GNGTS 2018 - 37° Convegno Nazionale

GNGTS 2018 S essione 1.1 23 minimum is compensated by the slip occurred on a N-S striking synthetic fault which contribute to maintain the westward downthrow definitely above 2 m. Nevertheless, it is noticeable that where CT values >150 m are observed, they seem to be partly counteracted by the significant slip occurred along a N-S striking, 2 km long, antithetic fault on which CT values up to ~ 60 cm, have been measured. Enlarging the 30 October ruptures analysis to the entire VBF, and considering the coseismic faulting due to all theM>5.5 events, the total cumulative surface rupture length (SRL), calculated along the principal fault is 30 km. Unfortunately, we could not solve the uncertainties about the SRL of the 26 and 30 October events separately. Most or the entire surface displacement on the Cupi-Ussita sections is due to 26 October shocks, but it is not clear if and how the 30 October event reactivated these sections. Our field measurements confirm that the 30 October event ruptured entirely the VRS and Bove-Porche segment, at least up to the latitude of Ussita, for a total SRL of ~22 km. This value can be considered the total (minimum) rupture length of the 30 October mainshock. The maximum cumulative displacement (MD), recorded on the Redentore fault section of the VRS, is 240 cm, while the maximum displacement of the. 30 October event is 222 cm. The latter was recorded close to, but not exactly on the same point of the cumulative MD. The average cumulative displacement (AD) calculated as both integral and arithmetic means are 36 and 37 cm, respectively. Discussion and Conclusions. The FSII, Redentore section of the VRS is characterized by a zone of very high maximum displacement (MD) localized in a ~700 m-long trace. Recently Boncio et al. (2018) propose that at least part of MD (~20%) can be due to hanging wall back- tilting. Back-tilting can accommodate volumetric problems in the hanging wall due to up-dip steepening of the fault. Considering that the fault is on limestone bedrock, the steepening should be due to dip variations of the bedrock fault surface. In the same area there is an antithetic hanging wall fault rupture located 400 - to - 650 m from the principal fault. Also antithetic faulting might result from volumetric adjustment in the hanging wall due to fault dip variations at depth. Interestingly, in the same trace of the Redentore FS the slip vectors, both long-term and coseismic slickenlines deviate of 20-40° from the average (N230°) slip vector, while the fault strike is nearly constant. In summary, the localized anomalously-high maximum displacement can be due to local tectonic factors due to fault irregularity. We compared the obtained SRL, AD and MD data of the 24 August, 30 October and “cumulative” surface ruptures with global coseismic data from the literature in the form of empirical relationships (Wells and Coppersmith, 1994; Wesnousky, 2008; Pavlides and Caputo, 2004) and normalized displacement profiles. For the cumulative rupture, we used a magnitude M w = 6.7, obtained by summing the seismic moment of the 24 August-to-30 October M>5.0 earthquakes. The comparison with empirical relationships shows that the 24 August rupture parameters are well within the values predicted by the empirical regressions. Instead, the parameters for the 30 October and cumulative rupture divert significantly from the empirical regressions for M w vs rupture parameter. The MD is much higher than the expected values and AD is significantly lower than the expected value. SRL approaches the empirical regressions better than MD and AD, but it is still higher than the expected values. Comparing the obtained displacement profile, normalized by the maximum displacement, D/MD, plotted against the normalized fault length, L/SRL, with a compilation of profiles for normal faulting earthquakes published by Wesnousky (2008), the 2016 profile locates close to the lower boundary of global normal faulting field. This is clearly an effect of the very high 2016 MD. As aforesaid, the MD value, exceeding the expected, might be due in part to localized tectonic phenomena. The AD value, lower than the expected, might be probably caused by an attenuation of coseismic displacement toward the surface, possibly due to the high segmentation of the fault system. We conclude that caution is needed in using regressions based on MD and AD because it can not be ruled out that other global earthquakes might be affected by comparable problems.