Observations for normalization?

Discussion about tangent linear and adjoint models, variational data assimilation, and other related issues.

Moderators: arango, robertson

Post Reply
Message
Author
jsaynisch

Observations for normalization?

#1 Unread post by jsaynisch »

Hello,

I'm setting up an assimilation system with ROMS. Now I want to compute the normalization factors and the system is asking for an observation file.
In "s4dvar.in" this file (OBSname) is marked as input/output. I do not have such a file and are not willing to create one (jet). Did I something wrong?
I thought that observations are not needed for the calculation of the normalization factors. Did I miss something?
How can I prevent ROMS from asking for this file?

Thank you,
Jan

jsouza
Posts: 31
Joined: Tue Apr 10, 2012 10:55 pm
Location: Metocean Solutions

Re: Observations for normalization?

#2 Unread post by jsouza »

The system is going to ask you for all the input files... you should just give it empty files (filled with zeros) and it should work. You will have the same problem witht he std files - and the solution is the same!

Cheers

User avatar
jivica
Posts: 172
Joined: Mon May 05, 2003 2:41 pm
Location: The University of Western Australia, Perth, Australia
Contact:

Re: Observations for normalization?

#3 Unread post by jivica »

You are right, you do not need obs file just to compute norms, they depend only on the grid. Whenever you change something with the grid you have to recompute norms. They are expensive, so be sure the grid is final before kick. To be more precise you need to define in addition error decorrelation scales as well.
In some version of ROMS I believe Hernan fixed that (not needing obs file).
If still struggled with must have one, then just create single observation file, dummy one. Here is example of cdl file:

Code: Select all

dimensions:
	survey = 1 ;
	state_variable = 7 ;
	datum = 1 ;
variables:
	char spherical(survey) ;
		spherical:long_name = "grid type logical switch" ;
		spherical:option_T = "spherical" ;
		spherical:option_F = "Cartesian" ;
	int Nobs(survey) ;
		Nobs:long_name = "number of observations with the same survey time" ;
	double survey_time(survey) ;
		survey_time:long_name = "survey time" ;
		survey_time:units = "day" ;
	double obs_variance(state_variable) ;
		obs_variance:long_name = "global time and space observation variance" ;
		obs_variance:units = "squared state variable units" ;
	int obs_type(datum) ;
		obs_type:long_name = "model state variable associated with observation" ;
		obs_type:option_1 = "free-surface" ;
		obs_type:option_2 = "vertically integrated u-momentum component" ;
		obs_type:option_3 = "vertically integrated v-momentum component" ;
		obs_type:option_4 = "u-momentum component" ;
		obs_type:option_5 = "v-momentum component" ;
		obs_type:option_6 = "potential temperature" ;
		obs_type:option_7 = "salinity" ;
	double obs_time(datum) ;
		obs_time:long_name = "time of observation" ;
		obs_time:units = "day" ;
	double obs_depth(datum) ;
		obs_depth:long_name = "depth of observation" ;
		obs_depth:negative = "downwards" ;
	double obs_Xgrid(datum) ;
		obs_Xgrid:long_name = "x-grid observation location" ;
		obs_Xgrid:units = "nondimensional" ;
	double obs_Ygrid(datum) ;
		obs_Ygrid:long_name = "y-grid observation location" ;
		obs_Ygrid:units = "nondimensional" ;
	double obs_Zgrid(datum) ;
		obs_Zgrid:long_name = "z-grid observation location" ;
		obs_Zgrid:units = "nondimensional" ;
	double obs_error(datum) ;
		obs_error:long_name = "observation error covariance" ;
		obs_error:units = "squared state variable units" ;
	double obs_value(datum) ;
		obs_value:long_name = "observation value" ;
		obs_value:units = "state variable units" ;
	int obs_provenance(datum) ;
		obs_provenance:long_name = "observation source" ;
	double obs_lat(datum) ;
		obs_lat:long_name = "observation latitude" ;
		obs_lat:units = "degrees" ;
	double obs_lon(datum) ;
		obs_lon:long_name = "observation longitude" ;
		obs_lon:units = "degrees" ;

// global attributes:
		:title = "Observations" ;
		:author = "Ivica Janekovic" ;
		:date = "06-Nov-2013 16:09:14" ;
		:type = "ROMS observations" ;
		:Conventions = "CF-1.0" ;
}
Good luck,
Ivica

jsaynisch

Re: Observations for normalization?

#4 Unread post by jsaynisch »

Hey you two,
thank you very much for your replies this helped a lot!!!

Cheers, Jan

Post Reply