Strange salinity issue using GOTM/FABM(ERGOM)

[Karsten Bolding]

Hello

Dennis and I are working on a lake-setup for a reservoir where we want to use GOTM (the lake branch) and using ERGOM via FABM.

We could not make it work until we switched on salinity in obs.nml and set the salinity to 0. Lesson learned ERGOM has processes depending on salinity - right?

Second thing - why does the simulation blow up when using s_prof_method=0 - that should initialise sprof and S to 0. Any ideas?

Karsten

[Ulf Graewe]

> We could not make it work until we switched on salinity in obs.nml and set
> the salinity to 0. Lesson learned ERGOM has processes depending on
> salinity
> - right?

Hi Karsten,
yes, ERGOM needs the salinity to compute the oxygen saturation of sea
water, which depends on the salinity and temperature.

Best, ulf


Ulf Gr�we

Leibniz Institute for Baltic Sea Research Warnemuende
Dept. for Physical Oceanography and Instrumentation
Seestrasse 15
D-18119 Rostock-Warnemuende
Germany

Tel. : +49 (381) 5197-358
Fax : +49 (381) 5197-440

[Karsten Bolding]

Hi Ulf

What I don't understand (and that is more a GOTM question) is why there is a differens between setting s_prof_method=0 (which actually sets sprof and S to 0) and s_prof_method=1 and set the initial constant to 0.

I've tried to run and print out S during the simulation - and it is 0 in both cases during the entire simulation.

Karsten

On Wed, Apr 24, 2013 at 8:45 AM, Ulf Graewe <ulf.g...@io-warnemuende.de> wrote:

> We could not make it work until we switched on salinity in obs.nml and set
> the salinity to 0. Lesson learned ERGOM has processes depending on
> salinity
> - right?

Hi Karsten,
yes, ERGOM needs the salinity to compute the oxygen saturation of sea
water, which depends on the salinity and temperature.

Best,  ulf

Ulf Gräwe

Leibniz Institute for Baltic Sea Research Warnemuende
Dept. for Physical Oceanography and Instrumentation
Seestrasse 15
D-18119 Rostock-Warnemuende
Germany

Tel. : +49 (381) 5197-358
Fax  : +49 (381) 5197-440

--
You received this message because you are subscribed to the Google Groups "FABM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to fabm-users+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.

[Karsten Bolding]

Correction - the model actually blows up if the salinity is set to 0. Setting it to 0.0001 - and it works.

Here is the code in ERGOM that uses salt:

   osat_weiss=exp(aa1+aa2/tk+a3*log(tk)+a4*tk    &

              +s*(b1+(b2+b3*tk)*tk))*mol_per_liter

to me it seems s=0 is valid.

Karsten

On Wed, Apr 24, 2013 at 8:45 AM, Ulf Graewe <ulf.g...@io-warnemuende.de> wrote:

> We could not make it work until we switched on salinity in obs.nml and set
> the salinity to 0. Lesson learned ERGOM has processes depending on
> salinity
> - right?

Hi Karsten,
yes, ERGOM needs the salinity to compute the oxygen saturation of sea
water, which depends on the salinity and temperature.

Best,  ulf

Ulf Gräwe

Leibniz Institute for Baltic Sea Research Warnemuende
Dept. for Physical Oceanography and Instrumentation
Seestrasse 15
D-18119 Rostock-Warnemuende
Germany

Tel. : +49 (381) 5197-358
Fax  : +49 (381) 5197-440

[Gennadi Lessin]

Hello Karsten,

But does it work if You set newflux to other value than 1? In this case osat_weiss will not be used and saturation depends only on temperature.

Actually, do You use "older" or "newer" ergom version?

Gennadi

--
You received this message because you are subscribed to the Google Groups "GOTM-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to gotm-users+...@googlegroups.com.

[Jorn Bruggeman]

Hi Gennadi,

 

That reminds me: if you pick a model name for your new ERGOM model (to go under iow, right?) and send me the latest code, I’ll get it into the FABM repository right away. We’ll leave the old ERGOM model the same (i.e., its name will stay “ergom”), but feel free to use any name you like for your own model.

 

Cheers

 

Jorn

[Jorn Bruggeman]

Hi Karsten,

 

Can you take a look at gotm_fabm.nml? If you have salinity_relaxation_to_freshwater_flux=.true. in there, that would explain the problems; that caused division by zero if the whole column is free of salt. I’ve just fixed that in the gotm repository. After the fix (or after setting salinity_relaxation_to_freshwater_flux=.false.), I can use ERGOM without problems if salinity is 0, for both s_prof_method=0 and s_prof_method=1.

 

It might be an idea to remove salinity_relaxation_to_freshwater_flux from fabm/testcases/GOTM/gotm_fabm.nml, or to set it to .false. by default, since the resulting behaviour is more or less a hack that is not always wanted…

 

Cheers,

 

Jorn

[Dennis Trolle]

Hi Jorn,

 

Thanks, that was it. We had salinity_relaxation_to_freshwater_flux=.true.

By setting this to .false. we can run ERGOM if salinity is 0 using both s_prof_method=0 and s_prof_method=1.

 

cheers

Dennis

 

---

From: fabm-...@googlegroups.com [fabm-...@googlegroups.com] on behalf of Jorn Bruggeman [jo...@bolding-burchard.com]
Sent: 24 April 2013 15:02
To: fabm-...@googlegroups.com; gotm-...@googlegroups.com
Subject: RE: fabm-users:261 Strange salinity issue using GOTM/FABM(ERGOM)

[Karsten Bolding]

Hi Gennadi

I'm - unfortunately - using the old version as this is a collabotate project and we rely on using git to keep our code in sync.

Karsten

[Karsten Bolding]

Hello

Should we not consider adding a new institution as well. Gennadi did not work at IOW when he did. In the long run we can hope for a merge of the 2 versions - but I guess that will depend on how ERGOM as a community model takes of.

Karsten

[Ulf Graewe]

Hi all,
I think that Genadi should place his ERGOM version in a folder named by
his own institute.

I will work in the next three years with ERGOM in the Baltic Sea and try
to keep the FABM code in sync with the IOW ERGOM version. Thus, I would
prefer to work in iow/ergom.
It might be possible that we will have for some time two or three versions
of ERGOM floating around, but I will hope that we can sort that out.

Best, ulf

Am Do, 25.04.2013, 02:40, schrieb Karsten Bolding:
> Hello
>
> Should we not consider adding a new institution as well. Gennadi did not
> work at IOW when he did. In the long run we can hope for a merge of the 2
> versions - but I guess that will depend on how ERGOM as a community model
> takes of.
>
> Karsten
>
>
> On Wed, Apr 24, 2013 at 12:34 PM, Jorn Bruggeman
> <jo...@bolding-burchard.com>wrote:
>

>> Hi Gennadi,****
>>
>> ** **

>>
>> That reminds me: if you pick a model name for your new ERGOM model (to
>> go

>> under iow, right?) and send me the latest code, I嚙締l get it into the
>> FABM
>> repository right away. We嚙締l leave the old ERGOM model the same (i.e.,
>> its
>> name will stay 嚙箴rgom嚙�, but feel free to use any name you like for your
>> own model.****
>>
>> ** **
>>
>> Cheers****
>>
>> ** **
>>
>> Jorn****
>>
>> ** **
>>
>> *From:* fabm-...@googlegroups.com [mailto:fabm-...@googlegroups.com]
>> *On
>> Behalf Of *Gennadi Lessin
>> *Sent:* 24 April 2013 10:41
>> *To:* gotm-...@googlegroups.com; FABM-users
>> *Subject:* Re: [gotm-users:1354] Re: fabm-users:259 Strange salinity
>> issue using GOTM/FABM(ERGOM)****
>>
>> ** **
>>
>> Hello Karsten,****
>>
>> ** **

>>
>> But does it work if You set newflux to other value than 1? In this case
>> osat_weiss will not be used and saturation depends only on

>> temperature.***
>> *
>>
>> ** **
>>
>> Actually, do You use "older" or "newer" ergom version?****
>>
>> ** **
>>
>> Gennadi****
>>
>> ** **
>>
>> ** **
>>
>> ** **

>>
>> On Wed, Apr 24, 2013 at 10:46 AM, Karsten Bolding <

>> kar...@bolding-burchard.com> wrote:****

>>
>> Correction - the model actually blows up if the salinity is set to 0.

>> Setting it to 0.0001 - and it works.****
>>
>> ** **
>>
>> Here is the code in ERGOM that uses salt:****
>>
>> osat_weiss=exp(aa1+aa2/tk+a3*log(tk)+a4*tk &****
>>
>> +s*(b1+(b2+b3*tk)*tk))*mol_per_liter****
>>
>> to me it seems s=0 is valid.****
>>
>> ** **
>>
>> Karsten****
>>
>> ** **

>>
>> On Wed, Apr 24, 2013 at 8:45 AM, Ulf Graewe
>> <ulf.g...@io-warnemuende.de>

>> wrote:****

>>
>>
>> > We could not make it work until we switched on salinity in obs.nml and
>> set
>> > the salinity to 0. Lesson learned ERGOM has processes depending on
>> > salinity
>> > - right?
>>
>> Hi Karsten,
>> yes, ERGOM needs the salinity to compute the oxygen saturation of sea
>> water, which depends on the salinity and temperature.
>>
>> Best, ulf
>>
>>

>> Ulf Gr嚙緩e

>>
>> Leibniz Institute for Baltic Sea Research Warnemuende
>> Dept. for Physical Oceanography and Instrumentation
>> Seestrasse 15
>> D-18119 Rostock-Warnemuende
>> Germany
>>
>> Tel. : +49 (381) 5197-358
>> Fax : +49 (381) 5197-440
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups
>> "FABM-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an
>> email to fabm-users+...@googlegroups.com.
>> For more options, visit https://groups.google.com/groups/opt_out.
>>

>> ****
>>
>> ** **

>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups
>> "GOTM-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an

>> email to gotm-users+...@googlegroups.com.****

>>
>>
>> For more options, visit https://groups.google.com/groups/opt_out.
>>

>> ****
>>
>> ** **

>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups
>> "FABM-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an
>> email to fabm-users+...@googlegroups.com.
>> For more options, visit https://groups.google.com/groups/opt_out.
>>

>> ****

>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups
>> "FABM-users" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an
>> email to fabm-users+...@googlegroups.com.
>> For more options, visit https://groups.google.com/groups/opt_out.
>>
>>
>>
>
> --
> You received this message because you are subscribed to the Google Groups
> "FABM-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to fabm-users+...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>

Ulf Gr嚙緩e

[Ulf Graewe]

Hi all,
this is a follow up of a discussion I had with Gennadi.

Currently, several ERGOM versions exist. The current version in the
iow/ergom path is quite old fashioned and will no longer be maintained.
Our ideas are as follows: move the current version of ERGOM into the
gotm/ergom path, since the current version is simply a copy of the
GOTM-BIO implementation. Maybe also port it to F2003.
After that is done, I will place a state of the art ERGOM version into the
iow/ergom path and I will also be responsible to maintain it.

So, what do you think, should we keep the old ERGOM version for
historical/pedagogical reasons or immediately switch to the new version?
In the case someone needs the old version, I can be pulled out of git.

Cheers, ulf

>>> under iow, right?) and send me the latest code, I�ll get it into the
>>> FABM
>>> repository right away. We�ll leave the old ERGOM model the same (i.e.,
>>> its
>>> name will stay �ergom�), but feel free to use any name you like for

>>> Ulf Gr�we

> Ulf Gr�we

>
> Leibniz Institute for Baltic Sea Research Warnemuende
> Dept. for Physical Oceanography and Instrumentation
> Seestrasse 15
> D-18119 Rostock-Warnemuende
> Germany
>
> Tel. : +49 (381) 5197-358
> Fax : +49 (381) 5197-440
>
> --
> You received this message because you are subscribed to the Google Groups
> "FABM-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to fabm-users+...@googlegroups.com.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>

Ulf Gr�we