QUANTITATIVE UNCERTAINTY ESTIMATES FOR COSMOGENIC CHLORINE-36 RESULTS

Cosmogenic nuclides commonly provide exposure ages in many geologic investigations, including moraines, lava flows, and landslides, among others. When using cosmogenic nuclides, most researchers report only the analytical uncertainty for the age, which includes only uncertainties calculated from the accelerator mass spectrometer results. This artificially reduces the apparent uncertainty of the result by neglecting uncertainties from laboratory processing, scaling models, geologic situation, production rates, and other physical parameters. Without the large datasets necessary to provide the statistical basis for calculating quantitative uncertainties, reporting the analytical uncertainty is the only real option. The CRONUS-Earth Project produced datasets appropriate for this assessment. Using those datasets, the uncertainty of the chlorine-36 technique was bounded using the Root Mean Square Error (RMSE) of the primary and secondary chlorine-36 datasets, yielding a technique uncertainty between 7.5-20.0%. The individual production rate uncertainties were calculated using a camel-plot analysis to look at the variation in sample uncertainty with changes in chlorine content. Production rate uncertainties for spallation and low-energy neutron pathways were calculated based on the results and yielded a 4% uncertainty for spallation pathways and 20% uncertainty for the low-energy neutron production pathway. These quantitative uncertainty estimates provide realistic assessments of the uncertainty and can be used to help researchers more accurately compare their cosmogenic chlorine-36 results to other cosmogenic nuclide results as well as results from other dating methods.