Climatology

Overview

The NCL climatology functions listed below can be replicated using xarray and/or geocat.comp

• calcDayAnomTLL

• calcMonAnomTLL

• clmDayTLL

• clmMonTLL

• month_to_season

• rmMonAnnCycTLL

• stdMonTLL

calcDayAnomTLL

calcDayAnomTLL calculates daily anomalies from a daily data climatology

Grab and Go

import xarray as xr
import geocat.datafiles as gdf
from matplotlib import pyplot as plt

ds = xr.open_dataset(gdf.get("applications_files/inputs/CMIP6_sea_ice_daily_subset.nc"))
aice = ds.aice_d
DayTLL = aice.groupby(aice.time.dt.dayofyear)
clmDayTLL = DayTLL.mean(dim="time")
calcDayAnomTLL = DayTLL - clmDayTLL

calcDayAnomTLL = calcDayAnomTLL.assign_attrs(long_name="sea ice anomaly")

calcDayAnomTLL[0, :, :].plot();

Downloading file 'applications_files/inputs/CMIP6_sea_ice_daily_subset.nc' from 'https://github.com/NCAR/GeoCAT-datafiles/raw/main/applications_files/inputs/CMIP6_sea_ice_daily_subset.nc' to '/home/runner/.cache/geocat'.

calcMonAnomTLL

calcMonAnomTLL calculates monthly anomalies by subtracting the long-term mean from each point

Grab and Go

import xarray as xr
import geocat.datafiles as gdf

ds = xr.open_dataset(
    gdf.get("applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc")
)
aice = ds.aice
MonTLL = aice.groupby(aice.time.dt.month)
clmMonTLL = MonTLL.mean(dim="time")
calcMonAnomTLL = MonTLL - clmMonTLL

calcMonAnomTLL = calcMonAnomTLL.assign_attrs(long_name="sea ice anomaly")

calcMonAnomTLL[0, :, :].plot();

Downloading file 'applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc' from 'https://github.com/NCAR/GeoCAT-datafiles/raw/main/applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc' to '/home/runner/.cache/geocat'.

clmDayTLL

clmDayTLL calculates long-term daily means (daily climatology) from daily data

Grab and Go

import xarray as xr
import geocat.datafiles as gdf

ds = xr.open_dataset(gdf.get("applications_files/inputs/CMIP6_sea_ice_daily_subset.nc"))
aice = ds.aice_d
DayTLL = aice.groupby(aice.time.dt.dayofyear)
clmDayTLL = DayTLL.mean(dim="time")

clmDayTLL[:, 10, 10].plot()
plt.title("daily climatology")
plt.xlabel("day of year")
plt.ylabel("sea ice area");

clmMonTLL

clmMonTLL calculates long-term monthly means (monthly climatology) from monthly data

Grab and Go

import xarray as xr
import geocat.datafiles as gdf

ds = xr.open_dataset(
    gdf.get("applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc")
)
aice = ds.aice
MonTLL = aice.groupby(aice.time.dt.month)
clmMonTLL = MonTLL.mean(dim="time")

clmMonTLL[:, 10, 10].plot()
plt.title("monthly climatology")
plt.xlabel("month of year")
plt.ylabel("sea ice area");

month_to_season

month_to_season computes a user-specified three-month seasonal mean (DJF, JFM, FMA, MAM, AMJ, MJJ, JJA, JAS, ASO, SON, OND, NDJ)

Note

You can do something similar with directly with Xarray as shown in this example in the Xarray documentation. However, it requires substantially more code and doesn’t have as much flexibility with respect to how the seasons are defined.

Grab and Go

import xarray as xr
import geocat.datafiles as gdf
from geocat.comp import month_to_season

ds = xr.open_dataset(
    gdf.get("applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc")
)
aice = ds.aice
mon_to_season = month_to_season(aice, "ASO")

mon_to_season = mon_to_season.assign_attrs(long_name="sea ice area")

mon_to_season[0, :, :].plot()

plt.title("2010 seasonal mean");

rmMonAnnCycTLL

rmMonAnnCycTLL removes the annual cycle from monthly data

Grab and Go

import xarray as xr
import geocat.datafiles as gdf

ds = xr.open_dataset(
    gdf.get("applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc")
)
aice = ds.aice
MonTLL = aice.groupby(aice.time.dt.month)
clmMonTLL = MonTLL.mean(dim="time")
rmMonAnnCycTLL = MonTLL - clmMonTLL

rmMonAnnCycTLL[:, 10, 10].plot()
plt.title("annual cycle removed")
plt.ylabel("sea ice area");

stdMonTLL

stdMonTLL calculates standard deviations of monthly means

Grab and Go

import xarray as xr
import geocat.datafiles as gdf

ds = xr.open_dataset(
    gdf.get("applications_files/inputs/CMIP6_sea_ice_monthly_subset.nc")
)
aice = ds.aice
MonTLL = aice.groupby(aice.time.dt.month)
stdMonTLL = MonTLL.std(ddof=1)

stdMonTLL[0, :, :].plot();

---

Python Resources

• GeoCAT Applications climatology page

• Climatematch Academy Xarray Data Analysis and Climatology tutorial

• Project Pythia Foundations Computations and Masks with Xarray tutorial

• Xarray User Guide section on time series data