Note
Click here to download the full example code
Taylor Diagram¶
What it's for: The Taylor Diagram provides a concise statistical summary of how well a model (or multiple models) matches a reference dataset (usually observations). It summarizes three statistics in a single point: the correlation coefficient, the root-mean-square (RMS) error, and the standard deviation.
When to use: Use this diagram to compare multiple models or different versions of the same model against a single set of observations. It is a standard tool in climate and meteorological model evaluation for assessing spatial or temporal patterns.
How to read: * Radial Distance from Origin: Represents the Standard Deviation of the data. * Angular Coordinate (Arc): Represents the Pearson Correlation Coefficient (R). * Distance from the Reference Point (on the X-axis): Represents the Centered Root-Mean-Square (RMS) error. * Interpretation: A perfect model would be represented by the "Reference" point on the x-axis (Correlation = 1, same Standard Deviation as observations, RMS error = 0). The closer a model's point is to the reference point, the better it matches the observations.
Out:
Reference std: 1.1750591258156107
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from monet_plots.plots.taylor_diagram import TaylorDiagramPlot
# 1. Prepare model data
np.random.seed(42)
n_points = 1000
obs = np.random.normal(0, 1.2, n_points)
model1 = 0.9 * obs + np.random.normal(0, 0.5, n_points)
model2 = 0.7 * obs + np.random.normal(0, 0.8, n_points)
df = pd.DataFrame({"obs": obs, "Model A": model1, "Model B": model2})
# 2. Initialize and create the plot
# TaylorDiagramPlot calculates statistics from the DataFrame
plot = TaylorDiagramPlot(df, col1="obs", col2=["Model A", "Model B"], scale=1.5)
plot.plot()
plt.title("Model Comparison Taylor Diagram")
plt.show()
Total running time of the script: ( 0 minutes 0.346 seconds)
Download Python source code: plot_taylor_diagram.py