Van der Wal olivine piezometer bug

[marcoalopez]

B parameter in Van der Wal olivine piezometer must be 1355.4 instead of 4250

[marcoalopez]

The correction factor should be 1.2 instead of 1.5 (this was corrected in v3.0.1). Yet, I am not sure whether a correction should be applied to this piezometer as the original 1993 work mix different types of grain size measurement (linear intercepts* on the wet samples and max. diameters with a correction factor of 1.2 on the dry samples). Unfortunately, they do not clarify whether they applied a correction factor to the linear intercepts.

When providing equivalent circular diameters (ECD) the following options might be considered:

• Convert the ECD to linear intercepts (LI) using the conversion factor of De Hoff and Rhines (1968): LI = ECD / (sqrt(4/pi))
• Convert the ECD to LI and multiply by the correction factor 1.2 (the option implemented in the v3.0.1)
• Apply the correction factor to ECDs (without converting to LI)
• No correction at all (leaving the ECDs as they are)

*They did not specify what protocol for linear intercepts they used (and there are several out there) or whether they applied a correction factor in this case.

Possible solution:
Convert the ECD to LI and multiply by 1.5 (as implemented in previous versions of the script v3.0RC)

Rationale:
Most studies on olivine paleopiezometry using linear intercepts multiplied the average value by a correction factor of 1.5; e.g. papers by Kohlstedt et al. group or in Jung and Karato (2001) (see the caption on figure 6). It, therefore, seems reasonable that Van der Wal et al. (1993) applied the same correction factor of 1.5 in this case (even if they indicate nothing in the paper). Another hint in favour of this solution is the application of the correction factor 1.2 applied to the maximum diameters. In fact, if one applies the conversion from ECDs to LI (dividing by ~1.13) and then multiplies by 1.5, it is as if one directly multiplies the ECDs by 1.33, a value close to the correction factor 1.2 applied by these authors.

Resolution:
Van der Val piezometer will follow a two-step procedure from version 3.0.2 onwards:
(1) Convert from ECD to LI using the empirical equation of De Hoff and Rhines (1968).
(2) Multiply by 1.5