lair.noaa.CarbonTracker.sample#

CarbonTracker.sample(points)[source]#

Sample the molefraction field at points.

points is a DataFrame with columns datetime (UTC sample time), lati, long, zagl, as in a PYSTILT particle table (for example particles.stilt.endpoints()); any other columns (e.g. indx, receptor) are carried through for downstream grouping. Returns one row per input point with ct_<specie>_<units> and the CT cell/level used. Points whose UTC date has no molefraction file are dropped.

Each point takes the time step whose center is nearest: the steps are 3-hour means labeled by their centers (ct_time: 01:30, 04:30, … UTC), so a point takes the mean of the period it is in, and a point on the boundary of two periods (03:00) takes the later one.

Parameters:

points (pd.DataFrame) – The points, with the columns datetime (UTC), lati and long (degrees) and zagl (meters above ground).

Returns:

pd.DataFrame – The points, with ct_<specie>_<units> (ct_ch4_ppb) and the cell, time and level it came from (ct_latitude, ct_longitude, ct_time, ct_level). A point outside the grid gets NaN.

Return type:

DataFrame

See also

CarbonTracker.background

Their mean.

Examples

The CarbonTracker example samples four points west of Salt Lake City; it reads the folder of one downloaded day:

points = pd.DataFrame(
    {
        "datetime": pd.to_datetime(["2020-01-15 17:00"] * 2),  # UTC
        "lati": [41.5, 39.5],
        "long": [-114.5, -114.5],
        "zagl": [100.0, 100.0],
    }
)
ct.sample(points)[["lati", "ct_ch4_ppb", "ct_time", "ct_level"]]