Effect of Initialization on Diagnoses of NMC Large-Scale Circulation Statistics

Richard D. Rosen Atmospheric and Environmental Research, Inc., Cambridge, MA 02139

Search for other papers by Richard D. Rosen in
Current site
Google Scholar
PubMed
Close
and
David A. Salstein Atmospheric and Environmental Research, Inc., Cambridge, MA 02139

Search for other papers by David A. Salstein in
Current site
Google Scholar
PubMed
Close
Restricted access

Abstract

General circulation fields and statistics are derived for the months of January and July 1983 from different sets of daily global analyses produced by NMC. The two sets differ only in that one has been archived prior to an initialization step, whereas the other has been archived after initialization. To place our comparisons of these pre- and post-initialized NMC fields in perspective, we have also calculated circulation statistics from analyses produced by the ECMWF for the same months and from analyses of available rawinsonde data.

As suggested by earlier studies, the adiabatic normal mode initialization procedure used until recently at NMC is found to reduce the strength of the Hadley cell. However, the impact of this initialization is not large. In fact, the difference between the pre- and post-initialized NMC Hadley cells is generally smaller than that between the NMC and ECMWF Hadley cells. Other zonal mean circulation fields appear even less affected by NMC's initialization. Locally, the initialization does noticeably dampen the maxima in the vertical motion field, and it modifies the velocity potential field over certain regions of the globe, particularly in the western Pacific. For the most part, however, the post-initialized NMC fields, which have been the only NMC fields normally available to the scientific community, appear suitable for use in diagnostic studies of the circulation. More effort is needed to understand and reduce the differences between the NMC and ECMWF fields, if possible.

Abstract

General circulation fields and statistics are derived for the months of January and July 1983 from different sets of daily global analyses produced by NMC. The two sets differ only in that one has been archived prior to an initialization step, whereas the other has been archived after initialization. To place our comparisons of these pre- and post-initialized NMC fields in perspective, we have also calculated circulation statistics from analyses produced by the ECMWF for the same months and from analyses of available rawinsonde data.

As suggested by earlier studies, the adiabatic normal mode initialization procedure used until recently at NMC is found to reduce the strength of the Hadley cell. However, the impact of this initialization is not large. In fact, the difference between the pre- and post-initialized NMC Hadley cells is generally smaller than that between the NMC and ECMWF Hadley cells. Other zonal mean circulation fields appear even less affected by NMC's initialization. Locally, the initialization does noticeably dampen the maxima in the vertical motion field, and it modifies the velocity potential field over certain regions of the globe, particularly in the western Pacific. For the most part, however, the post-initialized NMC fields, which have been the only NMC fields normally available to the scientific community, appear suitable for use in diagnostic studies of the circulation. More effort is needed to understand and reduce the differences between the NMC and ECMWF fields, if possible.

Save