Geophysical Studies of Geodynamics and Natural Hazards in the Northwestern Pacific Region: Introduction

The Topical Issue of Pure and Applied Geophysics “Geophysical Studies of Geodynamics and Natural Hazards in the Northwestern Pacific Region” aims to communicate multidisciplinary research focused on one of the most dangerous territories of the world, where inhabitants and infrastructure are exposed to extreme natural hazards associated with unique tectonics and geodynamics of the region. Most of these hazards relate to rapid plate convergence together with tearing and long-distance under-thrusting of the Pacific plate. These have produced many great earthquakes and one of the largest, if not the largest, volcanic eruption of the last millennium. The intersection of the Aleutian and Kamchatka subduction zones forms a sharp cusp in the NW Pacific characterized with one of the highest levels of seismic activity. The Northwestern Pacific Region extends southwards to Philippine and Mariana Islands and eastward to Alaska. The tectonics of the Region is very complex with deep subduction and pull-apart basins. Formation of the gigantic linked dextral pull-apart basin system in the NW Pacific is due to NNE- to ENE-ward motion of east Eurasia in response to the Indo–Asia collision, which started about 50 Ma ago. The large amount of motion of the eastern Eurasia region contradicts any traditional rigid plate tectonic reconstruction, but agrees with the more recent concepts of non-rigidity of both continental and oceanic lithosphere over geological times. Often overlooked is that damage and deaths from earthquakes result more from induced landslides than they do from building collapse from the shaking itself. Other mass movements arise from general crustal instability and deglaciation.

This anthology of current research results from the original inspiration and efforts of Alexander Anatolievich Soloviev. Tragically, he died suddenly on 23rd of September, 2021 while walking with his granddaughter in Moscow. A new corresponding guest editor was assigned in November 2021 to proceed with the work started by Prof. Alexander A. Soloviev. The guest editors dedicate this Topical Issue to his memory. Contained herein are articles on the models, methods, and case studies related to the Northwestern Pacific Region and adjacent territories. In particular -

In Models, Kaban et al. (2022) present “A new Moho map for North-Eastern Eurasia based on the analysis of various geophysical data”. The key fields considered are residual gravity, topography and vertical gravity gradients from the Gravity Field and Steady-State Ocean Circulation Explorer (GOCE). A joint step-by-step analysis of these three fields enables improvement of the final results and better separation of the Moho signal. The new Moho map demonstrates several principal features that were not resolved in the previous studies and correspond well to observed tectonic fragmentation of the study area. New crustal patterns of different kinds have been found under the Verkhoyansk Range, in the continental part of the Laptev rift system, in the East Siberian Sea, in the offshore part of the Chukotka microcontinent, and in the Anadyr-Koryak fold system. The new Moho map is a significant improvement of the previous Moho maps of north-eastern Eurasia.

Zabarinskaya et al. (2022) review “Deep Mariana Island Arc: Highlights of the tectonosphere” in regard to its unique variety of geological features and events, including earthquakes, volcanoes, hydrothermal vents, cold seeps and the largest mud volcanoes on Earth. The Mariana Island Arc is a classic young island arc in the western Pacific Ocean. The authors consider its (i) Tectonic setting wedged between the Eurasian, Pacific and Australian Plates; (ii) Heat flow (up to 2000 mW m⁻²), which points to both high tectonic activity and an important role played by the convective component in the thermal regime; (iii) Gravity field, which varies substantially in all reductions, reflecting the structural and density inhomogeneities of the tectonosphere; (iv) Seismicity in the northwestward subduction of the Pacific Plate beneath the Philippine Sea Plate, which controls seismic activity within the Mariana Island Arc from aseismically subducting beneath the Philippine Sea Plate without causing large under-thrusting earthquakes at the plate interface due to the weak coupling between the down-going and overlying plates; (v) Volcanism, which are mostly seamounts (underwater volcanoes) with only nine peaks tall enough to form islands and enormous mud volcanoes in the area between two tectonic plates that have been colliding for over 50 Ma and are generating the most extreme fluid composition recorded in the oceans; (vi) Deep structure, which includes the active Mariana Trough, and also a typical active back-arc basin, and two crescent-shaped island arcs, the remnant West Mariana Ridge to the west and the active Mariana Arc to the east.

Bykov et al. (2022) in “Stress transfer and migration of earthquakes from the Western Pacific subduction zone toward the Asian continent” explore the impact of subduction on the geodynamics of the Asian continent by analyzing migration of slow strain and earthquakes from the Nankai, Japan and Kuril-Kamchatka segments deep into mainland. The estimations on the profiles, crossing the Kuril Islands, the Japanese Archipelago and Sakhalin Island toward the Asian continent, have revealed the transverse migration of earthquakes from the Japan–Kuril-Kamchatka subduction zone. The velocities of hypocenter migration of M ≥ 4.5 earthquakes from the Kuril-Kamchatka Trench via northern and central Sakhalin vary from 6 to 17 km/year at different depths. The profiles crossing the islands of Hokkaido and Sakhalin show the M ≥ 4.earthquake migration from the Kuril and Japan trenches at velocities of 8–27 km/year.

Kim et al. (2022) in “Neotectonics at the SE continental margin of the Korean Peninsula: Implications for the back-arc region behind the SW Japan Arc” discuss the geological evolution of the SE continental margin of the Korean Peninsula resulting from crustal extension with back-arc rifting to spreading followed by crustal shortening with back-arc closing. In particular, the authors investigate the geological structure of the area near the largest event ever recorded here instrumentally, the Mw 5.0 earthquake that occurred in 2016. Seismic reflection profiles reveal abundant faults in the epicentral area that make up strike-slip fault systems. They suggest that the Mw 5.0 earthquake occurred due to the reactivation of an extensional fault created during back-arc rifting, which currently induces dextral slip under the ENE–WSW-oriented compressional stress field in and around the Korean Peninsula. The maximum magnitude of earthquakes expected at the margin is estimated as no higher than Mw 6.0. Restoration of seismic profiles indicates that the current stress field was established after 5.5 Ma. The S-wave velocity structure of the uppermost mantle shows asthenospheric upwelling elongated along the continental margin, which may be considered an important regional source of the current stress field by inducing convection in the uppermost mantle toward the Korean Peninsula lithosphere.

Didenko et al. (2022) in “A gravity-derived Moho model for the Sikhote Alin orogenic belt” have performed a two-dimensional power spectrum analysis of the Bouguer gravity field to calculate the crustal thickness (Moho depth) of the study region and adjust the regional model. The new model compared with the available structural and geological data shows that it is highly correlated with the development of Cretaceous–Early Eocene orogenic and post-orogenic granitoid massifs and Cretaceous–Pliocene extrusive igneous rocks. The former geographically coincide with two linear zones in the Moho relief with depths of more than 35 km, and the latter with the Mesozoic–Cenozoic sedimentary basins and the East Sikhote Alin volcano–plutonic belt.

Bergal-Kuvikas et al. (2022) in “Pleistocene-Holocene monogenetic volcanism at the Malko-Petropavlovsk zone of transverse dislocations on Kamchatka: Geochemical features and genesis” present new geochemical and isotopic results of monogenetic volcanism in the study area, based on whole rock and trace element geochemistry. Determinations of the pressure (9–11 kbar) and temperature (1160–1240 °C) conditions using a glass thermobarometer suggest that, prior to eruption, magma of monogenetic cinder cones resided near the Moho boundary. This observation correlates with the crustal discontinuity detected by seismic exploration and magnetotelluric sounding. Although eruptions have not been observed historically in the study area, continued quiescence in the future cannot be guaranteed. Taking into account the location of major population centers of Kamchatka (~ 250,000 people, i.e. ~ 80% of the entire population of the peninsula) the authors highlight an urgent need of continuous monitoring of the nearby volcanoes. More detailed studies regarding the age, volume, and pyroclastic vs extrusive of the monogenetic volcanoes will provide reliable assessment and reduction of potential risks for inhabitants and infrastructures.

Stepnova et al. (2022) in their “Predictive model of rainfall-induced landslides in high-density urban areas of the South Primorsky Region (Russia)” collected all available historical data about landslide incidents in the study area of Vladivostok City and surroundings and have derived a model of logistic regression analysis of antecedent, cumulative, and daily precipitation for forecasting eventual landslides. The advantage of the model is its simple mathematical expression. Despite rather crude reliable precipitation input data from a single meteorological station, accuracy of the model results shows a first approximation efficiency in assessing the probability of rainfall-induced landslides and suggests its further implementation for the purposes of early warning.

In Methods, Pisarenko and Pisarenko (2022) introduce “A modified k-nearest-neighbors method and its application to estimation of seismic intensity”, a new efficient nonlinear quantitative evaluation of the seismic activity based on locations of the observed earthquakes. It requires neither a preliminarily delineated area nor a normalization procedure. The method provides statistically justified output using an explicit form of the ‘‘uncertainty relation’’ between the spatial smoothing effect and random errors and a suggested objective procedure for choosing the nearest neighbors. The analysis of seismic activity in two regions surrounding the Kuril Islands and Japan, 1904–2011, identified “spots of increased seismic activity” along with their quantitative statistical characteristics. The proposed modified k-nearest-neighbors method might prove useful for seismic hazard and risk assessment applicable to locating critical infrastructures, hazardous waste repositories, etc.

Wang et al. (2022) in “Exploring magnitude estimation for earthquake early warning using available P-wave time windows based on Chinese strong-motion records” investigate directly the magnitude-scaling relationships within the available P-wave time window (APTW) defined as the window period starting from the trigger time of the P wave and ending at the arrival of the S wave. They apply this APTW method to explore real-time magnitude estimation. The results of an offline application demonstrate the good performance of the APTW method in terms of the stability of magnitude estimation for small to moderate earthquakes and improved estimation for large earthquakes. This new method can provide stable estimates of earthquake magnitude faster than the routine determinations available, providing an alternative choice for magnitude estimation in earthquake early warning systems.

Agayan et al. (2022) in “Fuzzy logic methods in the analysis of tsunami wave dynamics based on sea level data” present an algorithm for registering the arrival of tsunami waves based on the operational data of sea level measurements. The algorithm makes use of discrete mathematical analysis including a fuzzy logic approach and provides tools for expert assessment at the stage of adjustment and tuning. Its adaptive capabilities allow using real time input data preceding the arrival of the first tsunami wave. This could provide a universal tool for restructuring interpretation of different processes in real time.

Liu et al. (2022) report on “Probabilistic analysis of the landslide hazard in cold regions: Considering multiple triggering factors and their interdependence”. The authors analyze the performance the four machine learning technologies in assessing landslide probabilities under multiple triggering factors. Based on the Copula theory, they have developed the joint distribution of temperature difference and precipitation as triggering factors, determined the critical intensity and the intensity dependence curves of triggering factors under different return periods so as to illustrate their interdependence and influence on the landslide hazard. Making use of the Receiver Operating Characteristic curve (ROC), optimal national maps of landslide susceptibility and hazard were designed. The presented statistical test results allow concluding that, indeed, the landslide hazard probability and hazard intensity are coupled with each other. Therefore, neither the intensity of the hazards nor the precipitating factors can be ignored in a reliable risk assessment for population and infrastructure.

Tao et al. (2022) in their “Test of a PSHA map of China with fortification benefit evaluation” present an innovative methodology for testing probabilistic seismic hazard assessment (PSHA) maps by estimating the two indices of fortification, namely, economic benefit and safety benefit. The authors emphasize the importance of the effect of fortification intensity on the vulnerability of the infrastructure. The estimates of expected losses and casualties are calculated based on the total areas in each of the five damage states with local parameters. The case study of the 1990 PSHA map of China exemplifies its valuable role in earthquake disaster mitigation by the economic benefit of RMB 17.5, as well as by safety benefit of 20,838 fewer deaths and 77,801 fewer serious injuries.

Zhang et al. (2022a) in “Pattern Informatics (PI) of seismicity considering earthquake magnitude? An experiment in the central China North–South seismic belt” investigate the performance of intermediate-term earthquake forecasting of the central China north–south seismic belt for the last quarter century making use of a modified pattern informatics (PI) method. The modification uses the mean magnitude of earthquakes as an index instead of the number of earthquakes above the cutoff magnitude of the original PI analysis. The authors justify the alternative index, named PIm, by physics-based consideration that the mean magnitude of earthquakes is related to the maximum-likelihood estimation of b-values, a characteristic of seismicity associated with the regional level of tectonic stress. It is worth noting that PIm (i) assigns more weight to larger events, (ii) does not outscore the forecast performance of the original PI, (iii) while the superposition of PI × PIm, i.e. PIJ, significantly reduces the number of ‘hotspots’ and thus can reduce the number of false alarms.

Zhang et al. (2022b) in “Time-dependent seismic hazard assessment based on the annual consultation: A case from the China Seismic Experimental Site (CSES)” propose an interdisciplinary approach to time-dependent neo-deterministic seismic hazard assessment (T-NDSHA) for the China Seismic Experimental Site (CSES) at a one year time scale. T-NDSHA combines the neo-deterministic seismic hazard assessment (NDSHA) with the forecast defined by the Annual Consultation on the Likelihood of Earthquakes, organized by China Earthquake Administration (CEA). Since 1972, this interdisciplinary practice of the ‘alert regions’ with increased probabilities of strong earthquakes featured real forward forecasting. The authors take the year 2014 as a showcase example to illustrate how T-NDSHA may be conducted and evaluated. The results of evaluation of the T-NDSHA performance using confusion matrix and the Molchan’s error diagram suggests ready-to-use mapping of expected macroseismic intensities that outperforms random guessing. The combination with NDSHA provides substantial improvement to Annual Consultation. The T-NDSHA approach is applicable to other regions where intermediate-term, middle-range earthquake forecasts are available and where the needs of emergency preparation are duly considered.

Stark (2022) urges in “Pay no attention to the model behind the curtain” that many widely used models are rarely examined carefully to validate a reliable connection to real-world phenomena. Common steps in modeling to support policy decisions may conflict with reality but are convenient, customary, or familiar even in situations where the phenomena obviously violate the assumptions of the models. Impressive computer outputs and quantitative statements about probability, risk, health and/or economic consequences, etc., are often driven by a model behind the curtain—a model to which we are discouraged from paying attention. Not all costs and benefits can be put on the same scale, not all uncertainties can be expressed as probabilities, and not all model parameters measure what they purport to measure. These fundamental ideas are exemplified by considering widespread Probabilistic Seismic Hazard Assessment (PSHA), collisions of birds with wind turbines, statistical analysis of clinical trials, gender bias in academia, soccer penalty cards, climate models, and forecasting impacts of climate change. Multidisciplinary examples illustrate in a clear and understandable way both how not to model data and not to misinterpret statistical tests. In closing, the author re-emphasizes the key principles to help ensure that models serve society: i.e. (i) assess uncertainty and sensitivity; (ii) complexity can be the enemy of relevance; (iii) match purpose and context; (iv) quantification can backfire; and (v) acknowledge ignorance.

In Case studies, Safonov (2022) in the article “The earthquake of February 13, 2020, M = 7.0 and seismotectonic conditions at intermediate depths of the Southern Kuril Islands” performed the inversion of earthquake focal mechanisms of intermediate depth earthquakes for the southern part of the Kuril Islands. He found, in the upper layer, compression prevails along the slab subparallel to the Pacific Plate motion in the mantle, while the extension along the slab prevails in the lower layer and is mainly rotated by 20° clockwise from the direction of its dip. Such a pattern of earthquake focal mechanisms could indicate some additional mantle resistance to subduction. The most complex pattern of the stress is found near the hypocenter of the recent February 13, 2020, Mw 7.0 at about 95 km ENE of Kuril’sk (Russia), where the focal mechanisms in the upper layer more resemble those from the lower layer. The author reports “seismic quiescence” in the lower layer near the focus of this earthquake that might be useful for operational earthquake forecasting in the region.

Klausner et al. (2022) in “Ahead-of-tsunami magnetic disturbance detection using intrinsic mode functions: Tohoku-Oki earthquake case study” document magnetic disturbances that occurred during the Tohoku-Oki tsunami of 11 March 2011 using empirical mode decomposition (EMD) in a dataset derived from a network of ground-based magnetometers (INTERMAGNET and GIS). These disturbances, obtained by filtering the magnetic field data using the first intrinsic mode function (IMF1) of EMD, propagate ahead of the tsunami at a speed in the range of 600 to 1.6 km/s. Thus they appear at the magnetic observatories ahead of the tsunami arrival from about 3 min in the near-field to 2 h or more in the far-field. The authors named these disturbances as “ahead-of-tsunami magnetic disturbances” (ATMDs) and note that ATMDs commonly arrive about 10 min after the arrival of seismic Rayleigh waves. Monitoring of ATMDs in combination with seismic record analysis might become extremely useful for a reliable early warning of disastrous tsunami.

Soloviev et al. (2022) report “On the frequency distribution of geomagnetic K indices in the Northwestern Pacific Region over the 19–24 Solar Cycles” based on a huge collection of analog records that are the oldest measure of geomagnetic activity (K index) from observatories located in the Northwestern Pacific region. The records were digitized by a joint effort of the Russian and Japanese scientific teams, making it possible to study long-term evolution of geomagnetic activity in this region over 1954–2020. The authors reveal appropriate distribution laws, analyze the correlation between time-varying distribution features and sunspot numbers over the 19–24 solar cycles, and establish that the probability of K = 8 or larger events being detected simultaneously at all observatories in the region is less than one hundredth of a percent. The findings are of special importance for regional forecasting and early warning of geomagnetic storms.

Rodkin (2022) in “The variability of earthquake parameters with the depth: Evidences of difference of mechanisms of generation of the shallow, intermediate-depth, and the deep earthquakes” discusses variation of earthquake source characteristics with depth. He suggests that typical values of earthquake source parameters, such as normalized duration of seismic process, apparent stress, etc., vary substantially over depth. These variations are consistent with the presence of deep fluid that decreases the effective friction in rocks and/or by metamorphic processes occurring in down-going slabs including dehydration embrittlement and solid-state (phase) transformations.

Boginskaya and Kostylev (2022) in the article entitled “Change in the level of microseismic noise during the COVID-19 pandemic in the Russian Far East” have noticed a significant sharp decrease up to 30–50% in the daily background seismic noise during the period of the self-isolation. The pandemic’s restrictions on the operation of public institutions and the mobility of the population in large cities of the Russian Far East resulted in the improved quality of seismological observations. This is evident from analyzing seismograms in the period from March 23, 2020 to April 12, 2020 recorded at the seismic stations of Khabarovsk and Vladivostok located in busy parts of the cities, as well as those of the Yuzhno-Sakhalinsk seismic station located in a relatively quiet part of the city. Power spectra and temporal variations of microseismic noise levels based on the broadband seismometers records confirm a strong anthropogenic impact on seismic monitoring and earthquake hypocenter determination. Moreover, these analyzes allowed for identification of the main sources of the induced microseismic noise in or near the cities of Khabarovsk, Vladivostok, and Yuzhno-Sakhalinsk.

Kostylev et al. (2022) consider “Seismic activity in the focus of the Uglegorsk earthquakes, Sakhalin Island, related to intensive development of coal deposits”. The authors studied earthquakes that occur near the Solntsevskoye brown coal field, the most promising deposit on the Sakhalin Island. Active mining is on-going with blasting operations performed on a large scale. Earthquake recurrence graphs in 2000–2010 and 2011–2020 are evidently different and quite significant. Analyzing the spatiotemporal distribution of seismic event epicenters reveals an increase in seismic activity in the region during the past few years and a change in its character from natural to mixed natural and technogenic. In particular, the focal mechanisms of seismic events in 2020 were classified as strike-slip faulting, which is not characteristic of most earthquakes in the Sakhalin Island region. An attempt is made to determine some regularity in the parameters of the produced blasts and earthquakes, using dynamic parameters of the seismic event frequency content including corner frequency of the focal velocity spectrum. The study may be important for practical safety decisions related to the procedures and scale of blasting operations.

Konovalov et al. (2022) report on “Possible connection between recent seismicity and fluid injection in the offshore oil and gas field area of Sakhalin Island, Russia”. They observed that seismic activity on the northeast coast of Sakhalin in 2013–2014 coincides with the start of pilot operations at injection wells in nearby oil and gas fields. Hypothesizing that the recorded seismicity is entirely induced by the fluid disposal, and considering injection rates of 10⁵ m³/year, Konovalov et al. conclude that at least one induced earthquake with a magnitude M 5.5 is highly likely before 2041. The results obtained are of practical interest for developing seismic risk management strategies in the region.

Akiyama et al. (2022) describe “Spatiotemporal changes in fault displacements associated with seismovolcanic events in and around Miyakejima and Kozushima in 2000 inferred from GNSS Data”. This series of seismovolcanic events in central Japan began at the end of June 2000. Detailed analysis, inversion of the observed GNSS data, and location of microearthquakes identified three faults: Fault 1 to the northwest near Niijima; Fault 2 extending northwestward from Miyakejima volcano to Kozushima; and Fault 3 located beneath Miyakejima volcano. The data permitted considering several subfaults and performing more detailed inversion analyses to estimate the spatiotemporal dislocation field on the three faults. The inversion results show that (i) Fault 1 slipped significantly right-laterally by 18.9 m; (ii) the volume change is about 0.40 km³ and 0.74 km³ by opening on Faults 1 and 2, respectively; (iii) Fault 3 deflated by 0.57 km³. Therefore, since the total increase in volume on Faults 1 and 2 is greater than the total deflation on Fault 3 by a factor of 2, a magma supply from greater depths to Faults 1 and 2 is required for the volume overall balance.

Many authors who contributed to this Topical Issue continue to be inspired by Alexander Anatolievich Soloviev (21.10.1947–23.09.2021), a well-known Soviet-Russian geophysicist, former Director of Institute of Earthquake Prediction Theory and Mathematical Geophysics in Moscow (1998–2017), Corresponding Member of Russian Academy of Sciences (2000) and, in particular, by his integrity and a warm friendly attitude to people. Prof. Soloviev dedicated much of his attention to young researchers including participants of a series of Advanced schools on Nonlinear Dynamics and Earthquake Prediction, which he guided at the Abdus Salam Center for Theoretical Physics (Trieste, Italy, 1988–2011). Alexander Anatolievich (Sasha) made fundamental contributions to the development of methods to forecast locations of possible future strong earthquakes and to model dynamics of lithospheric blocks and faults. He developed approaches to universal description of situations preceding extreme events in complex systems of various natures including socioeconomic systems. Among his other scientific results are the discovery of the possibility of generating a magnetic field by the Couette-Poiseuille flow of a conducting fluid and the development of methods for calculating the movement of artificial satellites and other celestial bodies, taking into account perturbations of gravitational field, atmosphere and other factors. His scientific achievements were recognized by the global geophysical community and beyond.

Sasha left deep impression on everyone who met him as a kind gentleman as well as an accomplished scientist. We are devastated to have lost a wonderful man, talented scientist, wise mentor, colleague, and our friend…