Start AUS2200 documentation #32
Conversation
|
I'm not sure if I can get a review in time but it'll be the first thing I do tomorrow morning |
|
Thanks, this is only the start, so don't spend too much time on it. Its more about having something online in preparation for the meeting (3:30pm). I'll submit another pull request when its done and it can be properly reviewed then. |
There was a problem hiding this comment.
Overall is good, see comments about step vs task for consistency. I didn't add this before, make sure you have enough space between text and figure, it's ok for some but others don't have any space in between. Finally the page is quite long, I know some documentation just goes on indefinitely and so far it makes sense to keep all the steps in the one page. However for the last part i.e. ERA5-Land vs BARRA and netcdf output maybe you could put them in a separate page? It's really a matter of preference, so not a big deal. Quite a lot of work! :-)
Wiki/_toc.yml
Outdated
There was a problem hiding this comment.
I would probably put this under UM, as it's a UM configuration
There was a problem hiding this comment.
I thought about that, but the UM section is under ACCESS, which also has MOM and CABLE under it, so I'm thinking that's for coupled models?
There was a problem hiding this comment.
Then lets' pull that out and make it all the same level
ACCESS, UM, MOM, CABLE, then ACCESS can link to the others seems like a better structure
Wiki/models/models-aus2200.md
Outdated
| The tasks `install_cold_idl` and `install_cold_hpc` symlink the appropriate STASHmaster, ancil configurations and prebuilt software into the suite share directory. | ||
|
|
||
| ### `install_ec_startdata` | ||
| In this version of the suite, this task does nothing as the initial conditions are derived from data already present on disk. |
There was a problem hiding this comment.
It's a bit odd that a task which is not run has its own header , either you do the same for the first two tasks described in the sentence above, or Add another sentence to that paragraph to say this task is skipped
Wiki/models/models-aus2200.md
Outdated
| This step is used to carry over soil moisture from a previous run of the suite. When performing restart runs, it is advisable to carry the soil moisture through the entire run, as soil moisture has a much longer spin-up time than atmospheric processes. The restart run procedure is explained in more detail in the [Restart and Spinup](#restart-and-spinup) section below. This task is used to partially serialise restart runs, it can be submitted at any time and will wait until the correct dump file has been created from a previous run of the suite before creating the soil moisture ancillary file. The addition of this ancillary file into the input dump is controlled by the `1` entry under 'Configure ancils and initialise dump fields'. | ||
|  | ||
|
|
||
| The creation of this ancillary file is controlled by the `SOIL_MOISTURE_CARRYOVER` setting under 'Cycling options' under the 'Nesting Suite' area in the rose suite configuration. Note that if this is set to false, the addition of the soil moisture field must be removed from the input dump. This is done by right-clicking on the row in the 'Configure ancils and initialise dump fields' area and selecting 'Ignore this section'. |
There was a problem hiding this comment.
| The creation of this ancillary file is controlled by the `SOIL_MOISTURE_CARRYOVER` setting under 'Cycling options' under the 'Nesting Suite' area in the rose suite configuration. Note that if this is set to false, the addition of the soil moisture field must be removed from the input dump. This is done by right-clicking on the row in the 'Configure ancils and initialise dump fields' area and selecting 'Ignore this section'. | |
| The creation of this ancillary file is controlled by the `SOIL_MOISTURE_CARRYOVER` setting under 'Cycling options' under the 'Nesting Suite' area in the rose suite configuration. | |
| Note that if this is set to false, the addition of the soil moisture field must be removed from the input dump. This is done by right-clicking on the row in the 'Configure ancils and initialise dump fields' area and selecting 'Ignore this section'. |
Wiki/models/models-aus2200.md
Outdated
|
|
||
| ### `convert_to_nc` | ||
|
|
||
| This task converts the raw UM output from the previous cycle to netCDF format. It is advised to retain this step, as the converted netCDF output tales up around 1/6th the disk space of the raw UM output. This step can also act as a 'copy output to final location' step. This task is only run when 'Convert output to netCDF' is set to true under 'General Run Options' (`{{CONVERT_TO_NETCDF}}` in `rose-suite.conf`). The final location of the netCDF files is set in the 'Path to converted netCDF files' configuration option (`{{NC_OUTPUT_PATH}}` in `rose-suite.conf`) |
There was a problem hiding this comment.
| This task converts the raw UM output from the previous cycle to netCDF format. It is advised to retain this step, as the converted netCDF output tales up around 1/6th the disk space of the raw UM output. This step can also act as a 'copy output to final location' step. This task is only run when 'Convert output to netCDF' is set to true under 'General Run Options' (`{{CONVERT_TO_NETCDF}}` in `rose-suite.conf`). The final location of the netCDF files is set in the 'Path to converted netCDF files' configuration option (`{{NC_OUTPUT_PATH}}` in `rose-suite.conf`) | |
| This task converts the raw UM output from the previous cycle to netCDF format. It is advised to retain this step, as the converted netCDF output takes up around 1/6th the disk space of the raw UM output. This step removes the need to run the ACCESS-Archiver and can be used to move the output to its final location. This task is only run when 'Convert output to netCDF' is set to true under 'General Run Options' (`{{CONVERT_TO_NETCDF}}` in `rose-suite.conf`). The final location of the netCDF files is set in the 'Path to converted netCDF files' configuration option (`{{NC_OUTPUT_PATH}}` in `rose-suite.conf`) |
Wiki/models/models-aus2200.md
Outdated
| . | ||
|
|
||
| Spin up hours is only used in restart runs, and determines how many hours of the model run are estimated to be required to 'spin-up' the atmosphere model. Typically, the model is restarted every 24 hours, and run for `24 + SPINUP_HOURS` model hours, with the first `SPINUP_HOURS` discarded. In order for the soil moisture carryover to work, the suites must adhere to a strict naming convention that allows the `create_soil_moisture_ancil` task to derive the name of the previous restart suite. The convention is as follows: | ||
| ```{note} |
There was a problem hiding this comment.
| ```{note} | |
| ```{admonition} Naming convention |
There was a problem hiding this comment.
This will create a box like for the note but with a title, I think something like this, feel free to use a better title will work better for this case, as you're referring to it beforehand it's more an example/clarification than a "note"
| </terminal-window> | ||
| With the rose config editor window open, change `INITIAL_CYCLE_POINT` to match the date stamp of the directory, set `FINAL_CYCLE_POINT` to 20160602T1800. Re-enable `SOIL_MOISTURE_CARRYOVER` and ensure `SPINUP_HOURS` is set to 24. Ensure that item `1` in the 'Configure ancils and initialise dump fields' section of the `um` task has been re-enabled. | ||
|
|
||
|  |
There was a problem hiding this comment.
I'm just being fussy here but maybe slow this down a notch if possible?
There was a problem hiding this comment.
Can do, but will need to recreate the gif. The actual footage is 30 seconds, I'll try 50% speedup and see how that looks.
Actually... Turns out you can just embed mp4 in jupyter books, I'll give that a go.
There was a problem hiding this comment.
Not critical, wait and see what others think
Beginning of the AUS2200 documentation. I have a meeting with NRI tomorrow about this and want something to show.