Define a discordance cutoff to filter U--Pb data.
discfilter(option = 0, before = TRUE, cutoff)a list with the input parameters. Default values for
cutoff are
c(-48,140) if option=='t';
c(-5,15) if option=='r';
c(-0.36,0.96) if option=='sk';
c(-1.6,4.7) if option=='a'; and
c(-2,5.8) if option=='c'.
one of five options:
0: do not apply a discordance filter
1 or 't': the absolute age difference (Ma) between
the \(^{206}\)Pb/\(^{238}\)U and \(^{207}\)Pb/\(^{206}\)Pb
ages.
2 or 'r': the relative age difference (%) between the
\(^{206}\)Pb/\(^{238}\)U and \(^{207}\)Pb/\(^{206}\)Pb ages.
3 or 'sk': percentage of common Pb measured along a
mixing line connecting the measured composition and the
Stacey-Kramers mantle composition in Tera-Wasserburg space.
4 or 'a': logratio distance (%) measured along a
perpendicular line connecting Tera-Wasserburg concordia and the
measured composition.
5 or 'c': logratio distance (%) measured along a
line connecting the measured composition and the corresponding
single grain concordia_age composition.
Further details in Vermeesch (2021).
logical flag indicating whether the discordance
filter should be applied before (TRUE) or after
(FALSE) the common-Pb correction.
a two-element vector with the minimum (negative) and
maximum (positive) allowed discordance. Default values vary
between the different options. To view them, enter
discfilter(option) at the R command line.
The most reliable U--Pb age constraints are obtained from (zircon) grains whose \(^{206}\)Pb/\(^{238}\)U and \(^{207}\)Pb/\(^{206}\)Pb ages are statistically indistinguishable from each other. U--Pb compositions that fulfil this requirements are called `concordant'. Those that violate it are called `discordant'. The discordance of the \(^{206}\)Pb/\(^{238}\)U and \(^{207}\)Pb/\(^{206}\)Pb systems can be defined in five different ways. By setting a cutoff for any of these criteria, U--Pb data can be filtered for data quality.
Vermeesch (2021) ``On the treatment of discordant data in detrital zircon U--Pb geochronology'', Geochronology.
cad, kde,
radialplot
dscf <- discfilter(option='c',before=TRUE,cutoff=c(-1,1))
weightedmean(x=examples$UPb,exterr=FALSE,sigdig=2,
cutoff.disc=dscf,common.Pb=3)
Run the code above in your browser using DataLab