RE: [ferret_users] rho_un calculation

["JoshAB1995@xxxxxxxxxxx" <JoshAB1995@xxxxxxxxxxx>]

Hi William,

 

Thanks for the helpful info! So now I’ve come up with this script, is this what you meant by keeping a consistent reference pressure?

 

yes? use "./ECP85_sal.nc"

yes? use "./ECP85_temp.nc"

yes? let temp=ocn_temp[d=2]*1.00024

yes? let salt=ocn_sal[d=1]

yes? let potemp=THETA_FO(SALT,TEMP,Z[G=TEMP],2303.5) !Only define a single reference pressure to enable direct comparison between densities?

yes? let surfdens=rho_un(salt,potemp,0)

yes? let benthdens=rho_un(salt,potemp,2303.5)

yes? let y250strat= surfdens[k=1,l=250] - benthdens[k=12,l=250]

yes? list y250strat

 

Many thanks,

 

Josh

 

Sent from Mail for Windows 10

 

---

From: William Kessler - NOAA Federal <william.s.kessler@xxxxxxxx>
Sent: Wednesday, May 1, 2019 1:14:22 PM
To: JoshAB1995@xxxxxxxxxxx
Cc: ferret_users@xxxxxxxx
Subject: Re: [ferret_users] rho_un calculation

 

You second script is correct. First one is not. When comparing densities at different depths, always use a consistent reference pressure. Thus sigma-theta (what you call “surfdens” in your second script). That is valid for comparing depths above about 1000m:

strat_0_1000db=surfdens[z=1000]-surfdens[z=0]

To compare deeper, then you need to use sigma2 (your “benthdens” or similar). That would allow, say:

strat_1500_2500=benthdens[z=2500]-benthdens[z=1500]

The point here is to use THETA_FO (function of depth and reference pressure) and RHO_UN (at reference depth) to get values comparable across z-ranges. These need a consistent reference pressure.

Comparing density across very wide ranges (as you are trying to do with y250strat, using two different reference pressures) is harder, and generally not done. That garbles the combined effects of P,T,S in the nonlinear eqn of state to make values that are not comparable.

But try the alternatives and see how different they are! That will give you an estimate of the uncertainty of these calculations. Sigma-theta may be ok until you are getting into the abyss. I.e. how different are surfdens and benthdens at 2000db?

On Apr 30, 2019, at 10:33 AM, "JoshAB1995@xxxxxxxxxxx" <JoshAB1995@xxxxxxxxxxx> wrote:

Hi guys,

 

I’m currently calculating the density differences across the open ocean and have run into a little issue...

 

Slightly confused myself by finding two methods for calculating density on the expressions page, am I correct in thinking that:

 

yes? Use “./ECP85_sal.nc”

yes? Use “./ECP85_temp.nc”

yes? SET REGION/I=1:36/J=1:36

yes? LET salt = ocn_sal[d=1]
yes? LET t68 = 1.00024 * ocn_temp[d=2]
yes? LET dens = rho_un (salt[K=1], t68[K=1], Z[G=temp,K=1:10])

yes? LET y250strat = dens[k=1,l=250] – dens[k=10,l=250]

yes? List y250strat

 

Would be less accurate than calculating with the following script, because it’s generating the density based on surface water salinity and temperature values?:

 

yes? use "./ECP85_sal.nc"

yes? use "./ECP85_temp.nc"

yes? let temp=ocn_temp[d=2]*1.00024 !convert to different temp scale for accuracy

yes? let salt=ocn_sal[d=1]

yes? let potemp=THETA_FO(SALT, TEMP,Z[G=TEMP],0) !Define the potential temperature for surface ocean

yes? let potemp2=THETA_FO(SALT,TEMP,Z[G=TEMP],2303.5) !Define the potential temperature for benthic ocean

yes? let surfdens=rho_un(salt,potemp,0) !calculate density at surface 0 = dBar pressure

yes? let benthdens=rho_un(salt,potemp2,2303.5) !calculate density at 2303.5m deep, with 2303.5dBar pressure

yes? let y250strat= surfdens[k=1,l=250] - benthdens[k=12,l=250]

yes? list y250strat

 

Also is there anything obviously wrong with the script I’ve made? First time calculating ocean densities!

 

Cheers,

Josh

 

Sent from Mail for Windows 10