All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 122 22 3
PDF Downloads 15 5 1

Space-Time Spectral Structure of a GLAS General Circulation Model and a Comparison with Observations

David M. StrausLaboratory for Atmospheric Sciences, Modeling and Simulation Facility, NASA Goddard Space Flight Center, Greenbelt, MD 20771

Search for other papers by David M. Straus in
Current site
Google Scholar
PubMed
Close
and
J. ShuklaLaboratory for Atmospheric Sciences, Modeling and Simulation Facility, NASA Goddard Space Flight Center, Greenbelt, MD 20771

Search for other papers by J. Shukla in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

The wavenumber-frequency spectra of geopotential height have been computed from a winter simulation of a GLAS general circulation model, and are compared to the spectra obtained from 15 winters of observed analyses. The variances in several broadly defined space-time categories are presented, and their distribution with latitude and height discussed.

The low-frequency planetary waves (wavenumbers 1–4) in the model have substantially less variance than is observed and their latitudinal structure also differs from observations. The same is true for the medium-frequency planetary waves. The synoptic-scale waves (wavenumbers 5–10) with both low and medium frequencies are quite well reproduced by the model, with the total variances and their latitude-height structure comparing well to the observations. The net propagation tendency is examined for low-frequency planetary waves and the medium-frequency synoptic-scale waves. The model results show too much tendency for eastward propagation, the discrepancy being much larger in the case of the planetary waves. The model stationary waves were also examined. Their variance is considerably less than is observed, and their latitudinal structure incorrect. These findings are compared to the limited statistics available for the NCAR and GFDL models. The GLAS model's simulation of the synoptic-scale waves (with both low and medium frequencies) is apparently more realistic than that of the other models. The opposite is true for the stationary waves. All the models underpredict the low-frequency planetary-wave variance.

The differences between model and observed spectra are compared to (i) the range of interannual variability, and (ii) the estimated statistical uncertainties of the spectra. For the synoptic-scale waves the differences are generally smaller than either (i) or (ii), whereas for the planetary waves the differences are larger.

Possible causes for the GLAS model's underprediction of the stationary planetary wave variance are briefly discussed. It is suggested that the diabatic heating due to latent beat of condensation may play an important role. Further, evidence is presented indicating that the lack of variance in the model low-frequency planetary waves is closely related to the lack of variance in the stationary waves.

Abstract

The wavenumber-frequency spectra of geopotential height have been computed from a winter simulation of a GLAS general circulation model, and are compared to the spectra obtained from 15 winters of observed analyses. The variances in several broadly defined space-time categories are presented, and their distribution with latitude and height discussed.

The low-frequency planetary waves (wavenumbers 1–4) in the model have substantially less variance than is observed and their latitudinal structure also differs from observations. The same is true for the medium-frequency planetary waves. The synoptic-scale waves (wavenumbers 5–10) with both low and medium frequencies are quite well reproduced by the model, with the total variances and their latitude-height structure comparing well to the observations. The net propagation tendency is examined for low-frequency planetary waves and the medium-frequency synoptic-scale waves. The model results show too much tendency for eastward propagation, the discrepancy being much larger in the case of the planetary waves. The model stationary waves were also examined. Their variance is considerably less than is observed, and their latitudinal structure incorrect. These findings are compared to the limited statistics available for the NCAR and GFDL models. The GLAS model's simulation of the synoptic-scale waves (with both low and medium frequencies) is apparently more realistic than that of the other models. The opposite is true for the stationary waves. All the models underpredict the low-frequency planetary-wave variance.

The differences between model and observed spectra are compared to (i) the range of interannual variability, and (ii) the estimated statistical uncertainties of the spectra. For the synoptic-scale waves the differences are generally smaller than either (i) or (ii), whereas for the planetary waves the differences are larger.

Possible causes for the GLAS model's underprediction of the stationary planetary wave variance are briefly discussed. It is suggested that the diabatic heating due to latent beat of condensation may play an important role. Further, evidence is presented indicating that the lack of variance in the model low-frequency planetary waves is closely related to the lack of variance in the stationary waves.

Save