Error message on CF convention of standard_name in time axis attributes

[larsbuntemeyer]

Version of compliance checker running:
5.4.2

Describe the checker this affects:
CF checkers

Attach a minimal CDL or NetCDF file which is able to reproduce the issue

netcdf example_timeseries_cf_1_11 {
dimensions:
        time = 24 ;
variables:
        double air_temperature(time) ;
                air_temperature:_FillValue = -9999. ;
                air_temperature:standard_name = "air_temperature" ;
                air_temperature:long_name = "Air temperature" ;
                air_temperature:units = "K" ;
                air_temperature:coordinates = "time lat lon" ;
        int64 time(time) ;
                time:long_name = "Time of measurement" ;
                time:axis = "T" ;
                time:units = "seconds since 1970-01-01" ;
                time:calendar = "gregorian" ;
        double lat ;
                lat:_FillValue = NaN ;
                lat:standard_name = "latitude" ;
                lat:long_name = "Station latitude" ;
                lat:units = "degrees_north" ;
                lat:axis = "Y" ;
        double lon ;
                lon:_FillValue = NaN ;
                lon:standard_name = "longitude" ;
                lon:long_name = "Station longitude" ;
                lon:units = "degrees_east" ;
                lon:axis = "X" ;
        string station ;
                station:long_name = "Station Identifier" ;

// global attributes:
                :title = "Example CF-1.11 Time Series Dataset" ;
                :summary = "Synthetic single-station hourly air temperature time series for CF compliance checking." ;
                :institution = "Example Organization" ;
                :source = "Synthetic generation with xarray" ;
                :history = "Created 2025-09-21T18:21:58.371791Z by xarray" ;
                :references = "https://cfconventions.org/" ;
                :Conventions = "CF-1.11" ;
                :featureType = "timeSeries" ;
                :geospatial_lat_min = 42. ;
                :geospatial_lat_max = 42. ;
                :geospatial_lon_min = -71. ;
                :geospatial_lon_max = -71. ;
                :time_coverage_start = "2025-01-01T00:00:00Z" ;
                :time_coverage_end = "2025-01-01T23:00:00Z" ;
                :coordinates = "station" ;
}

To Reproduce:

# Create a simple CF-1.11 compliant time series NetCDF file with xarray
import numpy as np
import pandas as pd
import xarray as xr
from datetime import datetime, timezone

# Configuration
n_time = 24  # 24 hourly samples
start_time = "2025-01-01"
station_id = "station01"
output_path = "example_timeseries_cf_1_11.nc"

# Create a time coordinate (naive UTC datetimes; CF assumes UTC if no timezone)
time = pd.date_range(start_time, periods=n_time, freq="H")

# Generate example data (air temperature in Kelvin)
rng = np.random.default_rng(seed=42)
air_temperature_K = 273.15 + 5 + 2 * np.sin(np.linspace(0, 2 * np.pi, n_time)) + rng.normal(0, 0.3, n_time)

# Build Dataset
# We model a single-station timeSeries feature: dimensions (time, station)
ds = xr.Dataset(
    data_vars={
        "air_temperature": ("time", air_temperature_K, {
            "standard_name": "air_temperature",
            "long_name": "Air temperature",
            "units": "K",
            "coordinates": "time lat lon",
        })
    },
    coords={
        "time": ("time", time, {
          #  "standard_name": "time",
            "long_name": "Time of measurement",
            "axis": "T"
        }),
        # Represent the station as scalar lat/lon coordinates (1D with length 1 is also acceptable)
        "lat": ((), 42.0, {  # scalar coordinate
            "standard_name": "latitude",
            "long_name": "Station latitude",
            "units": "degrees_north",
            "axis": "Y"
        }),
        "lon": ((), -71.0, {
            "standard_name": "longitude",
            "long_name": "Station longitude",
            "units": "degrees_east",
            "axis": "X"
        }),
        "station": ((), station_id, {"long_name": "Station Identifier"}),
    },
    attrs={
        "title": "Example CF-1.11 Time Series Dataset",
        "summary": "Synthetic single-station hourly air temperature time series for CF compliance checking.",
        "institution": "Example Organization",
        "source": "Synthetic generation with xarray",
        "history": f"Created {datetime.utcnow().isoformat()}Z by xarray",
        "references": "https://cfconventions.org/",
        "Conventions": "CF-1.11",
        "featureType": "timeSeries",
        # ACDD geospatial + temporal summary (optional but useful)
        "geospatial_lat_min": 42.0,
        "geospatial_lat_max": 42.0,
        "geospatial_lon_min": -71.0,
        "geospatial_lon_max": -71.0,
        "time_coverage_start": time[0].isoformat() + "Z",
        "time_coverage_end": time[-1].isoformat() + "Z",
    }
)

# Explicit encodings for CF-compliant time (numeric since epoch) & fill values if needed
encoding = {
    "time": {"units": "seconds since 1970-01-01 00:00:00", "calendar": "gregorian"},
    "air_temperature": {"_FillValue": -9999.0}
}

# Write NetCDF file
ds.to_netcdf(output_path, format="NETCDF4", encoding=encoding)

print(f"Wrote CF-1.12 example file: {output_path}")
print(ds)

command

compliance-checker -t cf:1.11 example_timeseries_cf_1_11.nc

Describe the issue below:
The example above throws the following error:

                                     Errors                                     
--------------------------------------------------------------------------------
§5.1 Independent Latitude, Longitude, Vertical, and Time Axes
* Coordinate variable 'time' should have standard_name='time', found: 'None'

From my understanding of the CF convention regarding relative time axis, this should only be a warning, because

A reference time coordinate is identifiable from its units string alone.

it is only optional to provide a standard name for the time axis:

Optionally, the time coordinate may be indicated additionally by providing the standard_name attribute with an appropriate value, and/or the axis attribute with the value T.

[larsbuntemeyer]

Here are the lines:

compliance-checker/compliance_checker/cf/cf_1_6.py

Lines 2302 to 2306 in 9dd0ca3

	check_spatiotemporal_dims_coords.assert_true(
	std_name == expected_standard_names[dim],
	f"Coordinate variable '{dim}' should have standard_name='{expected_standard_names[dim]}', "
	f"found: '{std_name}'",
	)

Maybe i can make this an an additional test with Medium priority?

[ocefpaf]

Indeed it should be optional. Do you want to send a PR?

[larsbuntemeyer]

Yes, i'll do!

[Nesar976]

Thanks for the detailed analysis. I agree this should be optional per the CF conventions when the time reference is clear from the units and axis="T". Treating this as a warning instead of an error seems appropriate. Happy to help test or review if needed.

Perfect for a quick, professional contribution.

[ocefpaf]

@Nesar976 you have been sending way more PRs than we can review, many of those are AI generated and out of scope.

You are welcomed to send one for this issue, but please let's try to keep the AI to a minimum, better yet, disclose if you will use it or not.

[Nesar976]

Thanks for pointing this out — I understand your concern.
Sorry for the noise caused by my earlier PRs; that wasn’t my intention.

For this issue, I’ll focus only on what’s clearly in scope and keep the change minimal. I’ll also be more mindful about the use of AI tools going forward and will disclose it if I do use them.

Thanks for the clarification and for your time reviewing contributions.