Seasonal Nontectonic Loading Inferred From cGPS as a Potential Trigger for the M6.0 South Napa Earthquake

Abstract

We analyze crustal strain corresponding to transient continuous Global Positioning System (cGPS) horizontal displacements in Northern California, detecting a seasonal positive dilatational strain and Coulomb stress transient in the South Napa region peaking just before the 24 August 2014 M6.0 South Napa earthquake. Using data from 2007 to 2014, we show that average dilatational strain within a 500‐km2 region encompassing South Napa and northern San Pablo Bay peaks in late summer at 76 ± 17 × 10−9, accompanied by a Coulomb stress change of 1.9 ± 0.8 kPa. The situation reverses in winter, with an average dilatational strain of −51 ± 17 × 10−9 and Coulomb stress change of −1.4 ± 0.8 kPa. Within a smaller 100‐km2 area centered on the South Napa rupture, peak values are considerably higher, including a summer Coulomb stress peak of 5.1 ± 1.6 kPa. We examine regional seismicity but see no statistically significant correlation with seasonal Coulomb stressing in the declustered earthquake catalog. Using western U.S. vertical cGPS displacements, we estimate that strain from hydrologic loading explains ≤10% of the observed long‐wavelength strain and only 2–3% of peak strains around the South Napa rupture. Thermoelastic crustal strain estimated from temperature gradients between the San Francisco Bay and Sacramento Valley reaches values as high as 15% of the observed strain, but the strain patterns are not spatially consistent. Vertical deformation within the Sonoma and Napa Valley subbasins inferred from interferometric synthetic aperture radar explains large horizontal motions at nearby cGPS stations and suggests that seasonal changes in groundwater may contribute to observed strain and stress transients.

Spatiotemporal Correlation Between Seasonal Variations in Seismicity and Horizontal Dilatational Strain in California

“…They are so small partly because of our approach to focus on large‐scale strain signals that can be robustly resolved. Consequently, our Coulomb stresses are also at least an order of magnitude smaller than those reported by Kraner et al () for the Napa fault region in northern California. Our Coulomb stress variations are also at least an order of magnitude smaller than those predicted by hydrological loading (Amos et al, ; Johnson et al, , ).…”

“…Geophysical Research Letters, 45, 9559-9568. https://doi.org/10.1029/ 2018GL079536 1994), surface temperature (Prawirodirdjo et al, 2006), snow (Drouin et al, 2016;Heki, 2001) or water loading (Bettinelli et al, 2008;van Dam et al, 2001), or a combination therein (Johnson et al, 2017b;Tsai, 2011). The majority of the above studies used vertical GPS displacement time series, but the effect of surface loading has also been seen in horizontal time series (Argus et al, 2005;Chanard et al, 2014;Elósegui et al, 2003;Fu et al, 2013;Grapenthin et al, 2006;Kraner et al, 2018).…”

“…For example, seasonal displacements in GPS time series have been correlated with annual variations in atmospheric pressure (e.g., van Dam et al, ), surface temperature (Prawirodirdjo et al, ), snow (Drouin et al, ; Heki, ) or water loading (Bettinelli et al, ; van Dam et al, ), or a combination therein (Johnson et al, ; Tsai, ). The majority of the above studies used vertical GPS displacement time series, but the effect of surface loading has also been seen in horizontal time series (Argus et al, ; Chanard et al, ; Elósegui et al, ; Fu et al, ; Grapenthin et al, ; Kraner et al, ).…”

Crustal Strain Patterns Associated With Normal, Drought, and Heavy Precipitation Years in California

“…The horizontal spatial derivatives of the displacements in Equation (1) yield model estimates of horizontal strain on a 0.1° × 0.1° curvilinear grid (Kraner et al., 2018), where the Bessel form of bicubic spline interpolation (deBoor, 1978) is used to define the values and spatial derivatives of at grid nodes (Beavan & Haines, 2001; Haines et al., 1998). The curvilinear grid we use for this study is set up to study transients within the transform‐dominated plate boundary zone in California.…”

Stress Promotion of the 1958 Mw∼7.8 Fairweather Fault Earthquake and Others in Southeast Alaska by Glacial Isostatic Adjustment and Inter‐earthquake Stress Transfer

“…Luttrell and Sandwell (2010) and Neves et al (2015) assessed how post-LGM sea-level rise could have modulated the stress on the San Andreas Fault and strike-slip faults in Portugal, respectively, although those effects would have occurred over timescales much longer than the rapid post-1770 deglaciation in southeast Alaska. Seasonal hydrological loading has also been shown to influence seismicity in strike-slip regimes in central California (Amos et al, 2014;Johnson et al, 2017;Kraner et al, 2018), New Madrid (Craig et al, 2017), central and southwest Japan (Heki, 2003), and possibly in south-central Alaska (Johnson et al, 2020). However, the load changes in these cases are on the order of <1 m equivalent water height, in contrast to the >1 km cumulative ice loss in Glacier Bay.…”