What about the predecessors of the February 2023 earthquakes in Eastern Anatolia?

Abstract

When large earthquakes occur, it is natural enough to wonder about their likely predecessors (if any). This is why, after the earthquakes of February 2023, we began a review of the historical seismic record of Eastern Anatolia Our investigation concentrated on six major earthquakes/sequences occurred after 1000 AD (1003, 1114/1115, 1269, 1344, 1513/1514 and 1544). The earthquakes of 1822, 1872 and 1893 are better known and do not require priority investigation. The geopolitical history of this region is very complex, with many changes of rulers along the centuries, but its long-term seismic history is rather well known. Many earthquakes are on record both before and since the date - some 2100 years ago - when the region became a Roman province. Many of its main towns (Aleppo and Antakya for instance) have a long-term history that includes eyewitness observations of many strong earthquakes. Historical earthquake records for this area were collected and studied several times. The latest studies are Soysal et al. (1981), Ambraseys and Finkel (1995), Guidoboni & Comastri (2005), Sbeinati et al. (2005), Tan et al. (2008). N.N. Ambraseys wrote many papers on this subject and compendiated his work in Ambraseys (2009). However, time, epicentral location and size of many earthquakes are debatable and earthquake catalogues propose contrasting values for the same events. Our work consisted of: a) retrieving and analyzing the main historical sources for each earthquake; b) identifying the localities mentioned in the sources and assessing macroseismic intensities from the original information; c) determining earthquake parameters (Io, Mw and - whenever possible - source azimuth) with the “Boxer” method (Gasperini et al., 1999), after properly calibrating the relevant coefficient by using recent earthquakes of the Anatolian region. Fig. 3 presents the seismological results of this work: Mw (with uncertainty equal to 0.3) and the “boxes” obtained with the Gasperini et al. (1999) procedure and representing the surface projection of the possible earthquake sources: the epicentre is in the middle of the “box”. The epicentral location and Mw calculated for the main earthquake of 1114 are close to those of the first event of 6 February, 2023. The “box” seems to match the Pazarcik segment of the EAFZ (we refer hereafter to the fault definitions used by Duman and Emre, 2013 and by Duman et al., 2018). The 1269 earthquake was less energetic than the 1114 one, and its parameters are less well constrained. Its “box” suggests the Amanos segment as the likeliest source, with the Toprakkale segment as an alternative candidate. The 1344 earthquake is rather well known and was indeed a very large one. On account of its location it was not considered in the debate on the 2023 earthquake source. However, the identification of its source would be helpful for the understanding of seismicity in this region. As for the 1513/1514 earthquake, the first interpretation by Ambraseys (1988) was -and still is – considered as the absolute truth by literature, leading to a strong connection with the Pazarcık segment. Unfortunately, this interpretation is founded on poor information, as later stated by Ambraseys (2009) and confirmed by us. Though we cannot provide reliable epicentral location and magnitude estimates, we believe that the Toprakkale or Karataş segments could represent a more appropriate option for the source. Similar considerations can be proposed for the 1544 earthquake, whose informative background is also very weak. It could be located in the area where the recent M7.6 took place (Çardak fault), with Mw around 6.9.