TIFF file (247K)Principal Investigator: George Lake, University of Washington
(206) 543-7106, lake@hermes.astro.washington.edu
Credits: The simulations are by George Lake, Thomas Quinn, Neal Katz, Joachim Stadel, Ben Moore, University of Washington. The visualizations use the Theoretical Image Processing System (TIPSY) written by Katz and Quinn.
TIFF file (247K)
Galaxy Harassment
Through comparing supercomputer simulations and Hubble Space Telescope (HST) observations, Lake's Grand Challenge team has discovered a phenomenon they call galaxy harassment. As galaxies in clusters harass, or bombard, one another, some lose a large portion of their masses. The graphic depicts the smoothed surface brightness of the stellar tidal debris from a harassed galaxy after 4 billion years of evolution. The image is 6.5 million light years across, and the intensity of the color shows the logarithm of the smoothed stellar surface density plotted wherever the surface brightness is less than 30 magnitudes per square arcsecond. The white dots are individual particles from the galaxy's dark halo. The two long tails were stripped by strong encounters with other cluster members coupled with the mean tidal field of the cluster. At various positions along the orbital path, the stripped stars arrange themselves into long arcs following the galaxy's orbit. HST images should show the brightest features (where the color is brighter than orange).
TIFF file (803K)
The Current Configuration (two-panel comparison frame)
We live on the edge of a cluster of galaxies in the constellation Virgo. We use the dynamics of this cluster to try to answer basic cosmological questions such as the age of the universe and its ultimate fate: will it expand forever or recollapse?
The two panels show two simulations of a cluster of galaxies like Virgo. The 67 thousand-particle simulation was run on a CRAY C90. The 1.3 million-particle model was run on the KSR parallel supercomputer at the Cornell Theory Center. The higher resolution shows how rich the structure is even at this early time. In this particular model of galaxy formation, most structure forms after the universe is roughly 20-30 percent of its present age. However, in the high-density environments like clusters, our simulations show significant structure when the universe is only 4 percent of its current age. The panels shown here are at 20 percent of the current age of the universe and show a wealth of structure on galactic scales.
This simulation will be used to examine the final density profiles of the clusters and the dark halos of galaxies. We will examine the dynamics from multiple angles to see how our view of our own cluster might be biased. The mutual interactions of galaxies will be followed to see the effects of merging in low collisions and destruction by high-velocity passages.
We will resimulate this region adding gas dynamics to better follow the evolution of galaxies. Numerous surveys examine much larger slices of the universe. We would like to expand both resolution and scale to keep pace with observational surveys. This undertaking will require a dedicated teraFLOPS facility or the shared use of a petaFLOPS machine.
Principal Investigator: Richard Muntz, University of California, Los Angeles
(310) 825-3546, muntz@cs.ucla.edu
TIFF file (283K)
Density of Cyclopresence
Credits: E. Mesrobian, R. R. Muntz, E. C. Shek, J. R. Santos, J. Yi, K. Ng, S.-Y. Chien, Data Mining Laboratory, Computer Science Dept., UCLA; C. R. Mechoso, J. D. Farrara, Department of Atmospheric Sciences, UCLA; H. Nakamura, Dept. of Earth and Planetary Physics, University of Tokyo; P. Stolorz, Jet Propulsion Laboratory, California Institute of Technology
Cyclones are areas of minimum sea level pressure, hundreds of kilometers in size, that are the generators of most of the weather. The figures are density maps of cyclopresence during the northern spring (1985-1989) from datasets generated by two Atmospheric General Circulation Models (AGCMs) that are a part of the Atmospheric Model Intercomparison Project (AMIP): UCLA AGCM (upper figure) and ECMWF (European Center for Medium-range Weather Forecasts) AGCM (lower figure). Black represents the lowest value, while red (except for the continent outlines) represents the highest value. In both figures, most extratropical cyclones are formed and migrate within a few zonally elongated regions (i.e., "stormtracks") in the northern Atlantic and Pacific and around Antarctica. The extraction of cyclones and the computation of cyclopresence was performed using the CONQUEST Parallel Query Execution Environment running on a 24-node IBM SP-2.
Principal Investigator: Richard Rood, NASA/Goddard Space Flight Center
Contact: Peter Lyster, (301) 805-6960, lys@dao.gsfc.nasa.gov
TIFF file (611K)
Kalman Filter
Credits: P.M. Lyster, S.E. Cohn, R. Menard, L.P. Chang, S.J. Lin, R. Olsen, NASA/Goddard Space Flight Center
The goal of Four-Dimensional (space and time) Data Assimilation (4DDA) is to incorporate actual observations (satellite, ship, land surface, balloon) into mathematical and computational models in order to create a unified, complete description of the atmosphere. The Kalman filter uses the principles of Estimation Theory to formulate a consistent approach to 4DDA. Ultimately not only are accurate datasets produced, but strongly bound error bars are calculated. The latter are important to scientists who use the data, and who need to know how reliable those data are.
This image is a plot of N2O (nitrous oxide) on a latitude-longitude grid at about 35 kilometers in altitude (850 degrees kelvin isentropic surface). It was produced using a Kalman filter, which gives an estimate of the error in the analysis, something not done in traditional mapping schemes. The N2O is plotted as colored contours. The relative error (%) is plotted as lined black contours. The filter was run on an Intel Paragon supercomputer.
Principal Investigator: Robert Rosner, University of Chicago
Contact: Andrea Malagoli, (312) 702-0624, malagoli@liturchi.uchicago.edu
TIFF file (427K)
Convective Penetration in Stellar Interiors
Credits: Andrea Malagoli, Fausto Cattaneo, Anshu Dubey, University of Chicago
Temperature fluctuations (dark: cool; light: hot) in a layer of convectively unstable gas (upper half) overlying a convectively stable layer (lower half) within the deep interior of a Sun-like star. This simulation was carried out on the Argonne National Laboratory's IBM SP-1.
Principal Investigator: Max Suarez, NASA/Goddard Space Flight Center
(301) 286-7373, suarez@maxs.gsfc.nasa.gov
TIFF file (371K)
Large-Scale Cloudiness in a Global Climate Model
Credits: Max Suarez, Stephen Maher, Cathy Jones, NASA/Goddard Space Flight Center
By looking at the data through NASA/Goddard's virtual reality (VR) system, it became apparent that shallow clouds (large-scale) in the lower levels were maintained in several clumps off the western continental boundaries. The VR system showed that these clouds were maintained because of sea surface temperatures along the western coasts, not because of errors in the model's topography, as previously thought. The clouds are appropriate given the temperatures of the model, and they move north and south correctly with seasonal temperature fluctuations.
In the image, the aqua-colored objects represent the large-scale clouds, and the yellow/orange surface coloring represents the sea surface temperature. The clouds off the west coast of South America (Chile, etc.) are examples of these "permanent" clouds. The simulation date of this image is December 1987.
The investigators used the Scientific Visualization Laboratory's Silicon Graphics Onyx graphics supercomputer and Fakespace BOOM 3C+ (Binocular Omni-Orientation Monitor).
(From "Enabling Science," Space Data & Computing Division, July 1995, by Judy Laue, Hughes STX Corp.)
Principal Investigator: Wojciech Zurek, Los Alamos National Laboratory
Contact: Michael Warren, (505) 665-5023, msw@t6-serv.lanl.gov
TIFF file (997K)
Comparison of Four Popular Cosmological Models
Credits: Michael Warren, Wojciech Zurek, Los Alamos National Laboratory; John Salmon, Caltech; Peter Quinn, Australian National University
This figure shows a comparison of four popular cosmological models. We have simulated each model using 16.7 million particles. The upper right shows the standard Cold Dark Matter (CDM) model, while the upper left shows the same model with a lower normalization. The lower left shows a model using a cosmological constant and only 1/5 as much mass as standard CDM. The lower right shows the effect of replacing 30 percent of the dark matter with neutrinos.
Guest Computational Investigator: Marcus G. Pandy, The University of Texas at Austin, (512) 471-1273, pandy@mail.utexas.edu
TIFF file (611K)
Musculoskeletal Modeling Dynamic Simulations
Credits: Modeling-Marcus G. Pandy, Frank C. Anderson, Brian A. Garner, The University of Texas at Austin. Graphics-Brian A. Garner.
The ability to simulate human movement and accurately compute musculoskeletal loading histories is important to the space program, where exposure to different loading environments or gravitational fields can alter the morphology, biochemistry, and functional properties of muscle and bone tissue. The combination of optimal control theory and mathematical modeling has emerged as a powerful tool for determining musculoskeletal forces during human movement. With the emergence of high-speed parallel supercomputers together with the availability of fast, efficient computational algorithms, high-dimension dynamical models can be used to accurately simulate human movement and to determine musculoskeletal loading patterns during daily physical activity.
We have developed a very detailed three-dimensional, 23-degree-of-freedom, 54-muscle model of the human body and implemented it on Intel iPSC/860 and Thinking Machines CM-5 supercomputers. Each muscle in the model is represented by a three-element entity in series with tendon. Parameters describing the mechanical properties of each muscle were obtained from the literature and scaled to the strength of individual subjects.
Guest Computational Investigator: Andrew Pohorille, NASA/Ames Research Center
Contact: Michael Wilson, (415) 604-5496, mwilson@max.arc.nasa.gov
TIFF file (694K)
Alanine Dipeptide at a Water-Membrane Interface
Credits: Andrew Pohorille, Michael Wilson, NASA/Ames Research Center
Simple cell-like structures called vesicles may have been the first self-replicating systems in the world-marking the beginnings of life. Shown here is an alanine dipeptide at the interface between a glycerol-1-monooleate (GMO) membrane and water. In water, short peptides are usually random and do not form the ordered structures necessary to catalyze the life-sustaining reactions as proteins do in modern cells. However, Pohorille and Wilson's work demonstrates that short peptides can assume ordered structures if located at the interface between water and membrane. The oily GMO tails are blue, the polar GMO head groups are magenta, and the oxygen and hydrogen atoms of water molecules are red and white, respectively. In the alanine dipeptide, the carbon atoms are green, the nitrogen atoms are yellow, the oxygen atoms are green, and the hydrogen atoms are gray. Also available is a four-image sequence that depicts a sodium (Na+) ion crossing a GMO membrane.
TIFF file (1,735K)
The Rayleigh-Taylor Instability
Credit: Bruce Fryxell, George Mason University and NASA/Goddard Space Flight Center, (301) 286-8567, fryxell@neutrino.gsfc.nasa.gov
In the Rayleigh-Taylor Instability, a denser fluid forms fingers that drop into a lighter fluid below it, which in turn bubbles upwards. This phenomenon occurs in situations ranging from a supernova to laser fusion to coffee with cream. The graphic depicts a two-dimensional simulation run on NASA/Goddard's 16,384-processor MasPar MP-2, which took 60 hours and performed at 4 gigaFLOPS (billion floating-point operations per second). The model employed the PROMETHEUS code, an implementation of the Piecewise Parabolic Method that solves Euler's equations for compressible gas dynamics.
TIFF file (436K)
Particle Simulation of an Interacting Ring Galaxy
Credit: Kevin Olson, George Mason University and NASA/Goddard Space Flight Center, (301) 286-8707, olson@jeans.gsfc.nasa.gov
The images show before-and-after frames from a simulation of a disk galaxy interacting with a smaller galaxy. The smaller galaxy passes through the disk of the larger galaxy, causing ring structures to form. A particle technique is used for the simulation: an N-body tree code combined with smooth particle hydrodynamics. The dark blue particles represent the stellar component of the disk, and the green particles represent the gaseous component of the disk. Particles not shown here are also used to represent the bulge component of the galaxy and the dark matter halo. A total of 131,072 particles are used. This simulation was performed on the MasPar MP-2 at NASA/Goddard. Detailed comparisons will be made of these models and observations from the Hubble Space Telescope (see http://www.stsci.edu/pubinfo/gif/Cartwheel.gif) and other telescopes.
