offline and online MOC calculation for CESM/POP

[zdliu]

Dear,I am using cesm1.0.4 to simulate paleoclimate.
1) In the model, I turn off the calculation of MOC. Now I want to calculate it, but due to POP's non-standard ocean grids , I am running into difficulty with the MOC computation in POP, So I want to know whether there is some codes for offline MOC calculation.

2) And also I have read some old posts in CGD forum. There is one of them (https://bb.cgd.ucar.edu/node/1000043) which give some modified diagnostics file to get MOC computed online.In there njn01 has said that++++++++++++++++++++++++++++++In order to select the global-MOC-only option, you'll need
1) the modified diagnostics module
2) a modified version of the "pop_in" fileTo acquire the modified diagnostics file, go to
http://roskilde.eas.purdue.edu/~cdpl/phpwiki++++++++++++++++++++++++++++++But now I can not get the file from above website. So who can tell me where I can get the modified diagnostics file?Thanks. zedong

 

[bates]

Zedong,The modified code you refer to above is out of date, and the modifications have since been incorporated into the released code. The offline MOC code you want has been addressed in another bulletin board post, which you can find here:https://bb.cgd.ucar.edu/node/1001868#comment-1006891
Susan 

 

[bates]

Zedong,The modified code you refer to above is out of date, and the modifications have since been incorporated into the released code. The offline MOC code you want has been addressed in another bulletin board post, which you can find here:https://bb.cgd.ucar.edu/node/1001868#comment-1006891
Susan 

 

[bates]

Zedong,The modified code you refer to above is out of date, and the modifications have since been incorporated into the released code. The offline MOC code you want has been addressed in another bulletin board post, which you can find here:https://bb.cgd.ucar.edu/node/1001868#comment-1006891
Susan 

 

[bblack@berkeley_edu]

Hi Susan, I have a similar set of questions to Zedong. I have output from a number of paleo runs (cesm1_0_5) with a non-present-day ocean grid. I would like to calculate some measure of the strength of the MOC. There is an output variable, MOC, which I have turned on, and the model does compute values for MOC. Will this variable be calculated correctly even in my paleo configuration? In the history files, MOC is 4-D: time, moc_comp, moc_z, and lat_aux_grid. moc_comp has values of 0,1,2  which I assume are cross sections through the ocean at different longitudes. Where are these cross sections located? Is there some way to compute/record different cross sections? And is there a better/more flexible diagnostic for this available in the most recent POP diagnostics package? Thanks very much, Ben

 

[bblack@berkeley_edu]

Hi Susan, I have a similar set of questions to Zedong. I have output from a number of paleo runs (cesm1_0_5) with a non-present-day ocean grid. I would like to calculate some measure of the strength of the MOC. There is an output variable, MOC, which I have turned on, and the model does compute values for MOC. Will this variable be calculated correctly even in my paleo configuration? In the history files, MOC is 4-D: time, moc_comp, moc_z, and lat_aux_grid. moc_comp has values of 0,1,2  which I assume are cross sections through the ocean at different longitudes. Where are these cross sections located? Is there some way to compute/record different cross sections? And is there a better/more flexible diagnostic for this available in the most recent POP diagnostics package? Thanks very much, Ben

 

[bblack@berkeley_edu]

Hi Susan, I have a similar set of questions to Zedong. I have output from a number of paleo runs (cesm1_0_5) with a non-present-day ocean grid. I would like to calculate some measure of the strength of the MOC. There is an output variable, MOC, which I have turned on, and the model does compute values for MOC. Will this variable be calculated correctly even in my paleo configuration? In the history files, MOC is 4-D: time, moc_comp, moc_z, and lat_aux_grid. moc_comp has values of 0,1,2  which I assume are cross sections through the ocean at different longitudes. Where are these cross sections located? Is there some way to compute/record different cross sections? And is there a better/more flexible diagnostic for this available in the most recent POP diagnostics package? Thanks very much, Ben

 

[njn01]

Ben,
The standard POP2 model (including out-of-the-box cesm1_0_5) generates a 5D MOC: (time, transport_reg, moc_comp, moc_z, lat_aux_grid). These dimensions are further explained in the netCDF file by the following variables:moc_components =
  "Eulerian Mean",
  "Eddy-Induced (bolus)",
  "Submeso" ;

 transport_components =
  "Total",
  "Eulerian-Mean Advection",
  "Eddy-Induced Advection (bolus) + Diffusion",
  "Eddy-Induced (bolus) Advection",
  "Submeso Advection" ;

 transport_regions =
  "Global Ocean - Marginal Seas",
  "Atlantic Ocean + Mediterranean Sea + Labrador Sea + GIN Sea + Arctic Ocean + Hudson Bay" ;

You can also look through the model code diags_on_lat_aux_grid.F90 and follow the comments to get a better idea of how the MOC is computed.In a paleo case, you set n_transport_reg = 1 in the transports namelist, which means that your MOC would have transport_regions = 0 in the netCDF file, or Global Ocean - Marginal Seas.  So your output is a global average MOC, not "cross-sections through the ocean at different longitudes." As far as we know, the global MOC computation is working correctly for paleo applications.
As far as I know, the off-line diagnostics package does not contain any updates to the MOC computations.

 

[njn01]

Ben,
The standard POP2 model (including out-of-the-box cesm1_0_5) generates a 5D MOC: (time, transport_reg, moc_comp, moc_z, lat_aux_grid). These dimensions are further explained in the netCDF file by the following variables:moc_components =
  "Eulerian Mean",
  "Eddy-Induced (bolus)",
  "Submeso" ;

 transport_components =
  "Total",
  "Eulerian-Mean Advection",
  "Eddy-Induced Advection (bolus) + Diffusion",
  "Eddy-Induced (bolus) Advection",
  "Submeso Advection" ;

 transport_regions =
  "Global Ocean - Marginal Seas",
  "Atlantic Ocean + Mediterranean Sea + Labrador Sea + GIN Sea + Arctic Ocean + Hudson Bay" ;

You can also look through the model code diags_on_lat_aux_grid.F90 and follow the comments to get a better idea of how the MOC is computed.In a paleo case, you set n_transport_reg = 1 in the transports namelist, which means that your MOC would have transport_regions = 0 in the netCDF file, or Global Ocean - Marginal Seas.  So your output is a global average MOC, not "cross-sections through the ocean at different longitudes." As far as we know, the global MOC computation is working correctly for paleo applications.
As far as I know, the off-line diagnostics package does not contain any updates to the MOC computations.

 

[njn01]

Ben,
The standard POP2 model (including out-of-the-box cesm1_0_5) generates a 5D MOC: (time, transport_reg, moc_comp, moc_z, lat_aux_grid). These dimensions are further explained in the netCDF file by the following variables:moc_components =
  "Eulerian Mean",
  "Eddy-Induced (bolus)",
  "Submeso" ;

 transport_components =
  "Total",
  "Eulerian-Mean Advection",
  "Eddy-Induced Advection (bolus) + Diffusion",
  "Eddy-Induced (bolus) Advection",
  "Submeso Advection" ;

 transport_regions =
  "Global Ocean - Marginal Seas",
  "Atlantic Ocean + Mediterranean Sea + Labrador Sea + GIN Sea + Arctic Ocean + Hudson Bay" ;

You can also look through the model code diags_on_lat_aux_grid.F90 and follow the comments to get a better idea of how the MOC is computed.In a paleo case, you set n_transport_reg = 1 in the transports namelist, which means that your MOC would have transport_regions = 0 in the netCDF file, or Global Ocean - Marginal Seas.  So your output is a global average MOC, not "cross-sections through the ocean at different longitudes." As far as we know, the global MOC computation is working correctly for paleo applications.
As far as I know, the off-line diagnostics package does not contain any updates to the MOC computations.

 

[bblack@berkeley_edu]

As a follow-up to my previous question, I am posting an email exchange with njn01 regarding MOC diagnostics for paleo simulations:You mentioned that the MOC computed when n_transport_reg = 1 is a global average MOC. What do you mean, in practice, by "global average"? I assume that moc_z refers to depth and lat_aux_grid is latitude. So is the MOC variable a zonal average?Ultimately I'd like to have a single number that represents the strength of the MOC at any given time. Do you have any suggestions for the best metric to represent this?Reply from NJN01:Your interpretation of the global average MOC iscorrect: the MOC is averaged over all longitudes for eachlatitude; hence MOC on the lat,depth grid.As for a meaningful single-number representation of MOC, that willdepend on the characteristics of your paleo MOC.  In the present-day runs, we sometimes use max MOC, but really you need to baseyour metric on your paleo MOC.

 

[bblack@berkeley_edu]

As a follow-up to my previous question, I am posting an email exchange with njn01 regarding MOC diagnostics for paleo simulations:You mentioned that the MOC computed when n_transport_reg = 1 is a global average MOC. What do you mean, in practice, by "global average"? I assume that moc_z refers to depth and lat_aux_grid is latitude. So is the MOC variable a zonal average?Ultimately I'd like to have a single number that represents the strength of the MOC at any given time. Do you have any suggestions for the best metric to represent this?Reply from NJN01:Your interpretation of the global average MOC iscorrect: the MOC is averaged over all longitudes for eachlatitude; hence MOC on the lat,depth grid.As for a meaningful single-number representation of MOC, that willdepend on the characteristics of your paleo MOC.  In the present-day runs, we sometimes use max MOC, but really you need to baseyour metric on your paleo MOC.

 

[bblack@berkeley_edu]

As a follow-up to my previous question, I am posting an email exchange with njn01 regarding MOC diagnostics for paleo simulations:You mentioned that the MOC computed when n_transport_reg = 1 is a global average MOC. What do you mean, in practice, by "global average"? I assume that moc_z refers to depth and lat_aux_grid is latitude. So is the MOC variable a zonal average?Ultimately I'd like to have a single number that represents the strength of the MOC at any given time. Do you have any suggestions for the best metric to represent this?Reply from NJN01:Your interpretation of the global average MOC iscorrect: the MOC is averaged over all longitudes for eachlatitude; hence MOC on the lat,depth grid.As for a meaningful single-number representation of MOC, that willdepend on the characteristics of your paleo MOC.  In the present-day runs, we sometimes use max MOC, but really you need to baseyour metric on your paleo MOC.