Hematite (U-Th)/He thermochronometry detects asperity flash heating during laboratory earthquakes

Abstract

Evidence for coseismic temperature rise that induces dynamic weakening is challenging to directly observe and quantify in natural and experimental fault rocks. Hematite (U-Th)/He (hematite He) thermochronometry may serve as a fault-slip thermometer, sensitive to transient high temperatures associated with earthquakes. We test this hypothesis with hematite deformation experiments at seismic slip rates, using a rotary-shear geometry with an annular ring of silicon carbide (SiC) sliding against a specular hematite slab. Hematite is characterized before and after sliding via textural and hematite He analyses to quantify He loss over variable experimental conditions. Experiments yield slip surfaces localized in an ∼5–30-µm-thick layer of hematite gouge with <300-µm-diameter fault mirror (FM) zones made of sintered nanoparticles. Hematite He analyses of undeformed starting material are compared with those of FM and gouge run products from high-slip-velocity experiments, showing >71% ± 1% (1σ) and 18% ± 3% He loss, respectively. Documented He loss requires short-duration, high temperatures during slip. The spatial heterogeneity and enhanced He loss from FM zones are consistent with asperity flash heating (AFH). Asperities >200–300 µm in diameter, producing temperatures >900 °C for ∼1 ms, can explain observed He loss. Results provide new empirical evidence describing AFH and the role of coseismic temperature rise in FM formation. Hematite He thermochronometry can detect AFH and thus seismicity on natural FMs and other thin slip surfaces in the upper seismogenic zone of Earth’s crust.

Shallow Rupture Propagation of Pleistocene Earthquakes Along the Hurricane Fault, UT, Revealed by Hematite (U‐Th)/He Thermochronometry and Textures

“…Mirrors can exhibit triple-junction-forming (polygonal) grains that lack shape and crystallographic preferred orientation within ∼50 μm of the slip surface (Ault et al, 2015(Ault et al, , 2019McDermott et al, 2017), analogous to textures observed in high-temperature experiments (T > 800°C; Siemes et al, 2003Siemes et al, , 2011Vallina et al, 2014). Locally mirrored zones comprising sintered nanoparticles created during high-velocity hematite deformation experiments yield >70% He loss during slip, consistent with transient temperature rise >800°C (Calzolari et al, 2020). We lack direct constraints on Hurricane fault slip temperatures but suggest they must be markedly lower than those achieved on hematite fault mirrors.…”

“…Domain 1 contains void spaces that exceed the He stopping distance (∼14-16 μm), likely causing excess He loss and the anomalously young date (Huber et al, 2019;Huff et al, 2020). Domain 2 cataclastic hematite likely retained its He budget since formation because deformation experiments show that comminution during seismic slip does not induce substantial He loss (Calzolari et al, 2020). The ∼0.65-0.5 Ma domain 2 dates record the time of initial hematite precipitation in the coseismic or immediate post-seismic period (Nuriel et al, 2019;Williams et al, 2017).…”

Nanoscale Textural and Chemical Evolution of Silica Fault Mirrors in the Wasatch Fault Damage Zone, Utah, USA

“…For example, both comprise sintered polygonal nanoparticles and anhedral particle aggregates in the FM volume and textures vary parallel and perpendicular to the FM surface (Figures 3 and 4; Ault et al., 2015; McDermott et al., 2017). However, on hematite FMs, localized, polygonal hematite crystals immediately below the FM surface are commonly surrounded by cataclastic hematite fragments, and the recrystallized hematite grains are inferred to reflect flash heating at paleoasperities (Calzolari et al., 2020; McDermott et al., 2017). In contrast, the silica FM volume is fully transformed, albeit with variable resulting textures at the µm‐scales.…”

“…FMs are thin (mm‐scale) and comprise layers of sintered, colloidal, and(or) polygonal nanoparticles (e.g., Kirkpatrick et al., 2013; Power & Tullis, 1989; Siman‐Tov et al., 2013). FMs have been characterized in natural and experimental shale and sandstone (Borhara & Onasch, 2020; Kuo et al., 2016), carbonate (Collettini et al., 2013; Fondriest et al., 2013; Pozzi et al., 2018; Siman‐Tov et al., 2013; Smith et al., 2013), hematite (Ault et al., 2015, 2019; Calzolari et al., 2020; Evans et al., 2014; McDermott et al., 2017), basalt (Power & Tullis, 1989; Ujiie et al., 2007), and chert and novaculite (Kirkpatrick et al., 2013; Rowe et al., 2019). Some processes invoked for generating FMs include thermal decomposition (Collettini et al., 2013), melting (Ault et al., 2019; Kuo et al., 2016), gel lubrication (Kirkpatrick et al., 2013), brittle‐plastic deformation and recrystallization (Fondriest et al., 2013; Pozzi et al., 2018; Smith et al., 2013), and asperity flash heating (Calzolari et al., 2020; McDermott et al., 2017).…”

“…, and chert and novaculite (Kirkpatrick et al, 2013;Rowe et al, 2019). Some processes invoked for generating FMs include thermal decomposition (Collettini et al, 2013), melting (Ault et al, 2019;Kuo et al, 2016), gel lubrication (Kirkpatrick et al, 2013), brittle-plastic deformation and recrystallization (Fondriest et al, 2013;Pozzi et al, 2018;Smith et al, 2013), and asperity flash heating (Calzolari et al, 2020;McDermott et al, 2017). Many of these processes are dynamic weakening mechanisms that require coseismic heat, although laboratory FMs have also been generated at sub-seismic slip rates (Verberne et al, 2013(Verberne et al, , 2014.…”