search

CIRDLES / ET_Redux

EARTH-TIME.org's flagship data-processing and workflow automation software.
http://cirdles.org/projects/et_redux/
Apache License 2.0
12 stars 16 forks source link

Issue with uncertainties for uncorrected d234Ui for carbonates (v.3.6.86) #186

Open petermch opened 4 years ago

petermch commented 4 years ago

I've noticed a minor bug when calculating carbonate U-series ages where the program is not properly calculating the d234Ui analytical uncertainty.

noahmclean commented 4 years ago

Hi @petermch thanks for getting in touch! Can you send me more info? Just the input data you're using, as well as the uncertainty you're expecting for d234Ui should do it. Did the age and uncertainty come out as you'd expect?

Cheers, Noah

petermch commented 4 years ago

Hi Noah,

I was inputting coral U-series data from some samples I recently dated (so 234U/238U and 230Th/238U for the calculations). I had already calculated my ages using the internal lab spreadsheet, so that was how I knew something was off. When I checked the ET_redux output the age, age uncertainty, and d234Ui values all matched, but the d234Ui uncertainty did not and was very small (e.g. +/- 0.1 per mil). I also checked the detrital-corrected d234Ui uncertainty and that came out fine (~ +/- 1-2 per mil on average), which is what makes me think that it is an issue specifically with how the uncorrected d234Ui uncertainty is calculated. I cross-checked this with several of the example coral datasets and confirmed that the calculated d234Ui uncertainties are often not even the same magnitude as the published values.

Hope this helps, let me know if there is any other info I can provide!

Best,

Peter

bowring commented 2 years ago

@noahmclean - can you follow up on this? Thanks