Seafloor Geodetic Monitoring of the North Anatolian Fault in the Sea of Marmara: System Installation and its Initial Result

Abstract:

The North Anatolian Fault (NAF) get across the mainland of Turkey is known as a quite active strike slip fault. The earthquake recurrence period for individual segment is estimated roughly 300 years based on historical records. The Marmara Segment is the major seismic gap since the last earthquake in 1766, while the Murefte earthquake occurred in 1912 at its west side and the Izmit earthquake in 1999 at its east side. The relative motion across the NAF is ~22 mm/yr based on the data from space geodesy. Investigating how much degree of this displacement is released by aseismic creep or accumulated by slip deficit in the Marmara Segment is crucial to know the total seismic risk in this region. Because the NAF is submerged in the Sea of Marmara and is inaccessible by space geodesy, we employed seafloor geodetic technique using extensometers, which acoustically monitor baseline length across a strain-localized zone, such as surface trace of a fault. In 2014, we installed five extensometers at the Western High crossing the NAF one after the other, where the surface trace of the NAF is prominent and gas emission from the seafloor is reported in. Totally four beselines of ~1 km range are successfully formed and quality of initial test data was promising. Based on the initial data, detectable level of the baseline change is estimated to be ~2mm, which owing to quite stable seawater near the bottom due to strong density stratification in the Sea of Marmara. The extensometers are designed that data can be recovered via acoustic modem without disrupt the monitoring. Since the installation, we have visited the site twice and have recovered the data for ten months in total. Temperature measured by thermistor equipped on each extensometer showed coherent change and gradual increase by 0.007 degree during the period. This reflects apparent beseline shortening due to the corresponding increase of the sound speed. In the preliminary temperature correction, difference of the change in the length between NW-ward and NE-ward baselines is ~5 mm in the period. This indicates roughly 3 mm/yr of right-lateral displacement. To confirm the results, monitoring must be further continued and the temperature correction should be refined accounting for sub-resolution information (< 0.001 degree).