The global lead isotope system: Toward a new framework reflecting Earth's dynamic evolution

Abstract

The U-Th-Pb system is perhaps one of the most versatile isotopic systems in use by Earth scientists, widely applied to both date and compositionally trace geological events through Earth history. However, the Pb isotope systematics of the Earth are subject to two major paradoxes. Assuming our planet evolved uniformly from a chondritic composition, all the present-day Earth chemical reservoirs should plot on the 4.55 Ga meteorite isochron, also known as the geochron; but in fact, all known reservoirs are more radiogenic (having excess 206Pb and 207Pb) than the carbonaceous chondrites, constituting the first Pb paradox. The second Pb paradox (also called the kappa conundrum) is the apparent difference between the measured 232Th/238U ratio (κ) of oceanic basalts and the time-integrated 232Th/238U ratio (κPb) predicted from the Pb isotope ratios. While significant progress has been made since the realization of the two Pb paradoxes over 50 years ago, the persistence of these issues highlights the limitations in our current understanding of the Earth's evolution with respect to U, Th and Pb, as we are neither able to ascertain the composition of the present-day bulk silicate Earth (BSE; comprising Earth's mantle and its crust) nor to determine the starting time(s) of U/Pb and Th/U fractionations in the mantle. In this contribution, we first review the Pb paradoxes and their proposed solutions. We then discuss the previously proposed Pb evolution models and establish a new framework based on a reassessment of global data and current understanding of Earth dynamics. Our model invokes the presence of distinct Pb isotope evolution paths for a diverse range of segregated components of the BSE, with the present-day upper continental crust being one of the end members. The model also features a two-stage Pb evolution for the silicate Earth, with a data-defined ca. 3.2 Ga start time for major compositional differentiation and remixing, possibly due to the initiation of global plate tectonics at that time. We readdress the first Pb paradox through recognizing that, to the first order, the Pb isotopic values of present-day Earth materials lie on a Pb differentiation line defined by our re-estimated present-day BSE and continental crust, and proposing that the data plot mostly to the right of the geochron due to second-order complications caused by both source mixing and fractionation of Earth materials. We argue that rocks found on Earth's surface mostly originated from more radiogenic reservoirs (with HIMU being an end member) at shallower levels, where long-term gravitational differentiation and subduction-led mantle remixing preferentially concentrated more radiogenic materials. We also largely mitigated the second Pb paradox through an updated κ vs. κpb plot using modern global databases, which shows a general agreement between the mean κ and κpb values. We further demonstrate that the choice of Pb evolution models has potentially profound implications when applying non-radiogenic Pb corrections during U-Pb dating of Earth materials.