easyclimate.plot.taylor_diagram¶

Functions for Taylor Diagrams plotting.

Functions¶

`calc_TaylorDiagrams_values`(→ pandas.DataFrame)	Calculate the center root mean square error.
`calc_TaylorDiagrams_metadata`(f, r[, models_name, ...])	Calculating Taylor diagram metadata
`draw_TaylorDiagrams_base`(...)	Drawing Taylor Graphics Basic Framework
`draw_TaylorDiagrams_metadata`(→ dict)	Draw points to Taylor Graphics Basic Framework according to Taylor diagram metadata.

Module Contents¶

easyclimate.plot.taylor_diagram.calc_TaylorDiagrams_values(f: xarray.DataArray, r: xarray.DataArray, model_name: str, weighted: bool = False, lat_dim: str = 'lat', normalized: bool = True, r0: float = 0.999) → pandas.DataFrame¶

Calculate the center root mean square error.

where \(N\) is the number of points in spatial pattern.

Parameters¶

fxarray.DataArray, required.: A spatial array of models to be compared.
rxarray.DataArray, required.: A spatial array for model reference comparisons (observations).
model_name: str, required.: The name of the model.
weighted: bool, default False.: Whether to weight the data by latitude or not? The default value is False.
lat_dim: str, default lat.: The name of latitude coordinate name.
normalized: bool, default True, optional.: Whether the standard deviations is normalized, that is standard deviation of the model divided by that of the observations.
r0float, optional.: Maximum correlation obtainable.

Attention

f and r must have the same two-dimensional spatial dimension.

Returns¶

pandas.DataFrame.

Reference¶

Taylor, K. E. (2001), Summarizing multiple aspects of model performance in a single diagram, J. Geophys. Res., 106(D7), 7183-7192, doi:10.1029/2000JD900719.

easyclimate.plot.taylor_diagram.calc_TaylorDiagrams_metadata(f: xarray.DataArray, r: xarray.DataArray, models_name: list[str] = [], weighted: bool = False, lat_dim: str = 'lat', normalized: bool = True)¶

Calculating Taylor diagram metadata

Parameters¶

f: xarray.DataArray, required.: A spatial array of models to be compared.
r: xarray.DataArray, required.: A spatial array for model reference comparisons (observations).
model_name: list[str], required.: The list of the models’ name.
weighted: bool, default False.: Whether to weight the data by latitude or not? The default value is False.
lat_dim: str, default lat.: The name of latitude coordinate name.
normalized: bool, default True, optional.: Whether the standard deviations is normalized, that is standard deviation of the model divided by that of the observations.

Returns¶

pandas.DataFrame.

Examples¶

>>> import xarray as xr
>>> import pandas as pd
>>> import numpy as np
>>> import easyclimate as ecl
>>> da_a = xr.DataArray(
...:     np.array([[1, 2, 3], [0.1, 0.2, 0.3], [3.2, 0.6, 1.8]]),
...:     dims = ("lat", "time"),
...:     coords= {'lat': np.array([-30, 0, 30]),
...:              'time': pd.date_range("2000-01-01", freq="D", periods=3)
...:              }
...:)
>>> da_a
<xarray.DataArray (lat: 3, time: 3)>
array([[1. , 2. , 3. ],
    [0.1, 0.2, 0.3],
    [3.2, 0.6, 1.8]])
Coordinates:
* lat      (lat) int32 -30 0 30
* time     (time) datetime64[ns] 2000-01-01 2000-01-02 2000-01-03
>>>  da_b = xr.DataArray(
...:     np.array([[0.2, 0.4, 0.6], [15, 10, 5], [3.2, 0.6, 1.8]]),
...:     dims = ("lat", "time"),
...:     coords= {'lat': np.array([-30, 0, 30]),
...:              'time': pd.date_range("2000-01-01", freq="D", periods=3)
...:              }
...:)
>>>  da_b
<xarray.DataArray (lat: 3, time: 3)>
array([[ 0.2,  0.4,  0.6],
    [15. , 10. ,  5. ],
    [ 3.2,  0.6,  1.8]])
Coordinates:
* lat      (lat) int32 -30 0 30
* time     (time) datetime64[ns] 2000-01-01 2000-01-02 2000-01-03
>>>  da_obs = (da_a + da_b) / 1.85
>>>  da_obs
<xarray.DataArray (lat: 3, time: 3)>
array([[0.64864865, 1.2972973 , 1.94594595],
    [8.16216216, 5.51351351, 2.86486486],
    [3.45945946, 0.64864865, 1.94594595]])
Coordinates:
* lat      (lat) int32 -30 0 30
* time     (time) datetime64[ns] 2000-01-01 2000-01-02 2000-01-03
>>>  ecl.calc_TaylorDiagrams_metadata(
...:     f = [da_a, da_b],
...:     r = [da_obs, da_obs],
...:     models_name = ['f1', 'f2'],
...:     weighted = True,
...:     normalized = True,
...:)
item       std                   cc  centeredRMS       TSS
0  Obs  1.000000                  1.0     0.000000  1.002003
1   f1  0.404621  -0.4293981636461462     1.229311  0.003210
2   f2  2.056470    0.984086060161888     1.087006  0.600409

easyclimate.plot.taylor_diagram.draw_TaylorDiagrams_base(ax: matplotlib.axes.Axes = None, half_circle: bool = False, normalized: bool = True, std_min: float = 0, std_max: float = 2, std_interval: float = 0.25, arc_label: str = 'Correlation', arc_label_pad: float = 0.2, arc_label_kwargs: dict = {'fontsize': 12}, arc_ticker_kwargs: dict = {'lw': 0.8, 'c': 'black'}, arc_tickerlabel_kwargs: dict = {'labelsize': 12, 'pad': 8}, arc_ticker_length: float = 0.02, arc_minorticker_length: float = 0.01, x_label: str = 'Std (Normalized)', x_label_pad: float = 0.25, x_label_kwargs: dict = {'fontsize': 12}, x_ticker_length: float = 0.02, x_tickerlabel_kwargs: dict = {'fontsize': 12}, x_tickerlabel_pad: float | None = None, y_tickerlabel_pad: float | None = None, x_ticker_kwargs: dict = {'lw': 0.8, 'c': 'black'}, y_ticker_kwargs: dict = {'lw': 0.8, 'c': 'black'}) → matplotlib.collections.Collection¶

Drawing Taylor Graphics Basic Framework

Parameters¶

ax: matplotlib.axes.Axes, optional.: Axes on which to plot. By default, use the current axes, i.e. ax = plt.gca().
half_circle: bool, default False, optional.: Whether to draw the ‘half-circle’ version of the Taylor diagram.
normalized: bool, default True, optional.: Whether the standard deviations is normalized, that is standard deviation of the model divided by that of the observations. This parameter mainly affects the label x=1 on the x axis, if normalized to True, it is rewritten to REF.
std_min: float, default 0.0, optional.: Minimum value of x-axis (standard deviation) on Taylor diagram.

Note

The value of std_min shoud be 0 in the ‘half-circle’ version of the Taylor diagram.
std_max: float, default 2.0, optional.: Maximum value of x-axis (standard deviation) on Taylor diagram.
std_interval: float, default 0.25, optional.: The interval between the ticker on the x-axis (standard deviation) between the minimum and maximum values on the Taylor diagram.
arc_label: str, default ‘Correlation’, optional.: Label on Taylor chart arc, default value is ‘Correlation’.
arc_label_pad: float, default 0.2, optional.: The offset of the title from the top of the arc, based on x-axis based coordinate system.
arc_label_kwargs: dict, default {‘fontsize’: 12}, optional.: Additional keyword arguments passed on to labels on arcs, according to other miscellaneous parameters in`matplotlib.axes.Axes.text`.
arc_ticker_kwargs: dict, default {‘lw’: 0.8, ‘c’: ‘black’}, optional.: Additional keyword arguments passed on to tickers on arcs, according to other miscellaneous parameters in`matplotlib.axes.Axes.plot`.
arc_tickerlabel_kwargs: dict, default {‘labelsize’: 12, ‘pad’: 8}, optional.: Additional keyword arguments passed on to tickers on arcs, according to other miscellaneous parameters in`matplotlib.axes.Axes.tick_params`.
arc_ticker_length: float, default 0.02, optional.: Ticker length on arc.
arc_minorticker_length: float, default 0.01, optional.: Minor ticker length on arc.
x_label: str, default ‘Std (Normalized)’, optional.: Label on Taylor chart x axis, default value is ‘Std (Normalized)’.
x_label_pad: float, default 0.25, optional.: The offset of the title from the top of the x-axis, based on x-axis based coordinate system.
x_label_kwargs: dict, default {‘fontsize’: 12}, optional.: Additional keyword arguments passed on to labels on x-axis, according to other miscellaneous parameters in`matplotlib.axes.Axes.text`.
x_ticker_length: float, default 0.02, optional.: Ticker length on x-axis
x_tickerlabel_kwargs: dict, default {‘fontsize’: 12}, optional.: Additional keyword arguments passed on to tickers’ labels on x-axis, according to other miscellaneous parameters in`matplotlib.axes.Axes.text`.
x_tickerlabel_pad: float, optional.: The spacing between the horizontal axis and its tick labels in points when half_circle is False. Positive values move labels away from the horizontal axis. If None, use the original text-based spacing.
y_tickerlabel_pad: float, optional.: The spacing between the vertical axis and its tick labels in points when half_circle is False. Positive values move labels away from the vertical axis. If None, use the original text-based spacing.
x_ticker_kwargs: dict, default {‘lw’: 0.8, ‘c’: ‘black’}, optional.: Additional keyword arguments passed on to tickers on x-axis, according to other miscellaneous parameters in`matplotlib.axes.Axes.plot`.
y_ticker_kwargs: dict, default {‘lw’: 0.8, ‘c’: ‘black’}, optional.: Additional keyword arguments passed on to tickers on y-axis, according to other miscellaneous parameters in`matplotlib.axes.Axes.plot`.

Returns¶

matplotlib.collections.Collection.

easyclimate.plot.taylor_diagram.draw_TaylorDiagrams_metadata(taylordiagrams_metadata: pandas.DataFrame, marker_list: list | None = None, color_list: list | None = None, label_list: list | None = None, legend_list: list | None = None, ax: matplotlib.axes.Axes = None, normalized: bool = True, cc: str = 'cc', std: str = 'std', point_label_xoffset: float = 0, point_label_yoffset: float = 0.05, point_kwargs: dict | list[dict] = {'alpha': 1, 'markersize': 6.5}, point_label_kwargs: dict | list[dict] = {'fontsize': 14}) → dict¶

Draw points to Taylor Graphics Basic Framework according to Taylor diagram metadata.

Parameters¶

taylordiagrams_metadata: pandas.DataFrame, required.: Taylor diagram metadata generated by the function calc_TaylorDiagrams_metadata.
marker_list: list, optional.: The list of markers. The order of marker in marker_list is determined by the order in taylordiagrams_metadata. See matplotlib.markers for full description of possible arguments. If None, use “o” for all data points.
color_list: list, optional.: The list of colors. The order of color in color_list is determined by the order in taylordiagrams_metadata. If None, use Matplotlib’s default color cycle.
label_list: list, optional.: The list of data point labels (marked next to plotted points). The order of label in label_list is determined by the order in taylordiagrams_metadata. If None, do not draw text labels next to data points.
legend_list: list, optional.: The list of legend label. The order of label in legend_list is determined by the order in taylordiagrams_metadata. If None, use the item column in taylordiagrams_metadata.
ax: matplotlib.axes.Axes, optional.: Axes on which to plot. By default, use the current axes, i.e. ax = plt.gca().
normalized: bool, default True, optional.: Whether the standard deviations is normalized, that is standard deviation of the model divided by that of the observations.
cc: str, default ‘cc’, optional.: The name of correlation coefficient in taylordiagrams_metadata.
std: str, default ‘std’, optional.: The name of standard deviation in taylordiagrams_metadata.
point_label_xoffset: float, optional.: The offset of the labels from the points, based on x-axis based coordinate system.
point_label_yoffset: float, optional.: The offset of the labels from the points, based on y-axis based coordinate system.
point_kwargs: dict or list of dict, optional.: Additional keyword arguments passed on to data points, according to other miscellaneous parameters in`matplotlib.axes.Axes.plot`. If it is a list, each element is applied to the corresponding data point. If it is a dictionary whose values are dictionaries, each element is selected by data point item, legend, label, or row index.
point_label_kwargs: dict or list of dict, optional.: Additional keyword arguments passed on to the labels of data points, according to other miscellaneous parameters in`matplotlib.axes.Axes.text`. If it is a list, each element is applied to the corresponding data point. If it is a dictionary whose values are dictionaries, each element is selected by data point item, legend, label, or row index.

Returns¶

dict.: A dictionary keyed by data point item. Each value includes the point artist, label artist, point keyword arguments, and label keyword arguments.