Robust MFV analysis of geoelectrical data in heterogeneous subsurface media
ONLINE FIRST (pre-proof)
Keywords:
most frequent value, robust statistics, electrical resistivity, induced polarization, geophysical data processing, environmental geophysicsAbstract
Geoelectrical datasets in heterogeneous environments often show non-Gaussian distributions and outliers, limiting conventional averaging. This study evaluates the Most Frequent Value (MFV) method for robust characterization of vertically structured geophysical data. Results show that the fixed ε parameter, the scale parameter of the weighting function, provides stable representations, while the conventional iterative ε method enhances locally dominant structures. Quantitative comparison confirms that method performance depends on data variability and sampling density. The method offers a flexible alternative for interpreting complex subsurface conditions.
