A DEEP NEURAL NETWORK EXAMINATION OFMICROSEISMICITY IN THE NORTHERN PORTION OF THEMAIN RECENT FAULT

This research undertakes a detailed exploration of microseismic activity along the northern stretchof the Main Recent Fault (MRF) within the Zagros region, bracketed by the Main Zagros ReverseFault (MZRF) and the High Zagros Fault (HZF), with a particular concentration on the Ardal Faultsegment. We deployed a seven-station seismic network tasked with the surveillance of seismicoccurrences in the vicinity of the Rudbar dam. The PhaseNet deep neural network approach washarnessed for accurate earthquake detection and phase picking, enabling the analysis of ۵۸۸ seismicevents, including those with local magnitudes as low as ۰.۴. Our findings underscore a pronouncedlink between microseismic activities and the MRF. Comparatively, the HZF, located some ۱۵kilometers southwest and running parallel to the Ardal Fault, is associated with much fewer seismicactivities. Remarkably, minor seismic events were observed in connection with the MZRF, situatedapproximately ۱۰ kilometers northeast of the Ardal Fault, challenging its perceived status as asignificant active tectonic boundary between the Zagros region and Central Iran. The predominance ofthe MRF in accommodating the ongoing convergence between the Arabian and Eurasian plates isaccentuated as a pivotal aspect of this geographical segment.

Keywords:

Deep Learning , Main Recent Fault , Microseismicity , Zagros