m_kliphuis@phys_uu_nl
New Member
Dear sir/miss,
I am working on a project in which we do paleo simulations (for the Late Cretaceous) with CCSM3. We are trying to get the ocean in an equilibrium state by first doing a coupled run of 50 years. We then use mean values of the last 10 years of this run to drive an ocean only run (pop, datm, dlnd, dice, cpl) for about 1000 years. Then we couple for 50 years again, decouple etc.
The results of the coupled part are fine. However once we uncouple and
run ocean only, the SALT in the upper layers decreases fast. At some places
it drops 5 PSU in 5 years !!!! Within 200 years the salt even becomes negative!
You would think that the model should stop when negative values arise
(because what do negative values mean ?) but it doesn't. I found out that this is caused by the fact that we use state_range_opt = 'enforce' in the pop_in namelist. With this setting the model does check if SALT is negative (in state_mod.F90) but if this is true it ignores this and just uses a SALT value of zero throughout the rest of the code (e.g. for the calculation of the density in the equation of state etc.)
Furthermore I checked if the forcing files (= last 10 year average coupled run)
in the datamodels (dlnd, dice, datm) are correct and they look fine as far as the freshwater budget terms are concerned (runoff, precipitation, evaporation)
I hope that there are people who can give us some advise for our problem
because we really don't know what to do at the moment. Some people say that the salt will come back once we couple again after the 1000 years ocean only run but we don't understand how this will ever get us in an equilibrium state ? Is there someone who can help us out ?
Thank's for your time
Kind regards,
Michael Kliphuis
Here's some more information:
1.
I am very interested in the parameters that we should use. For instance
for our run we used sfwf_formulation = 'restoring' but on this forum I found
the following quote:
"Running ocean-standalone experiments in the CCSM context (compset "C"
integrations) requires use of the 'partially-coupled' option for sfwf
and shf_formulation. This configuration of CCSM3 is unsupported and
requires significant expertise to run correctly."
2.
Should we (in order to keep salt stable) use the two modifications on the
freshwaterflux that are discussed in the article
"The NCAR Climate System Model Global Ocean Component" of Peter Gent et all
http://ams.allenpress.com/perlserv/.../1520-0442(1998)011<1287:TNCSMG>2.0.CO;2&ct=1
And does this mean I should set:
lsend_precip_fact = .true.
ladjust_precip = .true. ?
Below are some extracts from this article:
"In POP only (CCSM3) runs, the freshwater surface flux needs to have a
term that restores to observations in order to keep the salinity field
stable, but the coefficient is quite small"
"The net ocean freshwater flux requires an estimate of precipitation.
Monthly estimates were computed from the Microwave Sounding Unit (MSU)
data; see Spencer (1993). There is considerable uncertainty in these
estimates and the equilibrium values of sea surface salinity are very
sensitive to the forcing because it has no local feedback on the
freshwater flux. To overcome this, we allowed two modifications to the
freshwater flux. The first was to multiply the precipitation P by a
factor, f, that made the globally averaged P balance the globally
averaged evaporation E. The second was to have a local restoring term to
observed salinity that has zero global mean and as long a restoring
timescale as possible. Thus the open-ocean freshwater flux is given by:
....etc."
3. I can only put 10000 characters in this thread so I will put the pop_in parameters that we use in a subthread.
I am working on a project in which we do paleo simulations (for the Late Cretaceous) with CCSM3. We are trying to get the ocean in an equilibrium state by first doing a coupled run of 50 years. We then use mean values of the last 10 years of this run to drive an ocean only run (pop, datm, dlnd, dice, cpl) for about 1000 years. Then we couple for 50 years again, decouple etc.
The results of the coupled part are fine. However once we uncouple and
run ocean only, the SALT in the upper layers decreases fast. At some places
it drops 5 PSU in 5 years !!!! Within 200 years the salt even becomes negative!
You would think that the model should stop when negative values arise
(because what do negative values mean ?) but it doesn't. I found out that this is caused by the fact that we use state_range_opt = 'enforce' in the pop_in namelist. With this setting the model does check if SALT is negative (in state_mod.F90) but if this is true it ignores this and just uses a SALT value of zero throughout the rest of the code (e.g. for the calculation of the density in the equation of state etc.)
Furthermore I checked if the forcing files (= last 10 year average coupled run)
in the datamodels (dlnd, dice, datm) are correct and they look fine as far as the freshwater budget terms are concerned (runoff, precipitation, evaporation)
I hope that there are people who can give us some advise for our problem
because we really don't know what to do at the moment. Some people say that the salt will come back once we couple again after the 1000 years ocean only run but we don't understand how this will ever get us in an equilibrium state ? Is there someone who can help us out ?
Thank's for your time
Kind regards,
Michael Kliphuis
Here's some more information:
1.
I am very interested in the parameters that we should use. For instance
for our run we used sfwf_formulation = 'restoring' but on this forum I found
the following quote:
"Running ocean-standalone experiments in the CCSM context (compset "C"
integrations) requires use of the 'partially-coupled' option for sfwf
and shf_formulation. This configuration of CCSM3 is unsupported and
requires significant expertise to run correctly."
2.
Should we (in order to keep salt stable) use the two modifications on the
freshwaterflux that are discussed in the article
"The NCAR Climate System Model Global Ocean Component" of Peter Gent et all
http://ams.allenpress.com/perlserv/.../1520-0442(1998)011<1287:TNCSMG>2.0.CO;2&ct=1
And does this mean I should set:
lsend_precip_fact = .true.
ladjust_precip = .true. ?
Below are some extracts from this article:
"In POP only (CCSM3) runs, the freshwater surface flux needs to have a
term that restores to observations in order to keep the salinity field
stable, but the coefficient is quite small"
"The net ocean freshwater flux requires an estimate of precipitation.
Monthly estimates were computed from the Microwave Sounding Unit (MSU)
data; see Spencer (1993). There is considerable uncertainty in these
estimates and the equilibrium values of sea surface salinity are very
sensitive to the forcing because it has no local feedback on the
freshwater flux. To overcome this, we allowed two modifications to the
freshwater flux. The first was to multiply the precipitation P by a
factor, f, that made the globally averaged P balance the globally
averaged evaporation E. The second was to have a local restoring term to
observed salinity that has zero global mean and as long a restoring
timescale as possible. Thus the open-ocean freshwater flux is given by:
....etc."
3. I can only put 10000 characters in this thread so I will put the pop_in parameters that we use in a subthread.