Coseismic slip of the 2020 Mw 6.4 Petrinja earthquake (Croatia) from dense geodetic benchmarks, optical image correlation and InSAR data

Abstract

The Mw 6.4 Petrinja earthquake (2020, Croatia) is among the strongest continental earthquakes that occurred in Eastern Europe for decades. In such low-strain contexts, the sparse terrestrial-monitoring (few seismic and geodetic stations) of rare but strong earthquakes often prevents a detailed analysis of their seismic source. Here, we take advantage of > 160 geodetic benchmarks and optical image correlation to obtain a dense near-field coverage of the coseismic surface displacements. The geodetic dataset is obtained by repeated measurements of benchmark networks designed for civilian purposes and constitutes a unique dataset of coseismic displacements in the near-field of the fault. The optical image correlation is based on pre-earthquake (December 2017) WorldView and post-earthquake (February 2021) Pleiades satellite images with a 50 cm resolution. We also complete these displacement fields with unwrapped coseismic interferograms based on Sentinel-1 products, except in the near field affected by decorrelation. These displacement fields are consistent and thus suitable for modeling the slip distribution of the Petrinja earthquake. The elastic inversion of the geodetic benchmarks revealed interesting characteristics of this event: the rupture occurred on a near-vertical strike-slip fault, at a shallow depth (< 10 km), with significant slip reaching the surface. It also suggests that the deformation was partly accommodated by a subparallel strand 2.5 km from the main source northward. The aim of this research is to improve the source model of Petrinja 2020earthquake sequence, with a joint inversion of the geodetic benchmarks, optical image correlation, and InSAR data. Nevertheless, the comparison of the geodetic and coseismic offsets measurement on the field, shows that > 70% of the slip is likely distributed at the surface. Moreover, the coseismic strain maps derived from the unique benchmark data set helped us to identify zones where deformation appears distributed. Finally, the new data raises questions about whether such moderate earthquakes are accompanied by subsurface off-fault deformation or residual elastic strain.