Toposplit #42
Merged
Toposplit #42
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Simplify the statement on which thermal distribution to use (standard vs HRRR).
Update the config to only allow one parallel run per PR and stop any previous one on a new push. Also remove the caching of GCC since it did not work and you can't cache OS level dependencies.
Centralize the file comparison to one method across all tests.
The check for a timeseries against a single date is flawed. More importantly, full WRF testing is yet to be done.
Logging the start and end date for SMRF seems unnecessary, given that we log each processed time step.
Also avoid using `setattr` when we can just straight assign to `self.` and update some docstrings.
Update the reference from topocalc that called `shade` to the recently renamed `illumination_angles` since that is the actual calculated property.
Will be reused soon in Toposplit. Also lints code and updates docstrings.
When configuring to only output net solar, the albedo needs precipitation, which requires air temperature and vapor pressure.
This is one of the most involved logic and needed by quite a few other variables.
jomey
force-pushed
the
toposplit
branch
3 times, most recently
from
9449b07 to
276c3f7
Compare
arobledano
reviewed
Nov 7, 2025
Member
Author
|
Thanks for checking @arobledano |
This implements the logic presented in Olson et al. to topographically downscale DSWRF from HRRR and calculate net solar.
The method previously returned the calculated horizon angles to save compute time on subsequent calls. This fix restores that.
Output and calculation variables were not reset when the sun was down, causing non-zero values from the last hour before sunset to be carried over.
This class was never executed with the missing __init__.py in the folder. Tests were also not passing, which are fixed with this as well.
The combination was not possible due to only checking albedo with standard solar.
Had a typo for GHI vis and the init of the class was also not setting up the correct names.
Identical to AWSM and eases troubleshooting.
And delete the existing ones.
Now that all variables get the fully parsed .ini dict passed, the value for threads can be read from there instead of copying it over.
Should speed up any topo domain of larger than 1000x1000
arobledano
approved these changes
Nov 14, 2025
Contributor
arobledano
left a comment
arobledano
left a comment
There was a problem hiding this comment.
3Cups seal of approval as well
Cythonize Toposplit
Now that we rely on Cython to build the C files for us, we no longer need the conditional logic in setup.py that looks for them.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds the first version of the toposplit #25 using the math presented in the recently submitted paper.
Note that the following models have not been implemented:
Other difference to the paper:
Important
Needs the recently upgraded version of Topocalc
Other things to note
The paper used
skyfieldto calculate the sun position.A follow up issue to get that added to SMRF is logged #41