The three aftershock sequences that occurred in Al Hoceima, Morocco, in May 1994 (Mw 6.0), February 2004 (Mw 6.4) and January 2016 (Mw 6.3) were stochastically modeled to investigate their temporal and energetic behavior. A form of the restricted trigger model known as the restricted epidemic type aftershock sequence (RETAS) was used for the temporal analysis of the selected series. The best-determined fit models for each sequence differ based on the Akaike information criteria. The revealed discrepancies suggest that, although the activated fault systems are close (within 10 to 20 km), their stress regimes change and shift across each series. In addition, a stochastic model was presented to study the strain release following a specific strong earthquake. This model was constructed using a compound Poisson process and depicted the progression of the strain release during the aftershock sequence. The proposed model was then applied to the data. After the RETAS model was used to evaluate the behavior of the aftershock decay rate, the best-fit model was obtained and integrated into the strain-release stochastic analysis. By detecting the potential disparities between the observed data and model, the applied stochastic model of strain release allows for a more comprehensive examination. Furthermore, comparing the observed and expected cumulative energy release numbers revealed some variations at the start of all three sequences. This demonstrates that significant aftershock clusters occur more frequently shortly after the mainshock at the start of the sequence rather than if they are assumed to occur randomly.

The problem of tsunami hazard assessment and water pollution appear during the last years in connection with the increased density of infrastructure and industrial facilities on the Black Sea coast. The paper deals with the main results of the tsunami vulnerability modelling in accordance with the internationally recognised methods and new data obtained during the last years. Data collection, new approaches for travel times and dynamic properties of the giant sea waves have been investigated in the previous research. Now the main tsunamigenic sources together with the most vulnerable areas are outlined on the Bulgarian part of the coast. Together with the tsunami energy distribution according to the new method of ray tracing and refraction, the results give the possibility to establish the concentration-dissipation areas, which are most important to the tsunami risk assessment.

We have conducted paleo- and archaeoseismological surveys in the Durankulak Eneolithic archaeological settlement in northeastern Bulgaria, identifying many earthquake-related deformations of different ages. The ubiquitous presence of counterfort walls added to damaged original (nearly north–south) walls on the eastern side provides evidence of the first significant earthquake in Durankulak that seems to have occurred during Middle to Late Eneolithic time (4650–4100 B.C.). This seismic event made the north–south walls tilt westward, so that the ancient residents had to build counterfort walls to prevent the original walls from collapsing. However, the upper excavations in the archaeological settlement show that the next, later, seismic deformation induced a tilting and collapse of remains of the original walls eastward, producing a gap of a few tens of centimeters between the original walls and the counterfort walls. Afterwards the settlement was abandoned. It may be surmised that the later earthquake was stronger than the earlier one, which left room for reconstruction work. The local seismic intensity of the earlier earthquake seems to have been at least I_l ≥ VIII, while the later event had I_l ≥ IX on the МSК-64 scale. Considering that the masonry of both walls, the original and the counterfort ones, is about the same age, the time interval between the two seismic events must have been short, one or two hundred years. Nevertheless, the epicenters of both of these earthquakes were to different directions from Durankulak. The seismic motion (compressive waves and shocks) went from the west during the first earthquake and from the east during the second. The rupture zone of the first seismic event was obviously related (1) to an earthquake-generating zone in continental Bulgaria situated west of the archaeological settlement in question, or (2) to the Intramoesian active fault that was responsible for the large earthquake of 1444. The rupture zone of the second seismic event was nearby, in the coastal part of the Black Sea; the responsible feature seems to have been the Shabla–Kaliakra seismogenic zone. Considerable seismic deformations were also identified in structures on the southern slope of the island in a dwelling dating back to the First Bulgarian Kingdom (the 9th century A.D.). This seismic event produced distortions in the walls (in map view), sigmoidal patterns, as well as fractures and rotations in the walls. The local seismic intensity due to this earthquake is supposed to be at least I_l ≥ VIII. The rupture zone of a third seismic event was likely to be nearby, in the coastal part of the Black Sea, and the responsible feature must have been the Shabla–Kaliakra seismogenic zone.

The article elaborates on two innovations developed for the purpose of protecting critical infrastructure facilities from two specific natural disasters – kinematic system for early warning of earthquakes applicable for the Bulgarian nuclear power plant (NPP), and a system for early warning of floods along the river beds near artificial lakes barrages. Both developments are part of the execution of p.1.1.6 from National Scientific Programme (NSP) “Security and Defence” – “Analysis of risk and threats, design and development of conceptual generating models and software for improvement of efficiency of control of forces and means for critical infrastructure influence during natural disasters, accidents and crises, on the territory of Republic of Bulgaria”.

—The ruins of Odessos are located on the territory of Varna, Bulgaria. Archaeoseismological studies carried out at the archaeological sites of ancient Odessos revealed some significant damage and destruction in the ancient building constructions. These are numerous breaks, tilts, protrusions and turns of wall parts, longitudinal asymmetric deformation of the above-entrance arch, lateral in plan bending of the extended wall, outward knock-out of the stone blocks and intensive cracking. We attributed traces of subsequent shoddy repairs, which sometimes involved using single expensive stone blocks taken out of the ruins, as circumstantial evidence of strong earthquakes that had occurred here. Interestingly, medieval builders already knew about the possibility of strong earthquakes in the region and put wooden longitudinal beams in the walls to absorb vertical seismic vibrations. The degree of destruction and damage in medieval buildings built on loose fill soils increased due to subsidence. Archaeological research established 6 periods of construction in ancient Odessos: Using this periodization, we calculated the age of past seismic events: (1) around the 1st century A.D.—the destruction of the Late Hellenistic fortress wall; (2) the early 5th century—the early 6th century (a cluster of at least 3 strong earthquakes, judging also by deformations in the Bishop’s Basilica of Odessos (Minchev et al., 2020)): dismantling of the fortress wall, discontinued use of both thermae. The gate of Odessos remains in operation after repairs. There is low quality construction of houses. Accumulation of tile debris; (3) the destruction of the Orthodox church in the 1444 earthquake. We were able to determine the average direction of the maximum aggregate seismic movements that destroyed the large and small thermae, as well as the excavated segment of the fortress wall and the Odessos gate complex. They propagated along the WNW-ESE axis, most likely from the Shabla-Kaliakra seismogenic zone. The deformations of the 13th-century Orthodox church were caused by seismic vibrations that propagated in the submeridional direction from one of the seismically active faults south or north of Varna, although they may have been affected by ground subsidence. Thus, as a result of the studies, we were able to detect traces of yet unknown strong earthquakes, to determine their age and direction to the ancient epicentral zones. The intensity of seismic vibrations during all these earthquakes was I_l = VIII—IX MSK-64. The data obtained can be used to improve the seismic hazard assessment of Varna and NE Bulgaria.