Support ILVIS2 #40
Support ILVIS2 #40
Conversation
Now need to figure out TODO related to there not being one single source of lat/lon/elevation for these data.
This replicates the behaivor in `valkyrie`: the GLAT, GLON, and ZG fields are used for lat/lon/elev. We may want to revisit supporting transformation of other lat/lon/elev combinations.
src/nsidc/iceflow/data/ilvis2.py
Outdated
There was a problem hiding this comment.
Most of this code is copied directly from valkyrie, with some minor modifications.
valkyrie did not have any tests defined to ensure the logic here is sound - it would be nice to add some! Right now, the end-to-end test asserts that data can be found, downloaded, and read with this code.
| # This ensures all columns are present, regardless of the date. Granules | ||
| # before 2017 use the V104 fields and anything after uses the v202b | ||
| # fields. The data type for all values must be `float` because the null | ||
| # value is `np.nan` - a float. |
There was a problem hiding this comment.
Maybe this class should be broken in two: one for pre-2017 and one for post-2017. The resulting df would just have the columns that are present in the source data.
The downside to this is that users would get dataframes with inconsistent columns for ILVIS2 data. At some point we want to look at harmonizing across all supported datasets, so that a user can easily compare elevations. Different datasets already have different sets of columns, so this is a bridge we'll have to cross soon
This PR adds support for ILVIS2 data.
Note that the behavior of
valkyrieis preserved here: theGLAT,GLON, andZGfields areused for lat/lon/elev. We may want to revisit supporting transformation of other
lat/lon/elev combinations (e.g.,
HLAT,HLON, andZH). See #41resolves #19 and #20
📚 Documentation preview 📚: https://iceflow--40.org.readthedocs.build/en/40/