1. High-resolution air-sea modeling of the Philippines winter monsoon
Julie Pullen Paul May Jim Doyle
NRL-Monterey
Cedric Chavanne Pierre Flament
University of Hawaii

2. Outline Model set-up: COAMPS & NCOM
Air-sea interaction features during winter monsoon
COAMPS-QuikSCAT comparison
COAMPS/NCOM February 2005 statistics
Global HYCOM & global NCOM error statistics
Plans

3. Model domain & features
The 3-km NCOM uses 8 km COAMPS forcing (nest 3). Global NCOM is available every 6 hours and global HYCOM is available every 24 hours. This was the frequency of update at the boundary for each simulation.
COAMPS with 72, 24, 8 km nests
3 km NCOM with initial and boundary conditions from 1.) global NCOM & 2.) global HYCOM

4. Wind jets are a persistent feature during winter monsoon conditions
Wind jets are a persistent feature during winter monsoon conditions. Frames every 3 hours; every 5/8 vector shown for atmosphere/ocean.

5. orographically-induced air-sea interaction

6. satellite observations
From cedric chavanne & pierre flament, university of hawaii

7. model/obs comparison
21 UTC 7 february 2005

8. model/obs comparison
21 UTC 11 february 2005

9. COAMPS 8-km february 2005 (20 days) mean standard deviation
Some wake areas contain areas of elevated variability

10. Frames every 3 hours; every 5/8 vector shown for atmosphere/ocean.

11. NCOM 3-km february 2005 (20 days) mean uses global HYCOM
uses global NCOM
Mean current and temperature patterns differ significantly between the two runs using global HYCOM and global NCOM, although both simulations present the double gyre at the mouth of Mindoro Straits that was not in the initialization fields

12. uses global HYCOM
uses global NCOM
february 2005 (20 days)
mean
standard deviation

13. Evaluation of SST from global HYCOM
This analysis uses NCODA to evaluate global HYCOM against available surface temperature obs (satellite, ship). Analyses were conducted using HYCOM as first guess every 24 hours (no cycling). HYCOM is too warm and has large errors, especially in the southeast/Mindanao Strait. In the Philippine Sea (Pacific) mean bias & rmse are low.

14. Comparison of global models for Philippines region
90 subsurface profiles (xbt’s and argo floats) during February 2005

15. Modeling studies planned
Evaluate global HYCOM/NCODA as a source of i.c.’s & b.c.’s
Study gap flow/ocean eddy interactions and tropical dynamics using 2-way coupling with global HYCOM i.c.’s & b.c.’s
Assess relative importance of atmospheric forcing, b.c.’s & i.c.’s in modeled ocean dynamics

16. Upcoming Conferences
Gordon Research Conference on Coastal Ocean Modeling, June 17-22, 2007, New Hampshire (space limited)
7th Conference on Coastal Atmospheric and Oceanic Prediction and Processes, September 2007, San Diego (joint with 7th Symposium on the Urban Environment): abstracts due May 11

17. Frames every 6 hours

18. Evaluation of SST for global NCOM

19. model/obs comparison
21 UTC 8 february 2005