Module gdess.operations.anomalies

Functions

def daily_anomalies(dataset: xarray.core.dataset.Dataset, varname: str = 'co2') -> pandas.core.frame.DataFrame

Parameters

dataset : xarray Dataset

 

varname : str, default 'co2'

 

Returns

pd.DataFrame

 

Expand source code

def daily_anomalies(dataset: xr.Dataset,
                    varname: str = 'co2'
                    ) -> pd.DataFrame:
    """

    Parameters
    ----------
    dataset : xarray Dataset
    varname : str, default 'co2'

    Returns
    -------
    pd.DataFrame
    """
    varlist = ['time', varname]
    tempds = dataset[varlist]

    # Get daily values, and add additional temporal label coordinates
    tempds = tempds.resample(time="D").mean()
    tempds = tempds.assign_coords(doy=tempds.time.dt.strftime("%j").str.lstrip("0").astype(int))
    tempds = tempds.assign_coords(year=tempds.time.dt.strftime("%Y"))
    ds_daily = tempds.assign_coords(year_month=tempds.time.dt.strftime("%Y-%m"))

    # Convert to pandas dataframe
    df_daily = ds_daily.to_dataframe()

    # --- Calculate Anomalies ---
    #
    """
    Daily anomaly of this data from the local monthly mean of the time-series. 
    That is, I want to take away the average of (eg) January 1979 from all the days in January 1979. 
    And I'd like to do this for every month of every year in my array."""
    df_daily['daily_anomaly_from_month'] = \
        (ds_daily.groupby("year_month") - ds_daily.groupby("year_month").mean("time"))[varname]
    # Get each datum's departure from it's year's mean
    df_daily['daily_anomaly_from_year'] = (ds_daily.groupby("year") - ds_daily.groupby("year").mean("time"))[varname]
    # Get each datum's departure from the entire dataset's mean
    df_daily['daily_anomaly_from_allmean'] = (ds_daily - ds_daily.mean("time"))[varname]
    # Get each datum's departure from it's year's mean
    df_daily['daily_anomaly_from_year'] = (ds_daily.groupby("year") - ds_daily.groupby("year").mean("time"))[varname]

    df_daily = df_daily.reset_index()

    return df_daily

def monthly_anomalies(dataset: xarray.core.dataset.Dataset, varname: str = 'co2') -> pandas.core.frame.DataFrame

Parameters

dataset : xarray Dataset

 

varname : str, default 'co2'

 

Returns

pd.DataFrame

 

Expand source code

def monthly_anomalies(dataset: xr.Dataset,
                      varname: str = 'co2'
                      ) -> pd.DataFrame:
    """

    Parameters
    ----------
    dataset : xarray Dataset
    varname : str, default 'co2'

    Returns
    -------
    pd.DataFrame
    """
    varlist = ['time', varname]
    tempds = dataset[varlist]

    # Get monthly values, and add additional temporal label coordinates
    tempds = tempds.resample(time="M").mean()
    tempds = tempds.assign_coords(moy=tempds.time.dt.strftime("%m"))
    tempds = tempds.assign_coords(year=tempds.time.dt.strftime("%Y"))
    ds_monthly = tempds.assign_coords(year_month=tempds.time.dt.strftime("%Y-%m"))

    # Convert to pandas dataframe
    df_monthly = ds_monthly.to_dataframe()

    # --- Calculate Anomalies ---
    #
    # Get each datum's departure from it's year's mean"""
    df_monthly['monthly_anomaly_from_year'] = \
        (ds_monthly.groupby("year") - ds_monthly.groupby("year").mean("time"))[varname]
    # Get each datum's departure from the entire dataset's mean"""
    df_monthly['monthly_anomaly_from_allmean'] = (ds_monthly - ds_monthly.mean("time"))[varname]

    df_monthly = df_monthly.reset_index()

    return df_monthly

def seasonal_anomalies(dataset: xarray.core.dataset.Dataset, varname: str = 'co2') -> pandas.core.frame.DataFrame

Parameters

dataset : xarray Dataset

 

varname : str, default 'co2'

 

Returns

pd.DataFrame

 

Expand source code

def seasonal_anomalies(dataset: xr.Dataset,
                       varname: str = 'co2'
                       ) -> pd.DataFrame:
    """

    Parameters
    ----------
    dataset : xarray Dataset
    varname : str, default 'co2'

    Returns
    -------
    pd.DataFrame
    """
    varlist = ['time', varname]
    tempds = dataset[varlist]

    # Get seasonal (quarterly, starting on December) values, and add additional temporal label coordinates
    tempds = tempds.resample(time="QS-DEC").mean()
    tempds = tempds.assign_coords(year=tempds.time.dt.strftime("%Y"))
    # add seasonal labels
    tempds = tempds.assign_coords(moy=tempds.time.dt.strftime("%m").astype(int))
    month_to_season_lu = np.array([
        None,
        'DJF', 'DJF',
        'MAM', 'MAM', 'MAM',
        'JJA', 'JJA', 'JJA',
        'SON', 'SON', 'SON',
        'DJF'
    ])
    tempds = tempds.assign_coords(season=('time', month_to_season_lu[tempds.moy]))
    ds_seasonal = tempds.assign_coords(year_month=tempds.time.dt.strftime("%Y-%m"))

    # Convert to pandas dataframe
    df_seasonal = ds_seasonal.to_dataframe()

    # --- Calculate Anomalies ---
    #
    # Get each datum's departure from the entire dataset's mean
    df_seasonal['seasonal_anomaly_from_allmean'] = (ds_seasonal - ds_seasonal.mean("time"))[varname]

    df_seasonal = df_seasonal.reset_index()

    return df_seasonal