Global versus local MJO forecast skill of the ECMWF model during DYNAMO

Jian Ling, Peter Bauer, Peter Bechtold, Anton Beljaars, Richard Forbes, Frederic Vitart, Marcela Ulate, Chidong Zhang

Research output: Contribution to journalArticle

39 Citations (Scopus)

Abstract

This study introduces a concept of global versus local forecast skill of the Madden-Julian oscillation (MJO). The global skill, measured by a commonly used MJO index [the Real-time Multivariate MJO (RMM)], evaluates the model's capability of forecasting global patterns of the MJO, with an emphasis on the zonal wind fields. The local skill is measured by a method of tracking the eastward propagation of MJO precipitation. It provides quantitative information of the strength, propagation speed, and timing of MJO precipitation in a given region, such as the Indian Ocean. Both global and local MJO forecast skills are assessed for ECMWF forecasts of three MJO events during the 2011-12 Dynamics of the MJO (DYNAMO) field campaign. Characteristics of error growth differ substantially between global and local MJO forecast skills, and between the three MJO quantities (strength, speed, and timing) of the local skill measure. They all vary considerably among the three MJO events. Deterioration in global forecast skill for these three events appears to be related to poor local skill in forecasting the propagation speed of MJO precipitation. The global and local MJO forecast skill measures are also applied to evaluate numerical experiments of observation denial, humidity relaxation, and forcing by daily perturbations in sea surface temperature (SST). The results suggest that forecast skill or errors of convective initiation of the three MJO events have global origins. Effects of local (Indian Ocean) factors, such as enhanced observations in the initial conditions, variability of tropospheric humidity and tropical SST, on forecasts of MJO initiation and propagation are limited.

Original languageEnglish (US)
Pages (from-to)2228-2247
Number of pages20
JournalMonthly Weather Review
Volume142
Issue number6
DOIs
StatePublished - 2014

Fingerprint

Madden-Julian oscillation
forecast
humidity
sea surface temperature
zonal wind

Keywords

  • Forecast verification/skill
  • Intraseasonal variability
  • Operational forecasting

ASJC Scopus subject areas

  • Atmospheric Science

Cite this

Ling, J., Bauer, P., Bechtold, P., Beljaars, A., Forbes, R., Vitart, F., ... Zhang, C. (2014). Global versus local MJO forecast skill of the ECMWF model during DYNAMO. Monthly Weather Review, 142(6), 2228-2247. https://doi.org/10.1175/MWR-D-13-00292.1

Global versus local MJO forecast skill of the ECMWF model during DYNAMO. / Ling, Jian; Bauer, Peter; Bechtold, Peter; Beljaars, Anton; Forbes, Richard; Vitart, Frederic; Ulate, Marcela; Zhang, Chidong.

In: Monthly Weather Review, Vol. 142, No. 6, 2014, p. 2228-2247.

Research output: Contribution to journalArticle

Ling, J, Bauer, P, Bechtold, P, Beljaars, A, Forbes, R, Vitart, F, Ulate, M & Zhang, C 2014, 'Global versus local MJO forecast skill of the ECMWF model during DYNAMO', Monthly Weather Review, vol. 142, no. 6, pp. 2228-2247. https://doi.org/10.1175/MWR-D-13-00292.1
Ling J, Bauer P, Bechtold P, Beljaars A, Forbes R, Vitart F et al. Global versus local MJO forecast skill of the ECMWF model during DYNAMO. Monthly Weather Review. 2014;142(6):2228-2247. https://doi.org/10.1175/MWR-D-13-00292.1
Ling, Jian ; Bauer, Peter ; Bechtold, Peter ; Beljaars, Anton ; Forbes, Richard ; Vitart, Frederic ; Ulate, Marcela ; Zhang, Chidong. / Global versus local MJO forecast skill of the ECMWF model during DYNAMO. In: Monthly Weather Review. 2014 ; Vol. 142, No. 6. pp. 2228-2247.
@article{7f5edec3de3f48cd8dd5b2366ece172b,
title = "Global versus local MJO forecast skill of the ECMWF model during DYNAMO",
abstract = "This study introduces a concept of global versus local forecast skill of the Madden-Julian oscillation (MJO). The global skill, measured by a commonly used MJO index [the Real-time Multivariate MJO (RMM)], evaluates the model's capability of forecasting global patterns of the MJO, with an emphasis on the zonal wind fields. The local skill is measured by a method of tracking the eastward propagation of MJO precipitation. It provides quantitative information of the strength, propagation speed, and timing of MJO precipitation in a given region, such as the Indian Ocean. Both global and local MJO forecast skills are assessed for ECMWF forecasts of three MJO events during the 2011-12 Dynamics of the MJO (DYNAMO) field campaign. Characteristics of error growth differ substantially between global and local MJO forecast skills, and between the three MJO quantities (strength, speed, and timing) of the local skill measure. They all vary considerably among the three MJO events. Deterioration in global forecast skill for these three events appears to be related to poor local skill in forecasting the propagation speed of MJO precipitation. The global and local MJO forecast skill measures are also applied to evaluate numerical experiments of observation denial, humidity relaxation, and forcing by daily perturbations in sea surface temperature (SST). The results suggest that forecast skill or errors of convective initiation of the three MJO events have global origins. Effects of local (Indian Ocean) factors, such as enhanced observations in the initial conditions, variability of tropospheric humidity and tropical SST, on forecasts of MJO initiation and propagation are limited.",
keywords = "Forecast verification/skill, Intraseasonal variability, Operational forecasting",
author = "Jian Ling and Peter Bauer and Peter Bechtold and Anton Beljaars and Richard Forbes and Frederic Vitart and Marcela Ulate and Chidong Zhang",
year = "2014",
doi = "10.1175/MWR-D-13-00292.1",
language = "English (US)",
volume = "142",
pages = "2228--2247",
journal = "Monthly Weather Review",
issn = "0027-0644",
publisher = "American Meteorological Society",
number = "6",

}

TY - JOUR

T1 - Global versus local MJO forecast skill of the ECMWF model during DYNAMO

AU - Ling, Jian

AU - Bauer, Peter

AU - Bechtold, Peter

AU - Beljaars, Anton

AU - Forbes, Richard

AU - Vitart, Frederic

AU - Ulate, Marcela

AU - Zhang, Chidong

PY - 2014

Y1 - 2014

N2 - This study introduces a concept of global versus local forecast skill of the Madden-Julian oscillation (MJO). The global skill, measured by a commonly used MJO index [the Real-time Multivariate MJO (RMM)], evaluates the model's capability of forecasting global patterns of the MJO, with an emphasis on the zonal wind fields. The local skill is measured by a method of tracking the eastward propagation of MJO precipitation. It provides quantitative information of the strength, propagation speed, and timing of MJO precipitation in a given region, such as the Indian Ocean. Both global and local MJO forecast skills are assessed for ECMWF forecasts of three MJO events during the 2011-12 Dynamics of the MJO (DYNAMO) field campaign. Characteristics of error growth differ substantially between global and local MJO forecast skills, and between the three MJO quantities (strength, speed, and timing) of the local skill measure. They all vary considerably among the three MJO events. Deterioration in global forecast skill for these three events appears to be related to poor local skill in forecasting the propagation speed of MJO precipitation. The global and local MJO forecast skill measures are also applied to evaluate numerical experiments of observation denial, humidity relaxation, and forcing by daily perturbations in sea surface temperature (SST). The results suggest that forecast skill or errors of convective initiation of the three MJO events have global origins. Effects of local (Indian Ocean) factors, such as enhanced observations in the initial conditions, variability of tropospheric humidity and tropical SST, on forecasts of MJO initiation and propagation are limited.

AB - This study introduces a concept of global versus local forecast skill of the Madden-Julian oscillation (MJO). The global skill, measured by a commonly used MJO index [the Real-time Multivariate MJO (RMM)], evaluates the model's capability of forecasting global patterns of the MJO, with an emphasis on the zonal wind fields. The local skill is measured by a method of tracking the eastward propagation of MJO precipitation. It provides quantitative information of the strength, propagation speed, and timing of MJO precipitation in a given region, such as the Indian Ocean. Both global and local MJO forecast skills are assessed for ECMWF forecasts of three MJO events during the 2011-12 Dynamics of the MJO (DYNAMO) field campaign. Characteristics of error growth differ substantially between global and local MJO forecast skills, and between the three MJO quantities (strength, speed, and timing) of the local skill measure. They all vary considerably among the three MJO events. Deterioration in global forecast skill for these three events appears to be related to poor local skill in forecasting the propagation speed of MJO precipitation. The global and local MJO forecast skill measures are also applied to evaluate numerical experiments of observation denial, humidity relaxation, and forcing by daily perturbations in sea surface temperature (SST). The results suggest that forecast skill or errors of convective initiation of the three MJO events have global origins. Effects of local (Indian Ocean) factors, such as enhanced observations in the initial conditions, variability of tropospheric humidity and tropical SST, on forecasts of MJO initiation and propagation are limited.

KW - Forecast verification/skill

KW - Intraseasonal variability

KW - Operational forecasting

UR - http://www.scopus.com/inward/record.url?scp=84901602295&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901602295&partnerID=8YFLogxK

U2 - 10.1175/MWR-D-13-00292.1

DO - 10.1175/MWR-D-13-00292.1

M3 - Article

VL - 142

SP - 2228

EP - 2247

JO - Monthly Weather Review

JF - Monthly Weather Review

SN - 0027-0644

IS - 6

ER -