Sorry about not explaining the depth variable on the 5th figure of the last post -- it is, you will note, the antilog equivalent of the sum of the log10 concentrations of each of the named elements. Originally, USSR geochemists used this simple data transformation for two reasons: 1) they found that mathematically combining the elemental concentrations in that way helped make up for analytical inaccuracies that were part and parcel of the spectroscopic analytical lab methods of the day (late 50s, 1960s), and 2) it helped statistically compensate for natural irregularities in characteristic distribution of the primary halo elements above and around various kinds of hydrothermal orebodies.
Now it’s used mostly for reason #2.
Further following the logical practices set by the USSR geochemists, the trace elements chosen for the purpose of detecting the presence of buried ore by surface sampling were those most consistently and reliably found in and around mineral deposits of a given type.