Use grided AVISO SLA data as observation data

[Arvin_Yin]

Hi All

I want to use grided AVISO SLA data as observation data in my 4D-Var run.

My question is should I use grided AVISO SLA data directly in the observation file
or should I add mean sea level data to grided AVISO SLA data to make it become SSH data first
and then write it into observation file?

I read some papers and they all use the second method. But it seems that the programs
in ROMS matlab toolbox (load_ssh_aviso.m and d_ssh_obs.m) use grided AVISO SLA data directly.

In my 4D-Var run, I use HYCOM SSHA data to serve as initial conditions and boundary conditions
since HYCOM SSH data are all positive in my domain (the South China Sea). So, what should I do?

Thanks.

[xwbin]

I have the same problem.And I find that zeta changes after modifying theta_s,theta_b,hc or depth.My roms domain is South China Sea too.I think we can have a disscussion personaly.My qq is 648673743.

[lmp4]

ROMS zeta is "absolute SSH" in the sense that it includes time mean and time varying SSH with respect to a datum (i.e. zeta=0) that is a geopotential surface, i.e. if zeta=constant there is no associated pressure gradient force.

Altimetrists do not say "absolute" SSH so let me clarify:

In altimetry the equivalent quantity to ROMS zeta is close to the sum of mean dynamic topography (MDT) and sea level anomaly (SLA). [John Wilkin: IMCS Rutgers University]

The above was taken from viewtopic.php?f=14&t=1809, and it really helped me clarify using AVISO data with ROMS.

So I think you want to add the mean sea level data (MTD) to the grided AVISO SLA data to make it become this "absolute SSH".

Since ROMS zeta is an absolute variable I am a little unsure why you would initialize with HYCOM SSHA data? This means you are manually subtracting/adding mass to the ocean which should be there?

Also I think I remember seeing somewhere that load_ssh_aviso.m and d_ssh_obs.m shouldn't be taken as absolute codes to use. Just recommended scripts to load data and you can change them accordingly.