Scientific Visualization Gallery

A collection of compelling science images from past and present collaborations between TACC and scientists

Visualization of E3SM global climate model data showing atmospheric, sea ice, and oceanic properties in the Southern Ocean.

Atmospheric, Sea Ice, & Oceanic Properties in the Southern Ocean

Science Behind the Image

In the past 20 years, climate science and computational modeling have been thrust into the national spotlight as climate change has become a pressing global issue. The Sixth Assessment Report by the Intergovernmental Panel on Climate Change (released in August 2021) makes extensive use of climate models to predict the effects of future emissions scenarios. One of those models, the Energy Exascale Earth System Model (E3SM), is featured in this visualization project.

The earth’s climate is complex with many interacting processes. Climate models mirror this complexity but allow researchers to analyze each variable at high resolution in space and time. Evaluation of these large data sets is important for climate science for verification, scientific understanding, and assessment of future emissions scenarios. The E3SM model was developed by researchers at the U.S. Department of Energy. The data in this visualization was generated on Argonne National Laboratory supercomputers by Mark Petersen of the Los Alamos National Laboratory.

Visualization Behind the Image

This visualization and its accompanying video demonstrates the use of the E3SM model to study the impact of polynyas. Polynyas are openings within the polar sea ice pack formed and sustained by atmospheric and oceanic processes. They occur in the Arctic and Southern oceans, last for many months, and function as a conduit for heat and water between the oceanic and atmospheric systems. In this image we see the extent of the Antarctic ice on a particular day in the Antarctic winter with a large polynya in the Weddell Sea. The ocean surface shows latent heat flux — the transfer of heat between the ocean and the atmosphere. Pathlines show surface winds over the 10-day period prior to the timestep. The visualization was created by Francesca Samsel and Greg Abram using the Stampede2 supercomputer at TACC using Paraview and custom code visualization tools.

This visualization was a finalist in the SC21 Visualization Showcase. Credits: Francesca Samel, Greg Abram, Stephanie Zeller at TACC. Mark Petersen, LeAnn Conlon, Prajvala Kurtakoti, Linnea Palstom, John Patchett, Andrew Roberts at Los Alamos National Lab.

Top view of storm anvil showing breaking wave and trailing cirrus plume.

Top View of Storm Anvil Showing Breaking Wave & Trailing Cirrus Plume

Science Behind the Image

A feature called an Above Anvil Cirrus Plume (AACP) has been observed in supercell thunderstorms by aircraft and satellite data and has been the subject of numerical study. The AACP is observed as an elongated elevated “tier” of ice cloud that extends downwind from the overshooting top. This feature is important to atmospheric researchers and forecasters because three-fourths of all supercells exhibiting AACP’s were associated with severe weather (large hail and/or damaging winds/tornadoes). Understanding the specific conditions by which AACP’s form in supercells is therefore of great interest as they have the potential to serve as a useful “nowcasting” device during severe weather events. Simulation credit: Leigh Orf from the University of Wisconsin, Madison. Completed on TACC’s Frontera supercomputer.

Visualization Behind the Image

This visualization used a combination of 3D techniques to visualize the top of the storm. The volume visualization provided a view of the cloud ice which revealed the presence of the AACP. Animations of the cloud like rendering revealed the mechanism of formation of the plume to be a wave like structure similar to that seen in a hydraulic jump. Paraview with OspRAY ray tracing was used to produce the visualization.

This visualization was a finalist in the SC21 Visualization Showcase. Credits: Greg Foss and Dave Semeraro, TACC.

This 3D model of a polymer desalination membrane shows water—channels moving from top to bottom—avoiding dense spots in the membrane and slowing flow.

Desalination Membranes to Maximize Flow, Clean More Water

Science Behind the Image

This 3D model of a polymer desalination membrane shows water—the silver channels moving from top to bottom—avoiding dense spots in the membrane and slowing flow.

Researchers from Iowa State, Penn State, and UT Austin found that creating a uniform membrane density down to the nanoscale of billionths of a meter is crucial for maximizing the performance of reverse-osmosis, water-filtration membranes. Their discovery was published online by the journal Science and was the cover paper of the Jan. 1, 2021, print edition.

Working with Penn State's transmission electron microscope measurements of four different polymer membranes used for water desalination, the Iowa State engineers used TACC supercomputers to predict water flow through 3D models of the membranes, allowing detailed comparative analysis of why some membranes performed better than others.

The researchers concluded that the key to better desalination membranes is figuring out how to measure and control at very small scales the densities of manufactured membranes. Manufacturing engineers and materials scientists need to make the density uniform throughout the membrane, promoting water flow without sacrificing salt removal.

Visualization Behind the Image

The Science cover image was created by Greg Foss, a visualization expert and artist at TACC, in collaboration with the Iowa State researchers. Foss also contributed to the manuscript illustrations. This visualization was generated using ParaView (Kitware, Inc.) and OSPRay (Intel, Inc.) without additional post-processing. This visualization realizes the promise of high-fidelity visualization: putting "cover quality" rendering in the tools used for immediate analysis, providing a single environment for exploratory analysis through final production render (i.e. no need to post-process in Maya, Blender, etc). The image was created running on TACC's Frontera and Stampede2 HPC systems.

Credit: Baskar Ganapathysubramanian Research Group/Iowa State University; Greg Foss/Texas Advanced Computing Center (TACC).

Atomically-Thin Layer of Hexagonal Boron Nitride

Science Behind the Image

A single atomically-thin layer of hexagonal boron nitride shows that thinner is better with regards to radio-frequency switching applications.

Hexagonal boron nitride (hBN) has a large bandgap, high phonon energies, and an atomically smooth surface absent of dangling bonds. As a result, it has been widely used as a dielectric to investigate electron physics in two-dimensional heterostructures and as a dielectric in the fabrication of two-dimensional transistors and optoelectronic devices.

Here we show that hBN can be used to create analogue switches for applications in communication systems across radio, 5G, and terahertz frequencies. Our approach relies on the non-volatile resistive switching capabilities of atomically thin hBN. The switches are composed of monolayer hBN sandwiched between two gold electrodes and exhibit a cutoff-frequency figure of merit of around 129 THz with a low insertion loss (≤0.5 dB) and high isolation (≥10 dB) from 0.1 to 200 GHz, as well as a high power handling (around 20 dBm) and nanosecond switching speeds, metrics that are superior to those of existing solid-state switches. Furthermore, the switches are 50 times more efficient than other non-volatile switches in terms of a d.c. energy-consumption metric, which is an important consideration for ubiquitous mobile systems. We also illustrate the potential of the hBN switches in a communication system with an 8.5 Gbit s–1 data transmission rate at 100 GHz with a low bit error rate under 10−10.

Visualization Behind the Image

The assembly was rendered using Autodesk Maya with some textures created with photography and edited in Photoshop. Some effects were created in post using Adobe Photoshop.

Visualization: Jo Wozniak, Texas Advanced Computing Center.

an almost-spherical cloud of cool blue and white fire

Volume Rendering of a Cosmological Simulation of the Universe

Science Behind the Image

This is an image from a cosmological simulation of the Universe performed by the Enzo multi-physics hydrodynamics code. Enzo is used to study the physical processes present in the evolution of the cosmos. To learn more, read "Ray Tracing the Cosmos" by Texascale.

Visualization Behind the Image

The visualization was performed using a modified version of the yt open source data analysis and visualization package. The yt-project grew out of the Enzo community to satisfy a need to visualize and analyze the large data resulting from the Enzo simulations. Collaborators at TACC and the National Center for Supercomputing Applications (NCSA) modified the yt framework to utilize a ray tracing back end called GraviT to perform volume rendering of the adaptive mesh refinement data. This image shows a volume rendering of density in which the color and opacity are mapped to the density distribution.

The visualization was performed by Dave Semeraro of TACC and Matthew Turk of NCSA.

Aggregate and monomers in equilibrium in a saturated aqueous solution

Aggregate and Monomers in Saturated Aqueous Solution

Science Behind the Image

Numerous membrane-less organelles (i.e. dense, liquid-like droplets) exist in eukaryotic cells and they afford spatial and temporal modes of cellular regulation through the functional organization or assembly of DNA, RNA, and protein constituents. Pathological changes in the properties of these liquid-like droplets towards more static, fibrous aggregates are thought to underlie the progression of neurodegenerative diseases. A fundamental understanding of the biophysical mechanisms that drive ‘normal', or non-pathological, phase separation is a critical first step towards understanding how they are altered in disease states. We have been particularly interested in the physicochemical properties of the protein backbone (i.e. the chemical scaffold common to all proteins) and the role these properties play in protein-mediated phase separation. Molecular dynamics simulations of a model protein backbone system in a super-saturated solution allows us to probe or investigate these properties at atomic resolution. The accompanying figure is a snapshot from such a simulation performed on Stampede2 and captures the approach of free polypeptides as they phase separate from solution to form liquid-like droplets.

Visualization Behind the Image

The image scene was staged using VMD and rendered using the Tachyon Ray Tracer with ambient light occlusion. A semi-transparent surface representation encapsulates the liquid-like droplet and the individual protein backbone model components can be seen within as van Der Waals spheres colored by atom types.

Authors of feature review article: Justin Drake, Texas Advanced Computing Center; Monte Pettitt, UT Medical Branch Galveston. Visualization: Anne Bowen and Justin Drake, Texas Advanced Computing Center. Link to article: https://pubs.acs.org/toc/jpcbfk/124/22

Spherically expanding turbulent flames in absence of gravity

Science Behind the Image

The image shows instantaneous flame surface at three different times (from left to right) as the flame expands radially outwards. Flame surface is coloured with the net local flame stretch (K, the normalized rate of growth of surface area) . Positive values (orange) indicate locally expanding surface and negative values (blue) indicate surface destruction due to curvature.

Visualization Behind the Image

Produced using Paraview.

Science: Fabrizio Bisetti, Tejas U. Kulkarni, The University of Texas at Austin Visualization: Greg Foss, Texas Advanced Computing Center

Arctic Ocean research - Ice interactions

Science Behind the Image

The image shows a daily average temperature field from a numerical model simulation of warm sub-surface water of subtropical origin carried with the Norwegian Atlantic boundary current into the Arctic Ocean, northwestward along the West Spitsbergen current and northeastward through Barents Sea opening. The view is northeastward, from Iceland toward the Fram Strait and the eastern Arctic, with Svalbard at the center, Greenland to the left, and Eurasia to the upper right. The horizontal extent covers 1320 x 1080 grid points, which corresponds to physical dimensions of ~5000 km x 3500 km. 33 vertical levels comprise the depth range between 0-1400 m. Temperature ranges from 0 (blue) to 6ºC (red). High performance computing is required to capture mesoscale eddies of length-scale less than 15 km shedding off and transporting heat and freshwater toward the ocean interior. Understanding how these currents transport heat into the high Arctic and its surrounding seas will enable the quantification of the impact of recent observed changes in the atmosphere and ocean heat content on Arctic sea ice and ecosystems, as well as Greenland's marine margins. Supported by NSF grants PLR-1603903 and PLR-1708289, and through continuing support from NASA for the "Estimating the Circulation and Climate of the Ocean (ECCO) project.

Visualization Behind the Image

The visualization was produced using ParaView and OSPRay.

Visualization: Greg Foss, Texas Advanced Computing Center Science: Nguyen, An T., Heimbach, P. and Vocaña, V., Institute for Computational Engineering and Sciences, The University of Texas at Austin

Water flowing through limestone

Science Behind the Image

This image shows the path of water through the karst limestone structure of a ground sample taken from south Florida. The photorealistic shading enhances the analysis of the stone porosity and the spatial arrangement of the flow traces.

Visualization Behind the Image

This image was generated using a ray-tracing framework supported in part by NSF grants ACI 13-39863 and ACI 1134872, and the image was generated on TACC Stampede, an NSF-funded cyberinfrastructure resource. Learn more about ray-tracing at TACC.

Data courtesy of Michael Sukop and Sade Garcia, Florida International University and Kevin Cunningham, U.S. Geological Survey. Visualization: Carson Brownlee, Aaron Knoll and Paul Navratil, Texas Advanced Computing Center

Molecular Dynamics Analysis of Ion Distributions Around a DNA strand

Science Behind the Image

The charged ions that surround DNA molecules can dramatically effect its conformation, folding, and dynamics. This visualization is from a Molecular Dynamics simulation of a DNA system where the electron density of the surrounding ions was critical to understanding how they might modulate the structure and interactions of the DNA double helix.

Visualization Behind the Image

This visualization was produced as part of an ECSS (https://www.xsede.org/for-users/ecss) collaboration where TACC HPC researcher Antonio Gomez assisted Daniel Roe to significantly optimize the parallelization and parallel IO of the trajectory analysis CCPTRAJ subroutine of the Molecular Dynamics package AMBER. AMBER is one of the most widely used Molecular Dynamics packages at TACC, and this optimization effort reduced the time for analyses that required the CCPTRAJ subroutine from days to minutes. The image is a single snapshot from a 20000 time-step trajectory output from AMBER on stampede, and was rendered using the VMD Tachyon renderer.

Researcher: Dan Roe of the University of Utah. AMBER/CPPTRAJ code optimizations by Antonio Gomez (TACC) Visualization by Anne Bowen (TACC)

Physical Signatures of Cancer Metastasis

Science Behind the Image

Metastasis is the development of secondary malignant growths at a distance from a primary site of cancer. In order to study physical signatures of metastasis, Abdul Malmi Kakkada, Xin Li, Himadri Samanta (post-docs in Dave Thirumalai's group - Department of Chemistry at the UT Austin) and Sumit Sinha (Graduate Student in Thirumalai group) have established a model for the proliferation behavior of tumor cells.This model is used to provide detailed analyses of individual cell trajectories and identify complex spatial/time dependent cell migration patterns. This provides insights into how cells are "prepared" for invasion into areas surrounding the tumor. The visualization shown depicts the velocity of each cell (colored by the magnitude of the velocity with the arrows representing the direction). In the tumor cross-section, it is clear to see that faster moving cells are concentrated at the outer periphery of the tumor. Arrows indicating the velocity direction show that cells in the periphery tend to move farther away from the center of the tumor as opposed to cells closer to the center of the tumor whose direction of motion is essentially isotropic. This prediction agrees well with the experiments, which showed that cells at the periphery of the tumor spheroid move persistently along the radial direction, resulting in polarized tumor growth.

Visualization Behind the Image

The models and simulations are done in MATLAB on the Stampede supercomputer at the Texas Advanced Computing Center. Most of the simulations require 1 node and 2 days of compute time to complete. In order to improve the visual fidelity of the 3D cell propagation data, the simulation data was visualized in Paraview 5.3 using the OSPRay renderer by Anne Bowen at the Texas Advanced Computing Center.

Science: Abdul Malmi Kakkada (Dave Thirumalai's group - Department of Chemistry at UT Austin) Visualization: Anne Bowen, Texas Advanced Computing Center.

Liquid K–Na Alloy Anode Enables Dendrite‐Free Potassium Batteries

Science Behind the Image

Schematic of making solid‐metal anodes into liquid‐alloy anode to suppress dendrite formation. Dendrites form on Na or K solid surface on charge and grow across the separator to the cathode to give an internal short‐circuit. In contrast, the mixing of Na and K results in a room‐temperature K–Na liquid alloy which allows a dendrite‐free large‐capacity battery. The immiscibility of the liquid K–Na in liquid organic electrolytes offers the possibility of a liquid–liquid anode–electrolyte interface. The three layers top‐to‐bottom represent cathode, electrolyte, and anode.

Visualization Behind the Image

Visualization: Jo Wozniak, TACC Science: Leigang Xue, Hongcai Gao, Weidong Zhou, Sen Xin, Kyusung Park, Yutao Li, John B. Goodenough

Tokamak Simulation: Toroidal Plasma Currents

Science Behind the Image

A tokamak's toroidal machine vessel is collapsed into a rectangular solid, front to back axis along the periodic. Radio frequency (RF) waves in MegaAmpere toroidal plasma currents are launched to confine solar temperature plasmas, and can be studied with visualization. Left shows the RF electric potential, right is electron density. White contours at the front define magnetic surfaces confining the core (top) and exhaust chamber (bottom) plasmas.

Visualization Behind the Image

Visualization: Gregg Foss, TACC Science: Wendell Horton, Lee Leonard, Institute for Fusion Studies, UT Austin

Updraft in a Hypothetical Supercell Simulation

Science Behind the Image

Streamlines follow Wind Velocity and identify a supercell's signature powerful updraft. Red shows highest magnitude values, and gold surfaces high levels of Vertical Vorticity. The grid divides the ground into 12 kilometer squares, and the sum of Ice and Mixed Cloud Water Ratio provide a translucent reference to the storm's volume roughly 17400 meters tall.

Visualization Behind the Image

Visualization: Greg Foss, Greg Abram, TACC Science: Amy McGovern, University of Oklahoma; Corey Potvin, University of Oklahoma, NOAA/National Severe Storms Laboratory

Plan View: Updraft in a Hypothetical Supercell Simulation

Science Behind the Image

An orange surface clipped at 12,000 meters elevation shows 50 m/s vertical Wind Velocity, grey is strong Vorticity Magnitude. Using a simplified weather map color key, the Radar Reflectivity background shows precipitation intensity at 500 meters above ground. The hook echo pattern at lower left often reveals severe weather.

Visualization Behind the Image

Visualization: Greg Foss, Greg Abram, TACC Science: Amy McGovern, University of Oklahoma; Corey Potvin, University of Oklahoma, NOAA/National Severe Storms Laboratory

Plasmonic Metasurfaces

Science Behind the Image

Graphene moiré metasurfaces are a new type of atomically thin metasurfaces. Consisting of a large number of graphene nanostructures of variable shapes and sizes in quasiperiodic arrays, the graphene moiré metasurfaces exhibit multiple resonance modes in mid-infrared regime. The tunable graphene moiré metasurfaces are fabricated by cost-effective and scalable moiré nanosphere lithography.

Visualization Behind the Image

Image was created as a submission for a journal cover of Advanced Optical Materials (it was selected as the inside cover) and was created to show moire pattern etching in graphene as well as highlight the hex graphene structure. Maya, Photoshop and Illustrator were used to create the composite.

Visualization: Jo Wozniak, TACC Science: Zilong Wu, Wei Liv, Maruthi Nagavalli Yogeesh, Seungyong Jung, Alvin Lynghi Lee, Kyle McNicholas, Andrew Briggs, Seth R. Bank, Mikhail A. Belkin, Deji Akinwande, Yuebing Zheng, UT Austin

Energy-efficient Electronics

Science Behind the Image

There has been a constant effort in the semiconductor industry to strive for energy-efficient electronic devices at ultra-scaled dimensions. This image demonstrates electronic devices based on van der Waals heterostructures of novel atomically thin two-dimensional (2D) semiconductors, such as molybdenum disulphide (MoS2) and molybdenum ditelluride (MoTe2), capable of such ultra-low-power operation. The possibility of realizing two separate device concepts using the same heterostructure device platform adds an element of uniqueness to this project.

Visualization Behind the Image

Image was created using Autodesk Maya and Adobe Photoshop based on an orthogonal front-view paper sketch by the researchers. Initially every component was in block form, followed by atomic structures in lieu of the original blocks.

Visualization: Jo Wozniak, TACC Science: Amritesh Rai and Maruthi Nagavalli Yogeesh, UT Austin

Cosmic Collision

Science Behind the Image

Visualization of a two black holes colliding, and the resulting generated gravitational waves using GR-Chombo simulation for the February 2015 announced event detection by the LIGO experiment.

Visualization Behind the Image

Created using ParaView with OSPRay.

Visualization: Carson Brownlee, Intel GR-Chombo simulation data: Pau Figueras, Markus Kunesch, Saran Tunyasuvunakool, Juha Jäykkä. Stephen Hawking Centre for Theoretical Cosmology.

Electronic tattoo

Science Behind the Image

Tattoo-like epidermal sensors are an emerging class of truly wearable electronics, owing to their thinness and softness. While most of them are based on thin metal films, a silicon membrane, or nanoparticle-based printable inks, we report submicrometer thick, multimodal electronic tattoo sensors that are made of graphene. The graphene electronic tattoo (GET) is designed as filamentary serpentines and fabricated by a cost- and time-effective "wet transfer, dry patterning" method. It has a total thickness of 463 ± 30 nm, an optical transparency of ∼85%, and a stretchability of more than 40%. The GET can be directly laminated on human skin just like a temporary tattoo and can fully conform to the microscopic morphology of the surface of skin via just van der Waals forces.

Visualization Behind the Image

Conceptual image created using Autodesk Maya and Adobe Illustrator and Photoshop to show how the sensor pattern is separated from the substrate after printing, as well as the flexibility of the tattoo.

Research team: Shideh Kabiri Ameri, Rebecca Ho, Hongwoo Jang, Li Tao,Youhua Wang,∥ Liu Wang, David M. Schnyer, Deji Akinwande, and Nanshu Lu. Visualization: Jo Wozniak, Texas Advanced Computing Center

Asteroid Impact

Science Behind the Image

Shown here are two time-steps of a simulation of a 250m asteroid impacting the ocean surface. Water vapor arising from the impact is shown in light blue, asteroid fragments in orange.

Visualization Behind the Image

Volume rendering combined with calibrated opacity levels, enables visualization of the structures of both the water vapor and asteroid fragments. Enabling scientists to see multiple volumetric structures facilitates understanding of the interaction of the physical materials.

Visualization: Francesca Samsel, Greg Abram, UT Austin Science: G.Gisler, J. Patchett, LANL

Hurricane Ike

Science Behind the Image

Throughout the 2008 hurricane season, the Texas Advanced Computing Center was an active participant in a NOAA research effort to develop next-generation hurricane models. Teams of scientists relied on TACC's Ranger supercomputer to test high-resolution ensemble hurricane models, and to track evacuation routes from data streams on the ground and from space.

Visualization Behind the Image

Using up to 40,000 processing cores at once, researchers simulated both global and regional weather models and received on-demand access to some of the most powerful hardware in the world enabling real-time, high-resolution ensemble simulations of the storm. This visualization of Hurricane Ike shows the storm developing in the gulf and making landfall on the Texas coast.

Science: Fuqing Zhang and Yonghui Weng, Pennsylvania State University; Frank Marks, NOAA; Visualization: Gregory P. Johnson, Romy Schneider, John Cazes, Karl Schulz, Bill Barth, The University of Texas at Austin

Going Nano: Big Innovations at Small Scales

Science Behind the Image

For the next big breakthroughs, scientists are thinking small – really small. Nanotechnology offers enormous new opportunities for innovation in materials science, electronics, medicine and more. Computational resources are critical to advancing this burgeoning field. At the University of Texas at Austin, researchers are applying nanotechnology to develop new uses for molybdenum disulphide, a compound used as a lubricant and in petroleum refining. This image is a close-up view of multilayered molybdenum disulphide stacked in a diamond anvil cell. Running electric leads through the structure, researchers can trigger electronic transitions and manipulate the mechanical, electrical and optical properties of this innovative layered nanomaterial.

Visualization Behind the Image

The image has been rendered with the perspective of one of the atoms inside the environment. The complex structure of the stacked assembly is visible in the highly reflective surface of the molecules.

Research Team: Avinash P. Nayak, The University of Texas at Austin; Swastibrata Bhattacharyya, Indian Institute of Science; Jie Zhu, The University of Texas at Austin; Jin Liu, The University of Texas at Austin; Xiang Wu, The University of Texas at Austin; Tribhuwan Pandey, Indian Institute of Science; Changqing Jin, Chinese Academy of Sciences; Abhishek K. Singh, Indian Institute of Science; Deji Akinwande, The University of Texas at Austin; Jung-Fu Lin, The University of Texas at Austin; Illustration: Jo Wozniak, Texas Advanced Computing Center (TACC)

Plasma Turbulence

Science Behind the Image

These computer generated graphics are visualizations of data from a simulation of plasma turbulence in Earth's ionosphere. These same physics are also applied to the research team's investigations of turbulence in the tokamak, a device used in nuclear fusion experiments.

Visualization Behind the Image

Wendell Horton and Lee Leonard use TACC's Stampede computing system to run their simulation which produces the data in these visualizations. Some of the visualizations are also produced with Stampede.

Greg Foss, Anne Bowen, Greg Abram,Texas Advanced Computing Center

H1N1 Epidemic

Science Behind the Image

The image shows a visualization of the H1N1 (swine flu) epidemic spreading throughout North America. In this test case, the epidemic begins in Mexico City. The visualization classifies individuals into three groups: susceptible (blue), infected (red), and recovered (green). Available antivirals are shown in purple. Cities and transportation links are highlighted in red to indicate large numbers of infected individuals and infectious travelers.

Visualization Behind the Image

A custom OpenGL application was developed that combines geospatial data with epidemic simulations computed on TACC's Lonestar supercomputer to produce time-series visualization of epidemic spread across North America. Users are able to interactively explore the simulations in real-time, visualizing overall epidemic spread or focusing on spread from a single city.

Lauren Meyers, Ned Dimitrov, Sebastian Goll, Section of Integrative Biology, The University of Texas at Austin. Visualization by Greg P. Johnson, Texas Advanced Computing Center (TACC).

Streamwise Vorticity in a Supercell Thunderstorm

Science Behind the Image

This wild-looking image compares variables from the dataset of a simulated hypothetical thunderstorm that spawned a tornado. The plot shows vortex lines, the gold streamlines modeled from vorticity, compared with wind velocity as grey streamlines with red representing highest speed values. The scene illustrates rotation in the updraft, the hallmark characteristic of a supercell. Featured is a vortex ring at the top of the storm, which forms as the strong updraft punches into the stable stratosphere and the air subsequently curls downward. A translucent volume rendering of cloud water combined with ice provides reference to the surrounding storm structure. In a real storm, you can see the clouds billowing upward and corkscrew striations in the rotating cloud, but the relationship between the wind and rotation isn't exactly clear. Plots like this help to illuminate this relationship giving a better sense of how the vorticity in the environment is tilted, stretched, and intensified in the updraft to make the storm rotate. The image illustrates how vorticity in the environment aligns with the winds feeding into the storm to enhance storm rotation.

Visualization Behind the Image

The visualization illustrates a structure in the dataset that the researcher wasn't aware of — the vortex ring. The meteorologist plans to incorporate this image in an upcoming seminar, and TACC will use the image in venues to promote the NSF XSEDE initiative, HPC, and visualization resources. The visualization was created with VisIt software (Lawrence Livermore National Lab) on Longhorn at the Texas Advanced Computing Center. The simulation was run on Kraken (National Institute for Computational Sciences). More specifically, TACC staff had to translate the netcdf data from the simulation into hdf to incorporate the irregular spaced Z coordinates that the user provided. The 1500x1500x50 grid is regular in x & y and the dataset includes several variables including wind velocity, reflectivity, perturbated potential temperature, cloud water mixing ratio, ice, vorticity, and others. TACC staff built several animations and stills.

Greg Foss, image, Texas Advanced Computing Center (TACC); Brittany Dahl, science content/research, School of Meterology, University of Oklahoma; Amy McGovern, science content/research, School of Computer Science, University of Oklahoma; Greg Abram, software support, Texas Advanced Computing Center (TACC).

Graphene and Electronics and Thermal Management

Science Behind the Image

When graphene is placed on an amorphous substrate like SiO2, thermal conductivity of graphene is strongly suppressed. In this work, the researchers measured how thermal conductivity depends on the number of graphene layers stacked on top of each other. They developed a theoretical model, and found that phonons in graphene hit the graphene-SiO2 interface, and interaction of phonons with interface is the major reason for thickness-dependent thermal conductivity of graphene samples. The result is relevant for the application of graphene for electronics, thermal management, and other applications

Visualization Behind the Image

TACC staff created the scientific illustration in Maya

Mir Mohammad Sadeghia, Insun Joa, and Li Shi, Department of Mechanical Engineering and Texas Materials Institute, The University of Texas at Austin. Edited by Eric Pop, Stanford University. Accepted by the Editorial Board August 26, 2013. Scientific illustration: Jo Wozniak and Greg Foss, Texas Advanced Computing Center (TACC).

TACC SuperComputer Queue Visualization

Science Behind the Image

SupercomputingQueueVis is a Processing sketch based off of Paul Bourke's HPC queue statistics visualization. It converts the CommnQ server output into pairs of spheres and cylinders arranged in a helical pattern that represents the status of the supercomputer's queue.

Visualization Behind the Image

Our goal was to develop a visualization that could give systems administrators and users a complete understanding of the queue's state with nothing more than a quick glance. In our implementation, · Each job is represented by a cluster of same-colored, consecutive spheres · Each sphere is a node · Sphere size is proportional to the number of nodes per job · Each cylinder represents allocated time · Color along cylinder represents time used now using Hanlon's API to access Stampede queue data in realtime

Heri Nieto and Matthew Hanlon, Texas Advanced Computing Center (TACC)

Spinning Star

Science Behind the Image

Solar and stellar observations reveal diverse and vibrant manifestations of magnetic activity, including the dramatic eruptions known as flares and coronal mass ejections. Yet, elucidating the origins of solar and stellar magnetism is a formidable task, requiring an understanding of the subtle nonlinear interplay between turbulent convection and rotation on a vast range of spatial and temporal scales.

Visualization Behind the Image

Facing this complexity requires sophisticated modeling and visualization on high-performance computing platforms. These images were made from a simulation of a rapidly spinning star with mass similar to our sun. The images show a cross section through the turbulent convection zone where rapid rotation distorts the convection patterns, forming elongated flow structures that play an essential role in the generation of the magnetic field. Such structures exhibit intense vorticity, swirling columns of plasma that wrap up and amplify magnetic fields. These 3D views allow scientists to identify and characterize the type of structures that may exist in turbulent stellar interiors, which are important to investigating energy transport. TACC staff built the animation with VisIt on Longhorn and the Ranger supercomputer as a computation resource. Several variables were modeled and combined in different ways over the course of the project to end up with: 1) velocity, 2) radial component of velocity modeled as a scalar colored with enstrophy; and 3) the radial component velocity colored by negative and positive values showing upwelling and downwelling. Users Ben Brown (University of Wisconsin at Madison) and Mark Miesch (University Corporation for Atmospheric Research) are investigating energy transfer.

Research: Ben Brown, University of Wisconsin at Madison; Mark Miesch, University Corporation for Atmospheric Research; Juri Toomre, University of Colorado. Simulation: Anelastic Spherical Harmonic, originally developed at the University of Colorado. Visualization/Software support/Poster Design: Greg Foss, Greg Abram, Jo Wozniak, respectively, Texas Advanced Computing Center (TACC).

Breaking Waves

Science Behind the Image

The simulation of complex free-surface phenomena such as breaking waves, spray sheets, and air entrainment play a key role in the design and operation of naval combatants. As a ship progresses through water, steep breaking waves are formed; spray is shed along the crests of the breaking waves and near the ship's bow where thin sheets of water form. Air is also entrained along the turbulent face of breaking waves and along the contact line where the free surface, which is the interface between the air and water, intersects the hull. These phenomena are among the most challenging problems in computational fluid dynamics today. The image shows the resulting turbulence as water flows past a NACA 0024.

Visualization Behind the Image

The visualization was created using a Virtual Rheoscopic Fluid plot. The plot uses the fluid velocity data to create and orient virtual rheoscopic particles to obtain an image similar to those used in bench experiments. This technique enables novel vector plots in VisIt on a wide variety of data and the visual comparison of the quality of different technique variations.

Simulation by: Doug G. Dommermuth, Science Applications International Corporation. The visualization technique was published in LDAV 2013 by Paul A. Navrátil, William L. Barth, and Hank Childs: Virtual Rheoscopic Fluids for Dense Large-Scale Fluid Flow Visualizations; proceedings of IEEE Symposium on Large Data Analysis and Visualization (LDAV) 2012; and featured in the SC12 visualization showcase.

Drug Delivery Systems to Aid Heart Disease

Science Behind the Image

The vast majority of heart attacks occur when there is a sudden rupture of vulnerable plaques forming clots that cause blockages in coronary arteries. The diseased arteries can be treated with drugs intravascularly injected into these rupture-prone plaques. In designing the local drug delivery devices, important issues regarding drug distribution and targeting need to be addressed to ensure therapeutic efficacy. In this case, a computational toolset was developed to support the design and analysis of a catheter-based local drug delivery system that uses nanoparticles as drug carriers to treat vulnerable plaques and diffuse atherosclerosis. Simulations were run on a 3D patient-specific diseased coronary artery segment obtained directly from CT-imaging data. The visualization depicts a cross-section of the artery taken right through the vulnerable plaque with a large lipid core and a thin fibrous cap that is formed near the coronary artery bifurcation region. Results show the drug (in red) accumulating in the lipid core of the vulnerable plaque, which is highly encouraging from a therapeutic point of view.

Visualization Behind the Image

Using TACC's HPC resources, a 3D mathematical model of coupled transport of drug and drug-encapsulated nanoparticles was developed and solved numerically using isogeometric finite element analysis. The visualization component of this project, a 13-minute animation, explains the motivation of the research, the physical mechanism of the proposed solution concept, the mathematical models used, the computational methodology applied for simulation, and the scientific visualization of the results. This tool is now poised to be used in the medical device industry.

First image: Karla Vega, Texas Advanced Computing Center (TACC). Credits for animation: Ben Urick, Jo Wozniak, Texas Advanced Computing Center (TACC); Shaolie Hossain, Thomas J.R. Hughes, Institute for Computational Engineering and Sciences, The University of Texas at Austin. Erik Zumalt and Juan Diaz, Faculty Innovation Center, The University of Texas at Austin.

Global Seismic Wave Propagation

Science Behind the Image

Modeling propagation of seismic waves through the Earth helps assess seismic hazard at regional scales and aids in interpretation of the Earth's inter structure at global scales. In this simulation of global wave propagation from the 2011 Tohoku earthquake, the Earth is modeled using hexahedral elements that are adaptively sized based on the local length of seismic waves. A discontinuous Galerkin method is used for the numerical solution of the seismic wage propagation partial differential equations.

Visualization Behind the Image

The simulation was run on the TACC Lonestar IV system and produced ~3.4 terabytes of data for 330 time steps. The visualization was run on 32 nodes and 64 GPUs of the TACC Longhorn system.

Simulation by Carsten Burstedde, Omar Ghattas, James R. Martin, Georg Stadler and Lucas C. Wilcox of the Institute for computational Sciences and Engineering, Jackson School of Geosciences and the Department of Mechanical Energy at the University of Texas at Austin. The visualization was created by Greg Abram of the Texas Advanced Computing Center at the University of Texas at Austin. Support from NSF grants OCI-0749334, DMS-0724746 CMMI-102888 (CDI), OCI-0906379 (ETF) and OCU-1042124 and from AFOSR grant FA9550-09-1-0608 (Computational Math)

Large Scale Distributed GPU-Based Visualization Framework

Science Behind the Image

Project Summary: No high-level project summary provided.

Visualization Behind the Image

TACC staff enhanced NVIDIA's CUDA Isosurfacer to perform twice as fast by using one-third of the GPU memory, and enabling efficient overlapping of GPU operations with I/O operations and sort-last image compositing to achieve high throughput, in-core rendering. The enhanced isosurfacer achieves an approximate speedup for 4.5x over CPU-based visualization methods on a 2048^3 scalar volume. The images show enstrophy data, isosurfaced using the modified NVIDIA CUDA isosurfacer. It was created using 64 nodes of Longhorn. It took 2.5 minutes to isosurface 935M triangles, 68G cells, which correspond to 256 GB of data. Additionally, it took 9 seconds to render the images on 128 GPUs. The two frames show the assignment of data to process by color.

Greg Abram, Byungil Jeong, Greg P. Johnson, Paul Navratil, Kelly Gaither, Texas Advanced Computing Center (TACC). Work in collaboration with Diego Donzis, Texas A&M University and P.K. Yeung, Georgia Tech.

Ozone Concentration in East Texas Due to Emissions in Houston

Science Behind the Image

This visualization shows simulations of ozone concentrations in eastern Texas. Emissions in Houston cause ozone concentrations to rise in other parts of the state, particularly in central and north Texas; thus, it is apparent that emissions from one city can affect ozone levels in an entire region. These simulations use various sources of data including satellite observations and land-based measurements.

Visualization Behind the Image

TACC staff developed a custom OpenGL application to visualize the data sets. This application combines geospatial data with the simulations to produce time-series animations of the ozone concentration across Texas. Cutting planes and isosurfaces show the ozone concentration in units of parts-per-billion.

Mariana Dionisio, Elena McDonald-Buller, David Allen, Center for Energy and Environmental Sciences; Visualization by Gregory P. Johnson, TACC.

Visualization of NARA Record Collections

Science Behind the Image

The National Archives and Records Administration (NARA) is responsible for ensuring continuous access to government records. To preserve and provide access to electronic records collections, archivists need to first conduct a series of analysis to discover their structure and content and to make decisions about long-term preservation needs. TACC's research examines information visualization for archival analysis and long-term preservation planning of terabyte size collections.

Visualization Behind the Image

This type of visualization work is intended to help digital archivists at NARA process their Federal Electronic Records collection for public access and long-term preservation. In response, TACC developed a program for NARA that visualized a test-bed collection of approximately 40,000 files. The visualization was rendered by a visual analytic application developed in Java. The application leveraged several existing information visualization packages, including Prefuse, JFreeChart, and OpenCloud. It includes publicly available data provided by Federal Agencies or harvested from their websites. Each record group corresponds to all the records of a small Federal Agency or some of the records of a larger Federal Agency and is represented as a node that includes child nodes. In turn, each record group may have different types of digital objects bearing different arrangements and a variety of file formats. The sample from the test-bed collection contained 1,031,118 files in 200 different formats with up to 12 levels of hierarchical nesting. Each square represents a directory within the file system with larger squares containing a higher number of files. The colors in the first image (casc_filetype) indicate the types and percentage of file formats present in those directories: green = web files; blue = images; coral = pdf files; light blue = video files; and red = word processing files. This view infers that this test-bed collection contains a majority of web pages including photographs. The color black indicates that there are file formats that current software could not identify.
The second picture (casc_datamining) presents the results of a data mining analysis showing that the files are organized by similar naming conventions (light green); in sequential order (orange); by date (brown); or by geographical location (bordeaux).

Maria Esteva, Weijia Xu, Texas Advanced Computing Center (TACC); Suyog D. Jain, Department of Computer Sciences, The University of Texas at Austin. Acknowledgments: This work was supported through a NARA supplement to the National Science Foundation (NSF) Cooperative Agreement: OCI-0504077. Access to the NARA test-bed collections in the Transcontinental Persistent Archives Prototype courtesy of The Center for Advanced Systems and Technologies at NARA.

Building Affinity for Effective Drugs

Science Behind the Image

Increasing the strength of binding between a molecule and a receptor is an important technique in the design of effective drugs. Binding affinity synthesizes molecules that are already in the shape that they will take when bound to a receptor. This technique works because it decreases the binding entropy, which increases the overall binding affinity. The research addresses biological and medical challenges from single molecules to the genome with high-performance computing and theory. In collaboration with other experimental groups, the researchers use computer modeling and simulations to understand these complex biomolecular systems and to discover molecules for treating disease and improving human health. Image Caption 1: Frame from the molecular dynamics (MD) simulation of constrained and unconstrained molecule bound to receptor. These calculations revealed that the origin of the constrained molecule (purple) and unconstrained molecule (pink) have similar conformations when bound to the receptor. Image Caption 2: Frame from the molecular dynamics (MD) simulation of the constrained and unconstrained molecule free in solution. In solution, the unconstrained molecule (pink) is capable of forming more compact structures that allow hydrogen bonds (green dotted line) to be formed with the phosphate group (represented by the yellow and red atoms).

Visualization Behind the Image

Yue Shi of the The Ren lab at The University of Texas at Austin probed the origin of this entropy paradox with molecular dynamics simulations run on the Lonestar and Ranger supercomputers at TACC. TACC staff used VMD to load the molecular dynamics data and to set up the model and the Tachyon Parallel Ray Tracer to render it. Their group used approximately 2 million CPU hours on Ranger and about 1 million CPU hours on Lonestar.

Simulation: Yue Shi, Biomedical Engineering, Ren Lab, The University of Texas at Austin. Visualization: Anne Bowen, Texas Advanced Computing Center (TACC).

Cosmic Web

Science Behind the Image

Science is entering the era of precision cosmology, where it is becoming possible to measure cosmological parameters with less than 1% error. Large numerical simulations are used to model the main physical processes to establish a 'standard ruler' with which to measure universal phenomena. The mysterious Dark Energy component of the Universe is of particular interest, as it accounts for more than 70% of known energy density ― it is responsible for causing cosmic expansion to accelerate, and yet its nature is still completely unknown.

Visualization Behind the Image

The visualization was completed in VisIt using both point plots and direct volume ray casting to show the particle field and particle density. This visualization shows particle density from a 3072^3 (29 billion particle) N‐body simulation of dark matter in a computational volume of (1/h Gpc)^3 (100 billion cubic light years) on a computational grid of 6144^3 cells and force softening length of 16/h kpc (74 thousand light years), evolved to the present epoch (z=0).

lian T. Iliev, Department of Physics and Astronomy, University of Sussex; Paul R. Shapiro, Department of Astronomy, The University of Texas at Austin; Vincent Desjacques and Robert Smith, Institute for Theoretical Physics, University of Zürich, Switzerland; Ue-Li Pen, Canadian Institute for Theoretical Astrophysics, University of Toronto, Canada. Paul A. Navratil, Texas Advanced Computing Center (TACC). The visualization was featured in the TeraGrid 2009 visualization showcase.Advanced Computing Center (TACC). Acknowledgments: Funding provided by the National Science Foundation TeraGrid initiative and a National Science Foundation RAPID Grant, Office of Cyberinfrastructure. Also, the Department of Homeland Security Science & Technology Directorate through the Center of Excellence for Natural Disasters, Coastal Infrastructure and Emergency Management.

BP Deepwater Horizon Oil Spill

Science Behind the Image

The animation depicts wind velocity, water levels, inundation, and passive particle movement obtained from a simulation of past hurricanes computed using the ADCIRC coastal circulation model coupled to the unstructured SWAN wave model. The initial particle distributions vary from simulation to simulation. Some points are hypothetical; others reflect the position of the oil on the surface and vary in date from 06/20/2010 to 06/23/2010 to 06/27/2010. Positions were derived from the NOAA/NESDIS Experimental Marine Pollution Surveillance Daily Composite Product (http://www.ssd.noaa.gov/PS/MPS/deepwater.html). Particles in the animation move with the depth-averaged water velocity and most accurately represent water movement in shallow estuarine, near shore, and continental shelf waters that were strongly mixed during the storm. Particle motion beyond the continental shelf is not reliable. During the simulation, particles do not disburse, stick, or degrade in any way. They may not accurately represent the movement of oil. These results should not be used to forecast the movement of material at the sea surface or in the water column during any future event.

Visualization Behind the Image

The visualization effort by TACC focused on the overlay of particle movement and satellite or aerial imaging data. The particles in the visualization representing the oil spill and their position is either hypothetical or reflects the position of the oil on the surface. The data was visualized using TACC's Longhorn visualization system and MINERVA, an open source geospatial software. The data was generated daily and is approximately 100 GB in size.

Univ. North Carolina at Chapel Hill, Institute of Marine Sciences; Univ. Notre Dame, Computational Hydraulics Laboratory Computational Hydraulics Group, ICES, The University of Texas at Austin; Center for Space Research, The University of Texas at Austin Seahorse Coastal Consulting. Web support provided by the Renaissance Computing Institute, University of North Carolina at Chapel Hill. Visualization: Adam Kubach and Karla Vega, Texas Advanced Computing Center (TACC). Acknowledgments: Funding provided by the National Science Foundation TeraGrid initiative and a National Science Foundation RAPID Grant, Office of Cyberinfrastructure. Also, the Department of Homeland Security Science & Technology Directorate through the Center of Excellence for Natural Disasters, Coastal Infrastructure and Emergency Management.