Held suarez docs #189
Held suarez docs #189
Conversation
…, and diagnsotics
|
@dennissergeev: Ignore the comment in the description. The deleted files have left no trace on the P/R. |
dennissergeev
added
docs
idealised
There was a problem hiding this comment.
This is really great @ntlewis - very impressively thorough! I've only made a few small edits/suggestions here.
I've made a bunch of additions to hs_forcing.F90 over here (Polvani-Kushner stratosphere, polar centred heating, QBO relaxation) and I'll be sure to update these docs when I eventually get around to a PR.
| The subroutine ``rayleigh_damping`` contains options to apply a linear relaxation to the zonal and meridional winds. | ||
|
|
||
| | **Held and Suarez linear drag** | ||
| | The default option for the ``rayleigh_damping`` subroutine (selected with ``relax_to_specified_wind=.FALSE.``) applies a linear drag to the zonal and meridional wind within the model boundary layer. The drag takes the form |
There was a problem hiding this comment.
I think the .FALSE. here (and elsewhere) might be better as a more pythonic False
|
|
||
| where :math:`\mathbf{u}_{r}` is the specified relaxation wind profile, and the overline denotes a zonal average. :math:`k_{f}` is the same as above. Note that in this instance, the relaxation is applied at all model levels, not just in the boundary layer. | ||
|
|
||
| The name of the netCDF files are specified with the namelist options ``u_wind_file`` and ``v_wind_file``, and should exclude the ``.nc`` suffix. Note: in the netCDF file, the :math:`u` and :math:`v` variables must be the same name as their respective file names, i.e., ``u_wind_file`` and ``v_wind_file``. The path to the file should be configured in the model's python run script (see the test cases for examples of how to do this). |
There was a problem hiding this comment.
I don't think there is actually a test case which takes in wind specified in a netCDF file.
|
|
||
|
|
||
|
|
||
| Top-down Newtonian damping |
There was a problem hiding this comment.
As a note, I have found the stability of the top-down routine to be very dependent on the choice of sigma levels. I think I only managed to get it to converge using even_sigma (as it is in the test case). I'm not sure if this is more widely known, but perhaps it should be noted here if so?
|
|
||
| **Other** | ||
|
|
||
| There are also optins for a tracer source and sink, ``trflux`` and ``trsink``. These are currently undocumented, however, as Isca only uses the ``hs_forcing`` when the model is dry and there are no tracers. |
There was a problem hiding this comment.
| There are also optins for a tracer source and sink, ``trflux`` and ``trsink``. These are currently undocumented, however, as Isca only uses the ``hs_forcing`` when the model is dry and there are no tracers. | |
| There are also options for a tracer source and sink, ``trflux`` and ``trsink``. These are currently undocumented, however, as Isca only uses the ``hs_forcing`` when the model is dry and there are no tracers. |
|
|
||
| **Other** | ||
|
|
||
| There are also optins for a tracer source and sink, ``trflux`` and ``trsink``. These are currently undocumented, however, as Isca only uses the ``hs_forcing`` when the model is dry and there are no tracers. |
There was a problem hiding this comment.
Perhaps change there are no tracers to there are no active tracers, since it is possible to add passive tracers in the dry model.
| **Top down** | ||
|
|
||
| :sigma_b: Top-of-boundary layer pressure, as for the Rayleigh damping. The same namelist variable is used for each of the ``rayleigh_damping``, ``newtonian_damping`` and ``top_down_newtonian_damping`` subroutines. | ||
| :lapse: Lapse rate :math:`\Gamma`. Units are :math:`\text{K}\,\text{km}^{-1}`. Default :math:`6\,\text{K}\,text{km}^{-1}`. |
There was a problem hiding this comment.
Missing \ on final units
| :lapse: Lapse rate :math:`\Gamma`. Units are :math:`\text{K}\,\text{km}^{-1}`. Default :math:`6\,\text{K}\,text{km}^{-1}`. | |
| :lapse: Lapse rate :math:`\Gamma`. Units are :math:`\text{K}\,\text{km}^{-1}`. Default :math:`6\,\text{K}\,\text{km}^{-1}`. |
|
|
||
| Authors | ||
| ---------- | ||
| This documentation was written by Neil Lewis, peer reviewed by X, and quality controlled by Y. |
There was a problem hiding this comment.
| This documentation was written by Neil Lewis, peer reviewed by X, and quality controlled by Y. | |
| This documentation was written by Neil Lewis, peer reviewed by Will Seviour, and quality controlled by Y. |
|
|
||
| Summary | ||
| ------- | ||
| This module is contains the 'physics options' used by Isca when the flag ``idealised_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. |
There was a problem hiding this comment.
Do we also want to make clear that if would be preferable to dofrom isca import DryCodeBase for this, not just setting the idealised_moist_model flag?
|
|
||
| Summary | ||
| ------- | ||
| This module is contains the 'physics options' used by Isca when the flag ``idealised_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. |
There was a problem hiding this comment.
I think we need American spelling for this one...
| This module is contains the 'physics options' used by Isca when the flag ``idealised_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. | |
| This module is contains the 'physics options' used by Isca when the flag ``idealized_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. |
| Namelist options | ||
| ---------------- | ||
|
|
||
| ``hs_forcing_mod`` is used when ``idealised_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. |
There was a problem hiding this comment.
| ``hs_forcing_mod`` is used when ``idealised_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. | |
| ``hs_forcing_mod`` is used when ``idealized_moist_model`` in ``atmosphere_nml`` is set to ``FALSE``. |
rosscastle
removed
priority:high
This PR addresses issue #155 by adding documentation for hs_forcing_mod.
Geoff, you're down to review this. Brett, you're down for quality control.
Denis: I have requested you as a reviewer also as there are a couple of 'deleted' files in the commit that were carried over from my master, and I'd like to remove them so the P/R is tidy. Can you assist with this?