Earthquake Seismic Moment, Rupture Radius, and Stress Drop From P-Wave Displacement Amplitude Versus Time Curves

Abstract

The reliable determination of earthquake source parameters is a relevant task of seismological investigations which ground nowadays on high-quality seismic waveforms collected by near-source dense arrays of ground motion sensors. Here, we propose a parametric modeling technique which analyzes the time-domain P-wave signal recorded in the near-source range of small-to-large-size earthquakes. Assuming a triangular moment rate function, a uniform speed, and circular rupture model, we develop the equations to estimate the seismic moment, rupture radius, and stress drop from the corner time and plateau level of the average logarithm of the P-wave displacement versus time curves (LPDT). The constant- <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> , anelastic attenuation effect, is accounted by a postprocessing procedure that evaluates the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q$ </tex-math></inline-formula> -unperturbed moment rate triangular shape. The methodology has been validated through application to the acceleration records of the 2016–2017 Central Italy and 2007–2019 Japan earthquake sequences covering a wide moment magnitude range (Mw 2.5–6.5) and recording distance <100 km. After correcting for the anelastic attenuation function, the estimated average stress drop and the confidence interval ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\langle \Delta \sigma \rangle =0.60$ </tex-math></inline-formula> (0.42–0.87) MPa and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\langle \Delta \sigma \rangle =1.53$ </tex-math></inline-formula> (1.01–2.31) for crustal and subcrustal events of Japan and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\langle \Delta \sigma \rangle =0.36$ </tex-math></inline-formula> (0.30–0.44) MPa for Central Italy) show for both regions a self-similar, constant stress drop scaling of the rupture duration/radius with seismic moment. The smaller sensitivity of the spatially averaged, time-varying peak displacement amplitude to the radiation from localized high slip patch on the fracture surface could explain the retrieved smaller average stress drops for subcrustal earthquakes in Japan and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$M &gt;5.5$ </tex-math></inline-formula> events in Central Italy relative to previous estimates using spectral methods.