Main Swarm Research Articles

A hovering swarm particle swarm optimization algorithm based on node resource attributes for hardware/software partitioning

Hardware/software (HW/SW) partitioning is a vital aspect of HW/SW co-design. With the development of the design complexity in heterogeneous computing systems, existing partitioning algorithms have demonstrated inadequate performance in addressing problems relating to large-scale task nodes. This paper presents a novel HW/SW partitioning algorithm based on node resource attributes hovering swarm particle swarm optimization (HSPSO). First, the system task graph is initialized via the node resource urgency partitioning algorithm; then, the iterative solution produced by HSPSO algorithm yields the partitioning result. We present new initialization by combining node resource attribute information and introduce two improvements to the learning strategy of HSPSO algorithm. For the main swarm, a directed sample set and the addition of perturbation particles are designed to direct the main swarm’s particle search process. For the secondary swarm, a dynamic particle update equation is formulated. Iterative updates are performed based on previous rounds’ prior information using adaptive inertia weight. The experimental results illustrate that, in large-scale systems task graph partitioning with more than 400 nodes, when compared with mainstream partitioning algorithms, the proposed algorithm improves partitioning performance by no less than 10% for compute-intensive task graphs and no <5% for communication-intensive task graphs, with higher solution stability.

Interaction Between Aseismic Slip and Fluid Invasion in Earthquake Swarms Revealed by Dense Geodetic and Seismic Observations

AbstractRecent seismic and geodetic observations suggest the involvement of pressurized fluid and aseismic slip during earthquake swarms. However, the interaction between these two factors is not fully understood. In this study, we show geodetic signals induced by aseismic opening and shear dislocation on a fault, accompanied by hypocenter migration during an earthquake swarm in the Hakone volcano, central Japan. The hypocenters were concentrated on a vertical plane, and the focal mechanisms were strike‐slip faults whose nodal planes were consistent with the hypocenter distribution. Precursory seismicity occurred 1 week before the main swarm. We observed a complex pattern of hypocenter migration during the seismic sequence; namely, the lower limit of the seismicity appears to expand downward during the precursory seismicity, followed by an upward hypocenter diffusion. We also observed tilt changes that could be explained by opening and shear dislocation on the fault plane inferred from the planar hypocenter distribution. Furthermore, we identified several groups of repeating earthquakes. The average aseismic slip history inferred from the repeating earthquakes indicated a minor aseismic slip of 0.4 mm during the precursory seismic activity, which accelerated by up to 3 mm during the main swarm. The total cumulative slip was consistent with the geodetic model results. These observations may suggest that fluid intrusion caused the aseismic slip, and the complex migration of the hypocenters reflects the aseismic slip propagation. We hypothesize that aseismic deformation acts as the driving force for the earthquake swarms together with the intrusion of pressurized fluid.

The seismicity of Campi Flegrei in the contest of an evolving long term unrest

One of the most effective approaches to identifying possible precursors of eruptions is the analysis of seismicity patterns recorded at volcanoes. Accurate locations of the seismicity and the estimate of source mechanisms can resolve fault systems and track fluid migrations through volcanoes. We analysed the six main swarms recorded at Campi Flegrei since 2000, using them as a proxy of the processes involved in the long-term-unrest of this densely populated caldera. We re-located the earthquakes comprised in these swarms and estimated the focal mechanisms, which appear in agreement with the fault systems of the caldera and with tomographic images. The focal mechanisms are in agreement with the tensional stress induced by the caldera uplift. Most of the swarms and remaining seismicity delineate a highly fractured volume extending vertically below the Solfatara/Pisciarelli vents, where gases find preferential paths to the surface triggering earthquakes. The main swarms are located below this volume where the presence of a rigid caprock is still debated. We interpreted the current unrest in term of a gradual increment in the activity of a wide hydrothermal system whose most evident manifestation is the enlargement of the fumarolic-field of Pisciarelli and the increment of the earthquakes occurrence rate.

A Novel Particle Swarm Optimizer and Its Application to the Yield Curve Estimation Problem

Particle swarm optimization (PSO) has been considered as one of the main swarm intelligence algorithms for solving single-objective optimization problem. How to update the velocities and positions of a swarm of particles is key to its optimization performance. In this paper, we propose to use the moving average of the local best positions visited so far as a key information to update the velocities and positions. Further, a central learning strategy is proposed in which the center of the local best positions is computed and used to update the global best position. Combining these two strategies with the updating formulas which are inspired by the free electron model in metal conductors placed in an external electric field, we name the proposed PSO algorithm as PSO with moving average and central learning strategies (dubbed as MAPSO). We test the performance of MAPSO on the CEC 2017 test problems with 10 and 30 dimensions. Experimental results show that MAPSO significantly outperforms some well-known PSOs in general. We then apply MAPSO on a very challenging real-world problem, i.e. the yield curve estimation problem, in macroeconomics. Our experimental study on the yield curve estimation problem with Shanghai interbank offered rates shows that MAPSO can effectively solve the problem and achieve the state-of-the-art performance.

Campi Flegrei, Vesuvius and Ischia Seismicity in the Context of the Neapolitan Volcanic Area

Studying seismicity in a volcanic environment provides important information on the state of activity of volcanoes. The seismicity of the Neapolitan volcanoes, Campi Flegrei, Vesuvius, and Ischia, shows distinctive characteristics for each volcano, covering a wide range of patterns and types. In this study we relocated some significant volcano-tectonic earthquake swarms that occurred in Campi Flegrei and Vesuvius. Moreover, we compared the earthquake occurrence evolution, the magnitude and the seismic energy release of the three volcanoes. Also, we considered the results of seismic analysis in the light of geochemical and ground deformation data that contribute to defining the state of activity of volcanoes. In Campi Flegrei, which is experiencing a long term unrest, we identified a seismogenic structure at shallow depth in Pisciarelli zone that has been activated repeatedly. The increasing seismicity accompanies an escalation of the hydrothermal activity and a ground uplift phase. At Vesuvius a very shallow seismicity is recorded, which in recent years has shown an increase in terms of the number of events per year. Earthquakes are usually located right beneath the crater axis. They are concentrated in a volume affected by the hydrothermal system. Finally, Ischia generally shows a low level of seismicity, however, in Casamicciola area events with a moderate magnitude can occur and these are potentially capable of causing severe damage to the town and population, due to their small hypocentral depth (typically &amp;lt; 2.5 km). After the seismic crisis of August 21, 2017 (mainshock magnitude M = 4), the seismicity returned to a low level in terms of occurrence rate and magnitude of earthquakes. The seismicity of these three different volcanic areas shows some common aspects that highlight a relevant role of hydrothermal processes in the seismogenesis of volcanic areas. However, while the main swarms in Campi Flegrei and most of the Vesuvian earthquakes are distributed along conduit-like structures, the seismicity of Ischia is mainly located along faults. Furthermore, the temporal evolution of seismicity in Neapolitan volcanic area suggests a concomitant increase in the occurrence of earthquakes both in Campi Flegrei and Vesuvius in recent years.

Visual victim detection and quadrotor-swarm coordination control in search and rescue environment

We propose a distributed victim-detection algorithm through visual information on quadrotors using convolutional neuronal networks (CNN) in a search and rescue environment. Describing the navigation algorithm, which allows quadrotors to avoid collisions. Secondly, when one quadrotor detects a possible victim, it causes its closest neighbors to disconnect from the main swarm and form a new sub-swarm around the victim, which validates the victim’s status. Thus, a formation control that permits to acquire information is performed based on the well-known rendezvous consensus algorithm. Finally, images are processed using CNN identifying potential victims in the area. Given the uncertainty of the victim detection measurement among quadrotors’ cameras in the image processing, estimation consensus (EC) and max-estimation consensus (M-EC) algorithms are proposed focusing on agreeing over the victim detection estimation. We illustrate that M-EC delivers better results than EC in scenarios with poor visibility and uncertainty produced by fire and smoke. The algorithm proves that distributed fashion can obtain a more accurate result in decision-making on whether or not there is a victim, showing robustness under uncertainties and wrong measurements in comparison when a single quadrotor performs the mission. The well-functioning of the algorithm is evaluated by carrying out a simulation using V-Rep.

Hovering Swarm Particle Swarm Optimization

PSO is a simple and yet powerful metaheuristic search algorithm widely used to solve various optimization problems. Nevertheless, conventional PSO tends to lose its population diversity drastically and suffer with compromised performance when encountering the optimization problems with complex fitness landscapes. Extensive studies suggest the needs of preserving high population diversity for PSO to escape from the local optima in order to solve complex optimization problems effectively. Inspired by these ideas, a hovering swarm PSO (HSPSO) is proposed in this paper, where a computationally efficient diversity preservation scheme is first introduced to divide the population of HSPSO into a main swarm and a hovering swarm. An exemplar construction scheme is subsequently proposed in the main swarm of HSPSO to generate a universal exemplar by considering the promising directional information contributed by the other non-fittest particles. The proposed universal exemplar is envisioned to suppress the negative impacts of global best particle, while remain effective to guide all particles of main swarm converging towards the promising solution regions. While hovering around the main swarm, an intelligent scheme is introduced to dynamically adjust inertia weights of all hovering swarm members to achieve proper balancing of exploration and exploitation searches at swarm levels. Extensive performance analyses are conducted by using the proposed HSPSO to solve 30 benchmark functions of CEC 2014 and five real-world engineering applications. Simulation results reveal that the HSPSO is able outperform the state-of-art optimizers when solving most tested functions due to its excellent diversity preservation capability.

Triggered crustal earthquake swarm across subduction segment boundary after the 2016 Pedernales, Ecuador megathrust earthquake

Megathrust ruptures and the ensuing postseismic deformation cause stress changes that may induce seismicity on upper plate crustal faults far from the coseismic rupture area. In this study, we analyze seismic swarms that occurred in the north Ecuador area of Esmeraldas, beginning two months after the 2016 Mw 7.8 Pedernales, Ecuador megathrust earthquake. The Esmeraldas region is 70 km from the Pedernales rupture area in a separate segment of the subduction zone. We characterize the Esmeraldas sequence, relocating the events using manual arrival time picks and a local a-priori 3D velocity model. The earthquake locations from the Esmeraldas sequence outline an upper plate fault or shear zone. The sequence contains one major swarm and several smaller swarms. Moment tensor solutions of several events include normal and strike-slip motion and non-double-couple components. During the main swarm, earthquake hypocenters increase in distance from the first event over time, at a rate of a few hundred meters per day, consistent with fluid diffusion. Events with similar waveforms occur within the sequence, and a transient is seen in time series of nearby GPS stations concurrent with the seismicity. The events with similar waveforms and the transient in GPS time series suggest that slow aseismic slip took place along a crustal normal fault during the sequence. Coulomb stress calculations show a positive Coulomb stress change in the Esmeraldas region, consistent with seismicity being triggered by the Pedernales mainshock and large aftershocks. The characteristics of the seismicity indicate that postseismic deformation involving fluid flow and slow slip activated upper plate faults in the Esmeraldas area. These findings suggest the need for further investigation into the seismic hazard potential of shallow upper plate faults and the potential for megathrust earthquakes to trigger slow-slip and shallow seismicity across separate segments of subduction zones.

Beekeeping features in the Cameroon Adamawa grasslands

Socioeconomic and technical factors influencing beekeeping were assessed in Mayo Banyo Division, Cameroon (Adamawa Highlands agroecological zone) from April and July 2015. A total of 113 beefarmers were surveyed using semi structural questionnaire, interviews and direct observations. Bees and beehive products were also described using classical features. Results showed that beekeepers were exclusively men, with 100% of respondents. They are between 30-50 years and married for more than 80% and all Muslim in Mayo Banyo. Less than 87% respondents have attended formal school. Nevertheless majority of them have been keeping bees for at least 15 years with main production objective being market and also own consumption. Bee species encountered in colonies is Apis mellifera adansonii, the African common bee, generally yellowish (in more than 89% colonies). The production system is predominantlyextensive, fixed hives being the main used, with number varying from 1 to 800 in the survey. Hives are placed on trees, at 1m at least above the soil. Bee wax is used as the main swarm attractant. Major beehives’ products are: honey (5 to 20 litres/hive) which is harvested yearly once in Mayo Banyo, wax, and rarely propolis and pollen. The main constraints identified are lack of training, colony absconding and pathologies, made up of predators and particularly hives’ moth. Beekeeping in Mayo Banyo division has socioeconomic and environmental potentials for communities’ livelihoods, though serious actions for improvement are required.

A modified symbiotic organisms search algorithm for unmanned combat aerial vehicle route planning problem

Route planning is the core component of unmanned combat aerial vehicle (UCAV) systems and the premise for implementation of airborne reconnaissance, surveillance, combat, and other tasks. The purpose is to find the optimal flight route under certain constraints, and its essence is a multi-constrained global optimisation problem. This paper presents a modified symbiotic organisms search algorithm based on the simplex method (SMSOS) to solve the UCAV path planning problem. In addition to the flight environment for the fixed threat area, this paper tested the flight environment of the randomly generated threat area because of the complexity of the actual battlefield threat area. After many simulation tests, it was concluded that SMSOS can find the shortest flight path while avoiding the threat areas. The experimental results show that SMSOS has faster convergence speed, higher precision, and stronger robustness than the other main swarm intelligence algorithms for solving the UCAV flight planning problem.

Mechanical interaction between volcanic systems in Libya

The spatial distributions of monogenetic volcanoes, primarily volcanic craters, within the four principal volcanic provinces of Libya are examined and presented on a volcano-density map. Six main volcanic clusters have been identified, referred to as volcanic systems. Remarkably, the Al Haruj (AHVP) and Nuqay (NVP) volcanic provinces have double-peak volcano-density distributions, while the Gharyan (GVP) and As Sawda (SVP) volcanic provinces have single-peak volcano-density distributions. We interpret each volcano-density peak as corresponding to a separate volcanic system, so that there is a total of six systems in these four provinces. There was an overlap in volcanic activity in these provinces with at least three simultaneously active. We propose that each of the 6 volcanic systems was/is supplied with magma from a large sill-like reservoir – similar in lateral dimensions to the systems/clusters themselves. Numerical results show zones of high tensile and shear stresses between the reservoirs that coincide roughly with the main swarms of extension (dykes and volcanic fissures) and shear (faults) fractures in the areas. The most recent volcanic eruptions in Libya fall within the modelled high-stress concentration zones, primarily eruptions in the volcano Waw an Namus and the Holocene Al Mashaqaq lava flow. There are no known eruptions in Libya in historical time, but some or all the volcanic systems may have had one or more arrested historical dyke injections. In particular, part of the recurrent seismic events in the Hun Graben in the northwest Libya may be related to dyke propagation and arrest. If some of the inferred magma reservoirs are still fluid, as is likely, they pose earthquake and volcanic hazards to parts of Libya, particularly to the city of Gharyan and Zallah town, as well as to many oil-field operations.

Wastewater disposal and earthquake swarm activity at the southern end of the Central Valley, California

AbstractFracture and fault zones can channel fluid flow and transmit injection‐induced pore pressure changes over large distances (&gt;km), at which seismicity is rarely suspected to be human induced. We use seismicity analysis and hydrogeological models to examine the role of seismically active faults in inducing earthquakes. We analyze a potentially injection‐induced earthquake swarm with three events above M4 near the White Wolf fault (WWF). The swarm deviates from classic main aftershock behavior, exhibiting uncharacteristically low Gutenberg‐Richter b of 0.6, and systematic migration patterns. Some smaller events occurred southeast of the WWF in an area of several disposal wells, one of which became active just 5 months before the main swarm activity. Hydrogeological modeling revealed that wastewater disposal likely contributed to seismicity via localized pressure increase along a seismically active fault. Our results suggest that induced seismicity may remain undetected in California without detailed analysis of local geologic setting, seismicity, and fluid diffusion.

A Mobile Sensor Network Tracking Moving Targets in a Dynamic Environment

This paper presents a novel approach to control a mobile sensor network to track moving targets in a dynamic environment. In this approach, we solve two main issues: the sensor splitting and merging control when the number of the targets is changed. The subgroup merging into the main group is controlled by the invariable attractive force field of the virtual leader in the main swarm. In other words, under the effect of this attractive force field, free-sensors can easily reach the main group and become new members in formation of this group. In contrast, subgroups will be split from the existing group to track new targets that appear. The splitting algorithm is built based on the geometry between targets and group position. The members in a split subgroup will connect with their neighbors in order to generate a robust formation without collision while tracking their target. In addition, when a target disappears, the sensors that are tracking this target will automatically distribute into the nearest existing subgroup. The effectiveness of this proposed approach has been verified in simulations.

Precursory characteristics of the seismicity before the 6 August 2012 eruption of Tongariro volcano, North Island, New Zealand

The 6 August 2012 eruption from the Upper Te Maari crater of Tongariro volcano followed approximately three weeks of precursory seismic activity. Earthquake relocations including data from extra temporary stations indicated that nearly all events were in a small area very close to Upper Te Maari. Most of these relocated events were very shallow, with nearly all events being between 1000 and 1500m below the ground surface.The pre-eruption seismicity occurred in three main swarms. During the first swarm on 12–13 July 2012, all the earthquakes had consistent inter-event times of 71±8min, while in the later swarms (17–20 and 29–30 July) many events had a similar pattern of consistent inter-event times. The stationary quasi-periodic (“clockwork”) earthquake process suggests that a single fracture point was excited by a nearly constant rate flux process.The dominant type of earthquake observed in these swarms had a sharp onset and a broad spectrum, with strong energy from 2 to 10Hz. Most events seen had a local magnitude of 1.5 to 2.5, with virtually no smaller events. Most of these earthquakes appeared to belong to a main earthquake family whose characteristics included a strong spectral component at about 2Hz and three bursts of energy spaced at intervals of about 1.5s. Of the 116 located earthquakes, 75 had a correlation coefficient greater than 0.70 with a master event. The spectra of these events did not change with size, with matching frequency peaks for all the events with a high correlation. The last event of this type was the day before the 6 August 2012 eruption, none have been seen since and there has been very little seismicity under Tongariro.This seismicity alerted GNS Science and other organisations to the unrest of Tongariro, and the Volcanic Alert Level and Aviation Colour Code were raised to publicise this. GNS Science also increased its monitoring of Tongariro, and discovered that the magmatic gas concentrations had increased compared to previous measurements in May 2012. However, the seismicity did not show any accelerating trend that suggested an immediate eruption threat, indicating the difficulty of predicting small eruptions in Tongariro and similar volcanoes.

Structure of regional dykes and local cone sheets in the Midhyrna-Lysuskard area, Snaefellsnes Peninsula (NW Iceland)

This paper provides the first detailed structural description of 48 vertical dykes, 384 inclined sheets and two large intrusions and the geometry (strike, dip direction and dip) of 1116 fractures in the central area of the Snaefellsnes peninsula, NW Iceland. Our data show a more complex setting than that depicted by the WNW-ESE en-echelon trend of the volcanic structures at the surface. In the Miocene basement lavas, dykes dominantly strike N50–100°E whereas other directions are also present with a higher dispersion. Two main swarms of centrally dipping sheets have also been recognized, focussing towards two areas. Sheet dips range from 2 to 75° with the higher frequency between 10 and 45°. In section view, there is no systematic variation of sheet dip with distance from the focus area. Gabbro and granophyre laccoliths are present in the studied area but cross-cutting relations indicate that most of the inclined sheets are younger. Comparison with regional tectonics suggests that the N50–80°E-striking dykes are coherent with emplacement under the stress field of the pre-6 Ma Snaefellsnes Rift dominated by a NNW-SSE-directed least principal stress (σ3). The N80–100°E dykes and the late Quaternary WNW-trending sub-aerial volcanic features are instead consistent with the development of a more recent E-W, right-lateral shear zone affecting the Snaefellsnes peninsula. Coherent sets of fractures have also been found. Within the inclined sheet swarms, the stress tensor rotated in response to an excess magma pressure linked to two underlying magma chambers of lobate shape, located at an estimated depth of about 400 and 500 m below sea level. This local magmatic stress also produced the centrally inclined fracture swarms that have been found in this area.

A competitive clustering particle swarm optimizer for dynamic optimization problems

Optimization in dynamic optimization problems (DOPs) requires the optimization algorithms not only to locate, but also to continuously track the moving optima. Particle swarm optimization (PSO) is a population-based optimization algorithm, originally developed for static problems. Recently, several researchers have proposed variants of PSO for optimization in DOPs. This paper presents a novel multi-swarm PSO algorithm, namely competitive clustering PSO (CCPSO), designed specially for DOPs. Employing a multi-stage clustering procedure, CCPSO splits the particles of the main swarm over a number of sub-swarms based on the particles positions and on their objective function values. The algorithm automatically adjusts the number of sub-swarms and the corresponding region of each sub-swarm. In addition to the sub-swarms, there is also a group of free particles that explore the environment to locate new emerging optima or exploit the current optima which are not followed by any sub-swarm. The adaptive search strategy adopted by the sub-swarms improves both the exploitation and tracking characteristics of CCPSO. A set of experiments is conducted to study the behavior of the proposed algorithm in different DOPs and to provide guidelines for setting the algorithm’s parameters in different problems. The results of CCPSO on a variety of moving peaks benchmark (MPB) functions are compared with those of several state-of-the-art PSO algorithms, indicating the efficiency of the proposed model.

Clustering algorithm based on density function and nichePSO

This paper introduces niching particle swarm optimization (nichePSO) into clustering analysis and puts forward a clustering algorithm which uses nichePSO to optimize density functions. Firstly, this paper improves main swarm training models and increases their ability of space searching. Secondly, the radius of sub-swarms is defined adaptively according to the actual clustering problem, which can be useful for the niches' forming and searching. At last, a novel method that distributes samples to the corresponding cluster is proposed. Numerical results illustrate that this algorithm based on the density function and nichePSO could cluster unbalanced density datasets into the correct clusters automatically and accurately.

A Hybrid Algorithm Based on ACO and PSO for Capacitated Vehicle Routing Problems

The vehicle routing problem (VRP) is a well‐known combinatorial optimization problem. It has been studied for several decades because finding effective vehicle routes is an important issue of logistic management. This paper proposes a new hybrid algorithm based on two main swarm intelligence (SI) approaches, ant colony optimization (ACO) and particle swarm optimization (PSO), for solving capacitated vehicle routing problems (CVRPs). In the proposed algorithm, each artificial ant, like a particle in PSO, is allowed to memorize the best solution ever found. After solution construction, only elite ants can update pheromone according to their own best‐so‐far solutions. Moreover, a pheromone disturbance method is embedded into the ACO framework to overcome the problem of pheromone stagnation. Two sets of benchmark problems were selected to test the performance of the proposed algorithm. The computational results show that the proposed algorithm performs well in comparison with existing swarm intelligence approaches.

A seismic swarm and regional hydrothermal and hydrologic perturbations: The northern Endeavour segment, February 2005

The February 2005 swarm at the overlapping spreading center (OSC) on the northern end of the Endeavour segment is the first swarm on the Juan de Fuca Ridge recorded on a local seafloor seismic network. The swarm included several larger earthquakes and caused triggered seismicity and a hydrothermal response in the Endeavour vent fields as well as regional‐scale hydrologic pressure perturbations. The spatial and temporal pattern of over 6000 earthquakes recorded during this seismic sequence is complex. Small‐magnitude events dominate, and seismicity rates wax and wane, indicating a magmatic process. The main swarm initiates at the northern end of the Endeavour ridge. However, most of the moment release, including six strike‐slip events, occurs in the southwest Endeavour Valley, where the swarm epicenters generally migrate south. The main swarm is contemporaneous with a hydrologic pressure response at four sealed seafloor boreholes, ∼25–105 km away. We infer that the seismic sequence is driven by a largely aseismic magma intrusion at the northern Endeavour axis. Resulting stress changes trigger slip on tectonic faults and possibly dike propagation at the opposing limb of the Endeavour OSC in the southwest Endeavour Valley, consistent with the eventual decapitation of the Endeavour by the West Valley segment. Furthermore, 2.5 days after the start of the main swarm, seismicity is triggered beneath the Endeavour vent fields, and temperature increases at a diffuse vent in the Mothra field. We infer that this delayed response is due to a hydrologic pressure pulse that diffuses away from the main magma intrusion.

Temporal and spatial history of the 1999–2000 Endeavour Segment seismic series, Juan de Fuca Ridge

Two large clusters of earthquake activity in June of 1999 and January of 2000 have dominated recent seismicity along the Endeavour Segment of the Juan de Fuca Ridge. The impacts of the June 1999 sequence on the hydrologic system, which include changes in vent temperature and chemistry within the Main Endeavour Vent Field, have been well documented previously. Analysis of seismic and hydroacoustic data indicates that both sequences exhibit a swarm‐like behavior, characterized by the absence of a dominant main shock event. The epicentral locations of events within the two swarms overlap spatially, with centroid positions near 47°49′ and 47°46′N latitude. During the June 1999 swarm, the initial activity spans the along‐axis region where a shallow axial magma chamber reflector was later imaged. The epicenters then migrate ∼12 km to the south at a rate of 0.3 m/s, consistent with lateral dike propagation. A distinct subcluster of events also occurred in the vicinity of Surveyor Volcano on the overlapping portion of the Cobb Segment. Given its distance from the main swarm, this activity may represent a triggered response to dynamic shaking. The January 2000 swarm has a more limited along‐axis extent, relative to the June 1999 swarm, with no indication of lateral migration. Much of this activity is concentrated in a region predicted to have undergone extension due to dike propagation in 1999. Although it contains fewer total events and is of shorter duration, relative to the June 1999 swarm, the January 2000 activity exhibits a higher peak rate of seismicity and greater mean event magnitude. As in situ temperature monitoring was not in place during January 2000 and vent fluids were not sampled until June 2000, the impacts of this swarm on the hydrothermal system are unknown. The southernmost tip of the Endeavour Segment also is found to be a region of repeating swarm activity. Although morphologic evidence indicates the Cobb Segment has been propagating northward recently, this seismic activity suggests that the western limb of the Endeavour‐Cobb overlap zone remains active.