Abstract

This study documents a detailed analysis on the Madden-Julian Oscillation (MJO) simulated by the National Centers for Environmental Prediction (NCEP) using the Global Forecast System (GFS) model version 2003 coupled with the Climate Forecast System model (CFS) consisting of the 2003 version of GFS and the Geophysical Fluid Dynamics Laboratory (GFDL) Modular Ocean Model V.3 (MOM3). The analyses are based upon a 21-year simulation of AMIP-type with GFS and CMIP-type with CFS. It is found that air-sea coupling in CFS is shown to improve the coherence between convection and large-scale circulation associated with the MJO. The too fast propagation of convection from the Indian Ocean to the maritime continents and the western Pacific in GFS is improved (slowed down) in CFS. Both GFS and CFS produce too strong intraseasonal convective heating and circulation anomalies in the central-eastern Pacific; further, the air-sea coupling in CFS enhances this unrealistic feature. The simulated mean slow phase speed of eastward propagating low-wavenumber components shown in the wavenumber-frequency spectra is due to the slow propagation in the central-eastern Pacific in both GFS and CFS. Errors in model climatology may have some effect upon the simulated MJO and two possible influences are: (i) CFS fails to simulate the westerlies over maritime continents and western Pacific areas, resulting in an unrealistic representation of surface latent heat flux associated with the MJO; and (ii) vertical easterly wind shear from the Indian Ocean to the western Pacific in CFS is much weaker than that in the observation and in GFS, which may adversely affect the eastward propagation of the simulated MJO.

Highlights

  • Since the discovery of the tropical Madden-Julian Oscillation (MJO) over three decades ago (Madden and Julian 1971, 1972), most of the studies on the MJO have considered it a result of internal atmospheric dynamics involving the interaction between the convection and largescale circulation

  • Errors in model climatology may have some effect upon the simulated MJO and two possible influences are: (i) Climate Forecast System model (CFS) fails to simulate the westerlies over maritime continents and western Pacific areas, resulting in an unrealistic representation of surface latent heat flux associated with the MJO; and (ii) vertical easterly wind shear from the Indian Ocean to the western Pacific in CFS is much weaker than that in the observation and in Global Forecast System (GFS), which may adversely affect the eastward propagation of the simulated MJO

  • This study examines the fidelity of the MJO simulation by the National Centers for Environmental Prediction (NCEP) uncoupled atmospheric Global Forecast System (GFS) model and the coupled Climate Forecast System (CFS) model based on a 21-year simulation of AMIP-type with GFS and CMIP-type with CFS

Read more

Summary

INTRODUCTION

Since the discovery of the tropical Madden-Julian Oscillation (MJO) over three decades ago (Madden and Julian 1971, 1972), most of the studies on the MJO have considered it a result of internal atmospheric dynamics involving the interaction between the convection and largescale circulation. A new global coupled atmosphere-ocean Climate Forecast System model (CFS) has recently been developed at the National Centers for Environmental Prediction (NCEP) (Saha et al 2006; Wang et al 2005). It consists of the NCEP atmospheric Global Forecast System model (GFS) and the Geophysical Fluid Dynamics Laboratory (GFDL) Modular Ocean Model V.3 (MOM3). We diagnose the characteristics of the MJO simulated by the uncoupled atmospheric GFS and coupled atmosphere-ocean CFS to compare the simulations with observation and document the role of the air-sea interaction. For an analysis on the boreal summer intraseasonal oscillation, refer to Seo et al (2007)

The Models
The Simulations and Observations
CLIMATOLOGY
Precipitation Rate
Atmospheric Equatorial Zonal Winds
INTRASEASONAL VARIABILITY
THE SIMULATED MJO
Power Spectra of Tropical 850 hPa Zonal Velocity
EOF Modes of Combined Fields
Propagation and Relationship among Associated Fields
Findings
SUMMARY
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call