Critical evaluation of chemical mohometry 2: Correlated changes in geochemistry, fO2, and temperature in the Gangdese Arc

Crustal thickness represents a key parameter for understanding the geodynamic behavior and tectonic evolution of the continental crust, but its estimation remains challenging. Several trace element geochemical proxies in arc rocks have been proposed to infer crustal thickness, yet they do not account for other factors that also influence the chemistry of magmas, including oxygen fugacity (f_O2), water fugacity, temperature and accessory mineral fractionation. Proxy calibrations also require significant averaging of data before robust correlations appear, leading to some preference for detrital zircon as a geochemical proxy because many data can be collected quickly.

The geochemistry and geochronology of the Gangdese arc in southern Tibet has been extensively studied, and several mohometry proxies have been tested using analyses in this region. In this study, we tested correlations among trace elements, temperature, and f_O2 through analysis of detrital zircon in modern sands from the central region of the Gangdese arc. The geochemistry of detrital zircon has been shown previously to parallel whole-rock changes, so its geochemistry should serve as a mohometry proxy. LA-ICP-MS spot analyses simultaneously resolved U-Pb ages and trace element concentrations. Temporal changes in europium anomaly, Th/Yb, and Sm/Yb – all proposed as mohometers – were compared with previous zircon and whole-rock data, as well as with geochemical proxies for temperature (Ti proxy) and f_O2 (Ce proxy).

With decreasing age from ~90 to ~15 Ma, zircon data show correlated changes with decreasing temperature and increasing Th/Yb, Eu anomaly, and f_O2. Sm/Yb shows no clear trend through time, although all measurements correlate positively with temperature. Previously published whole-rock La/Yb shows large scatter in ~90 Ma samples, and a significant increase between ~55 and ~15 Ma, in parallel with Th/Yb, Eu anomaly, and f_O2, and opposite temperature.

Although increases in La/Yb and Eu anomaly have previously been interpreted to indicate an increase in crustal thickness between at least ~60 Ma and ~15 Ma, our new data show these trends cannot be decoupled from correlated trends in f_O2 and temperature. Thus, as yet, a trend in crustal thickness through time is not resolvable for the Gangdese arc and likely not for other locations, too. These observations highlight the need for methodological and theoretical improvements that correct for effects of other parameters that influence magma chemistry besides crustal thickness.