The North Anatolian Fault (NAF) is a right‐lateral plate boundary fault that arcs across northern Turkey for ∼1500 km. Almost the entire fault progressively ruptured in the 20th century, its cascading style indicating that stress from one fault rupture triggers fault rupture of adjacent segments. Using published paleoseismic investigations, this study integrates all of the existing information about the timing of paleoearthquakes on the NAF. Paleoseismic investigation data are compiled into a database, and for each site a Bayesian, ordering‐constrained age model is constructed in a consistent framework. Spatial variability of recurrence intervals suggests a spatial pattern in the behavior of earthquakes on the NAF that may correspond to the tectonic provinces within the Anatolian plate. In the west, the shear stress associated with the escaping Anatolian plate interplays with the tensile stress associated with the Aegean extensional province. Along this western transtensional section we recognize short recurrence intervals and switching between the furcated fault strands. The central section of the NAF is translational with little influence of fault‐normal stresses from other tectonic sources. This section tends to rupture in unison or close succession. The eastern section of the NAF is transpressional due to the compressional fault‐normal stress associated with the indenting Arabian plate. Along this section the recurrence intervals are bimodal, which we attribute to variable normal stress, although there are other possible causes.

This paper summarizes developed guidance on the utilization of earthquake-induced ground motion simulations for engineering practice. Attention is given to the necessary verification, validation and utilization documentation in order for confidence in the predictive capability of simulated motions to be established. The construct of a ground motion simulation validation matrix is developed for assessing the appropriateness of a particular suite of simulated ground motions from the perspective of region-to-site-specific application and for different specific engineering systems. Appropriate validation metrics and “pass” criteria, the consideration of modeling uncertainties, and limitations associated with a relative lack of validation data are also addressed. An example is utilized in order to demonstrate the application of the guidance. This document is intended to be bidirectional in the sense that it provides guidance for earthquake engineers on the appropriateness of a suite of ground motion simulations for utilization in a site-specific context, as well as ground motion simulators to understand the context in which their results will be utilized.

The Hazar Basin is a 25 km‐long, 7 km‐wide and 216 m‐deep depression located on the central section of the East Anatolian Fault zone (eastern Turkey) and predominantly overlain by Lake Hazar. This basin has been described previously as a pull‐apart basin because of its rhombic shape and an apparent fault step‐over between the main fault traces situated at the southwestern and northeastern ends of the lake. However, detailed structural investigation beneath Lake Hazar has not been undertaken previously to verify this interpretation. Geophysical and sedimentological data from Lake Hazar were collected during field campaigns in 2006 and 2007. The analysis of this data reveals that the main strand of the East Anatolian Fault (the Master Fault) is continuous across the Hazar Basin, connecting the two segments previously assumed to be the sidewall faults of a pull‐apart structure. In the northeastern part of the lake, an asymmetrical subsiding sub‐basin, bounded by two major faults, is cross‐cut by the Master Fault, which forms a releasing bend within the lake. Comparison of the structure revealed by this study with analogue models produced for transtensional step‐overs suggests that the Hazar Basin structure represents a highly evolved pull‐apart basin, to the extent that the previous asperity has been bypassed by a linking fault. The absence of a step‐over structure at the Hazar Basin means that no significant segmentation boundary is recognised on the East Anatolian Fault between Palu and Sincik. Therefore, this fault segment is capable of causing larger earthquakes than recognised previously.

No abstract

The North Anatolian Fault (NAF) is a ~1500-km-long, arcuate, dextral strike-slip fault zone in northern Turkey that extends from the Karliova triple junction to the Aegean Sea. East of Bolu, the fault zone exhibits evidence of a sequence of large (Mw>7) earthquakes that occurred during the 20 th century that displayed a migrating earthquake sequence from east to west. Prolonged human occupation in this region provides an extensive, but not exhaustive, historical record of large earthquakes prior to the 20 th century that covers much of the last 2000 years. In this study, we extend our knowledge of rupture events in the region by evaluating the stratigraphy and chronology of sediments exposed in a paleoseismic trench across a splay of the NAF at Destek, ~6.5 km east of Lake Ladik (40.868°N, 36.121°E). The trenched fault strand forms an uphill-facing scarp and associated sediment trap below a small catchment. The trench exposed a narrow fault zone that has juxtaposed a sequence of weakly-defined paleosols interbedded with colluvium against highly-fractured bedrock. We mapped magnetic susceptibility variations on the trench walls and found evidence for multiple visually unrecognized colluvial wedges. This technique was also used to constrain a predominantly dip-slip style of displacement on this fault splay. Sediments exposed in the trench were dated using both charcoal and terrestrial gastropod shells to constrain the timing of the earthquake events. While the gastropod shells consistently yielded 14 C-ages that were too old (by ~900 years), we obtained highly reliable 14 C-ages from the charcoal by dating multiple components of the sample material. Our radiocarbon chronology constrains the timing of seven large earthquakes over the past 3000 years prior to the 1943 Ladik earthquake, including -1788AD, 1034-1321AD, 549-719AD, 17-585AD (1-3 events), 351BC-28AD, 700-392BC, 912-596BC. Our results indicate an average inter-event time of 385±166yrs