The 2020-12-29 (Mw 6.4) Croatia Earthquake

The central segment of a SE-striking right-lateral fault that passes west of Petrinja ruptured at approximately 12:20 PM CET (11:20 UTC) on 29 December 2020 (Figure 1). In this study, we investigate the coseismic slip distribution of the event, which sheds light on its kinematic rupture process.

Figure 1. The colored dots show the seismicity from 1971 to the end of 2020, as reported by the USGS. The white box shows the bounds of Figure 4 and encompasses the rupture area of the 29 December 2020 earthquake. Labels A, B, C, and D represent the endpoints of profiles AB (Figure 5) and CD (Figure 6). R and V represent a river to the northwest and a valley to the southeast of the main slip area, MSP. The upper focal mechanism is found in this study, and the lower one is reported by the Global CMT Project (Ekströem et al., 2012). The red lines show faults, and the black boxes represent towns.

1. Slip Inversion

To derive the slip distribution, we used the teleseismic body-waveform slip inversion method of Kikuchi et al. (1993). P-recordings of 148 seismic stations in the epicentral distance of 20° to 95° were used. The recordings were filtered in the frequency range of 0.015 to 0.5 Hz to capture details of the rupture process. 

Figure 2 shows the synthetic and observed seismograms. The inverted coseismic slip reveals two slip patches at depths less than 15 km (Figure 3), a major one above the hypocenter and a weaker one in the southeast. The main patch, MSP, has 15 km along-strike and 15 km along-dip dimensions, and its maximum slip is about 1 meter. The second patch, SSP, has a dimension of about 3 km by 3 km with a maximum slip of about 0.4 m. This secondary deeper patch is observed in other frequency bands and appears to be not an artifact of the modeling.

The observed SH seismograms showed a relatively poor match with the synthetic waveforms and therefore were removed from the inversion process.


Figure 2. The synthetic seismograms are shown in red and the observed waveforms are in black. The number to the right of each seismogram shows the station number on the map in the middle. The four parameters to the left of each seismogram represent wave-type/amplitude in µm, station name, azimuth, and epicentral distance. The moment rate graph in the top-right indicates a total rupture time of 13 s.
Figure 2. Continued

Figure 3. Slip distribution of the 29 December 2020 earthquake. R and V correspond to the river and valley marked in figures 1, 4-7. MSP stands for “Main Slip Patch” while SSP represents “Secondary Slip Patch”. The geographical orientations are added to the upper corners of the figure. The red star shows the hypocenter, and the black arrows show the slip vectors on the hanging-wall (the southwestern block in figures 1, 4-7).

2. Source parameters of the earthquake

The computed fault parameters for the event are consistent with those found by other studies, notably with the Global CMT solution, which reported a moment magnitude (Mw) of 6.4. Using the body-waveform slip inversion, we determined a focal mechanism of strike 132⁰, dip 82⁰, and rake 173⁰, which runs parallel to an existing fault (Figure 4). The spatial distribution of aftershocks confirms the SW-striking nodal plane as the causative fault plane (Figure 7). A depth of 11 km resulted in a minimum misfit solution, consistent with the 10 km depth reported by different seismological observatories. Source time function for the event shows a total rupture time of 13 s with the significant energy release during seconds 1 to 6 (Figure 2, top-right).

Figure 4. The black contours, 0.2 m intervals, show the slip distribution of the 29 December 2020 earthquake projected on the Earth’s surface. Figure 3 shows the slip distribution on the fault plane. The colored circles show the two foreshocks (in orange), the mainshock, and the early aftershocks, reported by the USGS. The events are colored based on date. However, the locally-recorded seismicity indicates a clustered foreshock activity (Figure 7, frames 1 to 3).

3. Implications of the slip distribution

Transpressional fault systems cause positive topography along the associated fault (Fossen et al., 1993; Dewey et al., 1998). The minor reverse component of the 2020/12/29 earthquake, as inferred from the rake angle of

Transpressional fault systems cause positive topography along the associated fault (Fossen et al., 1993; Dewey et al., 1998). The minor reverse component of the 2020/12/29 earthquake, as inferred from the rake angle of 173⁰, causes positive topography in the long-term (Figures 4, 5, and 6). A fault-normal compressive component increases the coupling on the fault plane; therefore, it may develop a barrier against rupture propagation. The following points can be stated on the connection of seismicity, slip distribution, and the geomorphological features in the macro-seismic area:

  1. The main slip patch, MSP, is delimited in the NW by a river R, and in the SE by a valley, V (Figure 3). This portion of the fault correlates with a hill (Figure 6). 
  2. The secondary slip patch, SSP, partially ruptured the down-dip of the fault segment to the southeast of the valley V. Its shallow counterpart did not rupture due to its higher shear strength.
  3. The area between V and the northwestern end of SSP did not rupture during the earthquake, most likely due to high shear strength. This portion corresponds with a prominent positive topography, which asserts the strength of the associated fault segment.
  4. The hill corresponding to SSP has a smaller elevation compared to its immediate northwestern neighbor. It may indicate a weaker fault segment and explain the rupture of SSP (Figures 4 and 6).
  5. The majority of the aftershocks fall in the hanging block, to the southeast of the fault plane (Figures 5 and 7).
  6. Some of the aftershocks cover the northwestern continuation of the causative fault for a length of about 45 km. Provided that the whole 45 km fails in a single earthquake, a magnitude of about 6.9 is expected.
  7. Aftershocks located to the southeast of valley V, as well as the rupture of SSP, indicate that the area to the southeast of valley V is not in the early stages of its earthquake cycle; otherwise, the SSP should not have ruptured. An earthquake cycle starts right after a characteristic earthquake and ends with the next characteristic earthquake at the same location. An earthquake cycle may take centuries.
  8. Continuity of the fault-parallel topography between valley V and the border river W may infer a single characteristic event for the VW segment (Figure 6). 
  9. Assuming the area under a fault-associated topography profile (Figure 6) corresponds with the causative seismic moment, the future earthquake confined within V and W will be 0.3 units larger (i.e. Mw 6.7) than the 2020/12/29 earthquake. The area under the VW profile is three times larger than that under the RV profile.
Figure 5. The fault-normal profile AB passes through the main slip area, MSP. The shallow part of the slip contours coincides with the red line representing the fault. The fault separates the hills snd the plain.
Figure 6. Profile CD passes through the crest of the fault-parallel hills. The extent of the main slip patch, MSP, and the secondary slip patch, SSP, are demarked by red lines above the profile. River R, Valley V, and the border river W delimit sections of the fault system.
Figure 7. Colored circles show the locally recorded foreshocks and aftershocks of the 2020/12/29 earthquake. Seven frames, with no geographical coordinates, are taken from here. Features such as the fault line and the black slip contours are carefully added. The red circle in frame 3 shows the Mw 6.4 mainshock. The chosen frames show the jump in seismicity to the neighboring segments.

Acknowledgments

Some of the figures are created using GMT (Wessel, et al., 2019). Figures 5 and 6 are created using Google Earth. We appreciate Hadi Ghofrani for peer-reviewing the note.

References

Ekströem, G., M. Nettles, and A. M. Dziewonski (2012), The global CMT project 2004-2010: Centroid-moment tensors for 13,017 earthquakes, Phys. Earth Planet. In., 200, 1–9, doi: 10.1016/j.pepi.2012.04.002.

Dewey, J. F., Holdsworth, R. E., and Strachan, R. A. (1998). Transpression and transtension zones. Geological Society, London, Special Publications, 135(1), 1-14.

Fossen, H., and Tikoff, B. (1993). The deformation matrix for simultaneous simple shearing, pure shearing and volume change, and its application to transpression-transtension tectonics. Journal of Structural Geology, 15(3-5), 413-422.

Kikuchi, M., Kanamori, H., and Satake, K. (1993). Source complexity of the 1988 Armenian earthquake: Evidence for a slow after‐slip event. Journal of Geophysical Research: Solid Earth, 98(B9), 15797-15808.

Wessel, P., Luis, J. F., Uieda, L., Scharroo, R., Wobbe, F., Smith, W. H. F., and Tian, D. (2019). The generic mapping tools version 6. Geochemistry, Geophysics, Geosystems, 20(11), 5556-5564.

Last Updated on September 1, 2021