Production rate calibration

Here a global compilation of 10Be and 26Al production rate calibration data published 2009-2019 is presented in tab-delimited text files.


Calibration methodology:

All data has been divided into calibration sites including samples from one location with a single deglaciation age.
Calibration ages are derived from the original publications with radiocarbon ages generally recalibrated against IntCal13, Marine13 (Reimer et al. 2013), and SHCal13 (Hogg et al. 2013) using OxCal 4.2.
The 10Be and 26Al concentrations from single sample repeat measurements have been combined as uncertainty-weighted means. To account for interlaboratory measurement uncertainties (Jull et al. 2015), the highest of the reported (or uncertainty-weighted) concentration uncertainty or 2.9% (10Be) and 4.9% (26Al) of the concentration is used for the calibration.
For all calibration sites with at least three separate samples, an uncertainty-weighted site reference production rate is calculated for the expage calculator (prodrate.m). The calibration age uncertainty, which is not an independent variable for each sample of a calibration site, is added in quadrature to the initial site reference production rate.
To avoid data affected by prior and/or incomplete exposure, well-clustered site production rates are combined into a global 10Be and 26Al reference production rate. For scattered site production rates, with chi-square statistics p-value < 0.05 for the hypothesis that all sample production rates are drawn from one normal distribution, a maximum of 1/3 of the samples are excluded as outliers (still requiring a minimum of three separate samples). The site production rates that cannot pass this clustering test after outlier rejection are not included in the global reference production rate.
The reference 10Be production rate of calibration site 2 is significantly higher than all other site reference production rates with no overlap within 2σ. With both Peirce’s criterion and Chauvenet’s criterion this reference production rate is identified as an outlier, and it is therefore excluded from the global reference 10Be production rate.
For global average production rates, the arithmetic mean of the well-clustered site reference production rates are used. To estimate the production rate uncertainty the root-mean-square deviation of all single sample production rates included in the well-clustered sites is used. This yields the global reference 10Be (3.98 ± 0.22 atoms g–1 yr–1) and 26Al (28.44 ± 1.90 atoms g–1 yr–1) production rates.

Global production rate kml: prodrate.kml

Global production rate log: P-log.txt

The production rate calibration data is presented in the following text files:

P-sites – Calibration site and global average reference 10Be and 26Al production rates.

P-calib – Data used for calibration of reference 10Be and 26Al production rates.

P-origin – Original data with individual reported 10Be and 26Al measurements.

P-comments – Comments on the data.

P-refs – References for the production rate calibration.

To display special characters in the text files correctly, use UTF-8 (unicode) encoding.