Precursory Swarms Research Articles

Earthquake Swarm and Quiescence Precursory to the NE Himalaya 2011 Sikkim Earthquake (M 6.9)

Abstract The September 18, 2011 earthquake of M 6.9 has been critically examined and found to be associated with episodes of precursory swarms, quiescence, mainshock and aftershocks. This earthquake was clearly preceded by precursory swarm activity for almost one year followed by quiescence for one year. The precursory swarm and quiescence period consists of four earthquake swarms and one foreshock event of magnitude (mb ≥ 4.5) in the epicenter preparatory zone of the 2011 Sikkim earthquake. The 2011 Sikkim earthquake recorded about five aftershocks of magnitude (mb) ≥ 4.5 between 2011 and 2014 for the same region. This study shows that the 2011 Sikkim earthquake followed the same sequence of activities as observed for other major earthquakes in the northeast Himalaya.

Analysis on the predicting significance of mid-strong earthquake swarms in Gansu and its surrounding areas

In order to obtain the accurate prediction efficiency of earthquake swarms in Gansu province, the statistics of minor-moderate earthquake swarms that have occurred in Gansu since 1980 are performed. The earthquake swarm parameters are calculated and the precursory earthquake swarms are extracted based on the criteria of distinguishing the abnormal earthquake swarm. The precursory swarms obtained were analyzed correspondingly with the earthquakes with M≥5 in and around Gansu, and the prediction efficiency of them was evaluated. The accurate precursor index of the earthquake swarms in Gansu were summarized, which provided a judging basis for the seismic trending and made a certain contribution to earthquake prevention and disaster reduction.

Anomalous Seismicity and Earthquake Forecast in Western Nepal Himalaya and its Adjoining Indian Region

Precursory swarms associated with major earthquakes in the Western Nepal Himalaya and its adjoining region (bounded by 28.0°–31.0°N and 79.5°–82.2°E) have been studied using seismicity data from 1963 to 2006. The delineation of preparation zones for future seismic disturbances is carried out using the temporal and the spatial distribution of earthquakes, considering the events with cutoff magnitude mb ≥ 4.3 in four anomalous episodes: normal/background (N); anomalous/swarm (A); precursory gap (G) and main shock sequence (M), respectively. Five cases of anomalous seismicity have been identified, including two cases for which quiescence episodes still continue. Three moderate earthquakes of 1980 (mb 6.1, Bajhang), 1984 (mb 5.6, Bajura) and 1999 (mb 6.6, Chamoli) in Western Nepal and its adjoining Indian region were preceded by well-defined patterns of anomalous seismicity/precursory swarm. Two additional cases of anomalous seismicity patterns were observed: (1) 1999–2006, and (2) 2003–2006. In these two cases no main shock has yet occurred. However, the seismicity from 1999 onwards has fluctuated from low to high to low, as in the precursory sequences for previous earthquakes. The occurrence of the swarm sequence followed by a quiescence phase, which is still continuing, is an indication of a precursory seismicity gap in the region. From the predictive equations developed for the Himalayan frontal arc, it is estimated that an earthquake of M 6.5 ± 0.5 may occur at any time up to 2011 in an area bounded by 29.3°–30.5°N and 81.2°–81.9°E, in the focal depth range 10–30 km.

Lessons and Questions from Thirty Years of Testing the Precursory Swarm Hypothesis

In 1976 Frank Evison identified the first examples of earthquake swarms as long-term precursors of main-shock events, and thereby discovered the predictive scaling relations of long-term seismogenesis. From this time on, forecasting became the main focus of his research. After learning from an early attempt to communicate forecasts confidentially to government, he recognised the importance of hypothesis testing, and the precursory swarm hypothesis was cast in a form similar to a regional likelihood model. Tests of its performance relative to a stationary Poisson model at M ≥ 5.8 in New Zealand were begun in 1977. The initial hypothesis was that of a 1–1 relation between swarms and main-shock events. Following a study of the Japan catalogue, the generalised swarm hypothesis, in which multiple swarms were precursory to multiple main-shock events, was formulated. Tests of this form of the hypothesis at M ≥ 6.8 were initiated in a region of surveillance east of Japan in 1983. Eventually the generalised hypothesis was adopted in New Zealand also. In 1999, tests were begun in a region of Greece. In 1994–1995, several main-shock events favourable to the swarm hypothesis occurred, however four main-shock events near Arthur’s Pass, New Zealand, occurred without precursory swarms. Subsequent analysis showed that events called “quarms”, which were similar to swarms but more protracted in time, had preceded these events. This led to the proposal of a qualitative physical process to account for swarms, quarms and the predictive relations: A three-stage faulting process, in which a major crack induces aftercracks in its neighbourhood, just as a main shock induces aftershocks. An inference from this process was that the most general long-term precursor should be an increase of seismicity at similar magnitudes to the eventual aftershocks. It turned out that such a precursory scale increase nearly always occurs before major earthquakes and conforms to the predictive scaling relations. Setting aside the problem of identifying the scale increase before the major earthquake, the EEPAS (Every Earthquake a Precursor According to Scale) forecasting model was formulated. The success of this relatively weak model in forecasting major events in New Zealand, California, Japan and Greece shows that the predictive scaling relations are ubiquitous in earthquake catalogues. Although none of the formal tests of the swarm hypothesis were successful in their own terms, they were beneficial in identifying shortcomings in its formulation, thereby leading to improved understanding of long-term seismogenesis and a better forecasting model. Some puzzling aspects of the scaling relations are whether they vary regionally, and why the precursor area and aftershock area scale differently with magnitude. A more practical question is whether the EEPAS model can be strengthened, by making use of the clustering of some precursors in swarms and quarms, to bring us nearer to the original goal of forecasting individual major earthquakes.

The precursory earthquake swarm in Greece

The Hellenic subduction region displays the same precursory swarm phenomenon as has been found in comparable regions of New Zealand and Japan. In the earthquake catalogue of the Aristotle University of Thessaloniki, 10 past sequences of precursory swarms and related major mainshock events have been identified. These correlate, in respect of location, magnitude and time, with the 9 sequences previously identified in New Zealand, and 9 in Japan, bringing the total of sequences to 28, and the totals of related events (allowing for clustering) to 56 precursory swarms and 42 mainshock events. The results add strength to the hypothesis that swarms are long-range predictors of mainshock events. A close similarity between the swarm and aftershock magnitudes in a given sequence is also confirmed in Greece, supporting the proposal that swarms are an integral part of the seismogenic process in subduction regions. Further, the modelling of swarms as part of an overall increase in seismicity, the onset of which marks the onset of seismogenesis, is well illustrated from past sequences in Greece. Formal tests are being carried out in Greece, in parallel with New Zealand and Japan, to ascertain the performance of the hypothesis as a basis for long-range synoptic forecasting.

Precursory scale increase and long-term seismogenesis in California and Northern Mexico

A sudden increase in the scale of seismicity has occurred as a long-term precursor to twelve major earthquakes in California and Northern Mexico. These include all earthquakes along the San Andreas system during 1960-2000 with magnitude M* 6.4. The full list is as follows: Colorado Delta, 1966, M 6.3; Borrego Mt., 1968, M 6.5; San Fernando, 1971, M 6.6; Brawley, 1979, M 6.4; Mexicali, 1980, M 6.1; Coalinga, 1983, M 6.7; Superstition Hills, 1987, M 6.6; Loma Prieta, 1989, M 7.0; Joshua Tree, 1992, M 6.1; Landers, 1992, M 7.3; Northridge, 1994, M 6.6; Hector Mine, 1999, M 7.1. Such a Precursory Scale Increase (s ) was inferred from the modelling of longterm seismogenesis as a three-stage faulting process against a background of self-organised criticality. The location, onset-time and level of s are predictive of the location, time and magnitude of the future earthquake. Precursory swarms, which occur widely in subduction regions, are a special form of s ; the more general form is here shown to occur frequently in a region of continental transform. Other seismicity precursors, including quiescence and foreshocks, contribute to or modulate the increased seismicity that characterises s . The area occupied by s is small compared with those occupied by the seismicity precursors known as AMR, M 8 and LURR. Further work is needed to formulate s as a testable hypothesis, and to carry out the appropriate forecasting tests.

Medium-term forecast of the 1988 north-east India earthquake

Northeast India is seismically one of the most active intra-continental regions in the world. It has been a site of 10 earthquakes of magnitude ( M) ≥7.5 during the last 100 years. Gupta and Singh [J. Geol. Soc. India, 28 (1986) 367–406] systematically analysed the time series of earthquakes associated with several main shocks in an area bound by 20°N and 32°N latitude and 87°E and 100°E longitude and concluded: (1) Moderate magnitude to great earthquakes in the north-east India region are found to be preceded, generally, by well defined earthquake swarms and quiescence periods. (2) On the basis of an earthquake swarm and quiescence period, an area bound by 21°N and 25.5°N latitude and 93°E and 96°E longitude is identified to be the site of a possible future earthquake of M=8±0.5 with a focal depth of 100±40 km. This earthquake should occur any time from now onwards. Should it not occur till the end of 1990, this forecast could be considered as a false alarm. This medium-term earthquake forecast came true with the occurrence of the M=7.3 earthquake on August 6, 1988, within the specified spatial and temporal parameters. The forecast of this earthquake was based on the concept of precursory swarms and quiescence preceding main-shocks. It is worthwhile noting that in the entire region under study, the August 6, 1988 earthquake was the largest since the August 17, 1952 earthquake of M=7.5. This lends support to the worldwide effort to study and identify spatial and temporal variation of seismicity and recognise patterns that precede major earthquakes.

The precursory earthquake swarm and the inferred precursory quarm

The earthquake swarm as a long‐term precursor to major earthquakes is being studied empirically by the method of hypothesis formulation and testing. Results suggest that swarms occur as an early stage of the seismogenic process for major earthquakes, but that the phenomenon is limited to shallow subduction regions and other environments where high fluid pressure provides a triggering mechanism for swarms. An inference from a proposed seismogenic model is that, if swarms are not triggered, the same precursory earthquakes should occur, though in a more protracted manner. A cluster of four major earthquakes (ML 6.7, 6.3, 6.0, 5.8) occurred without precursory swarms in the area around Arthur's Pass, New Zealand, during 1992–95. This area is in the region of continental collision which lies between the Hikurangi and Fiordland subduction regions. A prior cluster of seismicity anomalies has now been identified which displays the same precursory relations as have been established for swarms. These anomalies are termed quasi‐swarms, or quarms, and the swarm hypothesis is now being extended to include the precursory quarm.

The precursory earthquake swarm in New Zealand: Hypothesis tests

A series of tests is being carried out with the object of determining whether the relations evident in New Zealand and Japan, between precursory swarms and major earthquakes, are of value for long‐range, synoptic earthquake forecasting. The first New Zealand test (completed in 1990) showed that clustering, both of swarms and of mainshock events, should be allowed for, and the hypothesis was reformulated accordingly. The second New Zealand test, now completed, confirms the importance of clustering. It also reveals that the applicability of the swarm/mainshock relations is strongly affected by large‐scale tectonics. Further, a simulation study shows that, at the end of the test, the performance of the swarm hypothesis relative to the Poisson model lay between what would be expected if the hypothesis and the Poisson model, respectively, were correct in general. These results support a further reformulation of the hypothesis, which is now taken to be strongly applicable in the Hikurangi and Fiordland subduction zones, not at all in the region of Quaternary volcanism, and weakly elsewhere. This change and the provision for clustering made previously bring the present New Zealand formulation close to that under test in the Japan region. The third New Zealand test commenced on 20 December 1995 and is still in progress.

Comparison of precursory and non-precursory swarm activity in peninsular India

Earthquake swarms in peninsular India can be classified into two types namely: (1) precursory type which are followed by a larger earthquake; and (2) non-precursory type implying sudden increase of felt tremors which die off after some time. The characteristics of these tremors have been examined in this paper. Earthquake catalogues have been prepared for both type of swarms. It was found that the non-precursory swarms have shallower focal depths, up to 2 km or so, while precursory swarms/foreshocks are associated with greater focal depths. Also, the maximum magnitude of non-precursory swarms has rarely exceeded 4.0 which is lower than that observed in precursory swarms. The time t (in days) from the earliest felt earthquake to the main earthquake of magnitude M is related by M = 1.88 + 1.50log t. This relationship and the distinctive criteria for both type of swarms presented in this study could be refined when additional data becomes available. The occurrence of earthquakes according to all the three types of the model of Mogi, namely I, II and III, expresses the complexity in intraplate stress regimes over peninsular India, which is supported by the focal mechanism solutions.

Global volcanic earthquake swarm database and preliminary analysis of volcanic earthquake swarm duration

Global data from 1979 to 1989 pertaining to volcanic earthquake swarms have been compiled into a custom-designed relational database. The database is composed of three sections: 1) a section containing general information on volcanoes, 2) a section containing earthquake swarm data (such as dates of swarm occurrence and durations), and 3) a section containing eruption information. The most abundant and reliable parameter, duration of volcanic earthquake swarms, was chosen for preliminary analysis. The distribution of all swarm durations was found to have a geometric mean of 5.5 days. Precursory swarms were then separated from those not associated with eruptions. The geometric mean precursory swarm duration was 8 days whereas the geometric mean duration of swarms not associated with eruptive activity was 3.5 days. Two groups of precursory swarms are apparent when duration is compared with the eruption repose time. Swarms with durations shorter than 4 months showed no clear relationship with the eruption repose time. However, the second group, lasting longer than 4 months, showed a significant positive correlation with the log10 of the eruption repose period. The two groups suggest that different suites of physical processes are involved in the generation of volcanic earthquake swarms.

Application of PP cluster method in the earthquake swarm analysis

Taking 98 earthquake swarms occurred in Xinjiang during 1972–1992 as examples, and 8 parameters (e. g.U, K, ρ and the maximum energy rate of earthquake sequenceetc.) as the characteristic quantity in earthquake swarm pattern observation, the author made a numerical cluster by PP cluster analysis method. The results indicate that those 98 earthquake swarms can be divided into 4 types as A, B, C, D. There are 24 swarms in type A, among which strong shocks occur nearby after 18 swarms in the coming 12 months. Among 61 earthquake swarms in type C and D, strong shocks occur nearby only after 7 swarms in the same time period. The occurrence rate of strong shocks only takes 3/11 in type B swarms. No doubt, PP cluster analysis method can effectively distinguish precursory swarms (type A) and correctly judge the short- and medium-term trend in the area around the earthquake swarms. Being a new and useful classification, PP cluster provides a wide application to the identification of the type of earthquake sequence.

Statistical forecasting of repetitious dome failures during the waning eruption of Redoubt Volcano, Alaska, February–April 1990

The waning phase of the 1989–1990 eruption of Redoubt Volcano in the Cook Inlet region of south-central Alaska comprised a quasi-regular pattern of repetitious dome growth and destruction that lasted from February 15 to late April 1990. The dome failures produced ash plumes hazardous to airline traffic. In response to this hazard, the Alaska Volcano Observatory sought to forecast these ash-producing events using two approaches. One approach built on early successes in issuing warnings before major eruptions on December 14, 1989 and January 2, 1990. These warnings were based largely on changes in seismic activity related to the occurrence of precursory swarms of long-period seismic events. The search for precursory swarms of long-period seismicity was continued through the waning phase of the eruption and led to warnings before tephra eruptions on March 23 and April 6. The observed regularity of dome failures after February 15 suggested that a statistical forecasting method based on a constant-rate failure model might also be successful. The first statistical forecast was issued on March 16 after seven events had occurred, at an average interval of 4.5 days. At this time, the interval between dome failures abruptly lengthened. Accordingly, the forecast was unsuccessful and further forecasting was suspended until the regularity of subsequent failures could be confirmed. Statistical forecasting resumed on April 12, after four dome failure episodes separated by an average of 7.8 days. One dome failure (April 15) was successfully forecast using a 70% confidence window, and a second event (April 21) was narrowly missed before the end of the activity. The cessation of dome failures after April 21 resulted in a concluding false alarm. Although forecasting success during the eruption was limited, retrospective analysis shows that early and consistent application of the statistical method using a constant-rate failure model and a 90% confidence window could have yielded five successful forecasts and two false alarms; no events would have been missed. On closer examination, the intervals between successive dome failures are not uniform but tend to increase with time. This increase attests to the continuous, slowly decreasing supply of magma to the surface vent during the waning phase of the eruption. The domes formed in a precarious position in a breach in the summit crater rim where they were susceptible to gravitational collapse. The instability of the February 15–April 21 domes relative to the earlier domes is attributed to reaming the lip of the vent by a laterally directed explosion during the major dome-destroying eruption of February 15, a process which would leave a less secure foundation for subsequent domes.

Seismological aspects of the 1989–1990 eruptions at redoubt volcano, Alaska: the SSAM perspective

SSAM is a simple and inexpensive tool for continuous monitoring of average seismic amplitudes within selected frequency bands in near real-time on a PC-based data acquisition system. During the 1989–1990 eruption sequence at Redoubt Volcano, the potential of SSAM to aid in rapid identification of precursory Long-Period (LP) event swarms was realized, and since this time SSAM has been incorporated in routine monitoring efforts of the Alaska Volcano Observatory. In particular, an eruption that occurred on April 6 was successfully forecast primarily on the basis of recognizing the precursory LP activity on SSAM. Of twenty-two significant eruptions that occurred between December 14 and April 21, eleven had precursory swarms longer than one hour in duration that could be detected on SSAM. For individual swarms, the patterns of relative spectral amplitudes are distinct at each station and remain largely stationary through time, thus indicating that one source may have been preferentially and repeatedly activated throughout the swarm. Typically, a single spectral band dominates the signal at each seismic station: for the vigorous one-day swarm that preceded the first eruption on December 14, signals were sharply peaked in the 1.9–2.7 Hz band at the closest station, located 4 km from the vent, but were dominated by 1.3–1.9 Hz energy at three more distant stations located 7.5–22 km from the vent. The tendency for the signals from different swarms recorded at the same station to be peaked in the same frequency band suggests that all of the sources are characterized by a predominant length scale. Signals from the precursory LP swarms became weaker as the eruption sequence progressed, and swarms that occurred in March and April could only be detected at seismographs on the volcanic edifice. Onset times of precursory LP swarms prior to eruptions ranged from a few hours to about one week, but after the initial vent-clearing phase that ended December 19 these intervals tended to become progressively shorter for successive swarms. These trends in the relative onset times and intensities of successive precursory LP swarms are consistent with an overall depressurization of the magmatic system through time. In general, each of the swarms had an emergent onset, but the intensities did not always increase steadily until the eruptions. Instead, as the time of an eruption approached the intensity usually increased more rapidly before peaking and then declining prior to the eruption; for three of the swarms, two distinct peaks in intensity were apparent. The time intervals between final peaks in swarm intensity and ensuing eruptions ranged from about 2 hours to almost 2 days, but the peaks always occurred closer to the eruptions than to the swarm onsets. Both the onset of LP swarm activity and a decline in intensity prior to an eruption may represent critical points in the process of pressurization that drives the flow of fluids and gas in a sealed magmatic system. A notable exception to this pattern is the eruption of March 9 which lacked a detectable precursory LP swarm, but was followed by an unusually long period of strong LP seismicity that may have been stimulated by a depressurization of the magmatic system resulting from dome failure. On both December 14 and January 2, the spectra of early syn-eruptive signals have peaked signatures much like those of the spectra of precursory LP activity from shortly before the eruptions; these similarities may indicate that the source of precursory seismicity continued to be active during at least the early part of each eruption. In syn-eruptive signals from March and April recorded at stations on the volcanic edifice, the dominant spectral energy progressively shifts with time during the eruption to lower frequencies; at least part of the energy in these signals may have been generated by the debris flows associated with dome failures.

Precursory swarms of long-period events at Redoubt Volcano (1989–1990), Alaska: Their origin and use as a forecasting tool

During the eruption of Redoubt Volcano from December 1989 through April 1990, the Alaska Volcano Observatory issued advance warnings of several tephra eruptions based on changes in seismic activity related to the occurrence of precursory swarms of long-period (LP) seismic events (dominant period of about 0.5 s). The initial eruption on December 14 occurred after 23 years of quiescence and was heralded by a 23-hour swarm of LP events that ended abruptly with the eruption. After a series of vent-clearing explosions over the next few days, dome growth began on December 21. Another swarm, with LP events similar to those of the first, began on the 26th and ended in a major tephra eruption on January 2. Eruptions continued over the next two weeks and then ceased until February 15, when a large eruption initiated a long phase of repetitive dome-building and dome-destroying episodes that continued into April. Warnings were issued before the major events on December 14 and January 2, but as the eruptive sequence continued after January 2, the energy of the swarms decreased and forecasting became more difficult. A significant but less intense swarm preceded the February 15 eruption, which was not forecast. This eruption destroyed the only seismograph on the volcanic edifice and stymied forecasting until March 4, when the first of three new stations was installed within 3 km of the active vent. From March 4 to the end of the sequence on April 21, there were eight eruptions, six of which were preceded by detectable swarms of LP events. Although weak, these swarms provided the basis for warnings issued before the eruptions on March 23 and April 6. The initial swarm on December 13 had the following features: (1) short duration (23 hours); (2) a rapidly accelerating rate of seismic energy release over the first 18 hours of the swarm, followed by a decline of activity during the 5 hours preceding the eruption; (3) a magnitude range from −0.4 to 1.6; (4) nearly identical LP signatures with a dominant period near 0.5 s; (5) dilatational first motions everywhere; and (6) a stationary source location at a depth of 1.4 km beneath the crater. This occurrence of long-period events suggests a model involving the interaction of magma with groundwater in which magmatic gases, steam and water drive a fixed conduit at a stationary point throughout the swarm. The initiation of that sequence of events is analogous to the failure of a pressure-relief valve connecting a lower, supercharged magma-dominated reservoir to a shallow hydrothermal system. A three-dimensional model of a vibrating fluid-filled crack recently developed by Chouet is found to be compatible with the seismic data and yields the following parameters for the LP source: crack length, 280–380 m; crack width, 140–190 m; crack thickness, 0.05–0.20 m; crack stiffness, 100–200; sound speed of fluid, 0.8–1.3 km/s; compressional-wave speed of rock, 5.1 km/s; density ratio of fluid to rock, ≈0.4; and ratio of bulk modulus of fluid to rigidity of rock, 0.03–0.07. The fluid-filled crack is excited intermittently by an impulsive pressure drop that varies in magnitude within the range of 0.4 to 40 bar. Such disturbance appears to be consistent with a triggering mechanism associated with choked flow conditions in the crack.

Relocation of teleseismically recorded earthquakes near Tennant Creek, Australia: Implications for midplate seismogenesis

The three MS > 6 Tennant Creek, Australia, earthquakes of January 22, 1988, were preceded by small and moderate earthquakes in 1986 and 1987. We have used the method of joint hypocenter determination to estimate the positions of teleseismically recorded 1986–1987 earthquakes, the 1988 main shocks, and aftershocks with respect to a system of surface fault scarps. The 1986–1987 shocks and the 1988 main shocks nucleated near the center of the zone of surface scarps, where fault‐segment boundaries at depth are implied by complexities in the distribution of scarps at the surface. This suggests that the fault segments that ruptured in 1988 were already in existence in 1986–1987, which is consistent with the hypothesis that strong midplate earthquakes occur on preexisting faults. The redetermined 1986–1987 hypocenters are, however, also consistent with the hypothesis that midplate seismicity is localized by stress concentration due to bulk rheological heterogeneity of the crust, because they are situated on a regionally prominent gravity anomaly. The teleseismically recorded seismicity of the Tennant Creek region prior to the 1988 main shocks has the temporal pattern of a swarm followed by a lull. The concentration of swarm earthquakes between two scarps is consistent with models in which precursory‐swarm earthquakes correspond to faulting that is spatially distinct from the site of primary main‐shock faulting. The time interval between swarm and main shocks is similar to intervals between intermediate‐term precursory swarms and main shocks in regions that have much higher rates of tectonic loading; this similarity suggests that the time intervals between precursory swarms and subsequent main shocks are not strongly influenced by the rate of tectonic loading, but are determined primarily by time dependence of the failure process. The spatial distribution of teleseismically recorded aftershocks is in most respects like that of aftershocks recorded by a local network of portable stations in the half year following the main shocks. The set of teleseismically recorded aftershocks, like the set of locally recorded aftershocks, includes some events that occurred well away from the causative faults of the main shocks. At the length scale of the 1988 main‐shock rupture, the distribution of aftershocks occurring more than one year after the main shocks is not representative of the distribution of earlier aftershocks.

Earthquake swarms precursory to moderate to great earthquakes in the northeast India region

The northeast India region has seen ten M ⩾ 7 1 2 earthquakes since 1897, including two great earthquakes of M = 8.7 in 1897 and 1950. The last M ⩾7 1 2 earthquake occurred on August 17, 1952. With the exception of three earthquakes others are found to be associated with periods of background/normal seismicity, precursory swarms, quiescence and mainshocks (and the associated foreshocks and aftershocks). The dataset has been critically examined for completeness considering the current capabilities for defining locations and also considering other parameters. Regression equations relating the mainshock magnitude ( M m ), the average magnitude of the largest two events in the swarm ( M p ) and the time interval ( M p ) between the beginning of the swarm and the mainshock have been developed. These are: M m = 1.37 M p − 1.41 and M m = 3 log 10 T p − 3.27 It is important to recognize precursory swarms and quiescence in real time; indeed, we believe we have recognized one such sequence in the vicinity of the Arakan Yoma Fold Belt. On the basis of M p− and T p− values, the lateral extent of swarm activity and the fact that no event of M > 6 has occurred since 1975 in the preparation zone defined by the 1963–1965 swarm, we estimate that an M~ 8 earthquake could occur at any time in the area bounded by 21° and 25 1 2 ° N and 93° and 96° E. The focal depth of this impending earthquake is estimated to be 100 ± 40 km in view of the focal depths of the other events in the swarm.

Seismicity of the North-East India Region Part II: Earthquake Swarms Precursory to Moderate Magnitude to Great Earthquakes

Abstract Since 1897, ten earthquakes of M ≥ 7½ have occurred in north-east India region and its vicinity including the great earthquakes of June 12, 1897 and August 15, 1950, both of M 8.7. The last earthquake belonging to this category, occurred on August 17, 1952. With the exception of three earthquakes, others are found to be associated with the epochs of background/normal seismicity, precursory swarm, quiescence and mainshock (and the associated foreshocks and aftershocks). In our investigations, the earthquake data have been critically examined and taking due care of prevalent location - capabilities and other parameters. The mainshock magnitude (Mm) is related to the magnitude of the largest events in the swarm (̅Mp) and the time interval Tp (in days) between the beginning of the swarm and the main shock. The regression equations are: Mm = 1.37 ̅Mp - 1.41 and Mm = 3 log10 Tp - 3.27 Installation of the World Wide Standard Seismograph Network during 1962-63 improved the earthquake detection and location capabilities considerably in the northeast India region and all earthquakes of mb ≥ 4.5 are being located. Consequently, for the period 1963 through 1985, we have investigated all events of mb ≥ 6.1 for occurrence of precursory swarms. Ten such events have occurred, four of which in a quick succession in the Arakan Yoma fold belt within a time span of six months only, and constitute a well defined swarm. Two other earthquakes are possibly gap events of Arakan Yoma swarm. Out of the remaining four, three are found to be associated with epochs of normal/background seismicity, precursory swarm, quiescence and main. shock (and the associated foreshocks and aftershocks). The latest killer earthquake in northeast India of Dec. 30, 1984 of mb 5.6, which occurred in the vicinity of the Arakan Yoma fold belt and Shillong Plateau, was preceded by a well-defined precursory swarm and seismic quiescence. The most important thing is to recognize the precursory swarms and quiescence in real time, before the occurrence of the main shock. We believe to have recognized one such sequence in the vicinity of Arakan Yoma fold belt. On the basis of Mp and Tp values, lateral extent of the swarm activity and the fact that no event of mb ≥ 6.1 has occurred since 1975 in the preparation zone defined by the 1963-65 swarm, we estimate that an earthquake of M 8 ½ could occur any time from now in an area bound by 21°N and 25½°N latitudes and 93°E and 96°E longitude, with a focal depth of 100±40km. We have also recognized a region in Eastern Syntaxis where repeated swarms have occurred during 1968, 1977 and 1980. The largest events in these swarms, however, were only of mb 5 and the swarms were very localized. These may have been caused by magmatic emplacement at shallow crustal depths.

Stress distribution pattern and swarm occurrences

At the Himalayan trench three consecutive swarms (i.e. the precursory swarm, the main-shock swarm and the aftershock swarm) have been reported to occur in small epicentral regions. In these swarms the magnitudes of the main-shock events are one to two units higher than the other two. These swarms appear to be related to each other. In order to establish the relationship between the first two swarms, focal mechanism studies for main shocks and the precursory swarms were carried out. The study of the precursory swarm demonstrates the converging nature of stresses in the epicentral region of the main-shock swarm. This suggests that the stresses responsible for the precursory swarms are also the cause of the main-shock swarms. The aftershock swarms are the result of the interaction of the converging stresses with the outward stresses generated due to adjustment of faulted blocks during large earthquakes. In view of the above, it can be asserted that precursory swarms are precursors to the main shocks.

Focal mechanisms and aftershock locations of the Songpan earthquakes of August 1976 in Sichuan, China

The precursory swarm, three mainshocks (M = 7.2,6.7, 7.2), and aftershocks of the Songpan earthquakes have been reanalyzed using both local and teleseismic data. The three mainshocks of this sequence occurred on the Huya fault over a 7‐day period. Relocations of the aftershocks using local arrival times show that three fault strands were activated during this sequence. Each mainshock occurred on a separate strand, each one south of the strand activated in the previous mainshock, and the aftershock zones of each mainshock appear to abut rather than overlap. Fault plane solutions determined by matching teleseismic P waveforms at World‐Wide Standard Seismograph Network stations with synthetic seismograms are consistent with the observed aftershock zones. The first and third mainshocks (M0 = 1.3 ×1 019 and 8.4 × 1018 N m, respectively) showed almost identical senses of motion, a combination of reverse and left‐lateral strike‐slip motion, on parallel strands, striking N15°W, that were separated by a large rightstepping en echelon offset. The second mainshock (M0 = 4.0 × 1018 N m), occurred in this offset on a fault at a steep angle (∼125°) to the other two strands and showed almost pure reverse motion. Differences in the orientations of the slip vectors of the three mainshocks show that the first mainshock increased the normal and shear stresses on the fault segment that moved in the second mainshock and that the second mainshock decreased the normal stress on the fault segment activated by the third mainshock. These changes in normal stresses may have given rise to the longer time between the first and second events (5 days) as compared with the time between the second and third events (30 hours). A precursory swarm that preceded the Songpan sequence by 3 years occurred in a volume that surrounded the northernmost part of the planar aftershock zone. The time between the start of the swarm and the mainshocks and the magnitude of the largest event in the swarm are similar to those seen for precursory swarms in Soviet Central Asia.