Normalizing Well Log Data Before Inversion

Before running a joint inversion on well log data, it's worth normalizing each curve so no single measurement dominates the objective function purely because of its units.

The basic transform

For a log curve x, the z-score normalization is:

x' = (x − μ) / σ

where μ is the sample mean and σ the sample standard deviation.

import numpy as np

def zscore(x):
    return (x - np.mean(x)) / np.std(x)

Why it matters

Without normalization, resistivity's huge dynamic range would swamp the inversion's sensitivity to the other curves.