Global Parallel Experiment Log

Experiment Name

Scheduled June 2000 Implementation

Parallel Slot

PRX

Control Slot

MRF

Start date of parallel experiment

2000/03/16/06

End date of parallel experiment

2000/06/01/06

Environmental Modeling Center scientists

Hua-Lu Pan, Ken Campana, Pete Caplan, Bill Collins, John Derber, Bob Kistler, Yu-tai Hou, Mark Iredell, Qing-Fu Liu, S. Moorthi, Suru Saha, Joe Sela, Russ Treadon, Glenn White, Wan-Shu Wu

Abstract (including Motivation, Hypothesis and Method)

A number of improvements to the operational global model are slated for implementation in June 2000. The motivations for these changes are

improved tropical storm track forecasts,

improved ozone analyses and forecasts,

improved use of rawinsoned data,

improved use of satellite data,

improved initial development of tropical depressions,

improved winter midlatitude general circulation,

improved representation of orography particularly over Antarctica,

improved model forecast efficiency, and

improved summertime forecast bias near the continental surface.
OBSERVATIONAL AND ANALYSIS CHANGES
A new program to relocate tropical storms in the guess fields before analysis will be ready for the next hurricane season. The complex quality control provides a new observation error diagnosis using a combination of Bayesian logic, modeled error structure and fuzzy logic to make decisions. The rawinsonde radiation correction has been upgraded, too. The impacts of rawinsonde balloon drift in both space and time are included. GOES-10 radiance data is now used. A new ozone analysis background error is applied to avoid spurious far-field effects in data sparse regions. Satellite bias correction, quality control and data selection algorithms are upgraded.
FORECAST AND POST PROCESSING CHANGES
The deep convection algorithm is modified so instead of parameterizing the deepest possible cloud, the cloud top is chosen randomly among allowed clouds. This change reduces spurious tropical bombs and improves tropical storm forecasts. The gravity wave drag is corrected allowing more drag in the upper atmosphere. The terrain fields are recomputed from new orography datasets. The orography is filtered in the highest wavenumbers to reduce "ringing". A reduced grid over polar regions is used in the MRF to gain efficiency. A reversible interpolation is used to keep the surface file on a full grid for compatibility. New albedo climatology is used and aerosols over land are used in the shortwave radiation to reduce the summertime continental near surface warm bias.

Experiment changes log

Implementation changes list

Product changes list

Background links

Summer 1999 storm relocation and deep convection experiments by Hua-Lu Pan

April 1999 ozone analysis diagnostics by Wan-Shu Wu and Mark Iredell

Complex quality control draft office note by Bill Collins

Evaluation of parallel results

Daily forecast maps by Pete Caplan

Verification of precipitation by Hua-Lu Pan

Verification against rawinsondes timeseries by Suru Saha

Verification against rawinsondes horizontal maps by Suru Saha

Verification against rawinsondes vertical maps by Suru Saha

Conclusion

Global Parallel Home Page

Disclaimer