Revealing the Earth’s mantle from the tallest mountains using the Jinping Neutrino Experiment

Abstract

The Earth’s engine is driven by unknown proportions of primordial energy and heat produced in radioactive decay. Unfortunately, competing models of Earth’s composition reveal an order of magnitude uncertainty in the amount of radiogenic power driving mantle dynamics. Recent measurements of the Earth’s flux of geoneutrinos, electron antineutrinos from terrestrial natural radioactivity, reveal the amount of uranium and thorium in the Earth and set limits on the residual proportion of primordial energy. Comparison of the flux measured at large underground neutrino experiments with geologically informed predictions of geoneutrino emission from the crust provide the critical test needed to define the mantle’s radiogenic power. Measurement at an oceanic location, distant from nuclear reactors and continental crust, would best reveal the mantle flux, however, no such experiment is anticipated. We predict the geoneutrino flux at the site of the Jinping Neutrino Experiment (Sichuan, China). Within 8 years, the combination of existing data and measurements from soon to come experiments, including Jinping, will exclude end-member models at the 1σ level, define the mantle’s radiogenic contribution to the surface heat loss, set limits on the composition of the silicate Earth, and provide significant parameter bounds for models defining the mode of mantle convection.

Reference Models for Lithospheric Geoneutrino Signal

“…The estimated BSE radiogenic heat production of TW is consistent within 1‐sigma of the low, medium, and high‐H BSE models with the central value coincident with the medium‐H model ( TW). Furthermore, the prediction given here using LITHO1.0 is consistent with that predicted by the Šrámek et al () model ( TW) when using the updated Watanabe () KamLAND measurement (personal communication). Based on Figure , an observed mantle signal of 8.5 TNU with uncertainty below some TNU is needed to discriminate between the different BSE compositional models.…”

“…Increased correlation causes an inverse effect on uncertainty for the bulk crustal K/U and Th/U, resulting in less uncertainty than described by Huang et al (). Some previous geoneutrino signal predictions did not include uncertainty on geophysical inputs, including crustal thickness (e.g., Enomoto et al, ; Fiorentini et al, ; Šrámek et al, ), meaning that these studies report smaller uncertainty on the geoneutrino signal. Similar to signal, the uncertainty on continental crustal heat production is 50% larger than that estimated by Huang et al () ( TW).…”

“…Discrepancy between model predictions of geoneutrino signal at KamLAND from Enomoto et al () (28.2 TNU), Fiorentini et al () (26.5 1.52 TNU), Huang et al ()(20.6 TNU), and Šrámek et al () (24.2 3.5 TNU). Values share the same measured total signal ( axis) from Watanabe () (34.9 TNU) but are offset for visibility.…”

“…Huang et al () is built on a combination of the physical models CRUST2.0, CUB2.0 (Shapiro & Ritzwoller, ), and GEMMA1.0 and has been used in the prediction of signals at SNO (Huang et al, ; Strati et al, ) and JUNO (Strati et al, ), among others. Finally, Šrámek et al () calculated the signal at Jinping with a model built upon CRUST1.0 (Laske et al, ).…”

Stochastic modeling of 3-D compositional distribution in the crust with Bayesian inference and application to geoneutrino observation in Japan

“…For rock composition models, the log-normal distribution has been widely used in geochemistry including geoneutrino modeling studies (Huang et al, 2013;Huang et al, 2014;Šrámek et al, 2016). Adopting a log-normal distribution for rock compositions was suggested by Ahrens (1954), based on their observation that igneous rocks commonly show skewed compositional distributions that do not fit to a normal distribution.…”