Date of Award


Degree Type


Degree Name

Master of Science (MS)


Earth, Atmospheric, and Planetary Sciences

Committee Chair

Darryl Granger

Committee Member 1

Nathaniel Lifton

Committee Member 2

Marc Caffee


Cosmogenic nuclides are widely used to assess catchment-averaged denudation rates, although the method has been largely limited to the nuclide 10Be in the target mineral quartz. This is because (a) quartz is resistant to weathering, (b) the production rate of 10Be in quartz has been extensively studied and is well-constrained, and (c) quartz may be reliably separated from other minerals and cleaned of meteoric 10Be. Denudation rates have remained difficult to measure in volcanic landscapes due to a lack of quartz and the rapid weathering of olivine and other alternative target minerals. This creates an interest in developing new target minerals that may be used to reliably determine denudation rates where quartz is absent.

Magnetite is a promising target mineral candidate in several regards: (a) it is present in a range of rock types, including many basic igneous rocks that lack quartz, (b) like quartz it is resistant to chemical weathering, and (c) it may be collected in bulk and purified using simple magnetic separation techniques. In this thesis, procedures intended to isolate the in situ-produced 10Be and 36Cl signals in magnetite and to prepare magnetite samples for 10Be and 36Cl measurement by AMS are presented. The production rates of 10Be and 36Cl in magnetite are then calibrated using late-Pleistocene age glacial erratics from the eastern Sierra Nevada, California and banded iron samples from a slowly eroding ridge in the Quadrilátero Ferrífero, Minas Gerais, Brazil. Finally, the calibrated production rates are applied to determine denudation rates at 12 catchments in the Sierra Nevada region that encompass a range of weathering environments.

The results produce production rate estimates of 2.13 +/- .08 atoms g-1 yr-1 for 10Be and 1.24 +/- .03 atoms g-1 yr-1 for 36Cl in pure magnetite at sea level and high latitude. Denudation rates from 10Be in magnetite are often biased low by residual meteoric 10Be, whereas denudation rates from 36Cl reproduce those from 10Be in quartz within the expected range of variability. Thus, 36Cl in magnetite may serve as a suitable nuclide/target mineral pair for determining denudation rates in the absence of quartz. However, more work is needed to precisely calibrate the production rate of 36Cl from titanium, which readily substitutes for iron in magnetite’s crystal structure, and to reduce the native chloride content of magnetite samples to negligible levels.