 |
|
|
|
|
|
| Universe Projects |
|
| A New Network for Gamma-Ray Bursts and other Immediate Astronomical Events (PI: Williams, Roy,
California Institute of Technology
) | |
| Objectives We plan to leverage the NASA Gamma-ray bursts Coordinates Network (GCN) with a new system, called VO-GCN,
layered on the VOEvent standard of the International Virtual Observatory Alliance (IVOA). The old GCN distributes alerts of
transient astronomical phenomena from a narrow range of NASA spacecraft to a small group of power-users, and has been very
successful over its 14 years, but now there many more event streams that should be distributed by multiple channels to multiple
communities, and a potential for exciting new science through automated decisions and correlation studies. Our proposal leverages
NASA assets with those of the National Virtual Observatory (NVO), Caltech, NOAO, and UC Berkeley to address this future. The
proposed system unifies many event streams from missions such as: GLAST, Pann-Starrs, LISA, IceCube, EXIST, LSST, NOAO
Surveys, Palomar-QUEST, OGLE, SDSSSS, and others; as well as incorporating archival data through NVO web services, and
allowing sophisticated studies of correlations in past event streams. Technology The proposed software will be based on the VOEvent
standard (voevent.org), a recommended standard of the IVOA. Event notices are represented in XML, a structured format that allows
the event author great flexibility, but also allows receiving software to understand and take action. The proposed system will have three
basic components: (1) A new version of GCN called VO-GCN, to run at NASA/GSFC, with access to the growing set of NASA events
and advertised to the GCN subscriber base; (2) Relay stations, aggregators, and repositories of events at Caltech, UC Berkeley, and
NOAO, allowing: subscription with sophisticated criteria, pooling additional event streams, correlation studies of past events;
Google-Sky presentation of event streams. (3) Client software and documentation so that subscribers can effectively utilize VO-GCN,
as well as authoring and disseminating followup events. The proposed system will allow subscribers to choose from several protocols:
eg, socket (improved from GCN), instant message, SMS text-message. email, RSS and web browser. Significance Transient events are
growing in importance to modern astronomy: instruments from NASA and other agencies are generating more and larger streams of
these, and many communities wish to subscribe to these event streams. We propose a general, flexible, peer-to-peer, robust, secure,
scalable solution for this infrastructure, to eventually replace GCN. Instead of a single source of gamma ray events for a few power
users delivered via boutique protocol, VO-GCN will be a part of the emerging global infrastructure of the time domain: a
standards-based vision of multiple federated event streams shared by peers and evaluated by decision support.
| |
| Application of Iterative Blind Deconvolution (PI: Young, Eliot,
Southwest Research Institute
) | |
| PROJECT GOALS: The overall goal of this project is to enhance images taken by NASA spacecraft or telescopic imaging systems.
More specifically, this project seeks (i) to test existing and experimental Iterative Blind Deconvolution (IBD) algorithms, (ii) to quantify
the limits and capabilities of different algorithms, and (iii) package the most useful algorithms in a form that is easily accessible to
NASA investigators. PROPOSED WORK: This project is a focused, two-year program consisting of four main tasks. 1. Test existing
IBD algorithms on various Hubble Space Telescope (HST) and Infrared Telescope Facility (IRTF) image data sets. 2. Implement and
test ideas that might improve IBD algorithms. 3. Incorporate appropriate IBD algorithms as library routines for an IRTF data
reduction pipeline. 4. Study the efficacy of new IBD algorithms for upcoming releases of AIDA (Adaptive Image Deconvolution
Algorithm) and build them into AIDA if appropriate. RELEVANCE OF PROPOSED WORK: Image enhancement is one arena where
the goals of the astronomer have immediate applications in many real-world scenarios. An enormous amount of work has been done in
the fields of medical imaging and military surveillance, to name two examples. In these examples, as with most astronomical imaging,
the common goals are to estimate the true image in the presence of noise and blurring (both sharpening features and detecting faint
details) while ensuring that no artifacts are produced by the image enhancement process. The products of this proposal will be
incorporated into AIDA, an open source IBD package (Hom et al. 2007).
| |
| Facilitated Access and Application of Computational Astrostatistics Algorithms (PI: Miller, Chris,
NOAO/CTIO
) | |
| Astronomy and Astrophysics are experiencing a revolution with the combination of huge data resources and the drive for high
precision measurements. A part of this revolution is new algorithms for the analysis of astronomical data which scale--up into the
regime of billions of sources with thousands of dimensions. To address this need, we have established an interdisciplinary research
team of astrophysicists, statisticians and computer scientists: the INternational Computational Astrostatistics (INCA) group. This
group is responsible for developing fast and efficient statistical algorithms for astrophysics. Key to the success of INCA is the
dissemination of these new algorithms to the research communities. This dissemination requires much more than making available a
tar-ball package of code and some README files. The algorithms and techniques being built to handle modern and future
astronomical datasets, are often complex in both design and implementation. Therefore, we propose here to build a set of
well-documented libraries for the most commonly used scripting languages available to astronomers and statisticians. This includes
exporting the INCA algorithms (which include KD-tree codes, fast mixture model algorithms and non--parametric fitting routines) to
IDL, R and Python. In particular, IDL has gained phenomenal popularity within the astronomical community over the last 10 years.
The goal is to provide a majority of astronomical researchers the opportunity to trivially and quickly apply these new techniques to
their data and rapidly visualize the results. Just as importantly, we will export common astronomical algorithms to the statistical R
scripting-language environment. In doing so, we can facilitate new advances from statisticians which will flow back quickly to the
astronomical community, as they can be developed against astronomical challenges. Finally, and perhaps most importantly, we will
deploy these algorithms as service-based web services (i.e., in a Service Oriented Architecture or SOA). This means that the analysis
tools can be placed where the large data volumes reside. This is a paradigm shift in astronomy where we "move the services to the
data", as opposed to moving the data to our desktops. Along with the web service protocols, we will provide browser-based user
interface mechanisms for utilizing these amazing new algorithms.
| |
| A High Performance Computing Framework for Cosmic Microwave Background Data Management (PI: Borrill, Julian,
Lawrence Berkeley National Laboratory
) | |
| The Cosmic Microwave Background (CMB) comprises the earliest possible photon-image of the Universe. Tiny anisotropies in the
CMB temperature and polarization encode a wealth of information, and offer a unique window onto cosmology and ultra-high energy
physics. However the faintness of the CMB signals means that extremely large data sets must be gathered and reduced to detect them
above the instrument noise. Moreover since we want to measure the correlations in the CMB signal across the celestial sphere, and
have to extract them from data whose noise is temporally correlated, simple divide-and-conquer approaches are insufficient and the
enormous CMB data sets have to be treated as single data objects. The size and complexity of CMB data sets has resulted in high
performance computing (HPC) becoming the cornerstone of CMB research. This proposal focuses on the challenges inherent in
managing and manipulating very large volumes of CMB data in an HPC environment, and in particular the need to optimize data flow,
from archival storage to spinning disk, and from disk to memory. These phases are treated separately, with the division reflecting both
their distinct issues and the cost-structure of national HPC facilities. Runtime data management is addressed with a data input
abstraction layer, while pre/post-processing data management is built around a database driven set of put and fetch tools tuned to an
HPC environment. This project seeks to improve the design and operation of CMB missions, and the analysis of their data, by enabling
the efficient and effective use of limited national HPC resources, including forthcoming peta-scale systems. As such it supports NASA's
goals of advancing scientific knowledge and understanding the Universe.
| |
| Visualization of Terascale Datasets with Impostors (PI: Quinn, Thomas,
University of Washington
) | |
| The interactive visualization of large astrophysical datasets is a formidable challenge. Cosmological simulations performed on current
terascale facilities typically generate data sets up to 100 gigabytes in size. In order to understand the complex three dimensional
structure within these simulations, interactive visualization with rapid rotation and zooming is required, but handling this much data is
well beyond the capability of even current high end graphics workstations. The rendering can be performed on a parallel computer,
even the same facility on which the simulation was done, but the user typically is using such facilities remotely. Hence both the latency
and the bandwidth between the users workstation and the parallel machine preclude the possibility of interactive rendering. We will
use techniques from the computer graphics community such as ``impostors'' to overcome these latency and bandwidth hurdles.
Interactive high quality rendering can be performed by using the power of the parallel machine to construct texture-map
representations of pieces of the simulation, which can then be shipped to the graphics workstation, where the graphics hardware is
optimize to rotate such texture-maps in real time. We will implement this application within our parallel visualization and analysis
framework developed with support from a previous AISRP grant. This framework is currently being actively used in research on
galaxy formation and evolution, and is sufficiently extensible that it can be used in a variety of applications relevant to NASA from
planet formation simulations to catalogs of galaxies from large scale structure surveys. Hence data from all these sources can be
rendered in real-time 3D.
| |
| Multi-Objective Optimal Scheduling for Space Science Applications (PI: Giuliano, Mark,
Space Telescope Science Institute
) | |
| We propose to develop and assess a new multi-objective multi-participant optimization application for scheduling space-based
observatories. Our approach will address how to optimize scheduling given a set of competing global objectives for a mission as well as
a large set of operational and scientific constraints. We will initially focus on James Webb Space Telescope (JWST) as it is under
development and is representative of a broad class of science scheduling problems. We will also consider the requirements of the Space
Interferometry Mission (SIM) to ensure and validate the generality of our approach. Multi-objective means retaining and exploiting
information about multiple, often competing objectives, during schedule optimization. This is in contrast to optimization techniques
that require the consolidation of all objectives into a single quantity to be optimized. Consolidation necessarily implies committing in
advance to the relative importance of different objectives, and hides information that can be used to make scheduling more effective.
Multi-participant means architecting and designing the system to allow multiple users to work with the system at one time, generally in
collaborative ways, but possibly independently. This is in contrast to approaches based on a single central scheduling process. By
incorporating both of these capabilities into a single framework we gain more than we would from either one separately. This is
because it is frequently the case that competing objectives arise directly from the various participants, and balancing these objectives
requires the participants to cooperate and compromise. This work will advance the state of the art for observatory scheduling in three
ways: (1) by providing simulation and trade study capabilities that will facilitate key early design choices; (2) by scheduling more
responsively in reaction to interesting events or system capability changes, and (3) by optimizing competing objectives of the mission so
that the overall science schedule is more efficient, and robust. The key elements of our approach are: a) a model-based representation
of objectives, constraints, and preferences, which will provide a basis for schedule optimization and for flexible modification as the
mission progresses b) an architecture to support multi-participant scheduling with both global and per-participant objectives c)
multi-objective optimization algorithms to find and characterize Pareto-optimal solutions d) an experiment plan to systematically
investigate and document our results e) the application of our approach to two distinct observatory missions, as a demonstration of its
generality and applicability. Our approach will leverage our experience and familiarity with science planning and scheduling in past
missions, including HST, Spitzer, Chandra, EUVE, FUSE, etc. from both operational and software perspectives. We will assess and
incorporate available tools and approaches, such as existing scheduling modeling languages, algorithms and technologies (e.g. as
embodied in Spike, ASPEN, and others), science simulators (e.g. JMS, the JWST Mission Simulator), and user interface elements. We
will explicitly consider and design for the future incorporation of constraints, preferences, and objectives not yet defined, by providing
component APIs and extension points. Improvements to scheduling applications have a proven track record for increasing science
return and lowering operations costs for space missions. For example, scheduling improvements allow the Hubble Space Telescope to
routinely achieve a weekly scheduling efficiency higher than thought possible at launch. Our proposal builds on this work and our
success criteria include the demonstration of a simultaneous up to 20% increase in scheduled science activities for JWST along with a
reduction of up to 20% resource consumption that can lead to a prolonged mission lifetime.
| |
| Novel Methods for the Analysis of Photon-Limited Data (PI: Scargle, Jeffrey,
NASA Ames Research Center
) | |
| tatistical techniques, and efficient algorithms to implement them, will be developed for analysis of event data (a.k.a. point data)
in astrophysics, space and Earth science, and other applications. These methods, integrated with existing standard ones, will be
described in a Handbook of Statistics for Event Data, providing scientists with practical guidance on use and interpretation. This
choice of problem area is based on my experience and expertise, and deliberately focuses on an important class of NASA missions.
Event data are largely photon detections: photometry, imaging, timing, and spectrophotometry. Most important in high-energy
astrophysics, where arrival direction and time, plus the energy of individual photons can be accurately measured, the methods will also
have broader applicability. The specific problems addressed include repeated measurements; detection, characterization, and upper
limits (for signals consisting of spatial sources, time variability, or spectral lines); and analysis of image and other structure. Algorithm
design will focus on the practical issues of user comprehension of inputs and processing, plus understanding and interpretation of the
outputs. Statistically rigorous modeling will be combined with robust applicability to diverse contexts, ease of use, transparency of
parameter adjustment, validity and applicability of approximations, unification of novel and old methods, and computation efficiency.
Matlab algorithms (with adaptations to other high-level systems in common use, such as IDL, R, S, and Python, as well as low-level
compiled languages) will posted in the AISRP code repository, with documentation there and in the accompanying Handbook.
| |
| Interferometric Mission Simulation Environment: An Essential NASA Mission Planning Tool (PI: Rinehart, Stephen,
NASA Goddard Space Flight Center
) | |
| We propose to develop the Interferometric Mission Simulation Environment (IMSE) whose purpose is to enable design optimization
and decrease the cost and technical risk associated with space-based interferometry missions. IMSE will have broad application to
interferometers currently under consideration in several of the Divisions of NASA¿s Science Mission Directorate, such as the Space
Infrared Interferometric Telescope (SPIRIT) and the Terrestrial Planet Finder Interferometer (TPF-I) in Astrophysics, the Earth
Atmospheric Solar-Occultation Imager (EASI) in Earth Science, and the L2 Mars Atmospheric Probe (L2-MAP) in Planetary Science.
These interferometers will provide valuable imaging and/or spectroscopic data vital to NASA¿s scientific mission. The IMSE includes
three components. The first component is a software package that generates realistic test scenes, or ¿truth images.¿ These scenes will
be used to assess both the performance of interferometer designs and the algorithms used to process and analyze simulated
interferometric data. The Scene Generator for Interferometry (SGI) will include the ability to generate simple test scenes, in order to
enable the verification of the IMSE functionality through analytic calculation. It will also provide the ability to generate realistic
spectral/spatial scenes of scientifically interesting targets, including extragalactic deep fields using the Far-Infrared Sky Simulator
(FSS) and Galactic sources such as debris disks. The second component of IMSE is a ¿virtual interferometer,¿ which captures in detail
the important elements of an interferometer¿s engineering design. The Virtual Interferometry Simulator (VIS) will be based on the
existing Simulator for Infrared/Submillimeter Space Interferometry (SISSI), which was developed primarily to evaluate alternative
SPIRIT mission design concepts. The virtual interferometer will simulate instrumental response to photon input and will realistically
model error sources, exposing the causes of uncertainty that dominate. This will enable mission designers to analyze the impacts of
their architecture and design choices, and guide data analysts whose task will be to interpret the interferometric data. The third
component of IMSE comprises a set of interferometric data reduction tools designed to convert simulated interferometric data into
spatial-spectral data cubes for comparison with the input ¿truth¿ data sets. All of the components of the IMSE will work together to
facilitate space interferometer mission design, and each component of the IMSE will be applicable to a number of missions and
facilities in which NASA has a vested interest.
| |
| New Methods in Data Mining for Time-Domain Astronomy (PI: Alcock, Charles,
Harvard University
) | |
| Space missions and giant astronomical synoptic surveys and will soon monitor much of the sky regularly, detecting vast numbers of
interesting, variable astronomical objects. These objects will range from small solar system bodies to distant quasars and extra-solar
planets. The objective of this proposal is to develop the tools necessary for this new treasure trove. Specifically, this project will: 1)
Develop new, fast and scalable techniques to identify extra-solar planets. 2) Develop new, model based techniques to distinguish
transiting extra-solar planets and other, presently undiscovered "oddities" from variable stars. 3) Create data mining methods for
anomaly detection in time series data in order to enable discovery. 4) Preserve novelty detection when the volume of data becomes so
large that the scientists are "removed" from the data. The two "specific" problems are: (1) the discovery and characterization of
transiting extra-solar planets and (2) the classification of and outlier detection among variable stars. The proposed work will explore
the scientific analysis and modeling of massive datasets of light-curves (66 million light-curves now available, growing to 100 billion in a
decade) from a broad range of perspectives. The unique complications posed by the data in this domain will drive research for data
mining techniques to allow these analyses. The specific project goal is to enable qualitatively new science specifically to enable profound
insights into the astrophysics of extra-solar planets and variable stars. More generally, the proposed work will provide new data
mining modeling and analysis tools that scale well to very large datasets and are fast for "real time" analysis. This interdisciplinary
proposal will lead to the development of new insights, while training leading members of a powerful new generation of scientists.
| |
| Framework for Fast Spatial Searches Using the Hierachical Triangular Mesh (PI: Szalay, Alexander,
Johns Hopkins University
) | |
| Spatial searches over the celestial sphere represent the most frequent queries on astrophysics data sets. The AISRP program funded
the original Hierarchical Triangular Mesh library which produced a very fast C++ search engine now used at more than 20 institutions
over the world, also in various Earth Sciences applications. With the recent advances in the worldwide Virtual Observatory effort, we
have now a standard XML data model for space-time data. The community is now ready to deploy sophisticated spatial search engines,
but this requires an API with multiple language bindings and interfaces to multiple database platforms. We seek funding to rewrite,
extend and refine our existing library, make it fully compliant with the VO interface standards, and enhance its functionality and
performance. The new implementation will consist of four layers: a kernel library with all the spatial algorithms, the code for the
hierarchical tree representation of the regions, an extensible interface to describe the spherical regions, and a flexible and fast interface
to relational databases. We will add several new algorithms to the core of the existing library, develop a new VO-compliant API, and
build several new language implementations that others can integrate with their own archives. We will develop a Web Services
implementation of the library, accessible through remote calls. We will integrate our toolkit with the OpenSkyQuery framework for
cross-matching large astronomical catalogs. We will provide a framework for critical analysis of the HTM toolkit, including built-in
unit testing modules. We will build Java, C++ and C# implementations of the library. We will also build a web-service based
visualization toolkit to display the spherical regions specified in a selection. All our code from our three year proposal is public domain
Open Source (FreeBSD) and made available to the public on-line.
| |
| ExGKS: Extragalactic Knowledge System (PI: Schombert, James,
University of Oregon
) | |
| This is a modest project to explore techniques and tools to enhance the final stages of data analysis prior to publication. There is a
growing gap between the size of ever increasing data archives and the capability of individual researchers to absorb and compare these
datasets. Our goal is to begin the development of a knowledge framework through a three prong program; 1) establish a data system
that allows for flexible use, in our case we will be adopting an XML format that introduces several unique features when combined
with the Python interpretive language, 2) develop a system of network data retrieval that is independent of the features of the data
archive (e.g. ranging from password protected websites to simple sftp areas) and 3) present these new features and techniques in a
flexible, yet sophisticated GUI framework that the researcher can easily define and expand to meet their own needs. Our plans are
novel, but realistic. We will propose some innovative ideas, and applied them to concrete tools as both a proof of concept to our ideas
and as examples for other developers. This project has the larger goal of enhancing the development of tools to exploit the data in
NASA archives. The framework, as it develops, will enable Virtual Observatory capability to existing applications, provide new
techniques to combining data and literature searches and offer a flexible, user designed GUI system in which to explore these new
ideas. Our immediate goal is to provide the community with three testbed tools to outline our XML procedures, script browsing and
GUI construction. These prototypes enable inclusion of VO-like power to data reduction and final analysis, new types of viewers to
expand individual resources and allow wider science goals from our space mission datasets.
| |
| Distributed and Peer-to-Peer Data Mining for Scalable Analysis of Data from Virtual Observatories (PI: Kargupta, Hillol,
University of Maryland Baltimore County
) | |
| Design, implementation, and archiving of very large sky surveys play a critical role in today's Astronomy research. However,
astronomers will be unable to tap the riches of this collection of gigabyte, terabyte, and (eventually) petabyte catalogs without a
computational backbone that includes support for queries and data mining across distributed virtual tables of de-centralized, joined,
and integrated sky survey catalogs. Moreover, use of local data management systems such as MyDB, MySpace in AstroGrid, and Grid
Bricks for storing and managing user's local data is becoming increasingly popular. This is opening up the possibility of constructing a
Peer-to-Peer (P2P) network for data sharing and mining. This document proposes research and development for a new generation of
scalable data analytic services for the NVO based on advanced distributed and P2P data mining capabilities across multiple data
repositories. This research will develop technology for supporting web services within the NVO that will allow astronomy researchers
to analyze data from multiple surveys using fundamentally distributed algorithms. It will also develop several distributed data mining
algorithms for analysis of distributed Astronomy catalogs without requiring the data to be downloaded and centralized. Specific
objectives include the following items: (1) The project will design and implement distributed algorithms for computing statistical
primitives, principal component analysis, and outlier detection from distributed Astronomy catalogs and their partial images stored in
users' local data management systems. These algorithms will be able to analyze data without requiring source catalogs to be
downloaded and centralized. (2) The project will develop a prototype system which will offer a rich collection of web-services based on
various DDM algorithms. This service offers a novel augmentation to the existing NVO environment and it will support a rich variety
of data mining tasks that will work in a distributed fashion. (3) The developed system will be tested using specific astronomical
research problems. In particular, we will explore the multi-dimensional multi-wavelength parameter space of astrophysical properties
of starbursting galaxies. We will search for unusual correlations, outliers, sub-clusters, and fundamental planes within the
multi-dimensional parameter space presented by several large surveys. To carry out this research we will set up a simulated distributed
Astronomy catalog environment in the lab using data from publicly available source catalogs. Our system will be benchmarked
according to speed, communication cost, and accuracy. Accuracy will be validated within the context of the Astronomy problem
described above. This research is directly relevant to the AISR program for several reasons. First, it enables increased productivity of
NASA's Science Mission Directorate (SMD) research endeavors through rapid multi-mission correlative analysis, such as distributed
and P2P mining of the large survey catalogs from GALEX, Spitzer, 2MASS, and eventually WISE. Second, this research involves an
interdisciplinary team of researchers in Astrophysics (Borne), Database Technologies(Kargupta, Giannella), Distributed Systems
(Kargupta), and Distributed Data Mining (all). Dr. Borne is a senior member of the NVO project. We also have strong support from
NASA's Space Science Data Operations Office (see attached supporting letter) regarding this proposed collaborative project and its
transition to the practice. Finally, this project explicitly demonstrates the relevance, applicability, and potential impact of emerging
information technologies to SMD missions and programs.
| |
| VOClass: Classification and Learning in the Virtual Observatory (PI: Connolly, Andrew,
U Pittsburgh
) | |
| Classification and outlier detection are fundamental to all areas of the observational sciences. In astrophysics the interrelations between observed parameters can give insight into the physical processes within the universe. Deviations from these relations can be used to identify new classes of source or astrophysical phenomena. We propose here to develop an automated classification facility for the VO, VOClass, that will integrate state-of-the-art data-mining algorithms with an extensive set of VO-enabled tools that are now available. Classifications will be undertaken utilizing a range of supervised and unsupervised techniques that are designed to scale to the size of current and future surveys. Using VOClass, astronomers will be able to upload images or catalogs, correlate their sources with other multifrequency data sets available through the VO or NASA data archives and return classifications of these sources and identifications of unusual objects based on the measured and correlated properties.
The primary goals of VOClass are: (a) to enable fast classification of large, distributed databases, (b) to provide robust statistical descriptions of the interrelation between observables that scale to the size and dimensionality of current and future data sets, (c) to integrate classification and visualization tools available through the Virtual Observatory, (d) to provide the ability to adaptively classify sources by learning from user defined inputs and thereby develop new classification schemes, (e) to interface these knowledge discovery tools with the Virtual Observatory through the application of Web Services which will provide a modular access enabling scientists to build other classifiers that are optimized to a particular task.
VOClass will provide a comprehensive facility for source extraction, cross-matching, and classification. Combining classification with state-of-the-art VO tools it will provide an environment for distributed analysis and knowledge discovery that is unprecedented in astronomy. It has, however, many other applications both within astrophysics and in the context of the broader NASA mission. Development of VOClass will directly address the AISR program goal of providing "advances in information science and technology to increase life cycle effectiveness and efficiency of the Science Mission Directorate (SMD)". Beyond astrophysics the techniques we propose to deliver have applications to telemetry data streams, quick look observations and verification of data from satellite images.
| |
| Bayesian Source Separation for Astrophysical Spectra: Application to PAHs (PI: Knuth, Kevin,
NASA Ames
) | |
| We propose to apply Bayesian source separation techniques to a very important subset of astrophysical spectra, the emission spectra of PAHs. The PAH species, ubiquitous in our Galaxy and in external galaxies, are one of the most important molecular species in the pathway to life. Their spectra are the result of the blending of the complex spectra of individual PAH species, possibly at different temperatures, and to date they have defied all attempts to unscramble them into individual contributors. We propose to apply Bayesian source separation for the first time in order to quantitatively determine which of the many hundreds of individual PAH species contribute most significantly to the combined emission spectra of PAHs. Our research will lead to powerful new tools for decomposing, analyzing and characterizing the spectra of these critical species. Furthermore, these tools will be readily generalizable to a wide array of spectral analysis problems ranging across astrophysics, Earth science and biology.
| |
| Large Scale on Demand Cross-Matching with Open SkyQuery (PI: Thakar, Aniruddha,
Johns Hopkins U
) | |
| Summary
This proposal seeks to develop an advanced and versatile all-sky cross-matching engine for large distributed astronomical datasets by bringing together and leveraging the results of two previously funded AISR projects. We wish to extend the present SkyNode and Open SkyQuery functionalities in the following ways:
1. Incorporate parallel data access with Zone partitioning into individual SkyNodes
2. Transport data between SkyNodes and the portal using the Mega Streaming transport protocol developed by Grossman et al.
3. Replace the current synchronous query execution with an asynchronous system that is compatible with emerging IVOA standards.
We will apply the Zone-based partitioning and parallelism that we developed with prior AISR funding to individual SkyNodes in Open SkyQuery (also developed with AISR funding) so that we can parallelize the data access and the cross-match computation by distributing the data and workload among a cluster of database servers.
In order to move the cross-match data quickly between SkyNodes, we will incorporate the mega-streaming network transport protocol for fast data movement developed at NCDM/UIC.
Asynchronous workflow is the third main enhancement that we will make. The size and complexity of the large scale cross-matches will make synchronous response back to the Web client impracticable and inefficient. Asynchronous query execution will require identification of users, tying into the IVOA single sign-on (SSO) protocol. The large outputs resulting from large cross-matches will require direct routing to storage destinations like VOStores.
In addition to producing a large-area cross-matching facility that will allow the NVO to truly federate large distributed archives, this project will also provide high-speed data access to all large datasets supported by the NVO and NASA. It will enable unprecedented science that can be used to constrain cosmological n-point correlation function and power spectrum calculations, determine the AGN luminosity function, find galaxy clusters, etc. These studies will increase the science return from current NASA missions and promote the NASA strategic objective of exploring the universe to understand its origin, structure, evolution, and destiny. A facility that allows for extensive correlation with other catalogs will be extremely useful for near-earth and moving object searches, and for TPF searches related to the NASA strategic objective of searching for Earth-like planets.
| |
| Parallel-Processing Astrophysical Image-Analysis Tools (PI: Mighell, Kenneth,
NOAO
) | |
| This proposal describes a three-year research plan to develop and implement several parallel-processing astrophysical image-analysis tools for two of NASA's Great Observatories: the Hubble Space Telescope and the Spitzer Space Telescope. These new data-analysis software tools will be written with the goal of enhancing the scientific return of these missions. This project will improve and extend the AISR-funded enabling image-processing technology of the Principal Investigator's MATPHOT algorithm for precise and accurate stellar photometry and astrometry with discrete Point Spread Functions. The PI has recently demonstrated (in the 11 August 2005 issue of the Monthly Notices of the Royal Astronomical Society) that the current C implementation of the MATPHOT algorithm can achieve millipixel relative astrometry and millimag photometric precision with complicated space-based discrete Point Spread Functions imaged onto imperfect detectors with large intrapixel quantum efficiency variations. This project combines state-of-the-art astrophysical image processing techniques with the enabling technology of Beowulf clusters which offer excellent cost/performance ratios for computational power. The PI will use the project software to investigate the possibility of significantly improving the current state-of-the-art of space-based optical and infrared stellar photometry and astrometry by analyzing existing archival imaging data from the NASA mission data archives for HST and SST. This project is structured into three one-year phases. The major software products will be posted (at the end of each one-year phase) on web sites dedicated to this project on the web servers of the National Optical Astronomy Observatory.
| |
| BioLingua - An Integrated System for the Discovery how Genes Regulate Cell Behavior (PI: Pohorille, Andrew,
NASA Ames
) | |
| In order to improve productivity in astrobiology we propose to develop an intuitive, easy to use programming environment that enables scientists to combine their data, knowledge, computational tools and previous, relevant results to make new biological discoveries. We call this environment �BioLingua/CD� (BioLingua for Collaborative Discovery). BioLingua/CD will have a number of fundamental features. It will contain an efficient, easy to use, general purpose programming language augmented with a rich set of biology-specific functionalities, and transparent access to a wide range of basic bioinformatics algorithms, advanced statistical tools and databases. They include genomic, proteomic, metabolic and gene expression data, as well as relevant higher-level knowledge, such as the Gene Ontology, and representations of metabolic and regulatory pathways. All these capabilities, which enable scientists to access and interpret relevant data and knowledge, will be available through a common framework. At the center of BioLingua will be state-of-the art tools for discovering causal relations in gene regulation and metabolic networks and modeling relations between gene expression and metabolism, subject to constraints obtained from previously acquired data and knowledge. The environment will be highly interactive, web-based and equipped with a flexible and efficient visual programming language. The environment will provide tools for collaboration among biologists and a means for sharing results with the community. Using these capabilities the biologist will be able to manipulate and analyze data and knowledge for biological problems with convenience that would be otherwise impossible to achieve.
At the end of this project, we will deliver a powerful, easy to use system that will be an essential tool for astrobiologists engaged in genomic, cellular biology and systems biology research. This system will provide technology for increasing productivity in all science programs sponsored by the Science Mission Directorate that involve studies of extant or extinct organisms. In particular, it will support missions and research devoted to understanding the evolutionary mechanisms and environmental limits of life by determining the molecular, genetic, and biochemical mechanisms that control and limit evolution, metabolic diversity, and the ability of life to survive in space, as defined in Goals 4, 5 and 6 of Astrobiology Roadmap.
BioLingua technology has the potential to increase productivity in SMD programs beyond biological domain though its advanced knowledge discovery and data synthesis capabilities, and by fostering interdisciplinary collaborations that span the space science and computer science disciplines.
| |
| SkyView: The Visualization Portal for the Virtual Observatory (PI: McGlynn, Thomas,
NASA Goddard
) | |
| Over the past decade the SkyView Virtual Telescope has pioneered concepts of common, simple access to multi-wavelength data that have now begun to mature in the Virtual Observatory (VO) efforts underway in many nations. In this proposal we discuss how SkyView can build upon the nascent capabilities and protocols of the VO to provide a whole new way for astronomers to organize, explore and visualize sky data.
We propose extending SkyView to unify our view of the heavens in fundamentally new ways. Today, astronomers treat catalogs of objects and images as very different products even though they both are representations of what is in the sky. We propose to enable astronomers to revisualize catalog data helping astronomers to comprehend the huge giga-object catalogs that are now coming on-line in the astronomy community. Today, transformations between representations of large-scale structure, especially for cosmic microwave background, and conventional maps are not easily available to most astronomers and the simple approaches commonly used do not preserve as much information as they might. We propose to build transformation tools for SkyView to efficiently and robustly transform data among all useful astronomical data projections. Today, transforming image geometries is treated as a black art: most users pick the simplest possible algorithms even when they could preserve much more information at little cost in computing power. We propose to build and distribute a reference library of resampling techniques and algorithms so that users in astronomer and throughout the sciences have easy access to a variety of resampling techniques. As both a consumer and provider of VO datasets, we shall document how other users can effectively build VO-compliant data services.
The proposal takes advantage of the confluence of huge new data resources coming on-line in astronomy and the new community standards being promulgated in the Virtual Observatory to enable astronomers to flexibly visualize astronomical data. With our new approaches astronomers can more effectively realize the science potential of NASA missions and projects. Reducing the barriers to understanding and manipulating images and catalogs, our program also enlarges the pool of scientists able to use NASA data for research and makes it possible for the non-science community to participate in NASA�s science programs.
| |
| HYDRA: A New Paradigm for Astrophysical Modeling, Simulation, and Analysis (contd) (PI: Houck, John C,
MIT
) | |
| Regardless of the source or mission, the analysis of astrophysical
data invariably relies on the interplay between predictive physical
models, a detailed understanding of the observing instruments, and
the discriminating comparison of predictions and actual observations.
With HYDRA, we are developing a new platform which will provide scientists
a more flexible and extensible way of constructing models for
astrophysical sources that include geometric information, physical
emission and absorption mechanisms, transport processes, and
projection effects. We will also provide an interface through which
users may link existing astrophysical models to HYDRA. Similarly,
modules describing the performance of the instrumentation, be it
ground-based telescope or orbiting satellite, will be definable in
a mission-independent way to allow realizations of these models in
the form of simulated observations. By combining source and
instrumentation models, HYDRA can serve as a tool for observational
planning, instrument design, and calibration. Simultaneously, HYDRA
will provide an analysis environment that allows source models to be
compared to existing observations and iteratively adjusted. As part
of its design, the HYDRA system will also include advanced
visualization capabilities that will provide users with additional
diagnostic ability as well as a potentially powerful educational tool.
| |
| Scalable Algorithms for Analysis of Megapixel CMB Maps and Large Databases (PI: Szapudi, Istvan,
U of Hawaii
) | |
| We propose to develop a set of algorithms for spatial statistical analyses of large astronomical data bases, such a
CMB maps, galaxy catalogs, or any point source catalog. We will build on the success of our previous AISR, and
tackle computational challenges such as
i) fast and accurate estimation of temperature,
polarization and lensing correlation functions and
power spectra, ii) fast estimation of higher order
correlation functions from large CMB maps,
angular, redshift and weak lensing surveys.
iii) fast estimation of covariance matrices using novel Monte Carlo techniques.
In our unique interdisciplinary approach for algorithmic
development, we reformulate the statistical problems arising
in space astronomy with special attention to computational needs. Our philosophy naturally blends the principles
of astronomy, statistics, computer and computational science.
Special attention will be payed
to the user interface and quality control to ensure wide
practical usage. The resulting software package will be useful for accurate analysis of MAP and Planck, or any subsequent
megapixel surveys, galaxy surveys, background light correlations, correlations in maps produced in other wavelengths, such as infrared background, as well as cross correlations of all the above.
| |
| Enabling Bayesian Inference for the Astronomy Masses (PI: Weinberg, Martin,
U Massachusetts
) | |
| The wealth of data being acquired by NASA missions promises detailed
tests of scientific theories. Bayesian analysis has the power to
efficiently combine information from these diverse sources, rigorously
establish confidence bounds on theoretical models, and provide
powerful probability-based methods for model selection and complex
hypothesis testing. This statistical approach is superior to those
commonly used in astronomy. However, it is not currently favored
owing to its computational difficulty. To remedy this deficiency, a
multi-disciplinary investigator team from the Departments of Astronomy
and Computer Science at UMass recently developed the Bayesian
Inference Engine (BIE). BIE's success is aided by advances in Markov
Chain Monte Carlo (MCMC) algorithms and the availability of commodity
parallel hardware. Most importantly, the system frees the scientist
from the database mechanics of the ongoing statistical investigation
without sacrificing rigor, while providing high-level interaction with
the inquiry.
To promote wider application and use, we propose key enhancements and
stand-alone applications to allow the astronomical community to take
immediate advantage of these statistical inference tools. First, we
will implement a persistence system. This allows the entire inference
to be stored in a research log to be later recalled and reused. A
persistence system enables what-if exploration, checkpointing, and
collaborative investigations. We will enhance the BIE with advanced
MCMC algorithms, non-parametric prior distributions and new graphical
tools. Second, we will develop three stand-alone "killer" applications
that will demonstrate the power of Bayesian inference as well as
provide an introduction to using BIE and Bayesian methods for more
specialized problems. The applications are: 1) Analysis of Structure
in Catalog Databases--a general set of star and galaxy counting
routines with I/O interfaces to SQL relational databases (e.g 2MASS
and SDSS); 2) Determining Galaxy Components from Images--a galaxy
classification tool based on the popular GIM2D package with a BIE
back-end that will allow sophisticated inference over image ensembles;
and 3) A Rigorous Statistical Basis for Semi-Analytic Models
(SAMS)--these widely used multiparameter phenomenological models for
galaxy formation make predictions for and use constraints from
multiple types of observations. The BIE back-end will enable
methodical exploration of the free parameters and allow sophisticated
hypothesis testing.
| |
| Developing Methods to Incorporate Calibration Uncertainties in Data Analysis (PI: Kashyap, Vinay,
Harvard/SAO
) | |
| The aim of this proposal is to develop and make publicly available a set of tools designed to incorporate knowledge of calibration errors into data analysis. We have developed a method to handle in a practical way the effect of uncertainties in instrument response on astrophysical modeling, with specific application to Chandra/ACIS instrument effective area. This groundbreaking work holds great practical promise for a generalized treatment of instrumental uncertainties in not just X-ray spectra but also imaging, or any kind of higher-dimensional analyses. We apply a combination of Principal Component Analysis and Markov chain Monte Carlo methods to calibration uncertainties and include these uncertainties directly into analyses. We propose to develop modular tools that can be easily extended to different instruments and calibration products, and will provide a systematic framework to describe the true errors in the estimated model parameters, even in non-Gaussian regimes.
Calibration is the foundation on which all data are analyzed and interpreted. Therefore, any efforts to improve handling of calibration data and to use it in a better fashion is immediately relevant to Space Science programs. Considerable effort will be saved by reducing the propensity of systematic calibration-based biases from skewing the results. Our proposed project will also have specific impacts. Firstly, because algorithm development will mostly take place in the context of Chandra data, it will lead directly to an improvement in the quality of the analysis of Chandra data. Secondly, it will result in the useful characterization of a number of calibration products from Chandra, Suzaku, and other current missions. Finally, we will set up groundbreaking standards in the manner and specification of calibration data for future missions.
| |
| On-the-fly and Grid Analysis of Astronomical Images for the Virtual Observatory (PI: Ptak, Andrew,
Johns Hopkins U
) | |
| We propose to improve upon two existing software packages, XAssist and WESIX. WESIX automatically performs source detection on images that are uploaded by users. XAssist automatically analyzes X-ray astrophysics data, but currently does not have an interface for the Virtual Observatory, and a major goal will be to provide a web service interface to XAssist. The main improvements to XAssist will be added functionality to only analyze a subset of a field (i.e., around the requested position), the capacity to distribute its processing among multiple machines or CPUs, and enhanced speed and reliability. WESIX will be improved to compute calibrated optical fluxes for the detected sources, particularly for Hubble Space Telescope data. The ability to compute upper-limits will be added to both systems, which is particularly important for minimizing bias in statistical studies. We will also work on the production of a common interface to both systems that will allow the processing of optical and X-ray data to be requested in the same way, and allow the software to be used in a grid computing environment. This will set a standard for the inclusion of similar systems for other wavelengths, such as radio and infrared data. This project will improve the utilization of NASA archival data.
| |
| Advanced Astrophysical Algorithms to Novel Supercomputing Hardware (PI: Brunner, Robert,
U of Illinois
) | |
| Astronomy has become a data-rich field, and the future promises to only increase the data volume. NASA has a long, and distinguished track record of funding scientifically important missions that are in large part contributing to the quantity and quality of data being analyzed to gain fundamental insights into everything from our own planet to the origins of our universe. Our ability as scientists, however, to keep pace with this data flood is not sufficient. In response, many researchers, some in proposals to this AISR program, have developed innovative, cutting-edge statistical and computer science techniques and tools to improve our ability to analyses the data in hand. Even these efforts have not been adequate; we somehow must find mechanisms for improving the performance and robustness of our applications; otherwise we will not only fail to capitalize on the richness of the data we currently have in hand, but we also will fall further behind with the imminent arrival of petabyte datsets. This problem is not tied to one particular branch of NASA�s operating strategy, but crosses all scientific and engineering disciplines.
In this proposal, we outline an interdisciplinary collaboration that we feel has the potential to revolutionize the way in which complex algorithms are applied to astrophysical datasets. Our collaboration of domain specialists have identified a means by which the analysis of large datasets can be increased by a factor of 100 or more over that provided by the traditional use of supercomputers alone. We will combine the expertise of the Laboratory for Cosmological Data Mining, the NCSA Innovative Systems Laboratory (ISL) and the NCSA Automated Learning Group (ALG) to analyze terascale datasets, such as that from the Sloan Digital Sky Survey (SDSS). As an initial proof-of-concept, we will produce the largest and most accurately classified catalog of objects to date, using instance-based learning, and, subsequently use this classified catalog to calculate cosmological statistics such as N-point correlation functions. Both of these data analyses, using the most accurate statistical approaches are otherwise intractable. The ISL has available cutting-edge hardware in the form of high performance reconfigurable systems which enable the speedup described. Crucially, given the experimental nature of current non-embedded applications which this technology, we also have the expertise and proven results to make this possible. Tests have shown that instance-based learning, in particular when combined with the results from other algorithms, results in superior object classification, but the method remains relatively unexplored in astronomy due to its intractability.
The PI and two Co-Is have joint appointments between the Department of Astronomy and NCSA at the University of Illinois, Urbana-Champaign, and have previously found that technological advances often benefit from the proximity of these institutions. The cyberinfrastructure to produce the catalogs is already in place, in the form of the Data-to-Knowledge Toolkit developed by the ALG. Indeed, it has already been used to classify the 141 million objects in the SDSS Data Release 3, a 70 GB dataset, but to realize the full potential of the available facilities and expertise requires the interdisciplinary collaboration outlined herein. The ISL is keen to gain partners who have real-world applications which may benefit from reconfigurable computing, either in industry or academia. The relevance to NASA and AISR goals is the advancement of computer hardware and software in the field of high performance computing, cyberinfrastructure and machine learning. The use of FPGAs/reconfigurable computing will make tractable other data intensive problems in astronomy such as Fast Fourier Transforms (e.g., for the analysis of Planck mission data), or object classification and statistical analysis within the anticipated petascale datasets of missions such as the LSST.
| |
| Enabling Massive Scientific Databases Through Automated Schema Design (PI: Gardner, Jeffrey,
Carnegie Mellon
) | |
| As massively parallel platforms continue to expand in processor count, simulations can now generate datasets of unprecedented size relative to the processing power of a serial workstation. NASA has pushed this envelope with the Columbia supercomputer at NASA/ARC, which has 10,240 processors with a combined capability exceeding 50 teraflops. With the ability to run on 10 to 100 times as many processors as we could a mere five years ago, one of the greatest challenges scientists face in using these impressive resources is extracting meaningful scientific knowledge out of the immense datasets that they can generate. Simulations are now being limited in scope not by the capabilities of the computers that run them, but by the scientist's inability to cope with the resultant data flow.
The most obvious solution when dealing with massive data volumes is to register it into a database. The difficulty lies in the amount of computing and access time required to execute database queries: a cosmology simulation that fully utilized Columbia, for example, would generate at least 20 TB of data. The trick to optimizing queries lies in designing an intelligent schema, i.e. a description how the data is actually arranged within the database. Databases of this size mandate optimal schema design. Unfortunately, simulation groups--which are often quite small--rarely have the resources to spend on manually designing efficient database schema. Furthermore, the schema design for simulations is typically much more difficult than for observational datasets: a simulation catalog not only incorporates the time dimension, but it also frequently employs more layers of relationships. Current automated design tools are not useful for our purposes as they rely on aggressively replicating indexes, a strategy that increases the size of the database by factors of 2 to 3.
We propose to overcome the barriers to efficient query execution by combining work and ideas of the database design community with the needs of the computational cosmology domain. We will design and implement AutoPart, a system that automatically designs scientific databases by partitioning the tables in the original database according to a representative workload. Compared to conventional index-based techniques, AutoPart removes the need for replication by pre-designing the tables according to the query requirements. AutoPart will maximize performance of scientific queries while eliminating the need for additional space and update overhead, thereby permitting any researcher to interact with their data in the most efficient way possible. Our initial focus will be on enabling the mining of massive cosmological simulations. The results of our work, however, will be applicable far beyond cosmology and will effect any NASA endeavor that can exploit massive datasets.
| |
| Statistical Inference for Scientific Instruments: Event Analysis for the Gamma-Ray Large Area Space (PI: Morris, Robin,
Ames/USRA
) | |
| Scientific instruments are becoming more complex, are making indirect measurements that require complex interpretation, and are returning enormous quantities of data that require powerful methods of analysis. The Large Area Telescope (LAT) instrument on the Gamma-ray Large Area Space Telescope (GLAST) satellite is a prime example of an instrument where powerful statistical methods are needed at all stages of the data analysis -- the instrument makes indirect measurements of the incident gamma-rays, requiring complex event reconstruction to estimate the directions and energies of the photons; classification is needed to determine whether the event was caused by a gamma-ray or a charged particle; the detection and characterization of celestial sources of gamma-rays requires complex statistical hypothesis testing which is dependent strongly on the statistical characterization of the reconstruction and classification stages. In this proposal we focus on the first of these stages, on which all the others are built. The first task is to estimate the directions and energies of the incoming photons. The tracker element of the LAT is a series of 18 tungsten foils, interleaved with silicon microstrip detectors. Incident gamma-rays interact with the tungsten and are converted into electron-positron pairs. These charged particles trigger the microstrip detectors. However, the electron and positron are affected by numerous physics processes as they traverse the detector. The primary process that blurs the response is multiple scattering, which causes the tracks of the electron and positron to deviate from the ideal straight line paths. Other processes result in the ejection of further electrons from the material of the detector, and the production of further gamma-rays by positron annihilation. These gamma-rays can be converted into electron-positron pairs later in the detector. These processes and others produce charged particles that also interact with the silicon microstrips and provoke a response in the detector. The current analysis methodology being developed by the LAT collaboration involves finding the straightest tracks through the microstrip responses, a Kalman filter to estimate the trajectories of the primary electron and positron, and from the trajectories estimating the direction and energy of the photon. We propose an alternative methodology, based on nonparametric estimation using particle filters, that can more effectively use the full physics distributions for multiple scattering and other physics processes, and should give more accurate estimates of the direction and especially the energy of the incident photons and the uncertainties of these estimates - essentially a psf per event, which will enable more powerful and accurate estimation methods to be used in later stages of the data analysis. Improving the estimate of the photon's energy from thetracker response has been identified as an important goal for the event analysis. Preliminary work, funded by the NASA Ames Director's Discretionary Fund has demonstrated the applicability of the methodology to the LAT event analysis problem. This proposal is to further develop the new analysis methodology, and to integrate it into the GLEAM software environment being developed by the GLAST collaboration. A successful outcome would result in a greater science return from the mission deemed of highest priority in its category by the NRC Decadal Review, and also a wider knowledge of a new, powerful statistical methodology amongst the astrophysics and particle physics communities.
| |
| Fault-Tolerant Parallel-Processing Astronomical Image-Analysis Tools (PI: Mighell, Kenneth,
National Optical Astronomy Observatories
) | |
| This proposal describes a three-year research plan to develop and implement several new fault-tolerant parallel-processing astronomical image-analysis tools suitable for the analysis of stellar imaging data from the current and future NASA astrophysical missions. One such tool will be a fault-tolerant version of the PI's CCD Circular Aperture Photometry (CCDCAP) code for the Hubble Space Telescope's Advanced Camera for Surveys (ACS) instrument. The PI will enhance his existing AISR-funded QLWFPC2 quick-look photometry data-mining code and the MATPHOT precision PSF-fitting CCD stellar photometry and astrometry code by making them fault tolerant as well as faster and more robust. These data-analysis software tools will be written with the goal of enhancing the scientific return of the the HST and the Spitzer Space Telescope missions as well as future planned missions such as the James Webb Space Telescope. Efforts will be made to implement some of the data-mining tools at NASA data archives or mirror sites for HST and SST in order to provide an interactive and more scientifically useful user experience while possibly making the serving of archival imaging data more efficient. The fault tolerance of these applications will be achieved by working closely with the LAM/MPI 10.0 team (at the Los Alamos National Laboratory and Indiana University) to implement state-of-the-art fault-tolerant techniques into parallel-processing applications suitable for space-based scientific analysis with a cluster of COTS-based fault-tolerant computers. The LAM/MPI 10.0 team will test these space science applications inside a software test environment for fault tolerance. The PI will provide advice to the LAM/MPI 10.0 team on application implementation issues and the software needs of space science image-analysis applications. The PI will attend the quarterly LAM/MPI 10.0 meetings. Expert optical and infrared astronomers will test the software, review the documentation and make suggestions for improvements. The project software products will be released to the general space-science and computing-science communities by placing them in the AISRP Code Archive Library. Project highlights and annual reports will be posted at the main AISR website. Additional project related material will be posted on websites dedicated to this project on the web server of the National Optical Astronomy Observatory.
| |
| Automated Classification of X-ray Sources for Very Large Datasets (PI: Hojnacki, Susan,
Rochester Institute of Technology
) | |
| The enormous amount of multidimensional data generated by increasingly higher quality observatories is resulting in very large astronomical databases. The growing data archives of the Chandra X-ray Observatory (CXO) and the X-ray Multi-Mirror Mission (XMM-Newton) provide excellent examples of this conundrum. Unbiased source classification methods would augment and enhance standard spectral and temporal analysis techniques that are routinely applied to CXO and XMM data. We are proposing to develop a novel statistical source clustering and classification algorithm to maximize the science return from these and other large, multidimensional astronomical datasets. Our initial emphasis is on spectral classification of hundreds of X-ray sources detected in CXO observations of the Orion Nebula Cluster. We will generalize our classification techniques to take into account temporal attributes of X-ray sources and thereby create a robust clustering algorithm. The algorithm also will be extended for use in blindly classifying X-ray sources in other regions of the sky. Inputs and procedures specific to X-ray wavelengths will be modularized so the algorithm may be used with data from other regions of the electromagnetic spectrum. The ultimate goal is to develop the capability to group sources in large fields, such as a sky survey, independent of the requirement of a model or a priori knowledge of the nature of the sources (i.e., young stars, binaries, active galactic nuclei). Methods to visualize the results will be explored. The resulting algorithm will be developed into a tool and made available to the established centers for astronomical data analysis. The proposed research is relevant to Goal II, Astronomical Search for Origins, RFA 2(a). The final algorithm will increase science return from observational data through advanced knowledge discovery. The algorithm will be of immediate use to X-ray astronomers studying the mechanisms underlying X-ray emission to improve our understanding of the nature and timescale of accretion onto young, solar-mass stars from protoplanetary disks.
| |
| Certificate based NVO wide Authentication and Authorization driven by provision of local database st (PI: Szalay, Alexander,
Johns Hopkins University
) | |
| Recently the International Virtual Observatory Alliance (IVOA) has published a proposal for the basic building block of the Virtual Observatory, called an OpenSkyNode. This specifies a simple, queryable interface to the underlying archives of astrophysical data. This standard interface also enables interoperable services and portals built on top of the OpenSkyNodes. After studying the usage patterns of several large archives available at JHU (SDSS, SkyQuery), we have designed an experimental queue management system for large queries that also enables users to maintain their own value-added data next to the main archives, and perform a step-wise analysis there. We have built an experimental Batch Query System with local user-owned databases (called MyDB). These ideas were presented at ADASS XIII in Strasbourg. We propose to further develop and extend our MyDB and batch query prototypes into multiple, distributed nodes of the Virtual Observatory. The introduction of asynchronous queries necessitates temporary user space as well as job submission and tracking facilities. Beyond building a reference application, we will also specify a standardized, generic SOAP interface for these tasks so that even if the engines for managing the jobs differ, a common interface would exist for submission and querying of jobs independent of platform and architecture. We would like to explore a distributed system whereby multiple nodes form a trusted network (Virtual Organization) and users with an account (i.e. authentication) on one node would be accepted at other nodes without needing to register separately. We foresee users having datasets in multiple ''spaces'' on multiple machines. We propose to build a portal and appropriate protocols to enable the display of a given user's collections and the status of the respective workflows. This will require nodes to support queries about the user databases and jobs held or submitted by users, and a certain level of authentication (this time of the portal) and authorization to make such queries. Interaction with other initiatives similar to MyDB, such as AstroGrids's MYSPACE will allow us to experiment with issues such as how exactly the Virtual Organization will work i.e. if I am authorized for MyDB by NVO should AstroGrid allow MYSPACE? In this area we shall collaborate with the AstroGrid community, a letter of support is included from Tony Linde project manager of AstroGrid. Savas Parastatidis of the University of Newcastle upon Tyne UK is particularly active in the field of security and web services and we are delighted to collaborate with him in this area. We are requesting funding for 3 years to cover approximately 3 person years of work. A personnel summary follows. All software generated through this research will be Open Source and fully available for anyone.
| |
| Stardust@home: A massively distributed public search for interstellar dust in the Stardust Interste (PI: Westphal, Andrew,
University of California Berkeley
) | |
| In January 2006, the Stardust mission will return the first samples from a solid solar-system body beyond the Moon. Stardust was in the news in January 2004, when it encountered comet Wild2 and captured a sample of cometary dust. But Stardust carries an equally important payload: the first samples of contemporary interstellar dust ever collected. Stardust uses aerogel collectors to capture dust samples. Identification of interstellar dust impacts in the Stardust Interstellar Dust Collector probably cannot be automated, but will require the expertise of the human eye. However, the labor required for visual scanning of the entire collector would exceed the resources of any reasonably-sized research group. Here we propose to develop a project to recruit the public in the search for interstellar dust, based in part on the wildly popular SETI@home project, which has five million subscribers. We call the project Stardust@home.
| |
| Knowledge Discovery in a Virtual Universe (PI: Gardner, Jeffrey,
Carnegie Mellon University
) | |
| Astrophysics is witnessing a flood of data from new ground and space based telescopes and surveys. With access to data sets spanning X-rays through to radio wavelengths we will soon be able to extract sources from any region of the sky and measure their properties across the full range of the electromagnetic spectrum. Individually, each of these data sets has the potential to advance our understanding of the processes that drive the formation and evolution of our Universe. It is, however, only when these data are combined that their full scientific potential will finally be realized; the scientific returns from the total will far exceed those from any one individual component. The recognition of this by the Astronomical community has led to a major new initiative: the National Virtual Observatory (NVO; http//www.us-vo.org). If we are to realize the full potential of the era of the NVO we must be able to explore, analyze and interact with these massive datasets. We need tools for knowledge discovery that not only make data delivery fast and easy to learn but also enable rapid data analysis on a massive scale. The resources required to analyze these terabyte and larger databases can be found if we can harness the power of the upcoming generation of distributed computing resources (i.e. the Grid). The challenge we face is how do we help a user to realize the scientific value of the data without being limited by the laborious and complicated efforts required to implement data analysis techniques that span many processors distributed across a grid? Can we not only make data delivery fast and easy to learn (as the NVO is doing), but can we also minimize the time that it takes an astronomer to use that data to actually find an answer? This proposal will address the challenge of knowledge discovery in the era of the NVO and the Grid; providing the final component necessary to link the massive astrophysical data sources to massive distributed computational resources. We will develop new parametric and non-parametric algorithms and techniques that solve common problems in astrophysical data analysis and that are optimized for distributed computing environments. This research program has the potential to impact not only space science and the NVO, but also any NASA missions that generate large multidimensional datasets.
| |
| Precision High-Speed Crowded-Field Image Analysis in Space (PI: Lehner, Matthew,
University of Pennsylvania
) | |
| We propose to develop a photometric reduction software package implementing the Expectation-Maximization (EM) Algorithm, an iterative Maximum Likelihood method. This algorithm is highly suited to PSF fitting on images of crowded fields where blending makes standard fitting routines difficult and time-consuming. Such a solution would be extremely important in cases such as analysis of crowded images with very large focal planes, or onboard analysis of images from space-based instruments where optimized useage of CPU cycles is paramount. The software will be capable of fitting galaxy shapes and a variety of PSF shapes.
| |
| Segmented Nonparametric Models of Distributed Data: From Photons to Galaxies (PI: Scargle, Jeffrey D.,
Ames Research Center
) | |
| A novel technique, and an efficient algorithm to implement it, provides piecewise-constant models for a variety of one dimensional data types common in high energy astrophysics and cosmology, as well as in physical simulations of astrophysical systems. We will improve this algorithm in various ways (speed, parameter selection, alternative fitness functions, etc.) However, the major task will be extension of this methodology to 2D data (such as galaxy surveys), 3D data (redshift surveys), and higher dimensional data spaces. We have recently discovered a way to transform higher dimensional problems into approximately equivalent 1D cases, solvable with the 1D algorithm. A major mathematical goal of this work is to find rigorous solutions in higher dimensions. We will simultaneously derive scientifically important results and hone the methods on analysis of GLAST photon maps, 3D galaxy positions derived from redshift surveys, and stellar infrared data useful for detection of molecular clouds. Ultimately these methods will be applicable to a wide range of astrophysical problems.
| |
| The Journey of the Sun-A Virtual reality Simulation: Date-Constrained Modelling and Visualization (PI: Frisch, Priscilla C.,
University of Chicago
) | |
| The goal of this research is to construct effective computer grpahics tools to create a navigable 3D data-constrained model of the galactic environment of the Sun, and render the models using state-of-the-art computer visualization techniques. Sophisticated modern methods of eveidence assimilation, probabilistic reasoning, and inference will be combined with extant multi-spectral data to reconstruct unknown portions of the measured data, which can then be modified by the expert user.
| |
| Development of an IUE Time Series browser (PI: Massa, Derck,
SGT, Inc.
) | |
| The International Ultraviolet Explorer (IUE) satellite operated successfully for 17 years. Its archive of more than 100,000 science exposures is widely acknowledged as an invaluable scientific resource that will not be duplicated in the foreseeable future. We have searched this archive for objects which were observed 10 or more times with the same wavelength coverage and spectral dispersion over the lifetime of the satellite. Using this definition of a time series, we demonstrate that roughly half of the more than 100,000 science exposures are members of such time series.identifying these datasets and developing a means to easily examine them in order to determine which objects varied over the lifetime of IUE would be an extremely useful scientific asset. We have already identified the datasets and developed a prototype for a browser which operates over the web. The browser enables the researcher to visually inspect the repeated observations for variability and to examine each member spectrum individually. Further, once the researcher ascertains whether a specific dataset is worthy of further investigation, the entire dataset can be downloaded through links on the browser page for further scrutiny on the researcher's home computer. This project will produce a fully operational version of the browser.
| |
| Integrated Visual Simulation Pipeline - an extension of the Scientist's Expert Assistant (PI: Koratkar, Anuradha P.,
University of Maryland Baltimore County
) | |
| We propose to research and prototype a fully integrated visual astronomical pipeline linking data archives, scientific analysis tools, and new proposal preparation systems.
NGST funded the initial development of The Scientist's Expert Assistant (SEA) to research new visual approaches to proposal preparation. In the last call, AISR provided the seed funds to extend
SEA with initial scientific modeling tools and generalizing the SEA model to multiple observatories. SEA's success to date has already led to its adoption by the Space Telescope Science Institute for
production use with Hubble's observing program. SOFIA is in the initial stages of adapting SEA for their observing programs. SIRTF has adopted parts of the code and the MIDEX mission KRONOS
is considering SEA as the baseline for its user support strategy.
Building on this experience, we plan to extend both the scientific simulation capabilities and the data archive visualization into an integrated pipeline. The goal is a simulation pipeline that will allow the user to manage the complex process of simulating and analyzing images without heroic programming effort. Tying this into SEA will allow astronomers to effectively come "full circle" from retrieving archival images, to data analysis, to proposing new observations.
This simulation pipeline will increase scientific return within limited resources by improving the quality of observations and reducing the number of unusable observations. This is not only a valuable tool for traditional observatories and archives; it is also a key building block to allow the future Virtual Observatory to achieve its potential.
| |
| Novel Approaches to Supervised and Unsupervised Data Exploration (PI: Bazell, David,
Eureka Scientific
) | |
| n this proposal we examine two novel approaches to the exploration and understanding of astronomical data: the use of unlabeled data for supervised classification and semi-supervised clustering. Because of the large and increasing volume of data from astronomical satellites and ground-based telescopes, researchers can no longer hope to examine the data by hand. Automated techniques are essential lest important discoveries be lost. Current automated classification methods rely on supervised learning algorithms, such as neural networks and decision tree inducers, that require training set containing large amounts of previously classified, or labeled data. While unlabeled data is often cheap and plentiful, using a human to classify the data is tedious, time consuming and expensive. We will develop methods whereby supervised classification techniques can make use of cheaply available, large volumes of unlabeled data to substantially improve their ability to classify objects. If the target classes are unknown, unsupervised clustering is a standard method of exploring unknown data and partitioning it into useful groups. We will also implement and explore several semi-supervised clustering methods. Finally, while classifier learning based on mixed labeled/unlabeled data and semi-supervised clustering can be viewed as separate problems, we will develop and evaluate a unified framework where the learned models directly provide clustering or classification solutions, or both, depending on the needs of the user. This approach allows assignment of astronomical data to predefined classes and will facilitate the discovery of new object classes. To demonstrate the utility of these methods, we will apply them to several test problems of current astronomical interest. Our primary scientific scenario will be aimed at identifying galaxy mergers in a variety of large (unlabeled) catalogs using labeled data from both simulations and observations. We will also examine morphological galaxy classification using data from existing galaxy catalogs and the Hubble Medium Deep Survey. These several data sets contain different types of features used for classification. Both structural (e.g. shape and texture) and photometric features will be tested. The existence of different types of features is directly relevant to the co-training method, one of the methods we will investigate for building classifiers based on labeled and unlabeled data.
| |
| Parallel-Processing Astronomical Image Analysis Tools for HST and SIRTF (PI: Mighell, Kenneth,
National Optical Astronomy Observatories
) | |
| This proposal describes a two-year research plan to develop and implement several parallel-processing astronomical image-analysis tools for stellar imaging data from the Hubble Space Telescope and the Space Infrared Telescope Facility. The new data analysis software tools will be written with the goal of enhancing the scientific return of the HST and SIRTF missions as well as future planned NASA astrophysical missions like the Next Generation Space Telescope. This project combines the enabling image-processing technology of the Principal Investigators new digital PSF-fitting MATPHOT algorithm for accurate and precise CCD stellar photometry with enabling technology of Beowulf clusters which offer excellent cost/performance ratios for computational power. Data mining tools to do quick-look stellar photometry and other scientific visualization tasks will also be written and used in order to investigate how such tools could be used at the data servers of NASA archival imaging data like the Space Telescope Science Institute. Such data mining tools have the potential of making the serving of archival imaging data more efficient while simultaneously providing a more rewarding user experience for visiting astronomers. This project is structured into two one-year phases. The major software products (including original source code and detailed documentation) developed as part of this research project will be tested by expert optical and infrared astronomers before being released to the general astronomical community. The project software products will be posted (at the end of each one-year phase) on web sites dedicated to this project on the web server of the National Optical Astronomy Observatory.
| |
| The Journey of the Sun -- A Virtual Reality Simulatin: Data-Constrained Modeling and Visualization of Interstellar Matter in our Galaxy (PI: Frisch, Priscilla C.,
University of Chicago
) | |
| The richness of objects in the two-dimensional sky overhead belies the sparceness of space. The objective of this proposal is ti give context to this empty space by reconstruction and visualizing the Galactic surroundings of the Sun and nearby stars, including both spatial variations and temporal changes, such as stellar motions and the movement of interstellar matter (ISM) through space. The proposed research provides a new approach to data-interpretation by integrating multispectral ISM data (e.g., absorption and emission lines) with astrometric data on stellar motions to create coherent data-constrained models for 3D volumetric fields of the ISM in the Galactic neighborhood of the Sun. State-of-the-art computer techniques are used to manipulate, interact with, and visualize these models. The proposed tools and software can be extended to reconstruct 3-dimensional models of the volume distribution of any material giving rise to absorption lines, provided the distances of the background target objects are known. Our strategy is to merge cloud positions and velocities from the absorption data with two-dimensional emission line data (e.g., the HI 21-cm line) to fill in regions not sampled by the absorption lines. We will develop model visualization techniques that will use virtual reality methods augmented to handle continuous space-time navigation over scale sizes ranging from 1 AU to many megaparsecs. A single simulation could show the orbits of near-earth asteroids at a few AU from the Sun, the space-trajectories of the stars in Orion, the distribution of galaxies measured by the Sloan Digital Sky Survey and a 3D visualization of the Hubble Deep Field objects for which redshifts are available. The proposed research supports the National Virtual Observatory concept by providing computer-assisted methods for interpretation of ISM emission and absorption data, and by producing new methods for visualizing these data using virtual reality simulations and navigation tools adapted to the large spacetime scale ranges required for astronomical applications.
| |
| A High-Speed Data Access Component for a National Virtual Observatory Data Grid Node (PI: Thakar, Ani,
Johns Hopkins University
) | |
| We propose to augment the Science Archive of the Sloan Digital Sky Survey and employ it as a testbed for the development of a high-speed data access layer for a National Virtual Observatory data grid node. Our goal is to produce a scalable, distributed archive framework that uses standardized software toolkits and interfaces so as to provide a prototype for grid-compatible astronomical archives. We aim to achieve this by developing a MPI-based parallel, distributed system that features loosely-coupled clusters of compute and data nodes. The current design of the SDSS Science Archive already features a high degree of parallelism and scalability. However, it uses a proprietary, non-portable communication protocol and data access primitives that are specific to the object-oriented DBMS that serves as its data repository. The proposed re-engineering aims to use MPI to provide high-speed communications and portability in order to obtain a data access layer that is optimized for parallel communication, independent of the underlying DBMS architecture, and conforms to the NVO standards and protocols.
The virtual data grid that is expected to be the backbone of the NVO data services infrastructure will require individual archives to be structured in a data-parallel paradigm and be able to respond to requests for data and metadata in a fast, efficient, and standardized fashion. The successful deployment of the proposed component will also provide the high-speed data access required for the fast cross-identification and matching algorithms using co-operative agent technology that we are currently in the process of developing as part of an AISRP 2000 funded project.
This proposal relates to the fabric layer of the proposed NVO architecture, and requests funding for two years for partial salary support and travel support for the PI, travel support for one Co-I, and full support for one post-doc at JHU reporting to the PI. The project plan also includes collaboration with the Royal Observatory of Edinburgh (ROE) and the Edinburgh Parallel Computing Centre (EPCC) for MPI expertise. Additional travel support required for this collaboration is included in the budget.
| |
| Scalable Algorithms for Fast Analysis of Megapixel CMB Maps and Large Astronomical Databases (PI: Szapudi, Istvan,
University of Hawaii, Manoa
) | |
| We propose to develop a suite of algorithms for spatial statistical analyses of future large astronomical data bases, such a megapixel CMB maps, galaxy catalogs, or any point source catalog. We will tackle hitherto unsolved computational challenges with high accuracy, such as: i) fast simulation of correlation functions and angular power spectra, ii) fast estimation of $n$-point correlation functions and iii) counts in cells statistics in from large CMB maps, angular, redshift and weak lensing surveys. We develop a unique interdisciplinary approach for algorithmic development, in which we reformulate the statistical problems arising in space astronomy with special attention to computational needs. This approach heavily relies on blending the principles of astronomy, statistics, computer and computational science, and balances statistical accuracy vs. practical computability.
| |
| Galaxy Photometry for NASA/IPAC Extragalactic Database (PI: Schombert, James,
University of Oregon
) | |
| This project's goal is to provide a set of galaxy photometry tools to NASA/IPAC Extragalactic
Database (NED). The system will allow new and experienced researchers to perform a full surface
photometry analysis of any 2D data in NED's holdings or, by downloading the package, on their own
optical or near-IR datasets. A heavy emphasis has been placed on interactive/visualization tools plus
substantial documentation for the researcher new to the field of galaxy photometry. This project's
modest budget will add NVO-type functionality to NED and act as a prototype to future expansions
for data centers.
| |
| Classifying the High Energy Universe: A Prototype of the National Virtual Observatory (PI: McGlynn, Thomas A.,
Goddard/USRA
) | |
| We propose to prototype the National Virtual Observatory, building the tools and protocols needed to integrate large, distributed datasets to do science infeasible with a single institution's resources. Our pilot research project is to build an automated classifier for X-ray sources and to use it to try to
distinguish the physical classes of all known X-ray objects. This project is carefully selected as challenging but feasible, excellent science in its own right, and exercising the key technologies for the
NVO.
Massive datasets from the HEASARC, ST-MAST, and Chandra archives along with information from VizieR, 2MASS, FIRST and other systems will all be needed within this effort. The deliverables will include systems for coherent access and integration of information from many
different astronomy data providers, the methodology and software for the creation of automated classifiers in the NVO regime, a realistic assessment of the current connectivity of astronomical sites,
and our X-ray classifier itself. Our clear science context will help ensure that the tools we develop are truly useful to the astronomy community.
The science goals are particularly timely given the recent launches of the Chandra and XMM-Newton observatories. The number of known X-ray sources will soon double, but our current tools for categorizing them are extremely tedious and resource intensive. Our classifier must handle quite heterogeneous data. We propose using a network of sub-classifiers of several different types to address the diversity of data available.
This proposal leverages substantial institutional commitments from the collaborating groups. Funding is requested only to support the actual software development activities involved in this project. The
science oversight and astronomical research activities are provided by the collaboration from other institutional resources.
| |
| Cosmic Microwave Background Analysis Tools (PI: Borrill, Julian,
Lawrence Berkeley National Laboratory
) | |
| The Cosmic Microwave Background radiation provides a unique picture of the early universe. To realize its scientific potential NASA is leading an international effort to obtain precise measurements
of the CMB temperature and polarization from ground-based, balloon-borne, and satellite observatories. This proposal supports the development and distribution of the novel computational algorithms and implementations needed to plan observing strategies, simulate observations, and analyze data.
The CMB temperature power spectrum shape is a strong constraint on cosmologies. Its measurement requires a large area of sky observed at high resolution by many detectors at many frequencies. The
volume of such data makes their analysis a serious computational challenge; our MADCAP software can analyze maps with O(100,000) pixels but new data will give up to O(10,000,000) pixels,
requiring new approximate algorithms. New methods are also needed to analyze multiple channels to account for systematic effects from their cross-correlation and their different frequencies and beam
sizes.
CMB polarization provides a new window into the very early universe. Currently in the detection phase, the challenge is to observe a signal much fainter than the temperature anisotropies, requiring
well-planned observations and novel analyses. We will develop polarization extensions to our FORECAST and WOMBAT flight-planning tools, as well as first-generation simulation and analysis
tools.
To compare the CMB temperature or polarization power spectrum with cosmological models we search a 10-20 dimensional space of correlated parameters to get both the maximum likelihood parameter combination and the full likelihood contours. Since the parameter-likelihoods are
non-Gaussian formally we need to calculate them throughout this infinite space. In practice we must restrict ourselves to a finite sampling of a finite subset of the space: our goal is to codify and optimize the appoximations in this restriction.
| |
| An Automated System for the Reduction and Analysis of X-ray Data from Galaxies (PI: Ptak, Andrew F.,
Johns Hopkins University
) | |
| Much an observer's time is spent on repetitive data reduction, and
likewise certain aspects of data analysis such as determining errors
for spectral fits is also repetitive. Often, particularly early on
in a mission's life, data reduction and analysis need to be repeated
as calibration data and tools are updated. We propose the
development of a system for the automation of the reduction and
analysis of X-ray data from galaxies for the missions ASCA, ROSAT,
AXAF and XMM. For this specific scope, the system will handle all of
the tasks that could appropriately be handled in an automated
fashion. This will free a considerable part of the astronomer's time
for more advanced analysis of the data products. Furthermore it is
likely that it will not be feasible for astronomers to complete many
interesting projects without such a system in place given the large
amount of time required and the volume of the data involved in X-ray
research. This problem will become increasingly acute as future
X-ray missions produce more voluminous and complex data sets.
The system will be based mostly on scripts that drive existing
software (such as FTOOLS and XSPEC) but will also consist of advanced
(also scriptable) software already in development at CMU for spatial
and temporal analysis of complex X-ray data. A large emphasis will
be placed on ease-of-use and extensibility. This system will be
useful for both "legacy" analysis of a given galaxy (i.e., a complete
spectral, spatial and temporal analysis of the publically available
data from the supported missions) and survey studies (analysis of
deep fields and searches for serendipitous sources). A primitive
early version of the system has already been successfully used to
analyze and re-analyze the ASCA data from a sample of galaxies. As
the availability of funding for researchers is limited, a system such
as this will be of considerable value in assuring that as much X-ray
data as possible in the archives and forthcoming from missions such
as AXAF and XMM are analyzed and published. Following completion of
the initial version of the system, the system will be expanded, e.g.,
to include other extragalactic data such as clusters and groups, and
to include tools for multi-wavelength data analysis. The successful
completion of these goals will be particularly useful to institutions
will a limited number of X-ray-proficient personnel who nevertheless
wish to take advantage of the X-ray data available from these (and
future) missions. Since the products of the system will include
"calibrated" images, spectra and lightcurves as well as "physical"
parameters from fits to these products, such as temperatures and
spatial extents, the system could be applied to entire archives to
produce databases for use with future data-mining projects. We will
also explore the visualization of the final results (particularly
spatial results), to aid in the scientific understanding of the data
and for public outreach.
| |
| Advanced Galaxy Cluster Detection Software (PI: Nichol, Robert,
Carnegie Mellon University
) | |
| The Sloan Digital Sky Survey (SDSS) is one of the first of many NASA sponsored missions to produce a trillion pixel image of the sky. A main objective of the SDSS is to determine the spatial clustering of galaxies in the universe. With the advent of large, high-precision surveys like the SDSS, it is now time to apply advanced algorithms taken from statistics, signal processing and computer science to these data to quantify the degree of clustering seen and thus robustly find clusters of galaxies within these large datasets. Over the past five years, the authors of this proposal have collaborated with statisticians, engineers and computer scientists to test a variety of new algorithms on astronomical data. We now come together to produce -- for the benefit of the community -- a suite of high performance cluster detection software that synthesizes the best of all our experiences.
In addition, we will use this suite of software to produce the definitive galaxy clustering database for the SDSS. This will be achieved as follows: First, we will develop the test bed for our detection code. This will consist of publicly available software for generating simulated data sets with known clustering properties. These simulations are critical for defining the selection function of any cluster survey. Second, we will compare the different detection algorithms that have been developed by the team members. Based on these comparisons, we will select the best combination of detection
algorithms, and implementation techniques, and use these to produce an open source software suite. Finally, we will apply the software we have developed to the SDSS galaxy database and produce a
robust, well-understood and well-parameterized catalog of many thousands of clusters. The broad applicability of this software is critical to its success and should be useful for the Large Synoptic
Survey Telescope, the "Virtual Observatory" and many other NASA missions like 2MASS, GALEX, Chandra.
| |
| WITS the WEB Infrared Tool Shed (PI: Wolfire, Mark G.,
University of Maryland College Park
) | |
| Using a previous NASA ADP type 2 grant, we created a well known and user tested WEB based tool for the analysis of dust continuum and PhotoDissociation Region (PDR) observations. The WEB Infrared Tool Shed (WITS) contains two "toolboxes", the PDR Toolbox (PDRT) and the Dust InfraRed Toolbox (DIRT) and are currently available at http://dustem.astro.umd.edu or http://wits.ipac.caltech.edu. These toolboxes provide an extensive data base of PDR and dust continuum models which can be "mined", displayed, and manipulated using a Java user interface.
This proposal seeks to enhance these tools with an expanded data base of model sources, with additional algorithms and user options to find the best fit models for an input data set, and with modules tailored for the spectral and spatial responses of specific NASA instruments.
| |
| Multifrequency, Multiresolution Image Detection (PI: Connolly, Andrew John,
University of Pittsburgh
) | |
| The
next decade will mark the start of a new era in multi-frequency
astronomy. New surveys and satellites are coming on line that will open
up the full electromagnetic spectra at higher precision and to fainter
flux levels than ever before (from X-rays through to the submillimeter
and radio). Together these multi-frequency surveys will provide a
bolometric view of the Universe from which we might fully reconstruct
the properties of Galactic and extra-galactic sources. Studying these
data sets in isolation will provide some insight into the processes
that drive the evolution of our Universe. It is only, however, if we
analyze the multi-frequency data in unison that we might fully exploit
their scientific potential. We have, therefore, developed a new and
novel technique for combining multi-frequency data that provides
objective source detection. We propose here to develop these algorithms
into a suite of software tools that will enable the astronomical
community to explore multi-frequency, multi-resolution data in an
optimal fashion.
| |
| A Parallel Genetic-Algorithm-based approach to White Dwarf Pulsation Models (PI: Winget, Donald E.,
University of Texas Austin
) | |
| White dwarf asteroseismology offers the opportunity to probe the
structure and composition of stellar objects governed by relatively
simple principles. The observational requirements of asteroseismology
have been addressed by the development of the Whole Earth Telescope
(WET) in conjunction with the HST, but the analytical procedures need
to be refined before this technique can yield the complete physical
insight that the data can provide. We propose to develop an
optimization method utilizing a genetic algorithm (GA) for fitting
white dwarf pulsation models to the observations. We will develop
software to use this global optimization approach in two distinct
ways: (1) to investigate the uniqueness and objectivity of the
solutions by combining existing models with a GA-based fitting
routine, and (2) to investigate the completeness and adequacy of our
understanding of the principles governing white dwarf interiors by
parameterizing the constitutive physics and using a GA to find the
family of solutions that produce observationally indistinguishable
behavior. We have configured a specialized computational instrument to
develop this software---a parallel metacomputer consisting of a
network of 64 minimal PCs running Linux. The structure of a GA is very
conducive to parallelization, so this metacomputer will allow us to
develop and run the code on a practical timescale.
| |
|
|
|
|
|
