Physics Lecture Series | Spring 2014

My personal journey towards a career in seismology began as a physics major at the University of Wisconsin. There I gained exposure to the field of seismology and had an opportunity to participate in research with my academic advisor. However, my horizons and opportunities were expanded even further after being accepted into the IRIS Undergraduate Internship Program to conduct seismological research during the summer of my sophomore year.

Non-volcanic tremor is a weak, extended duration seismic signal observed episodically on some major faults, often in conjunction with slow slip events. Such tremor may hold the key to understanding fundamental processes at the deep roots of faults, and could signal times of accelerated slip and hence increased seismic hazard. Since the discovery of deep, non-volcanic tremor many studies have attempted to locate it and understand its origin; however, tremor has proven difficult to study due to the lack of impulsive wave arrivals, such as those used to locate and constrain the mechanism of ordinary earthquakes. My current work at Cal-tech focuses on extracting low frequency earthquakes from Non-volcanic tremor in order to gain a precise idea of the mechanics of tremor and slow slip on faults prone to large ruptures.

Over the past two decades, planets, many massive ones like Jupiter, have been found in large numbers around middle-aged stars generally similar to the Sun. Many years earlier, astronomers had already determined the frequency of stellar companions to stars, finding large numbers of such binaries. Together, these studies have revealed a curious phenomena: the so-called brown dwarf desert. Relatively few stars have nearby companions that are brown dwarfs — objects too massive to be a planet but not massive enough to be a star. The cause of this brown dwarf desert is currently unknown. I will describe recent efforts to look for brown dwarfs and giant planets around very young stars in an effort to explore whether this desert represents a fundamental limit on nature's ability to form these objects, or whether the desert is the result of evolutionary processes at work in very young systems.

High power Solar Electric Propulsion (SEP) technology using VASIMR® engines can dramatically reduce mission cost, when factoring in the time cost of money, and duration for Near-Earth Asteroid (NEA) retrieval and deflection missions. The current paper compares the 2008 HU4 asteroid retrieval mission using a 40 kW Hall thruster array with SEP-VASIMR® missions. The capabilities of 400 kW SEP-VASIMR® system were also studied for a NEA deflection mission with an orbit similar to 99942 Apophis.

Astrophysics traditionally was only studied using ground based or space based observations. Over the past two decades, rapid advances in high-performance computing and intense lasers opened up new windows on simulating complex astrophysical phenomena on the computer as well as in the laboratory. In this talk I will highlight some of the recent developments in trying to model high energy astrophysical phenomena related to black holes, gamma-ray bursts, pulsars and AGN, using both computer simulations and laboratory experiments. In the long run, advances in astrophysics must rely on the interplay between observation, computer simulation and laboratory experiments.

The asteroid impact near the Russian city of Chelyabinsk on 15 February 2013 was the largest airburst on Earth since the 1908 Tunguska event, causing a natural disaster in an area with a population exceeding one million. Because it occurred in an era with modern consumer electronics, field sensors, and laboratory techniques, unprecedented measurements were made of the impact event and the meteoroid that caused it. Well-documented meteorite falls are permitting a revolution in our understanding of the nature and dangers presented by Earth-crossing asteroids.

Numerical simulations are becoming a more effective tool for conducting detailed investigations into the evolution of our universe. In this presentation, we show how the framework of numerical relativity can be used for studying cosmological models. The author is working to develop a large-scale simulation of the dynamical processes in the early universe. These take into account interactions of dark matter, scalar perturbations, gravitational waves, magnetic fields and a turbulent plasma. The codes described in this report are GRMHD codes based on the Cactus framework and is structured to utilize one of several different differencing methods chosen at run-time.

Breast cancer patients with high expression of Estrogen Receptor alpha (ERa) are associated with better prognosis and response to anti-estrogen treatment. Over 70% of primary breast cancer patients are of the ERa positive type. Gata3 is one member of the Gata transcription factor family containing zinc-finger motifs and promotes chromatin remodeling [Takemoto et al.,2002; Zhou and Ouyang, 2003; Yamashita et al., 2004]. Previous studies [Eeckhoute 2007, Marconett et al., 2010] suggest that ERa and GATA3 constitute a transcriptional regulatory network.

We use mathematical modeling and perturbation experiments to unravel the various regulatory connections in the network. Specifically we perform cell population-average, single-cell and single-nucleus measurements in the ERa-positive breast cancer cell line T47D to characterize the directivity and strengths of the ERa and GATA3 network. A third transcription factor FOXA1 was also implicated to be a key factor of the dynamic redistribution of binding sites on the ERa chromatin in the case of patients having metastatic status and resistance to drugs [Jason Caroll, 2012]. The Triad FOXA1-ERa-GATA3 have to work together to restore estrogen-responsive growth as in MCF7 Era +ve cell line (Ed. Liu et al., 2011).
Directions of regulation among the three TF were found as before in our laboratory and regions of stability due to such regulations were mapped out with respect to transcription amplification and protein degradation parameters.

In 1 sec, NCSA's PetaFLOPS Blue Waters Cray Supercomputing Cluster can do what a person performing simple math for 32,000,000 years. Following an NSF XSEDE Fellowship, I am continuing the current research collaborating momentum to help optimize the fastest earthquake simulation to run even faster as it moving forward higher precision/frequency in reduced complexity and cost on Blue Waters and other world's fastest supercomputer systems. Just returned from the 25th ACM Supercomputing Conference anniversary celebration at SC13 in Denver with 10,600 attendants, where the M8 (Magnitude 8) earthquake simulation/visualization of our southern California earthquake seismic hazard research collaboration team was highlighted as the sole largest ACM global HPC analysis achievement for 2010 in the 25th SC anniversary exhibit.

While the scientists are running out of high-performance computation tuning tactics known for CyberShake HPC achievement (including computation-communication overlapping, MPI-IO, SGT short-cut, memory striping, CPU-GPU co-scheduling,... etc.), I was honored to help last summer 2013 as the nation's first NSF XSEDE CC Fellow in the processor subnet topology tuning coordination to minimize network fragmentation issue in cluster subnet allocation. Our summer 2013 effort improved the fastest seismic wave propagation computation scalability from 52% to 81% and speedup by 35% further on 4,096 nodes of the $200M Blue Waters Cray cluster at NCSA UIUC (world's fastest academic supercomputer).

To sustain CyberShake research in the fast changing HPC field, I am looking into computation optimization by concentrating limited resource to where it matters, including input rupture prediction number reduction to only major rupture contributors, reducing building sites of interests to only strategic ones, and migrating tasks from GPU-CPU to hostless-GPU or FPGA in traditional Finite Difference FDTD and exploring transformative approaches with Cellular Automata, Artificial Neural Net, Quantum, Active Learning, etc. which will be detailed in the presentation.

Golden Spike is a company formed for the purpose of getting humans back to the Moon using private funding. President and CEO of Golden Spike is planetary scientist Dr. Alan Stern. The presentation will discuss how the Golden Spike project got started, and how the company leaders think they can get their first human crew to the Moon for $6 billion to $8 billion dollars, with succeeding flights for roughly $1.5 billion each. This is far lower than cost estimates for plans to establish lunar bases under President George H. W. Bush, or for the Constellation program under President George W. Bush. The presentation will also discuss what the Golden Spike leadership thinks are their potential markets, and who they think are their potential customers.

Rather than develop a large booster of the Saturn class, Golden Spike plans to use multiple launches of smaller boosters, with orbital rendezvous as needed. The presentation will sketch out some mission models.

The Macho Mengi program is for the design, build, and operation of an observatory system consisting of networks of both ground-based interferometric telescopes and space-based telescopes. This distributed system would combine the capability of independent telescopes to function as a large telescope. M2 unites educational outreach with research and development, where we hope to spark the interest and cultivate innovate thought in the minds of our youth.

One of the most popular questions that drives planetary science is: Where and when did life form in the Solar System? Estimates for the timing of the origin of life require analyses of the ancient geologic record not only to locate direct evidence of life itself but also to understand the context of life-promoting environments. This geologic record has been broadly erased by plate tectonics on Earth and has been somewhat concealed by volcanism on Mars. The Moon represents an optimal location to constrain the geologic environment present during the ancient Solar System, a time when planets were accreting and nascent life may have been developing. During this talk, we will identify destinations on the Moon where we might collect samples to characterize the environment of the ancient Solar System.

The Astromaterials Research and Exploration Science (ARES) Directorate at NASA Johnson Space Center combines fundamental planetary and exploration science activities with long-term curation of NASA's extraterrestrial sample collections. In this talk, I will give a brief overview of ARES activities, including sample- and experiment-based science research, preparation and support for robotic and eventual human exploration of the solar system, and the success of our Curatorial efforts in maintaining pristine sample collections for allocations to the science community.