Tectonic tremor, a weak and emergent seismic signal, presents challenges in understanding its generation mechanisms. This study investigates tremor signals along the Dead Sea Transform (DST), a major plate boundary with slow deformation, focusing on events remotely triggered by the powerful 2023 Kahramanmaraş earthquake in Turkey. The Kahramanmaraş earthquake pair, consisting of a Mw7.8 and a subsequent Mw7.6 event, significantly altered seismicity rates in Palestine and Israel. This presents a unique opportunity to study seismogenic processes along the DST, as large earthquakes are rare in this region. The study aims to understand the generation of these remotely triggered tremors along the DST, characterizing their seismic attributes, and evaluating existing models of tremor generation to explain the observed spectral variability. The intersection of the DST and the Carmel-Fari'a Fault (CFF) is particularly relevant, as it represents a potential zone of stress concentration and complex fault interaction. The research addresses the ongoing challenge in classifying and modeling tectonic tremor and provides a case study that may improve the understanding of remotely triggered seismic events in regions of slow deformation. The implications of the findings will improve our understanding of the dynamic processes occurring along plate boundaries, ultimately contributing to more accurate seismic hazard assessments.

Tectonic tremor spectra are notoriously difficult to model due to masking by ambient noise at low frequencies (<1 Hz) and obscuring by coda waves at higher frequencies (>1 Hz). Existing models attempt to reproduce the observed low-frequency enrichment and sharp high-frequency decay. These models fall into two categories: deterministic and stochastic. Some propose that tremor is a sequence of low-frequency earthquakes (LFEs), characterized by high-frequency depletion compared to regular earthquakes. Another model suggests that tremor results from inertial vibrations of a frictionally-controlled oscillator. While both models capture first-order temporal and spectral observations, they have different implications for the underlying physics. Distinguishing between spontaneous and remotely triggered tremor is crucial for understanding the tremorgenic fault behavior. Spontaneous tremors often correlate with slow-slip events (SSEs), whereas triggered tremors do not show geodetic signals. However, triggered tremors are generally larger in amplitude and have better-constrained stress conditions, facilitating model testing.

The study utilized acceleration and velocity seismograms from the TRUAA network in Israel and Palestine to identify remotely triggered tremor. Signals were filtered between 8-16 Hz and visually inspected for tremor-like signals correlated with surface wave arrivals from the Kahramanmaraş earthquakes. Tremor-like signals were detected in the Jordan Valley section of the DST. Ground acceleration and velocity were analyzed at station HMDT, revealing low-amplitude, emergent signals correlated with Love wave amplitude maxima. These signals differed significantly from regular DST earthquakes in duration and lack of clear P- and S-wave arrivals. Envelope cross-correlation was employed to locate the strongest tremor source near Kirbhet Samara, at a depth of 10-20 km. A triggered earthquake in the Jezre'el Valley was also identified and located using the same method. Ground-motion spectra of the DST tremor were analyzed and compared to local earthquakes and remotely triggered tremors from other locations (San Andreas Fault, San Jacinto Fault, and Hikurangi subduction zone). Spectral differences were assessed, focusing on high-frequency energy deficiencies. Two theoretical models were employed to explain the observed spectral variability: 1) inertial vibrations of a frictionally controlled oscillator, and 2) a swarm of LFEs with prescribed ω² source-time functions. The analysis involved generating moment-rate functions and spectra for both models, considering variations in oscillation periods, acceleration phases, LFE moment rates, and recurrence intervals. The models' ability to reproduce the observed spectral decay rates (f⁻¹, f⁻², f⁻³) was evaluated.

The study identified tremor along the DST in the Jordan Valley, triggered by the Mw7.6 Kahramanmaraş earthquake, but not the Mw7.8 event. This is surprising given the Mw7.8's larger magnitude. Analysis of Peak Ground Velocities (PGVs) and dynamic velocity gradients revealed that the Mw7.6 induced locally amplified long-period motion, despite its smaller magnitude. The tremor's location coincides with maxima in surface-wave induced deformation rate gradients. The tremor source was located near Kirbhet Samara at a depth of 10-20 km. Spectral analysis revealed that the DST tremor is characterized by a strong high-frequency energy deficiency compared to other triggered tremors, exhibiting a rapid spectral fall-off (inversely proportional to frequency cubed) in the 4-13 Hz band. Comparison with other triggered tremors (SAF, SJF, Hikurangi) indicated variability in spectral shapes, with the DST tremor being notably high-frequency depleted. The two proposed source models (inertial oscillation and LFE swarm) can both reproduce the observed high-frequency spectral decay, but differ in their low-frequency characteristics. The inertial oscillation model, with a constant slip rate and oscillating rupture area, produces a spectral decay proportional to f⁻¹ under certain conditions. The LFE swarm model, with appropriate parameter choices (LFE recurrence intervals and duration), also yields f⁻¹ and f⁻² spectral decays. The ratio between LFE recurrence intervals and durations influences the low-frequency spectral behavior, with LFE-poor tremors exhibiting less low-frequency enrichment.

The location of the triggered tremor and earthquake at maxima of long-period velocity gradients suggests that these locations are particularly susceptible to remote triggering. The amplification of the Mw7.6 surface waves relative to the Mw7.8 remains unexplained, but potential causes may include structural anomalies associated with the DST-CFF intersection or basin edge effects. The variability in triggered tremor spectral shapes highlights the complexity of tremor generation. The high-frequency depletion in the DST tremor is unique among continental transform faults, suggesting potential differences in the rupture process or fault properties compared to other locations. The observed spectral decay rates (f⁻¹, f⁻², f⁻³) can be explained by both the inertial oscillation and LFE swarm models, demonstrating the potential complexity of the underlying physics. Further research is needed to definitively distinguish between these models and explore the role of near-fault attenuation. The DST tremor's location near a seismic gap, last ruptured about 1000 years ago, indicates a potential link between remote triggering and the proximity to fault failure.

This study presents the first observation of instantaneous triggering of tremor along the DST and CFF, highlighting the complex interaction between the Kahramanmaraş earthquakes and the DST fault system. The variability in triggered tremor spectral shapes emphasizes the need for further investigation into tremor source physics and highlights the potential influence of local geological structures. Future work should focus on identifying the causes of local surface wave amplification, clarifying the relative contributions of the inertial oscillation and LFE swarm models, and investigating the possibility of spontaneous tremor along the DST. Continued seismo-geodetic monitoring of this region is critical given the observed remote triggering and its potential implications for seismic hazard.

The analysis was limited by the availability of data, particularly the clipping of broad-band velocity seismograms during the surface wave train. The low amplitude and emergence of the tremor signals resulted in relatively low signal-to-noise ratios, making detection challenging. The relatively linear geometry of the TRUAA seismic network may have introduced some uncertainty into depth resolution. The exact cause of the observed local amplification of Mw7.6 surface waves remains unknown and requires further investigation.