Search NASA⌕ Search

SEARCH · Search NASA

Results for “data clustering”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 757 records · Page 42

On the merging cluster Abell 578 and its central radio galaxy 4C+67.13

Here we analyze radio, optical, and X-ray data for the peculiar cluster Abell 578. This cluster is not fully relaxed and consists of two merging sub-systems. The brightest cluster galaxy (BCG), CGPG 0719.8+6704, is a pair of interacting ellipticals with projected separation ∼10 kpc, the brighter of which hosts the radio source 4C+67.13. The Fanaroff–Riley type-II radio morphology of 4C+67.13 is unusual for central radio galaxies in local Abell clusters. Our new optical spectroscopy revealed that both nuclei of the CGPG 0719.8+6704 pair are active, albeit at low accretion rates corresponding to the Eddington ratio ∼10{sup −4} (for the estimated black hole masses of ∼3×10{sup 8} M{sub ⊙} and ∼10{sup 9} M{sub ⊙}). The gathered X-ray (Chandra) data allowed us to confirm and to quantify robustly the previously noted elongation of the gaseous atmosphere in the dominant sub-cluster, as well as a large spatial offset (∼60 kpc projected) between the position of the BCG and the cluster center inferred from the modeling of the X-ray surface brightness distribution. Detailed analysis of the brightness profiles and temperature revealed also that the cluster gas in the vicinity of 4C+67.13 is compressed (by a factor of about ∼1.4) and heated (from ≃2.0 keV up to 2.7 keV), consistent with the presence of a weak shock (Mach number ∼1.3) driven by the expanding jet cocoon. This would then require the jet kinetic power of the order of ∼10{sup 45} erg s{sup −1}, implying either a very high efficiency of the jet production for the current accretion rate, or a highly modulated jet/accretion activity in the system.

79 ASTRONOMY AND ASTROPHYSICS↗

DID THE INFANT R136 AND NGC 3603 CLUSTERS UNDERGO RESIDUAL GAS EXPULSION?

Based on kinematic data observed for very young, massive clusters that appear to be in dynamical equilibrium, it has recently been argued that such young systems are examples of where the early residual gas expulsion did not happen or had no dynamical effect. The intriguing scenario of a star cluster forming through a single starburst has thereby been challenged. Choosing the case of the R136 cluster of the Large Magellanic Cloud, the most cited one in this context, we perform direct N-body computations that mimic the early evolution of this cluster including the gas-removal phase (on a thermal timescale). Our calculations show that under plausible initial conditions which are consistent with observational data, a large fraction (>60%) of a gas-expelled, expanding R136-like cluster is bound to regain dynamical equilibrium by its current age. Therefore, the recent measurements of velocity dispersion in the inner regions of R136, which indicate that the cluster is in dynamical equilibrium, are consistent with an earlier substantial gas expulsion of R136 followed by a rapid re-virialization (in Almost-Equal-To 1 Myr). Additionally, we find that the less massive Galactic NGC 3603 Young Cluster (NYC), with a substantially longer re-virialization time, is likely to be found to have deviated from dynamical equilibrium at its present age ( Almost-Equal-To 1 Myr). The recently obtained stellar proper motions in the central part of the NYC indeed suggest this and are consistent with the computed models. This work significantly extends previous models of the Orion Nebula Cluster which already demonstrated that the re-virialization time of young post-gas-expulsion clusters decreases with increasing pre-expulsion density.

79 ASTRONOMY AND ASTROPHYSICS↗

Nonnegative tensor decomposition with custom clustering for microphase separation of block copolymers

High–dimensional datasets are becoming ubiquitous in many applications and therefore unsupervised tensor methods to interrogate them are needed. Here, we report a new unsupervised machine learning (ML) approach (NTFk) based on nonnegative tensor factorization integrated with a custom k–means clustering. We demonstrate the ability of NTFk to extracting temporal and spatial features of phase separation of copolymers as they are modeled by self–consistent field theory. Microphase separation of block copolymers has been extensively studied both experimentally and theoretically. However, the interpretation of computer simulations and/or experimental data, representing temporal and spatial changes of molecular species concentration is still a challenging task. Thus, extracting the phase diagram from simulations or experimental data as well as the interpretation of data requires discernment of the model/experimental parameters (such as, temperature, concentrations, the number of molecular species and the interaction between species) impact on the microphase separation process. An attractive and unique aspect of the introduced ML method is that it ensures the nonnegativity of the extracted latent features. Nonnegativity is an essential constraint needed to obtain interpretable and sparse latent features that are parts–based representation of the data. The custom clustering in NTFk serves to estimate the number of latent features in the data.

97 MATHEMATICS AND COMPUTING↗

Exploring Memory Hierarchy to Improve Scientific Data Read Performance, In: 2015 IEEE International Conference on Cluster Computing

Improving read performance is one of the major challenges with speeding up scientific data analytic applications. Utilizing the memory hierarchy is one major line of researches to address the read performance bottleneck. Related methods usually combine solide-state-drives(SSDs) with dynamic random-access memory(DRAM) and/or parallel file system(PFS) to mitigate the speed and space gap between DRAM and PFS. However, these methods are unable to handle key performance issues plaguing SSDs, namely read contention that may cause up to 50% performance reduction. In this paper, we propose a framework that exploits the memory hierarchy resource to address the read contention issues involved with SSDs. The framework employs a general purpose online read algorithm that able to detect and utilize memory hierarchy resource to relieve the problem. To maintain a near optimal operating environment for SSDs, the framework is able to orchastrate data chunks across different memory layers to facilitate the read algorithm. Compared to existing tools, our framework achieves up to 50% read performance improvement when tested on datasets from real-world scientific simulations.

Computer Science↗

Anomaly detection for MPC forecast in Fleet of Water Heaters

Among residential devices, water heaters consume 20% of home energy use in the United States. Water heaters possess the capability to store energy within their reservoirs, enabling the ability to decouple energy use from hot water use. This capability can be used to reduce energy usage and costs while also supporting grid services. This requires accurate forecasting of the parameters of the water heater such as upper and lower temperatures. In this study, we analyzed the performance and behavior of a water heater model used in the real-world to predict a control mechanism that is implemented in a smart residential neighborhood. The model forecasts are accurate in most cases but not all. In such scenarios, error correction of the model is necessary to further improve model predictive control accuracy. Anomaly detection is the first step of error correction. This study complements existing research by grouping time series data into two clusters one with anomalies and another without anomalies. To achieve this task, we explored and compared multiple unsupervised machine learning algorithms to perform clustering. Among these algorithms, Ward clustering has the lowest running time and identified the highest number of anomalies for the upper temperature limit. The proposed approach is tested based on the data collected in a neighborhood with 46 townhomes located in Atlanta, GA.

Lebakula, Viswadeep↗

Toward IMRT 2D dose modeling using artificial neural networks: A feasibility study

Purpose: To investigate the feasibility of artificial neural networks (ANN) to reconstruct dose maps for intensity modulated radiation treatment (IMRT) fields compared with those of the treatment planning system (TPS). Methods: An artificial feed forward neural network and the back-propagation learning algorithm have been used to replicate dose calculations of IMRT fields obtained from PINNACLE{sup 3} v9.0. The ANN was trained with fluence and dose maps of IMRT fields for 6 MV x-rays, which were obtained from the amorphous silicon (a-Si) electronic portal imaging device of Novalis TX. Those fluence distributions were imported to the TPS and the dose maps were calculated on the horizontal midpoint plane of a water equivalent homogeneous cylindrical virtual phantom. Each exported 2D dose distribution from the TPS was classified into two clusters of high and low dose regions, respectively, based on the K-means algorithm and the Euclidian metric in the fluence-dose domain. The data of each cluster were divided into two sets for the training and validation phase of the ANN, respectively. After the completion of the ANN training phase, 2D dose maps were reconstructed by the ANN and isodose distributions were created. The dose maps reconstructed by ANN were evaluated and compared with the TPS, where the mean absolute deviation of the dose and the {gamma}-index were used. Results: A good agreement between the doses calculated from the TPS and the trained ANN was achieved. In particular, an average relative dosimetric difference of 4.6% and an average {gamma}-index passing rate of 93% were obtained for low dose regions, and a dosimetric difference of 2.3% and an average {gamma}-index passing rate of 97% for high dose region. Conclusions: An artificial neural network has been developed to convert fluence maps to corresponding dose maps. The feasibility and potential of an artificial neural network to replicate complex convolution kernels in the TPS for IMRT dose calculations have been demonstrated.

61 RADIATION PROTECTION AND DOSIMETRY↗

A polymer, random walk model for the size-distribution of large DNA fragments after high linear energy transfer radiation

DNA double-strand breaks (DSBs) produced by densely ionizing radiation are not located randomly in the genome: recent data indicate DSB clustering along chromosomes. Stochastic DSB clustering at large scales, from > 100 Mbp down to < 0.01 Mbp, is modeled using computer simulations and analytic equations. A random-walk, coarse-grained polymer model for chromatin is combined with a simple track structure model in Monte Carlo software called DNAbreak and is applied to data on alpha-particle irradiation of V-79 cells. The chromatin model neglects molecular details but systematically incorporates an increase in average spatial separation between two DNA loci as the number of base-pairs between the loci increases. Fragment-size distributions obtained using DNAbreak match data on large fragments about as well as distributions previously obtained with a less mechanistic approach. Dose-response relations, linear at small doses of high linear energy transfer (LET) radiation, are obtained. They are found to be non-linear when the dose becomes so large that there is a significant probability of overlapping or close juxtaposition, along one chromosome, for different DSB clusters from different tracks. The non-linearity is more evident for large fragments than for small. The DNAbreak results furnish an example of the RLC (randomly located clusters) analytic formalism, which generalizes the broken-stick fragment-size distribution of the random-breakage model that is often applied to low-LET data.

DNA/radiation effects↗

Applications of Fuzzy Set Theory to Satellite Soundings

The introduction of an appropriate fuzzy setting for satellite soundings and its application to clustering methods via unimodal fuzzy sets in the future is proposed. Methods of hard clustering analysis and fuzzy partitioned clustering were applied on simulated data with very encouraging results. The proposed clustering technique is discussed. The notion of a unimodal fuzzy set was chosen to represent the partition of a data set for two reasons: (1) it detects all the locations in the vector space where highly concentrated clusters of points exist; and (2) the notion is general enough to represent clusters that exhibit quite general distributions of points. The technique detects all of the existing unimodal fuzzy sets and realizes the maximum separation among them. It is economical in memory space and computational time requirements and also detects groups that are fairly generally distributed in the feature space.

Munteanu, M. J.↗

THE SUPERNOVA DELAY TIME DISTRIBUTION IN GALAXY CLUSTERS AND IMPLICATIONS FOR TYPE-Ia PROGENITORS AND METAL ENRICHMENT

Knowledge of the supernova (SN) delay time distribution (DTD)-the SN rate versus time that would follow a hypothetical brief burst of star formation-can shed light on SN progenitors and physics, as well as on the timescales of chemical enrichment in different environments. We compile recent measurements of the Type-Ia SN (SN Ia) rate in galaxy clusters at redshifts from z = 0 out to z = 1.45, just 2 Gyr after cluster star formation at z {approx} 3. We review the plausible range for the observed total iron-to-stellar mass ratio in clusters, based on the latest data and analyses, and use it to constrain the time-integrated number of SN Ia events in clusters. With these data, we recover the DTD of SNe Ia in cluster environments. The DTD is sharply peaked at the shortest time-delay interval we probe, 0Gyr < t < 2.2 Gyr, with a low tail out to delays of {approx}10 Gyr, and is remarkably consistent with several recent DTD reconstructions based on different methods, applied to different environments. We test DTD models from the literature, requiring that they simultaneously reproduce the observed cluster SN rates and the observed iron-to-stellar mass ratios. A parameterized power-law DTD of the form t {sup -1.2{+-}0.3} from t = 400 Myr to a Hubble time can satisfy both constraints. Shallower power laws such as t {sup -1/2} cannot, assuming a single DTD, and a single star formation burst (either brief or extended) at high z. This implies that 50%-85% of SNe Ia explode within 1 Gyr of star formation. DTDs from double-degenerate (DD) models, which generically have {approx}t {sup -1} shapes over a wide range of timescales, match the data, but only if their predictions are scaled up by factors of 5-10. Single-degenerate (SD) DTDs always give poor fits to the data, due to a lack of delayed SNe and overall low numbers of SNe. The observations can also be reproduced with a combination of two SN Ia populations-a prompt SD population of SNe Ia that explodes within a few Gyr of star formation, and produces about 60% of the iron mass in clusters, and a DD population that contributes the events seen at z < 1.5. An alternative scenario of a single, prompt, SN Ia population, but a composite star formation history in clusters, consisting of a burst at high z, followed by a constant star formation rate, can reproduce the SN rates, but is at odds with direct measurements of star formation in clusters at 0 < z < 1. Our results support the existence of a DD progenitor channel for SNe Ia, if the overall predicted numbers can be suitably increased.

79 ASTRONOMY AND ASTROPHYSICS↗

A cluster analysis method for identification of subpopulations of cells in flow cytometric list-mode arrays

A specialized program was developed for flow cytometric list-mode data using an heirarchical tree method for identifying and enumerating individual subpopulations, the method of principal components for a two-dimensional display of 6-parameter data array, and a standard sorting algorithm for characterizing subpopulations. The program was tested against a published data set subjected to cluster analysis and experimental data sets from controlled flow cytometry experiments using a Coulter Electronics EPICS V Cell Sorter. A version of the program in compiled BASIC is usable on a 16-bit microcomputer with the MS-DOS operating system. It is specialized for 6 parameters and up to 20,000 cells. Its two-dimensional display of Euclidean distances reveals clusters clearly, as does its 1-dimensional display. The identified subpopulations can, in suitable experiments, be related to functional subpopulations of cells.

Li, Z. K.↗

Chemisorption of CO and Mechanism of CO Oxidation on Supported Platinum Nanoclusters

Kinetic, isotopic, and infrared studies on well-defined dispersed Pt clusters are combined here with first-principle theoretical methods on model cluster surfaces to probe the mechanism and structural requirements for CO oxidation catalysis at conditions typical of its industrial practice. CO oxidation turnover rates and the dynamics and thermodynamics of adsorption-desorption processes on cluster surfaces saturated with chemisorbed CO were measured on 1-20 nm Pt clusters under conditions of strict kinetic control. Turnover rates are proportional to O2 pressure and inversely proportional to CO pressure, consistent with kinetically relevant irreversible O2 activation steps on vacant sites present within saturated CO monolayers. These conclusions are consistent with the lack of isotopic scrambling in C16O-18O2-16O2 reactions, and with infrared bands for chemisorbed CO that did not change within a CO pressure range that strongly influenced CO oxidation turnover rates. Density functional theory estimates of rate and equilibrium constants show that the kinetically relevantO2 activation steps involve directO2* (or O2) reactions with CO* to form reactive O*-O-C*dO intermediates that decompose to form CO2 and chemisorbed O*, instead of unassisted activation steps involving molecular adsorption and subsequent dissociation of O2. These CO-assisted O2 dissociation pathways avoid the higher barriers imposed by the spin-forbidden transitions required for unassistedO2 dissociation on surfaces saturated with chemisorbed CO. Measured rate parameters for CO oxidation were independent of Pt cluster size; these parameters depend on the ratio of rate constants for O2 reactions with CO* and CO adsorption equilibrium constants, which reflect the respective activation barriers and reaction enthalpies for these two steps. Infrared spectra during isotopic displacement and thermal desorption with 12CO-13CO mixtures showed that the binding, dynamics, and thermodynamics of CO chemisorbed at saturation coverages do not depend on Pt cluster size in a range that strongly affects the coordination of Pt atoms exposed at cluster surfaces. These data and their theoretical and mechanistic interpretations indicate that the remarkable structure insensitivity observed for CO oxidation reactions reflects average CO binding properties that are essentially independent of cluster size. Theoretical estimates of rate and equilibriumconstants for surface reactions and CO adsorption show that both parameters increase as the coordination of exposed Pt atoms decreases in Pt201 cluster surfaces; such compensation dampens but does not eliminate coordination and cluster size effects on measured rate constants. The structural features and intrinsic non-uniformity of cluster surfaces weaken when CO forms saturated monolayers on such surfaces, apparently because surfaces and adsorbates restructure to balance CO surface binding and CO-CO interaction energies.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

Elucidating the Roles of Local and Nonlocal Rate Enhancement Mechanisms in Plasmonic Catalysis

Plasmonic metal nanoparticles (e.g., Ag, Au, and Cu) constitute a class of materials that interact with light via the excitation of localized surface plasmon resonance (LSPR). Numerous studies have reported substantial enhancements in the rates of chemical reactions on illuminated plasmonic nanoparticle catalysts compared to corresponding systems in the absence of illumination. There are two mechanisms that have been proposed to explain the LSPR-induced chemical reactivity. One mechanism assumes a local plasmon-induced hot charge-carrier-mediated activation of the reactants, while the other assumes an LSPR-induced equilibrium heating of the catalyst, which leads to energy transfer to and chemical reaction of the adsorbed reactants. Here, in this contribution, we developed a setup amenable to accurate in situ catalyst temperature and kinetic reaction rate measurements. We employed this setup to study the LSPR-induced rate enhancement in a case study of the CO oxidation reaction on plasmonic, monometallic Ag nanoparticle catalysts supported on α-Al 2 O 3 . We explored various Ag loadings and clustering levels. Our data show that the equilibrium heating of the catalyst cannot fully explain the illumination-induced plasmonic rate enhancements. This is the case even for high loading and clustering of Ag nanoparticles, where the equilibrium heating significantly increases. Based on the analysis, we propose that local effects, related to the plasmon-induced activation of adsorbates (reactants) via electronic excitation of the reactant or photothermal heating of the reactants that is highly localized to the individual nanoparticles, play a critical role in driving LSPR-induced chemical reactions.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗

THE IMACS CLUSTER BUILDING SURVEY. III. THE STAR FORMATION HISTORIES OF FIELD GALAXIES

Using data from the IMACS Cluster Building Survey and from nearby galaxy surveys, we examine the evolution of the rate of star formation in field galaxies from z = 0.60 to the present. Fitting the luminosity function to a standard Schechter form, we find a rapid evolution of M{sub B}{sup *} consistent with that found in other deep surveys; at the present epoch M{sub B}{sup *} is evolving at the rate of 0.38 Gyr{sup -1}, several times faster than the predictions of simple models for the evolution of old, coeval galaxies. The evolution of the distribution of specific star formation rates (SSFRs) is also too rapid to explain by such models. We demonstrate that starbursts cannot, even in principle, explain the evolution of the SSFR distribution. However, the rapid evolution of both M{sub B}{sup *} and the SSFR distribution can be explained if some fraction of galaxies have star formation rates characterized by both short rise and fall times and by an epoch of peak star formation more recent than the majority of galaxies. Although galaxies of every stellar mass up to 1.4 Multiplication-Sign 10{sup 11} M{sub Sun} show a range of epochs of peak star formation, the fraction of ''younger'' galaxies falls from about 40% at a mass of 4 Multiplication-Sign 10{sup 10} M{sub Sun} to zero at a mass of 1.4 Multiplication-Sign 10{sup 11} M{sub Sun }. The incidence of younger galaxies appears to be insensitive to the density of the local environment; but does depend on group membership: relatively isolated galaxies are much more likely to be young than are group members.

79 ASTRONOMY AND ASTROPHYSICS↗

Scalable deep text comprehension for Cancer surveillance on high-performance computing

Background: Deep Learning (DL) has advanced the state-of-the-art capabilities in bioinformatics applications which has resulted in trends of increasingly sophisticated and computationally demanding models trained by larger and larger data sets. This vastly increased computational demand challenges the feasibility of conducting cutting-edge research. One solution is to distribute the vast computational workload across multiple computing cluster nodes with data parallelism algorithms. In this study, we used a High-Performance Computing environment and implemented the Downpour Stochastic Gradient Descent algorithm for data parallelism to train a Convolutional Neural Network (CNN) for the natural language processing task of information extraction from a massive dataset of cancer pathology reports. We evaluated the scalability improvements using data parallelism training and the Titan supercomputer at Oak Ridge Leadership Computing Facility. To evaluate scalability, we used different numbers of worker nodes and performed a set of experiments comparing the effects of different training batch sizes and optimizer functions.Results: We found that Adadelta would consistently converge at a lower validation loss, though requiring over twice as many training epochs as the fastest converging optimizer, RMSProp. The Adam optimizer consistently achieved a close 2nd place minimum validation loss significantly faster; using a batch size of 16 and 32 allowed the network to converge in only 4.5 training epochs.Conclusions: We demonstrated that the networked training process is scalable across multiple compute nodes communicating with message passing interface while achieving higher classification accuracy compared to a traditional machine learning algorithm.

Qiu, John X.↗

Variable Blue Straggler Stars in the Open Cluster NGC 6819 Observed in the Kepler “Superstamp” Field

Abstract NGC 6819 is an open cluster of age 2.4 Gyr that was in the NASA Kepler spacecraft’s field of view from 2009 to 2013. The central part of the cluster was observed in a 200 × 200 pixel “superstamp” during these four years in 30 minute cadence photometry, providing a unique, high-precision, long time-series data set. The cluster contains “blue straggler” stars, i.e., stars on the main sequence above the cluster turnoff that should have left the main sequence to become red giants. We present light curves and pulsation frequency analyses derived from custom photometric reductions for five confirmed cluster members—four blue stragglers and one star near the main-sequence turnoff. Two of these stars show a rich spectrum of δ Scuti pulsation modes, with 236 and 124 significant frequencies identified, respectively, while two stars show mainly low-frequency modes, characteristic of γ Doradus variable stars. The fifth star, a known active X-ray binary, shows only several harmonics of two main frequencies. For the two δ Scuti stars, we use a frequency separation–mean density relation to estimate their mean densities, and then use these values along with their effective temperature to derive their stellar masses and radii. For the two stars showing low frequencies, we searched for period-spacing sequences that may be representative of gravity-mode or Rossby-mode sequences, but found no clear sequences. The common age for the cluster members, considered along with the frequencies, will provide valuable constraints for asteroseismic analyses, and may shed light on the origin of the blue stragglers.

79 ASTRONOMY AND ASTROPHYSICS↗

Optimizing tertiary storage organization and access for spatio-temporal datasets

We address in this paper data management techniques for efficiently retrieving requested subsets of large datasets stored on mass storage devices. This problem represents a major bottleneck that can negate the benefits of fast networks, because the time to access a subset from a large dataset stored on a mass storage system is much greater that the time to transmit that subset over a network. This paper focuses on very large spatial and temporal datasets generated by simulation programs in the area of climate modeling, but the techniques developed can be applied to other applications that deal with large multidimensional datasets. The main requirement we have addressed is the efficient access of subsets of information contained within much larger datasets, for the purpose of analysis and interactive visualization. We have developed data partitioning techniques that partition datasets into 'clusters' based on analysis of data access patterns and storage device characteristics. The goal is to minimize the number of clusters read from mass storage systems when subsets are requested. We emphasize in this paper proposed enhancements to current storage server protocols to permit control over physical placement of data on storage devices. We also discuss in some detail the aspects of the interface between the application programs and the mass storage system, as well as a workbench to help scientists to design the best reorganization of a dataset for anticipated access patterns.

Chen, Ling Tony↗

The Double Galaxy Cluster A2465. III. X-Ray and Weak-lensing Observations

We report Chandra X-ray observations and optical weak-lensing measurements from Subaru/Suprime-Cam images of the double galaxy cluster A2465 ( z = 0.245). The X-ray brightness data are fit to a β model to obtain the radial gas density profiles of the northeast (NE) and southwest (SW) subcomponents, which are seen to differ in structure. We determine core radii, central temperatures, the gas masses within r {sub 500c}, and the total masses for the broader NE and sharper SW components assuming hydrostatic equilibrium. There is no large X-ray excess between the two components. The central entropy of the NE subcluster is about two times higher than the SW. Along with its structural properties and an apparent radio halo that is a sign of a merger, this suggests that the NE component has undergone merging on its own. The weak-lensing analysis gives virial masses for each substructure, which compare well with earlier dynamical results. The derived outer mass contours of the SW sub-component from weak lensing are more irregular and extended than those of the NE. Although there is a weak enhancement and small offsets between X-ray gas and mass centers from weak lensing, the lack of large amounts of gas between the two subclusters indicates that A2465 is in a pre-merger state. We discuss star formation enhancement in this system resulting from its dynamics and shock-induced star formation scenarios. A dynamical model that is consistent with the observed cluster data, based on the FLASH program and the radial infall model, is constructed, where the subclusters currently separated by ∼1.2 Mpc are approaching each other at ∼2000 km s{sup −1} and will meet in ∼0.4 Gyr.

79 ASTRONOMY AND ASTROPHYSICS↗