Quetions about solar radiation as ROMS input forcing

[chysun]

Hi,
I'm a little confused weather I should use net solar radiation or downward solar radiation as ROMS input forcing.Could anyone tell me which one is correct and why should we use it ?

Thanks advance.
Haiyun

[LeaF_7]

Hi,
For my experience, if you defined bulk_fluxes in your cppdefs, you should use downward solar radiation.

If not, you should use the net one.

Hope that makes sense.

[wilkin]

You should use net solar radiation. This is what enters the ocean and heats it.

Atmospheric models frequently output downward and upward solar radiation separately, and in that case you need to compute the difference to get the net. Using downward flux only would neglect the fact that some of that radiation is reflected and does not enter the water.

Looking at the atmospheric model products served at http://nomads.ncdc.noaa.gov/thredds/catalog.html you will find the vast majority offer output for Downward_short_wave_flux and and Upward_short_wave_flux.

Meteorological models carry these separately because they have explicit models for the earth's albedo - which can change with time e.g. due to seasonal vegetation cover changes.

However, in my experience, most of these models use a constant albedo over the ocean (close to 0.06) that is not dependent on any other conditions (like sea state), so you could in principle just take 0.94*Downward_short_wave_flux and be close to the right answer (certainly within the range of the large systematic errors in most radiation products that exceed 20 W/m2 - see Zhao et al. 2013, Correcting surface solar radiation of two data assimilation systems against FLUXNET observations in North America, JGR Atmos., 118, 9552–9564, doi:10.1002/jgrd.50697)

[kate]

If you do any sea ice modeling, you want the downward shortwave (and longwave) radiation because the ice model will compute its own albedo and reflect sunlight accordingly. My branch has various albedo options for both ice and ocean, of course with the ocean albedo of 0.06 as one option.

[chysun]

Hi John and Kate,
Follow your ideas I get a better understanding of the procedure of air-sea interchange.I'm not using sea ice model but I will take care when I deal with it in the future.
Thanks for your opinions！

[fagundesmo]

Hi all,

I have faced the same issue but now that I have the net radiation how should I set my .h file so ROMS can read and use it properly?

Best Regards,

Matheus

[wilkin]

If you wish the net shortwave radiation to penetrate into the ocean then you #define SOLAR_SOURCE

This effectively the causes the net shortwave to be separated from the total heat flux, and for it to be successively absorbed going down into the water column. The vertical scale over which the shortwave radiation is absorbed is controlled by the choice Jerlov water type in ocean.in via the keyword WTYPE (see the comments at the end of ocean.in or the WikiROMS entry at https://www.myroms.org/wiki/Jwtype for more about this).

The remainder of the heat flux (i.e. net heat flux minus net shortwave) is absorbed in the surface-most cell.

So to do this you must provide ROMS with both net shortwave and net heat flux, or net shortwave and activate #define BULK_FLUXES to have ROMS compute sensible and latent heat flux from marine boundary layer conditions, and add this to net radiation to get net heat flux.

If your net shortwave resolves the diurnal cycle reasonably well (say, you have 3-hourly data from a forecast or reanalysis atmospheric model) then that carries through to the surface forcing.

If you only have daily average net shortwave, but you think the diurnal cycle of solar heating might be important in your dynamics, then you can #define DIURNAL_SRFLUX and ROMS will distribute the daily heating over an idealized local solar day (according to latitude, year-day and solar angle) to achieve the same net input of heat. Only use DIURNAL_SRFLUX if your shortwave data are on a 24-hour or longer average or things get messed up.

[fagundesmo]

Hi John,

thank you for the information, I'll try this here.

Best Regards,

Matheus