A gravity reading is mostly the earth being round, spinning, and having you standing higher than the last station. The geology is what is left after all of that is taken off, and it is a thousand times smaller than the thing you are subtracting.
Theoretical gravity comes from Somigliana's formula on the GRS80 ellipsoid. The check that matters is at the poles: the formula is built from the equatorial value and a flattening term, and it must reproduce the polar gravity, which is a separately measured number. It does, to eleven decimal places — so neither figure is a lone recollection.
Reading from a ship or an aircraft, your own motion changes your weight — moving east adds to the earth's rotation and you get lighter. That is the Eötvös correction, and it is 2ωV cos φ sin α + V²/R, which is simply the centrifugal terms of the motion you have added.
A buried thing produces an anomaly whose width tells you its depth. For a sphere the anomaly falls to half its peak at a distance of 0.766 z from the centre, so the depth is 1.305 times that half width. For a long horizontal cylinder the half width IS the depth. Both fall straight out of the geometry — no table required.
These are maximum depths in the strict sense: any real body is more spread out than a point, and a more spread out body at the same depth makes a wider anomaly. So the depth you get is the deepest the source could be — which is exactly the useful direction to be wrong in.
The Bouguer slab is the one correction with geology inside it. Its whole content is 2πGρ — and with G known, that is 0.0419 mGal per metre for every 1000 kg/m³. Choose the density wrongly and the anomaly you produce will be a picture of the topography.
| material | density kg/m³ |
|---|
Nettleton's method settles the density in the field rather than from a table: reduce a profile over a hill with several densities and keep the one whose anomaly least resembles the hill.
A magnetic anomaly is almost always magnetite, and how much of it. Induced magnetisation is J = κH — susceptibility times the field it sits in — so a rock's response is read straight off its magnetite content, and the susceptibilities span six orders of magnitude where densities span less than one.
| rock | κ, SI × 10⁻³ |
|---|
A rock can also carry remanent magnetisation from when it cooled, pointing somewhere else entirely. Where the Koenigsberger ratio — remanent over induced — is above about one, modelling the anomaly as induced alone will put the body in the wrong place.