Surface ages and weathering rates from 10Be (meteoric) and 10Be/9Be: Insights from differential mass balance and reactive transport modeling

To reconstruct the dynamic evolution of weathered material at the Earth's surface requires techniques that record the fluxes of material into and out of the weathering zone. Cosmogenic isotopes, formed in situ within the Earth's surface or in the atmosphere (“meteoric”) are important chronometers for both time and fluxes. Meteoric 10Be has been used to evaluate surface ages, soil residence times and rates of sediment transport; however, the meteoric 10Be dating tool relies on complete retention of 10Be on mineral surfaces within the weathered regolith, an assumption that will not be valid where soil pH is low or mineral surface area available for sorption is lacking. Here we evaluate whether the stable isotope of 9Be that is released upon weathering (“reactive” Be) can be used to correct for 10Be mobility. This approach relies on the inventory of reactive 9Be in the weathering zone as a measure of the mass depletion, whereas the 10Be inventory is a function of age. To explore the spatial and temporal relationship between the Be isotopes, we use a multi-component reactive transport model framework to evaluate a set of differential mass balance equations for the 10Be/9Be composition of the bulk solid and sorbed reservoirs on non-eroding surfaces. We first find that the (10Be/9Be)reac is inversely proportional to the long-term chemical weathering rate, even in cases of incomplete Be retention. Second, we find that coupling 10Be and 9Be to provide a tool for surface age dating (10Be) is complex because incomplete retention affects each isotope to a different extent. Beryllium-10 is supplied at the surface of soil, where sorption onto secondary minerals is more extensive, while 9Be is released from primary minerals at the weathering front where secondary minerals are less abundant. If 9Be loss is less than about 15%, we find 9Be inventories can be used to reliably correct 10Be inventories to within 10%. The error associated with the correction also increases with increasing age, and thus the correction is more favorable for younger surfaces. Over longer timescales, the error on the age introduced by the loss correction using 9Be is magnified, whereas the error associated with uncorrected ages decreases because the 10Be is more strongly retained as chemical weathering progresses. We also find that the reactive transport model confirms a set of simple mass balance descriptions, providing guidance for future data collection. Collectively, the modeling results suggest that measurement of Be isotopes in weathered regolith provides a unique approach to determine both surface ages and chemical weathering rates.