Experimental replacement of zircon by melt-mediated coupled dissolution-precipitation causes dispersion in U-Pb ages

Zircon geochronology provides critical information on the rates and durations of geological processes and enables researchers to explore deep time. However, some zircon datasets show a continuum of concordant ages (‘smear’) without well-defined age populations. These age smears are typically interpreted to represent variable loss of radiogenic Pb or protracted geological events lasting tens of millions of years. Coupled dissolution-precipitation replacement of zircon has been suggested as one process that may produce these complex age datasets. Here, we react fragments of the well characterised Mud Tank zircon standard with natural intermediate and mafic melts (0.9 GPa, 1100–1180°C) to test if short-term exposure to a melt can modify the geochronological patterns of zircon. Our observations show that within a short duration (18 h to 3.5 days), most Mud Tank zircon fragments display microstructural and/or chemical evidence for modification by dissolution at fragment boundaries along with partial replacement by coupled dissolution-precipitation processes. The replaced zircon domains have U–Pb ages that smear over one hundred million years, between 764–647 Ma, illustrating variable mobility and redistribution of the U and Pb isotopes. Our experiments demonstrate that zircon modified by coupled dissolution-precipitation replacement may not faithfully record the age or duration of geological events and that investigation of zircon microstructure in high-resolution backscattered electron, cathodoluminescence imaging and/or Raman mapping is needed to better understand complex zircon geochronological datasets.