Data-driven localization mappings in filtering the monsoon-Hadley multicloud convective flows

Michèle De La Chevrotière, John Harlim

Research output: Contribution to journalArticle

Abstract

This paper demonstrates the efficacy of data-driven localization mappings for assimilating satellite-like observations in a dynamical system of intermediate complexity. In particular, a sparse network of synthetic brightness temperature measurements is simulated using an idealized radiative transfer model and assimilated to the monsoon-Hadley multicloud model, a nonlinear stochastic model containing several thousands of model coordinates. A serial ensemble Kalman filter is implemented in which the empirical correlation statistics are improved using localization maps obtained from a supervised learning algorithm. The impact of the localization mappings is assessed in perfect-model observing system simulation experiments (OSSEs) as well as in the presence of model errors resulting from the misspecification of key convective closure parameters. In perfect-model OSSEs, the localization mappings that use adjacent correlations to improve the correlation estimated from small ensemble sizes produce robust accurate analysis estimates. In the presence of model error, the filter skills of the localization maps trained on perfect- and imperfect-model data are comparable.

Original languageEnglish (US)
Pages (from-to)1197-1218
Number of pages22
JournalMonthly Weather Review
Volume146
Issue number4
DOIs
StatePublished - Apr 1 2018

Fingerprint

monsoon
Kalman filter
brightness temperature
simulation
radiative transfer
experiment
filter

All Science Journal Classification (ASJC) codes

  • Atmospheric Science

Cite this

@article{68fc1a4062a44696be57e428e0da9f8d,
title = "Data-driven localization mappings in filtering the monsoon-Hadley multicloud convective flows",
abstract = "This paper demonstrates the efficacy of data-driven localization mappings for assimilating satellite-like observations in a dynamical system of intermediate complexity. In particular, a sparse network of synthetic brightness temperature measurements is simulated using an idealized radiative transfer model and assimilated to the monsoon-Hadley multicloud model, a nonlinear stochastic model containing several thousands of model coordinates. A serial ensemble Kalman filter is implemented in which the empirical correlation statistics are improved using localization maps obtained from a supervised learning algorithm. The impact of the localization mappings is assessed in perfect-model observing system simulation experiments (OSSEs) as well as in the presence of model errors resulting from the misspecification of key convective closure parameters. In perfect-model OSSEs, the localization mappings that use adjacent correlations to improve the correlation estimated from small ensemble sizes produce robust accurate analysis estimates. In the presence of model error, the filter skills of the localization maps trained on perfect- and imperfect-model data are comparable.",
author = "{La Chevroti{\`e}re}, {Mich{\`e}le De} and John Harlim",
year = "2018",
month = "4",
day = "1",
doi = "10.1175/MWR-D-17-0381.1",
language = "English (US)",
volume = "146",
pages = "1197--1218",
journal = "Monthly Weather Review",
issn = "0027-0644",
publisher = "American Meteorological Society",
number = "4",

}

Data-driven localization mappings in filtering the monsoon-Hadley multicloud convective flows. / La Chevrotière, Michèle De; Harlim, John.

In: Monthly Weather Review, Vol. 146, No. 4, 01.04.2018, p. 1197-1218.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data-driven localization mappings in filtering the monsoon-Hadley multicloud convective flows

AU - La Chevrotière, Michèle De

AU - Harlim, John

PY - 2018/4/1

Y1 - 2018/4/1

N2 - This paper demonstrates the efficacy of data-driven localization mappings for assimilating satellite-like observations in a dynamical system of intermediate complexity. In particular, a sparse network of synthetic brightness temperature measurements is simulated using an idealized radiative transfer model and assimilated to the monsoon-Hadley multicloud model, a nonlinear stochastic model containing several thousands of model coordinates. A serial ensemble Kalman filter is implemented in which the empirical correlation statistics are improved using localization maps obtained from a supervised learning algorithm. The impact of the localization mappings is assessed in perfect-model observing system simulation experiments (OSSEs) as well as in the presence of model errors resulting from the misspecification of key convective closure parameters. In perfect-model OSSEs, the localization mappings that use adjacent correlations to improve the correlation estimated from small ensemble sizes produce robust accurate analysis estimates. In the presence of model error, the filter skills of the localization maps trained on perfect- and imperfect-model data are comparable.

AB - This paper demonstrates the efficacy of data-driven localization mappings for assimilating satellite-like observations in a dynamical system of intermediate complexity. In particular, a sparse network of synthetic brightness temperature measurements is simulated using an idealized radiative transfer model and assimilated to the monsoon-Hadley multicloud model, a nonlinear stochastic model containing several thousands of model coordinates. A serial ensemble Kalman filter is implemented in which the empirical correlation statistics are improved using localization maps obtained from a supervised learning algorithm. The impact of the localization mappings is assessed in perfect-model observing system simulation experiments (OSSEs) as well as in the presence of model errors resulting from the misspecification of key convective closure parameters. In perfect-model OSSEs, the localization mappings that use adjacent correlations to improve the correlation estimated from small ensemble sizes produce robust accurate analysis estimates. In the presence of model error, the filter skills of the localization maps trained on perfect- and imperfect-model data are comparable.

UR - http://www.scopus.com/inward/record.url?scp=85047071824&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047071824&partnerID=8YFLogxK

U2 - 10.1175/MWR-D-17-0381.1

DO - 10.1175/MWR-D-17-0381.1

M3 - Article

AN - SCOPUS:85047071824

VL - 146

SP - 1197

EP - 1218

JO - Monthly Weather Review

JF - Monthly Weather Review

SN - 0027-0644

IS - 4

ER -