Ouranosinc / xclim / 11485787478

90%

Build Type

push

github

Committed by web-flow

Commit Message

Fix conversion of units of multivariate DataArray in sdba (#1972)

<!--Please ensure the PR fulfills the following requirements! -->
<!-- If this is your first PR, make sure to add your details to the
AUTHORS.rst! -->
### Pull Request Checklist:
- [ ] This PR addresses an already opened issue (for bug fixes /
features)
    - This PR fixes #xyz
- [ ] Tests for the changes have been added (for bug fixes / features)
- [ ] (If applicable) Documentation has been added / updated (for bug
fixes / features)
- [ ] CHANGELOG.rst has been updated (with summary of main changes)
- [ ] Link to issue (:issue:`number`) and pull request (:pull:`number`)
has been added

### What kind of change does this PR introduce?

Conversion of units of multivariate DataArray is now properly handled
`sdba.TrainAdjust` and `sdba.Adjust`. There was a bug where the units
could be changed before a conversion of the magntitudes could occur.

e.g. 

```python
out = sdba.dOTC.adjust(sim=sim,ref=ref,hist=hist)
```
`hist` would be converted to `ref` units properly, and then, the units
of `sim` would be changed too, without any actual conversion taking
place. This may have something to do with the weirdness of Generators.



### Does this PR introduce a breaking change?

No 

### Other information:

Just to show where I'm getting at, you can test: 

```python
from xclim import sdba
from xclim.testing import open_dataset
ds = open_dataset('sdba/ahccd_1950-2013.nc').isel(location=0)
ds2 = open_dataset('sdba/CanESM2_1950-2100.nc').isel(location=0)
da = sdba.stack_variables(ds)
da2 = sdba.stack_variables(ds2)
ref = da.sel(time=slice("1950", "1979"))
hist = da2.sel(time=slice("1950", "1979"))
sim = da2.sel(time=slice("1980", "2009"))
out = sdba.dOTC.adjust(sim=sim.copy(), ref=ref.copy(), hist=hist.copy()) 
print(out["multivar"].attrs["_units"], out.sel(multivar="tasmax").mean().values)

>>> ['mm day-1', 'degC'] 14.25703882896165  # new xclim
>>... (continued)

Run Details

6 of 6 new or added lines in 1 file covered. (100.0%)

2 existing lines in 1 file now uncovered.

9373 of 10487 relevant lines covered (89.38%)

7.55 hits per line