'siteid' in add_data output

monetio/profile/gml_ozonesonde.py

@@ -136,7 +136,7 @@ def add_data(dates, *, location=None, n_procs=1, errors="raise"):
         For example 'Boulder, Colorado'.
         If not provided, all locations will be used.
         Valid options correspond to the directories in https://gml.noaa.gov/aftp/data/ozwv/Ozonesonde/
-        and may include data from more than one unique site (output column 'station').
+        and may include data from more than one unique site (output column 'siteid').
     n_procs : int
         For Dask.
     errors : {'raise', 'warn', 'skip'}
@@ -213,6 +213,9 @@ def func(fp_or_url):
     assert set(repl.values()) <= set(LOCATIONS)
     df["station"] = df["station"].replace(repl)
 
+    # Normalized station name as site ID
+    df = df.rename(columns={"station": "siteid"})
+
     # Add metadata
     if hasattr(df, "attrs"):
         df.attrs["ds_attrs"] = {"urls": urls}

tests/test_gml_ozonesonde.py

@@ -71,7 +71,7 @@ def test_add_data():
     assert 1 < latlon.nunique() <= 10, "multiple sites; lat/lon doesn't change in profile"
 
     # NOTE: Similar to the place folder names, but not all the same
-    assert df["station"].nunique() == latlon.nunique()
+    assert df["siteid"].nunique() == latlon.nunique()
 
 
 def test_add_data_location_sel():