so, as computing the simulations to be interpolated remains the biggest expense, the choice of these simulations deserves attention. When making this choice, two requirement should be considered: (i) the nature of the interpolation and (ii) the available information about input uncertainty. Examples comparing polynomial interpolation and Gaussian process interpolation are presented for three different views of input uncertainty.",
keywords = "Error propagation, Gaussian process, Ocean modeling, Optimal interpolation, Polynomial chaos, Uncertainty quantification",
author = "Thacker, {W. Carlisle} and Mohamed Iskandarani and Gon{\cc}alves, {Rafael C.} and Ashwanth Srinivasan and Knio, {Omar M.}",
year = "2015",
month = "11",
day = "1",
doi = "10.1016/j.ocemod.2015.09.001",
language = "English (US)",
volume = "95",
pages = "25--36",
journal = "Ocean Modelling",
issn = "1463-5003",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Pragmatic aspects of uncertainty propagation

T2 - A conceptual review

AU - Thacker, W. Carlisle

AU - Iskandarani, Mohamed

AU - Gonçalves, Rafael C.

AU - Srinivasan, Ashwanth

AU - Knio, Omar M.

PY - 2015/11/1

Y1 - 2015/11/1

N2 - When quantifying the uncertainty of the response of a computationally costly oceanographic or meteorological model stemming from the uncertainty of its inputs, practicality demands getting the most information using the fewest simulations. It is widely recognized that, by interpolating the results of a small number of simulations, results of additional simulations can be inexpensively approximated to provide a useful estimate of the variability of the response. Even so, as computing the simulations to be interpolated remains the biggest expense, the choice of these simulations deserves attention. When making this choice, two requirement should be considered: (i) the nature of the interpolation and (ii) the available information about input uncertainty. Examples comparing polynomial interpolation and Gaussian process interpolation are presented for three different views of input uncertainty.

AB - When quantifying the uncertainty of the response of a computationally costly oceanographic or meteorological model stemming from the uncertainty of its inputs, practicality demands getting the most information using the fewest simulations. It is widely recognized that, by interpolating the results of a small number of simulations, results of additional simulations can be inexpensively approximated to provide a useful estimate of the variability of the response. Even so, as computing the simulations to be interpolated remains the biggest expense, the choice of these simulations deserves attention. When making this choice, two requirement should be considered: (i) the nature of the interpolation and (ii) the available information about input uncertainty. Examples comparing polynomial interpolation and Gaussian process interpolation are presented for three different views of input uncertainty.

KW - Error propagation

KW - Gaussian process

KW - Ocean modeling

KW - Optimal interpolation

KW - Polynomial chaos

KW - Uncertainty quantification

UR - http://www.scopus.com/inward/record.url?scp=84945185360&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84945185360&partnerID=8YFLogxK

U2 - 10.1016/j.ocemod.2015.09.001

DO - 10.1016/j.ocemod.2015.09.001

M3 - Article

VL - 95

SP - 25

EP - 36

JO - Ocean Modelling

JF - Ocean Modelling

SN - 1463-5003

ER -