options for calculating wind stress from wind speed

[Haiyan]

Hi everyone,

I found the wind stress near the coast was sometimes weird and model would blow up at some points next to the land, so I download the wind speed and calculated the wind stress. The wind speed I downloaded is at 10 m.

Recently, I am confused about the calculation from wind speed to wind stress. There is several options to calculate the wind stress, such as Large and Pond,1981; Smith 1988; Wu 1982. The bulk flux in roms uses Liu et al(1979).

I am confused about:
1, which option is better than others?
2, for these options, it uses wind speed at 10m(or at other height) and get the surface wind stress?

[kate]

What I do in ROMS is to use a global wind product at some altitude (10 m, 2 m, doesn't matter as long as you set BLK_ZW correctly). ROMS then interpolates to all the ROMS ocean points. What can happen is that the ROMS land mask and the winds land mask are different near the coast, so that land-type winds are used over the coastal ocean. To prevent this, we pre-process the winds so that ocean values are extrapolated into the land. ROMS sees these winds after this step and only gets oceanic winds to use everywhere. Raf Dussin gave me his preprocessing tool (a Sosie fork).

You are asking a second question, though. The trunk ROMS supports the Fairall et al. bulk fluxes. Some people don't like them, though, so we have brought in a Large and Yeager, 2009 variant of bulk fluxes. I can't say which is better - I just use the latter.

[Haiyan]

Thanks very much, kate.

I agree with you about preprocessing the wind. But now I am not sure how to extrapolate properly.

You are asking a second question, though. The trunk ROMS supports the Fairall et al. bulk fluxes. Some people don't like them, though, so we have brought in a Large and Yeager, 2009 variant of bulk fluxes. I can't say which is better - I just use the latter.

'the Large and Yeager, 2009 variant of bulk fluxes' is provided in roms? I just find BULK_FLUXES and NL_BULK_FLUXES [Fairall et al, 1996] are provided in roms.

[detraceyb]

A Correction for Land Contamination of Atmospheric Variables near Land–Sea Boundaries

[kate]

'the Large and Yeager, 2009 variant of bulk fluxes' is provided in roms? I just find BULK_FLUXES and NL_BULK_FLUXES [Fairall et al, 1996] are provided in roms.

It's in my branch of ROMS, which is full of all sorts of bizarre things. Search on "roms github kshedstrom" and you should find it.

[Haiyan]

A Correction for Land Contamination of Atmospheric Variables near Land–Sea Boundaries

Thanks detraceyb. I will read this paper to find a method.

[Haiyan]

'the Large and Yeager, 2009 variant of bulk fluxes' is provided in roms? I just find BULK_FLUXES and NL_BULK_FLUXES [Fairall et al, 1996] are provided in roms.

It's in my branch of ROMS, which is full of all sorts of bizarre things. Search on "roms github kshedstrom" and you should find it.

Hi kate, I have downloaded your version from https://github.com/kshedstrom/roms. There is a 'CCSM_FLUXES' cpp option. I do not think I find the 'Large and Yeager, 2009' bulk_fluxes.?

[kate]

CCSM_FLUXES is the Large and Yeager, 2009, as used (more or less) by CESM.