Cheat Sheet

ds = nc.open_data("foo.nc")

ds = nc.open_data(["foo1.nc", "foo2.nc"])

ds = nc.open_data("data/*.nc")

ds = nc.open_thredds("https://foo.nc")

ds = nc.open_url("https://foo.nc")

ds = nc.from_xarray(xr_ds)

ds2 = ds.copy()

ds = nc.open_geotiff("foo.tif")

files = nc.glob("data/foo")

ds.variables

ds.contents
# a table with one row per variable and columns:
# variable, ntimes, npoints, nlevels,
# long_name, unit, data_type, fill_value

ds.contents.long_name

ds.contents.unit

ds.contents[["variable", "long_name"]]

ds.contents.data_type

ds.contents.ntimes

ds.contents.npoints

ds.contents.nlevels

ds.contents.fill_value

ds.years

ds.times

ds.levels

ds.current

ds.history

ds.size

ds.subset(month=6)

ds.subset(years=range(1980, 1990))

ds.subset(season="DJF")

ds.subset(time=0)

ds.subset(variables=["var1", "var2"])

ds.drop(variables="var1")

ds.subset(lon=[-10, 10], lat=[45, 60])

ds.mask_box(lon=[-10, 10], lat=[45, 60])

ds.tmean()

ds.tmean("year")

ds.tmean("month")

ds.tmean(["year", "month"])

ds.tmax()

ds.trange()

ds.tpercentile(90)

ds.tstdev()

ds.tcumsum()

ds.rolling_mean(10)

ds.annual_anomaly(baseline=[1970, 1979])

ds.monthly_anomaly(baseline=[1970, 1979])

ds.shift(hours=-1)

ds.tmin()

ds.rolling_max(10)

ds.cor_time(var1="var1", var2="var2")

nc.cor_time(x=ds1, y=ds2)

ds.spatial_mean()

ds.spatial_sum(by_area=True)

ds.spatial_max()

ds.zonal_mean()

ds.meridonial_mean()

ds.box_mean(x=2, y=2)

ds.cell_area()

ds.spatial_min()

ds.zonal_max()

ds.meridonial_max()

ds.box_max(x=2, y=2)

ds.cor_space(var1="var1", var2="var2")

nc.cor_space(x=ds1, y=ds2)

ds.top()

ds.bottom()

ds.vertical_mean(fixed=True)

ds.vertical_range()

ds.vertical_interp(levels=[10, 50, 100], fixed=True)

ds.vertical_cumsum()

ds.vertical_sum()

ds.vertical_integration(fixed=True)

ds.invert_levels()

ds.bottom_mask()

ds.regrid(ds2)

ds.regrid(ds2, method="nn")

ds.to_latlon(lon=[-180, 180], lat=[-90, 90], res=2)

ds.regrid(lon=-4.1, lat=50.3)

ds.regrid(lon=[-4.1, -5.5], lat=[50.3, 50.0])

ds.to_transect(start=[-30, 0], end=[-30, 60], nsteps=61)

ds.resample_grid(2)

ds.fill_na()

ds.time_interp(start="1990/01/01", end="1990/12/31", resolution="daily")

ds.timestep_interp(steps=6)

ds.assign(sst_k=lambda x: x.sst + 273.15)

ds.assign(sst_k=lambda x: x.sst + 273.15, drop=True)

ds + 273.15
ds.add(273.15)

ds * 100
ds.multiply(100)

ds - ds2
ds.subtract(ds2)

ds / ds2
ds.divide(ds2)

ds.abs()

ds > 20
ds.compare(">20")

ds.exp()

ds.sqrt()

ds.square()

ds.power(2)

ds.rename({"old": "new"})

ds.set_units({"sst": "degC"})

ds.set_longnames({"sst": "Sea surface temperature"})

ds.as_missing([-5, 0])

ds.missing_as(0)

ds.set_fill(-999)

ds.strip_variables(vars="sst2")

ds.sum_all()

ds.set_gridtype("lonlat")

ds.set_precision("F64")

ds.ensemble_mean()

ds.ensemble_mean(ignore_time=True)

ds.ensemble_range()

ds.merge("time")

ds3 = nc.merge(ds1, ds2)

ds.split("year")

ds.append("foo.nc")

ds.remove("foo.nc")

ds.collect()

ds.distribute(m=2, n=2)

ds.plot()

ds.pub_plot()

ds.pub_plot(out="map.png", dpi=300)

nc.panel_plot({"a": ds1, "b": ds2})

ds.to_nc("out.nc")

ds.to_xarray()

ds.to_dataframe()

ds.zip()

ds.format("nc4")

ds.check()

ds.is_corrupt()

ds.na_count()

ds.na_frac()

ds.no_leaps()

ds.run()

nc.options(lazy=False)

nc.options(cores=6)

nc.deep_clean()