On the seismicity of central Kamchatka before the 1997, M = 7.9 Kronotskii earthquake

We report results from a detailed study of seismicity in central Kamchatka for the period from 1960 to 1997 using a modified traditional approach. The basic elements of this approach include (a) segmentation of the seismic region concerned (the Kronotskii and Shipunskii geoblocks, the continental slope and offshore blocks), (b) studying the variation in the rate of M = 4.5–7.0 earthquakes and in the amount of seismic energy release over time, (c) studying the seismicity variations, (d) separate estimates of earthquake recurrence for depths of 0–50 and 50–100 km. As a result, besides corroborating the fact that a quiescence occurred before the December 5, 1997, M = 7.9 Kronotskii earthquake, we also found a relationship between the start of the quiescence and the position of the seismic zone with respect to the rupture initiation. The earliest date of the quiescence (decreasing seismicity rate and seismic energy release) was due to the M = 4.5–7.0 earthquakes at depths of 0–100 km in the Kronotskii geoblock (8–9 years prior to the earthquake). The intermediate start of the quiescence was due to distant seismic zones of the Shipunskii geoblock and the circular zone using the RTL method, combining the Shipunskii and Kronotskii geoblocks (6 years). Based on the low magnitude seismicity (M≥2.6) at depths of 0–70 km in the southwestern part of the epicentral zone (50–100 km from the mainshock epicenter), the quiescence was inferred to have occurred a little over 3 years (40 months) before the mainshock time and a little over 2 years (25 months) in the immediate vicinity of the epicenter (0–50 km). These results enable a more reliable identification of other types of geophysical precursors during seismic quiescences before disastrous earthquakes.     

The Pn wave velocities and the relief of Moho in the Tibetan Plateau

Eleven PASSCAL broadband digital seismic stations were employed in the Tibetan Plateau by the Sino-US team from September, 1991 to June, 1992. Seven of them were distributed along the Qinghai-Tibet highway, others in Maqin and Yushu of Qinghai Province, Linzhi and Xigatze of Tibet. The data included 31 local earthquakes recorded by these stations from July, 1991 to January, 1992. Considering the characters of digital data, we identified the seismic phases carefully in several methods using SAC softwares (Seismic Analysis Code) in SUN workstation. We compared the seismic phases after converting the seismograms of the single stations to the seismic profiles; analyzed the first arrivals of P waves in the incident planes by rotating 3 component seismic records; identified the seismic phases from the particle motion pictures. The Pn apparent velocities were calculated in the distance range of 230–1200 km from Linzhi earthquakes, western Changtang earthquakes, Xitieshan and Gonghe earthquakes and the earthquakes in India. The results show that the Pn velocities change slightly in the Tibetan Plateau (8.0–8.1 km/s). These values near the velocities at the uppermost mantle of the normal continents. The Moho undulation in the Tibetan Plateau was also studied by using Pn data by the time-term method. The primary results indicate that the Moho beneath the Tibetan Plateau is flat, and its depths are 67–70 km. The Chinese version of this paper appeared in the Chinese edition ofActa Seismologica Sinica,14, Supp., 593–600, 1992.     

A 3D model of the central asia mantle from dispersion of Rayleigh-wave phase velocities

Frequency-time analysis was applied to records of Rayleigh surface waves due to teleseismic earthquakes (M ≥ 6.0) to obtain dispersion curves of fundamental mode phase velocities for paths between 51 pairs of digital seismic stations in the Asian IRIS networks; the range of periods was 10–200 s. For each of these pairs we derived 1D-shear-velocity sections that fit the corresponding dispersion curves and image the integral earth structure down to depths of ∼650–700 km beneath each of the interstation seismic paths, which traverse the area of study in different directions. These structures were used by Yanovskaya (2001) to derive a 3D-shear-velocity model by 2D tomography, imaging the distribution of the larger horizontal inhomogeneities in the Central Asia mantle for depths of 50 to 600 km, and to construct 2D-velocity sections for five lines passing through the major tectonic features of Central Asia.     

Seismological evidence for the nucleation of two strong earthquakes

The structure of seismicity before the Kronotskii (1997) and Simushir (2006) earthquakes off Russian shores is studied. Different algorithms are applied to the identification of seismic quiescence, seismicity activation, and clustering of seismic events. It is established that the seismic process developed according to the same scenario in both cases. Seismic quiescence arose a few years before an earthquake near the epicenter in an area ～200 km in size. A 100-km area of subsequent activation included the epicenter. The activation stage ended more than one year before an earthquake. An abrupt rise in the number of clusters of seismic events was observed a few days before an earthquake near its epicenter.     

Topography of the Moho and Conrad discontinuities in the Kyushu district,Southwest Japan

The first P-arrival time data from local earthquakes are inverted for two-dimensional variation of the depths to the Conrad and Moho discontinuities in the Kyushu district, southwest Japan. At the same time, earthquake hypocenters and station corrections are determined from the data. The depths to the discontinuities are estimated by minimizing the travel time residuals of first P-arrival phases for 608 earthquakes observed at 57 seismic stations. In the land area of Kyushu, the Conrad and Moho discontinuities are located within the depth ranges of 16–18 and 34–40 km, respectively. The Conrad discontinuity is not as largely undulated as the Moho discontinuity. The depth to the Moho is deep along the east coast of Kyushu, and the deepest Moho is closely related to markedly low velocity of P wave. We regard the deepest Moho as reflecting the Kyushu–Palau ridge subducting beneath the Kyushu district, together with the Philippine Sea slab. In western Kyushu, the shallow Moho is spreading in the north–northeast–south–southwest direction in the Okinawa trough region. Based on the presence of low-velocity anomaly in three-dimensional velocity structure and seismogenic stress field of shallow crustal earthquakes, the shallow Moho is interpreted as being due to lower crustal erosion associated with a small-scale mantle upwelling in the Okinawa trough region. The velocity discontinuity undulation basically has insignificant effect on hypocenter determination of the local earthquakes, but the Moho topography makes changes in focal depths of some upper mantle earthquakes. The depth variation of the Moho discontinuity has a good correlation with the Bouguer gravity anomaly map; i.e., the shallow Moho of western Kyushu and the deep Moho of eastern Kyushu closely correlate with the positive and negative Bouguer gravity anomalies, respectively.     

A long-term earthquake forecast for the Kuril-Kamchatka arc for the period from September 2010 to August 2015 and the reliability of previous forecasts,as well as their applications

We describe results from the ongoing 2008–2010 work on long-term earthquake prediction for the Kuril-Kamchatka arc based on the patterns of seismic gaps and the seismic cycle. We provide a forecast for the next 5 years, September 2010 to August 2015, specified for all segments of the earthquake-generating Kuril-Kamchatka arc zone. For 20 segments we predict the phases of the seismic cycle, the normalized rate of small earthquakes (A₁₀), the magnitudes of moderate earthquakes to be expected with probabilities of 0.8, 0.5, and 0.15, the maximum possible magnitudes, and the probabilities of great (M ≥ 7.7) earthquakes. It is shown that the forecast given for the previous 5 years, from September 2005 to September 2010, was found to be accurate. We report the measures that were taken for seismic safety and retrofitting based on these forecasts.     

The deep structure of the Kulu earthquake-generating zone in the upper Kolyma highland from geophysical data

We measured and interpreted 30 physical magnetotelluric sounding sites using an SGS-E station and 20 km of electrical profiling observations using SDVR-4M instrumentation. We constructed a map of seismicity, an interpretation map, and four geoelectric sections, which give an idea of the deep structure for the Kulu earthquake-generating zone. A general geoelectric upper crustal model was developed for the zone down to depths of 20–22 km. Three nearly vertical conductive volumes were identified (thickness 3–5 km, depth 10–22 km), which provide the positions of seismically active deep-seated faults that pinpoint the Kulu earthquake-generating zone. The preliminary boundary of the zone was determined. It was found that earth-quake epicenters are confined to lithosphere volumes with increased concentrations of conductive layers and zones.     

Gravity anomaly, lithospheric structure and seismicity of Western Himalayan Syntaxis

A compiled gravity anomaly map of the Western Himalayan Syntaxis is analysed to understand the tectonics of the region around the epicentre of Kashmir earthquake of October 8, 2005 (Mw = 7.6). Isostatic gravity anomalies and effective elastic thickness (EET) of lithosphere are assessed from coherence analysis between Bouguer anomaly and topography. The isostatic residual gravity high and gravity low correspond to the two main seismic zones in this region, viz. Indus–Kohistan Seismic Zone (IKSZ) and Hindu Kush Seismic Zones (HKSZ), respectively, suggesting a connection between siesmicity and gravity anomalies. The gravity high originates from the high-density thrusted rocks along the syntaxial bend of the Main Boundary Thrust and coincides with the region of the crustal thrust earthquakes, including the Kashmir earthquake of 2005. The gravity low of HKSZ coincides with the region of intermediate–deep-focus earthquakes, where crustal rocks are underthrusting with a higher speed to create low density cold mantle. Comparable EET (∼55 km) to the focal depth of crustal earthquakes suggests that whole crust is seismogenic and brittle. An integrated lithospheric model along a profile provides the crustal structure of the boundary zones with crustal thickness of about 60 km under the Karakoram–Pamir regions and suggests continental subduction from either sides (Indian and Eurasian) leading to a complex compressional environment for large earthquakes.     

Crustal Structure of the Southern Korean Peninsula from Seismic Waves Generated by Large Explosions in 2002 and 2004

In order to investigate the velocity structure of the southern part of the Korean peninsula, seismic refraction profiles were obtained along a 294-km WNW-ESE line and a 335-km NNW-SSE line in 2002 and 2004, respectively. Seismic waves were generated by detonating 500–1000 kg explosives in drill holes at depths of 80–150 m. The seismic signals were recorded by portable seismometers at nominal intervals of 1.5–1.7 km. Separate velocity tomograms were derived from first arrival times using a series expansion method of travel-time inversion. The raypaths indicate several mid-crust interfaces including those at approximate depths of 2–3, 15–17, and 22 km. The Moho discontinuity with refraction velocity of 7.8 to 8.4 km/s has a maximum depth of 37–39 km under the southern central portion of the peninsula. The Moho becomes shallower as the Yellow Sea and the East Sea are approached on the west and east coasts of the peninsula, respectively. The depth of the 7.6 km/s velocity contour varies from 29.4 km to 36.5 km. The discrepancy in depth between the seismological Moho and the interpreted critically refracting interface may result from the presence of a gradual transition between the crust and mantle. The velocity tomograms show particular crustal structures including (1) the existence of an over 70-km wide low-velocity zone centered at 6–7 km depth under the Okchon fold belt and Ryeongnam massif, (2) existence of high-velocity materials under the Gyeongsang basin, and (3) the downward extension of the Yeongdong fault to depths greater than 10 km.     

Petrology and geochronology of basalt breccia from the 1996 earthquake swarm of Loihi seamount, Hawaii: magmatic history of its 1996 eruption

 Samples of basalt were collected during the Rapid Response cruise to Loihi seamount from a breccia that was probably created by the July to August 1996 Loihi earthquake swarm, the largest swarm ever recorded from a Hawaiian volcano. ²¹⁰Po–²¹⁰Pb dating of two fresh lava blocks from this breccia indicates that they were erupted during the first half of 1996, making this the first documented historical eruption of Loihi. Sonobuoys deployed during the August 1996 cruise recorded popping noises north of the breccia site, indicating that the eruption may have been continuing during the swarm. All of the breccia lava fragments are tholeiitic, like the vast majority of Loihi's most recent lavas. Reverse zoning at the rim of clinopyroxene phenocrysts, and the presence of two chemically distinct olivine phenocryst populations, indicate that the magma for the lavas was mixed just prior to eruption. The trace element geochemistry of these lavas indicates there has been a reversal in Loihi's temporal geochemical trend. Although the new Loihi lavas are similar isotopically and geochemically to recent Kilauea lavas and the mantle conduits for these two volcanoes appear to converge at depth, distinct trace element ratios for their recent lavas preclude common parental magmas for these two active volcanoes. The mineralogy of Loihi's recent tholeiitic lavas signify that they crystallized at moderate depths (∼8–9 km) within the volcano, which is approximately 1 km below the hypocenters for earthquakes from the 1996 swarm. Taken together, the petrological and seismic evidence indicates that Loihi's current magma chamber is considerably deeper than the shallow magma chamber (∼3–4 km) in the adjoining active shield volcanoes. Received: 21 August 1997 / Accepted: 15 February 1998     

Upper mantle structure in the ocean-continent transition zone: Kamchatka

We consider results from modeling the crustal and upper mantle velocity structure in Kamchatka by seismic tomography and compare these with gravity data and present-day tectonics. We found a well-pronounced (in the physical fields) vertical and lateral variation for the upper mantle and found that it is controlled by fault tectonics. Not only are individual lithosphere blocks moving along faults, but also parts of the Benioff zone. The East Kamchatka volcanic belt (EKVB) is confined to the asthenospheric layer (the asthenosphere lens) at a depth of 70–80 km; this lens is 10–20 km thick and seismic velocity in it is lower by 2–4%. The top of the asthenosphere lens has the shape of a dome uplift beneath the Klyuchevskoi group of volcanoes and its thickness is appreciably greater; overall, the upper mantle in this region is appreciably stratified. A low-velocity heterogeneity (asthenolith) at least 100 km thick has been identified beneath the Central Kamchatka depression; we have determined its extent in the upper mantle and how it is related to the EKVB heterogeneities. Gravity data suggest the development of a rift structure under the Sredinnyi Range volcanic belt. The Benioff zone was found to exhibit velocity inhomogeneity; the anomalous zones that have been identified within it are related to asthenosphere inhomogeneities in the continental and oceanic blocks of the mantle.     

Calibration of the Tibetan Plateau Using Regional Seismic Waveforms

We use the recordings from 51 earthquakes produced by a PASSCAL deployment in Tibet to develop a two-layer crustal model for the region. Starting with their ISC locations, we iteratively fit the P-arrival times to relocate the earthquakes and estimate mantle and crustal seismic parameters. An average crustal P velocity of 6.2–6.3 km/s is obtained for a crustal thickness of 65 km while the P velocity of the uppermost mantle is 8.1 km/s. The upper layer of the model is further fine-tuned by obtaining the best synthetic SH waveform match to an observed waveform for a well-located event. Green's functions from this model are then used to estimate the source parameters for those events using a grid search procedure. Average event relocation relative to the ISC locations, excluding two poorly located earthquakes, is 16 km. All but one earthquake are determined by the waveform inversion to be at depths between 5 and 15 km. This is 15 km shallower, on average, than depths reported by the ISC. The shallow seismicity cut-off depth and low crustal velocities suggest high temperatures in the lower crust. Thrust faulting source mechanisms dominate at the margins of the plateau. Within the plateau, at locations with surface elevations less than 5 km, source mechanisms are a mixture of strike-slip and thrust. Most events occurring in the high plateau where elevations are above 5 km show normal faulting. This indicates that a large portion of the plateau is under EW extension.     

Body wave velocity distribution in the Benioff zone of central Kamchatka during aftershocks of the Kronotskii earthquake of 1997 (M = 7.9)

The distribution of the P and S velocities in the Benioff zone of central Kamchatka during the period of aftershocks (1997–2004) of the disastrous Kronotskii earthquake of 1997 (M = 7.9, M_W = 7.7) has been determined. Based on the data for the foreshock period immediately preceding the earthquake (1991–1997), a sharp increase in the body wave velocities in the Benioff zone below the Kronotskii Peninsula (up to 9.5–9.7 km/s for V _P and 5.1–5.3 km/s for V _S) has been determined at depths of 55–140 km in the subvertical region. Based on observations during the period of aftershocks comparable with the last period of foreshocks (about 7 years), it has been established that the body wave velocities calculated for the Benioff zone below the Kronotskii Peninsula returned to the initial values typical of the beginning of that period. This indicates that stresses relaxed around the head part of the Kronotskii earthquake rupture zone after its origination. This conclusion is confirmed by a sharp decrease in the number of earthquakes with M = 2.3–4.9 in the Benioff zone below the Kronotskii Peninsula. Moreover, taking the velocity distribution during the period of aftershocks into account, it has been determined that a second stress relaxation zone is located at the southwestern flank of the Kronotskii earthquake rupture zone where the largest (M = 6.7) aftershock occurred. According to these data, it is concluded that two stress concentration centers could have existed during the preparation of the Kronotskii earthquake.     

Marine seismological observations in the Central Kurile segment before catastrophic earthquakes on November, 2006 (<Emphasis Type="Italic">M</Emphasis> = 8.3) and January, 2007 (<Emphasis Type="Italic">M</Emphasis> = 8.1)

The results of detailed seismological observations with bottom seismographs in the Central Kurile segment in August-September, 2006 are discussed. The system of six bottom seismographs was placed on the island slope of the Kurile deep-sea trench southeast of Urup Island and southwest of the Bussol Strait. Over 230 earthquakes with M _LH = 0.5–5.5 were registered in the area with a radius of 150 km around the center of the observation system at depths up to 300 km during 16 days. Records of 80 earthquakes with hypocenters in the earth crust (h = 0–30 km) beneath the island slope of the Kurile deep-sea trench were first obtained by bottom seismographs. These data are inconsistent with previous concepts of aseismicity of this zone. The discovery of the unique morphological structure of the Benioff zone beneath the central Kurile Arc represents the most important result of detailed seismological observations. The zone consists of an inner seismoactive subzone, which is located beneath the island slope of the arc at depths of 15–210 km, being characterized by an angle of incline of 50° under the latter and crosses the ocean bottom approximately 80 km away from the trench axis, and outer low-activity subzone. The latter is traceable beyond the trench almost parallel to the inner zone beginning from a depth of 50 km below the sea bottom up to a depth of approximately 300 km. Due to the slightly lower incline (∼45°) of the outer subzone, both subzones gradually converge downward. The integral thickness of the Benioff zone varies from 150 km in its upper part to 125 km at depths of 210–260 km. The medium sandwiched between these subzones is practically aseismic. The reality of this defined structure is confirmed by the distribution of aftershocks of the earthquake that occurred on November 15, 2006 (M = 8.3). These seismic events served as foreshocks for the subsequent strong earthquake of January 13, 2007 (M = 8.1) with the hypocenter located beyond the trench under the ocean bottom. Such a structure of this zone within the central Kurile Arc segment is unique, having no analogues either in the flanks of the Kurile-Kamchatka Arc or other arcs. The results of detailed seismological observations obtained two months before the first of the catastrophic Central Kurile earthquakes appeared to be typical for the period of foreshocks (the lower seismic activity of the Simushir block, which hosted the hypocenter of the earthquake that occurred on November 15, 2006, particularly at depths of 0–50 km, the gentler incline of the recurrence plot, and other features).     

天山中部的地震定位和地壳活动性分析

利用宽频带流动台网记录的地震P波和S波到时,根据一维和三维地壳速度模型,对天山中部及其邻近地区1997～1998年的地震进行了重新定位,以重新确定的震源参数为依据分析了地壳的活动性.震源分布表明,造山带边缘和内部的大部分断裂都显示出活动的迹象,它们对天山的地壳构造变动起到明显的作用;塔拉斯－费尔干纳断裂的活动具有分段特征：其东南段以及西南天山的部分断裂目前活动比较弱,西北段受周边断裂的影响活动较强;另外造山带边缘的构造活动有向山前盆地渗透（Penetration）的趋势.30～40km深度的地震活动表明,天山中部的地壳中下层仍然具有一定的破裂条件,它们与壳幔边界附近热扰动的驱动有关,暗示小尺度地幔对流或软流层上涌等动力作用仍在持续进行.     

Inversion of seismic and gravity data for the composition and core sizes of the Moon

We model the internal structure of the Moon, initially homogeneous and later differentiated due to partial melting. The chemical composition and the internal structure of the Moon are retrieved by the Monte-Carlo inversion of the gravity (the mass and the moment of inertia), seismic (compressional and shear velocities), and petrological (balance equations) data. For the computation of phase equilibrium relations and physical properties, we have used a method of minimization of the Gibbs free energy combined with a Mie-Gr@uneisen equation of state within the CaO-FeO-MgO-Al₂O₃-SiO₂ system. The lunar models with a different degree of constraints on the solution are considered. For all models, the geophysically and geochemically permissible ranges of seismic velocities and concentrations in three mantle zones and the sizes of Fe-10%S core are estimated. The lunar mantle is chemically stratified; different mantle zones, where orthopyroxene is the dominant phase, have different concentrations of FeO, Al₂O₃, and CaO. The silicate portion of the Moon (crust + mantle) may contain 3.5–5.5% Al₂O₃ and 10.5–12.5% FeO. The chemical boundary between the middle and the lower mantle lies at a depth of 620–750 km. The lunar models with and without a chemical boundary at a depth of 250–300 km are both possible. The main parameters of the crust, the mantle, and the core of the Moon are estimated. At the depths of the lower mantle, the P and S velocities range from 7.88 to 8.10 km/s and from 4.40 to 4.55 km/s, respectively. The radius of a Fe-10%S core is 340 ± 30 km.     

Geochemical constraints on the model of the composition and thermal conditions of the Moon according to seismic data

A new method of reconstruction of the temperature profile in the lunar mantle from the velocities of seismic P- and S-waves for different models of chemical composition is developed. The procedure of the solution of an inverse problem is realized with the help of the minimization of the Gibbs free energy and the equations of state of a mantle substance, taking into account phase transformations, anharmonicity, and the effects of inelasticity. The geophysical and geochemical constraints on composition and temperature distribution in Moon’s mantle are established. The upper mantle can be composed of olivine pyroxenite, depleted by low-volatile oxides (∼2 wt % of CaO and Al₂O₃). On the contrary, the lower mantle must be enriched by low-volatile oxides (∼4–6 wt % of CaO and Al₂O₃). Its composition can be represented by a mineral association of the olivine + clinopyroxene + garnet or olivine + orthopyroxene + clinopyroxene + garnet type, which is close in composition to pyrolite. The temperature distribution at depths 50–1000 km are approximated by the equation: T(°C) = 351 + 1718[1–exp (−0.00082H)]. The constraints inferred make it possible to conclude that the published values of the velocities of P- and S-waves for the lunar mantle, obtained by processing the data of seismic experiments of the Apollo lunar mission are inconsistent with each other at depths below 300 km. Otherwise, the variations in the velocities of P- and S-waves disturb the symmetry between the petrological model (composition), the temperature profile, and the seismic profile.     

A long-term earthquake forecast for the Kuril-Kamchatka arc for the period from September 2011 to August 2016. The likely location,time, and evolution of the next great earthquake with M ≥ 7.7 in Kamchatka

We consider the results from the ongoing 2010–2011 work on long-term earthquake prediction for the Kuril-Kamchatka arc based on the pattern of seismic gaps and the seismic cycle. We develop a forecast for the next 5 years, from September 2011 to August 2016, for all segments of the Kuril-Kamchatka arc earthquake-generating zone. For 20 segments we predict the appropriate phases of the seismic cycle, the normalized rate of small earthquakes (A₁₀), the magnitudes of moderate earthquakes to be expected with probability 0.8, 0.5, and 0.15, and the maximum possible magnitudes and probability of occurrence for great (M ≥ 7.7) earthquakes. This study serves as another confirmation that it is entirely necessary to continue the work in seismic retrofitting in the area of Petropavlovsk-Kamchatskii.     

Evolution of the seismic activity from the 1982 eruption of El Chichon Volcano, Chiapas, Mexico

 We analyzed more than 1700 earthquakes related to the 1982 eruption of El Chichon volcano in southern Mexico. The data were recorded at specific periods throughout the whole eruptive interval of March to April 1982, by three different networks. The seismic activity began several months before the first eruption on 28 March. During this period the seismicity consisted of hybrid and long-period shallow earthquakes most likely related to processes of faulting, fracturing, and fluid movement underneath the volcano. The foci of events occurring before the eruption circumscribe an aseismic zone from approximately 7 to 13 km below the volcano. After the eruption, the seismic activity consisted of tectonic-type earthquakes that peaked at 1200 events/h. This later activity occurred over a wide range of depths, mostly between 5 and 20 km, that includes the former aseismic zone and is roughly limited by the major tectonic faults in the area. Received: 19 May 1998 / Accepted: 13 June 1999     

Focal Depths of Earthquakes in the Crimea–Black Sea Region

The problem of determining focal depths of earthquakes in the Crimea–Black Sea region is considered. Based on the results of interannual studies, it is found that the focal depths of Crimean earthquakes are mainly crustal, with maximum values of up to 60 km. Some recent publications, however, have described deep-focus earthquakes with depths of up to 300 km which were “revealed” in the Crimean region. In this respect, there arose the need to study such a large difference in estimated focal depths. Convincing examples show that the sensational “revelation” of deep earthquakes in Crimea was caused by incorrect processing of the experimental data, in particular, due to (1) a sharp distortion in the recorded arrival times of body waves, (2) exclusion of data from stations nearest to a source, (3) unreasonable arbitrary selection of data from seismic stations, and (4) dropping of data from the worldwide seismological network, including those on deep seismic phases. Thus, the conclusions about the presence of deep mantle earthquakes in Crimea are erroneous. We have redetermined the parameters of hypocenters and verified that the focal depths of earthquakes in the Crimea–Black Sea region are no more than 60 km. Based on these data, we analyze the features of the spatial distribution of focal depths to show that earthquake sources are grouped along conduits that dip southeastward, from the continental part of Crimea toward the Black Sea Basin, in the case of grouping of sources in the Alushta–Yalta and Sevastopol areas. The seismic focal layer of the Kerch–Anapa area dips northeastward, from the Black Sea beneath the North Caucasus.