Transformations

Density Plots

The analytical precision of many geochronological datasets is proportional to the age. Therefore, the relative errors generally show less grain-to-grain variability than the absolute errors. To take advantage of this, it may be beneficial to subject the data to a logarithmic or square root-transform before plotting them on a KDE, PDP, or histogram. To do this, select the appropriate option in the Options→ Transformations menu.

Radial Plots

Data transformations are even more important for radial plots than they are for density plots. In the case of fission track dating with the external detector method, Galbraith (1990) proposed three transformations:

Linear:

Logarithmic:

Arcsine:

With λ the ²³⁸U decay constant (=1.55125x10^-10a^-1), g a geometric factor (=0.5), and all other parameters as defined before. The Logarithmic transformation is used by default, unless there are grains with zero spontaneous tracks, in which case the program defaults to Arcsine.

To plot 'Other' input data on radial plots, the transformations are as follows:

Linear:

Logarithmic:

Where a=0 if all the values are positive and a>0 if there are negative values (Galbraith, 2010).

Square root:

The default is set to Logarithmic.