 |
|
|
|
|
|
| Sun Projects |
|
| Development of an Adaptive Non-Ideal MHD Simulation Tool for Multiple Space Science Applications (PI: Gombosi, Tamas,
University of Michigan
) | |
| Our overarching goal is to develop a powerful, flexible, general use simulation tool applicable to a large variety of space science
problems. Over the last decade our group at the University of Michigan has developed a highly parallelized adaptive mesh refinement
scheme for solving the equations of ideal and resistive magnetohydrodynamics. The BATS-R-US code makes use of robust
high-resolution upwind finite-volume schemes and explicit and/or fully implicit time stepping to solve the time-dependent 3D MHD
equations in conservation form. A flexible block-based hierarchical tree data structure is used to facilitate automatic solution
adaptation on a logically Cartesian grid. With the support of a previous AISRP grant the proposing team has added the capability of
solving the Hall-MHD equations that enables a more realistic modeling of space science phenomena, such as magnetic reconnection.
Modern global simulation tools are now widely used to address a variety of space science problems. However, there are some extremely
important physics and performance limitations to our simulation capabilities. We plan to remove some of these limitations to extend
the physical validity and efficiency of space plasma simulation codes. Specifically: * We propose to develop a multi-fluid Hall MHD
code with anisotropic pressure. This ambitious goal will be approached through intermediate stages: multi-fluid MHD, single-fluid
MHD with anisotropic pressure, multi-fluid Hall MHD with isotropic pressure, and finally multi-fluid Hall MHD with anisotropic
pressure. * We propose to implement a new phenomenological resistivity model to efficiently model reconnection. * In order to mitigate
the computational cost we propose implement various subcycling algorithms into our code. This algorithmic improvement will benefit
many space physics applications that requires high resolution in a smallfraction of the computational domain. The BATS-R-US code is
already available to the space science community. The new algorithms will also be freely available to maximize the potential benefits.
| |
| Magnetogram Synthesis - A Vital Data Analysis Component of A Space Weather Prediction Infrastructure (PI: MacNeice, Peter,
NASA Goddard Space Flight Center
) | |
| We propose to develop a toolkit for synthesis of solar magnetogram data. The toolkit will take both line-of-sight and vector
magnetogram data from all available sources, both ground based and satellite. It will craft the optimal field and surface flow solutions
consistent with the data, for user defined surface grids at times identified by the user. The Space Weather community needs time
dependent models of solar active regions. A wealth of new data sources, particularly vector magnetogram data, is about to become
available to facilitate this effort. However no tool exist which allows the modelers to analyse all the available data and craft accurate
and physically consistent boundary conditions for their models. Our proposed tool will fill this gap. We will develop it as an application
of the KAMELEON data formating and interpolation toolkit which has been developed by the CCMC.
| |
| Extending Virtual Observatory Capabilities Through Grid Enabled Resources (PI: Szabo, Adam,
NASA Goddard
) | |
| Existing and forthcoming Virtual Observatories will enable unified access to many disparate data sources and services. As these systems expand they face issues in scalability. In particular, user contributed services and applications, which play a vital role in long-term success, are hindered by volume of use and lack of hosting facilities. We aim to address this issue by establishing the infrastructure for a VO-Grid. The Globus Toolkit, the de facto open source standard in Grid software, will be used to establish a mechanism for allowing user contributed submissions of services. Additionally, the infrastructure for allowing these services to be shared along with unused processor time and storage space will be set in place.
| |
| Building and Implementing a Virtual Sun-Solar System (Great) Observatory (VS3O) (PI: Roberts, Aaron,
NASA Goddard
) | |
| We propose to build on the success of the Virtual Space Physics Observatory (VSPO) and related efforts to realize the potential of Virtual Observatories to unite data and service providers in the Sun-Solar System domains into a new "Great Observatory," VS3O. Current plans at NASA and in the international, multi-agency Sun-Solar System Connection community include making data easily available from all missions relevant to the global problem of the effects of solar plasma dynamics on the Earth and other planets. To accomplish this, Virtual Observatories register data products and services from disparate providers using a common language that allows searching and use in a uniform way. Subfields may be organized along the lines of the now operational Virtual Solar Observatory, but more is needed to realize the full potential of these systems. We will solve the following critical problems in implementing VS3O: (1) developing a suite of software to allow resource providers to easily transform whatever system they have for accessing their data into a service for VOs; (2) providing an API and useful visual and other (e.g., IDL) front-ends to the system; (3) building on current efforts to provide a "mission independent data layer" to allow wide ranges of data and models to be dealt with uniformly; (4) integrating the independently developed "higher order query" service along with subsetting and other services we will develop into VS3O; (5) incorporating parallelism as needed to make the VS3O responsive to large numbers of complex queries; (6) developing community outreach, feedback, and oversight mechanisms that assure broad participation and high utility; and (7) beginning efforts to incorporate the resulting Observatory into public and educational outreach. This system will be flexible, extensible by others, and will greatly aid in doing focused research projects as well as facilitating large-scale research efforts. We will develop operational modules for various functions that will also serve as prototypes for others to use the VS3O in currently unforeseen ways. The PI and team we have assembled has strong connections to most of the current efforts in this area and has a proven track record of fundamental accomplishments.
| |
| Interactive out-of-core visualization of very large multiresolution time series data (PI: Bergeron, R Daniel,
U of New Hampshire
) | |
| One of the major challenges facing scientists today is effective analysis of the explosion of data that is being generated by current scientific research technology. This is particularly a problem in the simulation of time-dependent magnetohydrodynamics (mhd) phenomena that produces enormous data sets of many hundreds of gigabytes. It is also the case that interactive visualization of the data is still one of the most effective techniques for gaining insight into the physical phenomenon. The huge size of the data, however, severely limits the ability to visualize the data in an effective interactive environment. A promising solution to this dilemma is to represent the data at multiple resolution levels - the coarser resolutions are used for large-scale overview visualizations and increasingly finer resolutions are accessed as the user zooms in for more limited range views.
This project builds on previous work by the computer science principal investigators in the areas of multiresolution representation of scientific data and multidimensional caching and prefetching.
We propose to develop a software framework and a prototype visualization environment that will support the interactive visualization of three-dimensional time-dependent mhd simulations using multiresolution and adaptive resolution data representations. An integral component of this research will be the use of very efficient I/O caching and prefetching techniques that are critical for achieving interactive performance. Although the conceptual framework applies to a very wide range of scientific data, the prototype visualization environment will focus on space science applications, particularly magnetohydrodynamics simulation and computational fluid dynamics. These areas represent two extremely important science research domains within the NASA research mission. The proposed work specifically addresses the AISR goal of providing better tools to enhance the scientists' ability to do research.
| |
| Auroral Phenomenon Localization, Classification, and Temporal Evolution Tracking (PI: Newman, Timothy,
UA Huntsville
) | |
| In this project, techniques that will aid investigation of auroral phenomena will be developed. The techniques to be developed will be automated and shape model-based and will allow accurate localization of the auroral oval (in the presence of unfriendly background noise) from UVI and IMAGE FUV data. The techniques will provide useful input to downstream analysis tasks, including classification of auroral features and exploration of auroral feature temporal evolution. New, robust automated techniques for feature classification and for exploration of temporal evolution will also be developed. All of these techniques will be incorporated into a unified auroral phenomenon search tool.
| |
| Innovative Techniques for Producing Corrected Synoptic Maps for a UVCS Outflow Velocity Database (PI: Strachan, Leonard,
Harvard/SAO
) | |
| The primary objective of this project is to develop a set of software tools that will allow researchers to obtain empirically derived plasma parameters of the solar wind acceleration region of the corona. We will produce line-of-sight corrected synoptic maps of outflow speeds and kinetic temperatures for both hydrogen and O5+ ions in the extended solar corona. Such information is valuable for constraining theoretical solar wind models and it can be used in Space Weather models that predict ambient solar wind conditions. A set of visualization tools for displaying the derived coronal parameters will also be developed and made available to the public.
The derived outflow velocities will be based on well-proven UVCS/SOHO spectroscopic techniques. Synoptic maps of various parameters will be produced using software that has been developed for solar tomography applications. Extensive use will be made of open source software for developing visualization tools for displaying the data.
All software developed from the project will be applicable to other NASA programs that require the use of 3-D data rendering such as the STEREO mission.
| |
| Novel Higher-Order Statistical Method for Extracting Dependencies in Multivariate Geospace Data Sets (PI: Johnson, Jay,
Princeton U
) | |
| Understanding magnetospheric dynamics and
the relationship between the solar wind driver and the magnetospheric
response is of great practical interest because it could potentially
help to avert catastrophic loss of power and communications. In order
to build good predictive models it is necessary to understand the most
critical nonlinear dependencies among observed plasma and
electromagnetic field variables in the coupled solar
wind/magnetosphere system. We propose to develop and apply a novel
cumulant-based information-dynamical measure in conjunction with other
nonlinear techniques to magnetospheric data to characterize: (1) the
underlying dynamics, (2) the nonlinear behavior of the geospace
systems, (3) the solar cycle dependence of these properties, and (4)
the predictability of the systems. Because this nonparametric,
statistical approach assumes no intrinsic underlying dynamics, it is
possible to avoid some of the pitfalls involved in parametric modeling
and/or physics based models. Moreover, the information gained from the
cumulant-based information flow could be used to guide the development
of predictive models.
To illustrate the power of our method, we will examine the nonlinear
cross dependencies in a large database of geospace variables. We will
evaluate the underlying dynamics of the magnetosphere by examining the
time evolution of geomagnetic indices, Kp, Dst, and AE, as well as
more direct measures of the magnetospheric state, energetic electron
flux, tail stretching, and energetic electron precipitation. From
this data we will be able to estimate a predictability horizon which
will indicate the maximum ``look ahead'' for which the space climate
can be predicted. We will also use the information-dynamical measures
to identify variables derived from solar wind data that maximize the
information content about the magnetospheric response.
This work should have a significant impact on
space science, space/mathematical science education, and society at
large. First, we have developed a new technique for analyzing
nonlinear dependencies in large data set using cumulant-based
significance measures. Because the methods and techniques do not
presuppose an underlying dynamics, they are transportable across a
wide range of complex nonlinear systems which can broadly impact
fields ranging from geoscience to bioscience. Second, detection of
the most important nonlinear interactions in the systems could be used
to identify the most important physical processes that will aid
development of physical models and predictive capabilities for
understanding severe space weather. Third, we will submit an E/PO
proposal where we will develop web-based outreach
tools to teach high school students basic concepts in nonlinear
dynamics with applications to space science. Finally, the
results will be widely disseminated in mathematical and geospace
journals and presented at meetings and workshops.
| |
| Sun Earth Connection Distributed Data Services (PI: Cornillon, Peter,
OPeNDAP
) | |
| The goal of the work proposed herein is to integrate a number of current IT efforts into a suite of data services designed to facilitate the access to and use of data for Sun Earth Connection studies. The collection of services to be developed is based on the OPeNDAP Data Access Protocol (DAP), a protocol in heavy use in the Earth sciences. Although these services will be developed for Sun Earth Connection studies, we anticipate that they will be of use to other areas in the sciences as well. Specifically, we will make use of some of these enhancements in ongoing data system work in oceanography; i.e., this project will leverage work done in the Earth sciences to benefit space physics and the work done to benefit space physics will feed back to the Earth sciences. General Web services to be developed or enhanced as part of this effort: - High bandwidth data access - OPeNDAP-g, a Grid-enabled version of OPeNDAP, will be re-integrated into the OPeNDAP core such that all OPeNDAP core services are Grid accessible. - Catalog access integrated with content-based access - A hierarchical cataloging capability being developed at UCAR will be integrated into the OPeNDAP core, greatly facilitating access to large multi-file data sets. - Metadata services - The ability to augment data set descriptions at a central site for remotely accessible data sets will be enhanced to allow for the addition of variables and for data object type modification. This will make it possible to provide consistent, 'complete' descriptions of data sets of interest without requiring that the metadata at the data provider's site be modified. Specialized capabilities to be developed for Sun Earth Connections science: - Metadata - Data set descriptions will be completed for Sun Earth Connections data sets of broad interest to the community and made accessible via an OPeNDAP Ancillary Information Services server at HAO. - Data servers - An SPDML-OPeNDAP server will be written. This will provide a base for the metadata descriptions to be used in the project while at the same time greatly facilitating the provision of project related data via OPeNDAP. - Application clients - The IDL-OPeNDAP client will be rewritten to work on a broader range of CPUs and specialized IDL scripts to further facilitate access and use of key data sets will be developed. - Data server installation - OPeNDAP data servers will be installed at several sites for a targeted suite of Sun Earth Connections data sets.
| |
| PECASE: Applied Pattern Recognition Research (PI: Turmon, Michael,
NASA JPL
) | |
| NASA scientists are overwhwlmed by the vast amounts of data available
from instruments, experiments, and simulations, and all trends point to
continuing increases in the size of these datasets. Our ability to collect,
store, and process ever-greater volumes of data is guaranteed - but to
condense these data into scientific understanding, we will need new
analysis tools. We propose to enable pattern discovery in large
spatiotemporal datasets by developing new algorithms that find and analyze
objects, which we take to be spatially-coherent regions of interest that
evolve
through time. The notion of objects allows investigators to climb above a
pixel-level understanding of their data. These objects must therefore be found
in single images, tracked through image sequences, and their temporal
behaviours modeled quantitatively. In prior work, we have solved a large
class of object-identification problems with general statistical methods
that employ unsupervised learning or scientist-provided labelings to define
local assessments of activity. These pixel-level cues are linked into an
object-level scene description. Related time series models are used for
object tracking and analysis. Here, we propose to extend the underlying
methodology
to cope with much larger object sets, and to allow for non-Gaussian state
representations. This combination of capabilities will enable complex
scenes to be broken down in terms of interacting objects. The system is
field-tested on a vast multi-wavelength series of solar images from an
imager aboard SoHO. Strong collaborations with domain scientists ensure
successful deployment into other Sun-Earth connection missions, and NASA
application areas like planetarty science and geophysics. Expressing the
models in neutral, mathematical/statistical lanaguage allows
repeatability, model refinement over time, and model exchange among
investigators. Our use of open standards for structured data (e.g. object
models and datasets in XML with suitable schemata) ensures portability
across applications and extensibility.
| |
| Machine Learning and Data Mining for Automatic Detection and Interpretation of Solar Events. (PI: Poland, Arthur,
George Mason University
) | |
| We propose to use novel information technology (IT) methods, characteristic of machine learning and data mining, to address issues related to space science data. In particular, we will develop the capability to automatically detect and characterize solar events for purposes related to space weather. The detection and tracking of solar events is important for the Sun-Earth Connection study and space weather prediction. As the amount of data available becomes overwhelming and near-real time detection is important, there is a strong need to automate this process. The specific tasks addressed by this proposal are to use techniques already developed for other problems to automate the detection, tracking and characterization of solar events and to uncover meaningful relationships between different solar events. Towards that end we propose to develop automated and interdisciplinary computational means driven by machine learning, data mining, image processing, pattern recognition and computer vision. The main modules for our automated system include: (A) preprocessing to improve the quality of data using morphology and relaxation tools, (B) detection of solar objects using a combination of methods including boundary detection, region growing and wavelet analysis, (C) tracking using Condensation (Conditional Density Propagation) methods, (D) characterization of solar event, and (E) spatial/temporal data mining of associations between different type of solar events. The proposed methodology is iterative and the steps (ìmodulesî) reinforce each other in their findings. We will develop the software tools needed and test the proposed system using, among others, CME observations from LASCO (Large Angle Spectrometer and Coronagraph) and coronal observations from EIT (Extreme-ultraviolet Imaging Telescope) instruments on the SOHO spacecraft. This proposal investigates novel information technology and computational methods to increase productivity of the OSS missions and programs. It will be achieved through close collaboration between experts in computational sciences, computer science and space science.
| |
| Robust Localization, Classification, and Temporal Evolution Tracking in Auroral Data (PI: Newman, Tim,
University of Alabama, Hunstville
) | |
| A project to develop techniques that aid investigation of auroral phenomena is proposed. The project involves exploiting shape-based techniques that can enable accurate localization of the auroral oval (in the presence of unfriendly background noise), classification of auroral features, and auroral temporal evolution tracking from UVI and IMAGE FUV data, and wrapping these techniques into a web-based content-based image retrieval system. The work includes development of new, robust automated techniques for feature classification and for exploration of temporal evolution.
| |
| Solar Loop Mining to Support Studies of the Coronal Heating Problem (PI: Nasraoui, Olfa,
University of Louisville
) | |
| The Coronal Heating Problem is the longest standing unsolved mystery in astrophysics. Measurements of the temperature distribution along the loop length can be used to support or eliminate various classes of coronal temperature models. The temperature analysis of coronal loops is a state-of-the-art astronomy. In order to make progress, scientific analysis requires data observed by instruments such as EIT, TRACE, and SXT . The combination of EIT, TRACE, and SXT information provides a powerful data set that will yield unprecedented detail on the plasma parameters of a variety of coronal loop structures. The biggest obstacle to completing this project is putting the data set together. EIT has taken over 300,000 images during its 6-year (and counting) mission. The search for interesting images (with coronal loops) is by far the most time consuming aspect of this project. Currently, this process is performed manually, and is therefore extremely tedious, and hinders the progress of science in this field. The next generation ''EIT'' called MAGRITE, scheduled for launch in a few years on NASA's Solar Dynamics Observatory, should be able to take 300,000 images in about four days! and will no doubt need state of the art techniques to sift through the massive data to support scientific discoveries. We propose an approach based on data mining to quickly sift through massive data sets downloaded from the online NASA solar image databases and automatically discover the rare but interesting images with solar loops, which are essential in studies of the Coronal Heating Problem. The proposed solar loop mining scheme will rely on the following components: 1) Collection and labeling of a sample data set of images coming from both categories (with and without solar loops), 2) An optimal feature selection strategy that will facilitate the retrieval task, 3) A classification strategy to classify the transformed image into the correct class, and 4) Appropriate measures to validate the effectiveness of the loop mining process. This project will be implemented in three main phases that target the image databases collected by two different instruments, EIT aboard the NASA/European Space Agency spacecraft SOHO and NASAís TRACE. We will leave open the possibility of targeting the SXT database on the Japanese Yohkoh spacecraft if time permits. Each phase will involve collecting and labeling a small sample of the images from the targeted instrument, for which a solar loop mining strategy is to be designed. All the results of this project: literature, software, and outputs Semantic loop features and class labels (in \emph{XML} format) of the developed classification methods on tested portions of the different instrument databases will be made available to the public and other interested researchers via the World Wide Web.
| |
| Advanced MHD Algorithm for Solar and Space Science (PI: Schnack, Dalton D.,
SAIC
) | |
| The Living With a Star initiative aims to understand those aspects of the Sun-Earth system that affect our life and society. It will investigate the influence of magnetic reconnection on the global corona, the initiation and propagation of CMEs, the evolution of active regions, and the identification of the birthplace of transients. It promises to deliver images and magnetic field measurements with an unprecedented resolution. New advanced software algorithms are required to model these observations and assimilate observational data into models.
The magnetohydrodynamic (MHD) model is appropriate for describing the low-frequency motions of the coronal plasma. In order to be useful to investigate the science problems in the LWS program,, an MHD algorithm must satisfy several computational requirements due to:
*Extreme separation of time scales.
*Extreme separation of spatial scales.
For the algorithm to be widely adopted by the solar and space science community, it must:
*Use parallel computing methods.
*Use standard languages and software.
We propose to meet these requirements through an open source project to deliver a sophisticated MHD algorithm highly reliant on other free and/or open source software tools. The goal of the project is to develop MH4D: parallel, machine independent, code that solves the resistive MHS equations on an unstructured grid of tetrahedra. It will employ Adaptive Mesh Refinement (AMR) to resolve evolving small scale spatial structures, such as current sheets and shock waves. It will feature a fully implicit treatment of viscous and resistive dissipation, and a semi-implicit advancement of the momentum equation (the time step will be limited only by the advective CFL condition). MH4D will be developed in Fortran 90 and MPI, use publicly available libraries and software tools such as PETSc, Metis, and GMV, and be made accessible to the solar and space science community.
| |
| Inteball and Vega Data Archiving (PI: Sibeck, David,
JHU/Applied Physics Laboratory
) | |
| We propose to complete our efforts to calibrate and combine interplanetary, magnetosheath, and magnetospheric observations from the two Interball-1 magnetometers. We will serve the merged data sets from mirrored sites at our home institutions (IZMIRAN and JHU/APL) and work with NASA/NSSDC personnel to add them to the interactive CDAWeb server. The resulting data set will be an integral part of NASA.s ISTP mission. In addition, we propose to translate, calibrate, and despike the Vega-1 and -2 interplanetary magnetometer observations. We will establish a WWW server and work with NSSDC personnel to add the Vega observations to the interactive COHOWeb data server, thereby completing the archival of all relevant interplanetary observations from these heliospheric spacecraft.
| |
| The Analysis and Archival of HRDI data in the MLT region (PI: Skinner, Wilbert R.,
University of Michigan Ann Arbor
) | |
| The High Resolution Doppler Imager (HDRI) on the Upper Atmosphere Research Satellite (UARS) has measured the wind, temperature, and ozone density in the middle atmosphere regions for approximately 10 years. The HRDI data, along with the rest of the UARS data set, comprise a unique record of the middle atmosphere. As the UARS project winds down it is appropriate to pursue measures to preserve and make this data available to future researchers in ways that are most useful. We propose to enhance the value of this data set by: 1) Converting all levels of HRDI data to a common format (netCDF) and making them available to the scientific community over the internet and on DVD; and 2) incorporating the Tide-Mean Assimilation Technique (TMAT) model with the HRDI data to obtain meaningful estimates of the tides and mean conditions over short time periods. This data set in a convenient form will be very useful for investigators studying the mesosphere and lower thermosphere *MLT) and particularly for those working on the TIMED data set.
| |
| Terrestrial Plasmasphere Feature Tracking and Volume Visualization via Tomographic Backprojection (PI: Newman, Tim,
University of Alabama, Hunstville
) | |
| The proposed work will enable more accurate and useful exploitation of satellite data of the terrestrial plasmasphere. A recently-developed tomographic backprojection technique that allows reconstruction of plasma densities in3-space from a single satellite image will be enhanced by coupling with conservative models of plasma distribution in the ionosphere and inner plasmasphere. Integration of data from multiple viewpoints will also be utilized to produce derived volume data products suitable for scientific study. In addition to allowing the volumetric visualization of plasma distributions in the plasmasphere by traditional volume rendering methods, the work will enable plasma features to be manually observed. To aid the observation process, methods to track plasma features in the reconstructed images and volumes will also be developed.
| |
| Next Generation Transport Phenomenology Modeling Tool for Terrestrial and Planetary Applications (PI: Strickland, Douglas J.,
Computational Physics Incorporated
) | |
| In this project, previously developed and validated codes for modeling photoelectron and auroral electron transport and related optical emissions for the upper atmosphere are being reengineered and incorporated into a middleware framework based on the CORBA Component Model (CCM), which allows coarse grain parallelization and distributed computing, among other things. Specifications have been written for each of the transport solutions, which provide descriptions of the solution techniques at a level of detail not previously available. The reengineered versions are written primarily in Java, while the original legacy versions were written in FORTRAN 77. Although there have been concerns that Java is slower than FORTRAN, these concerns have been allayed, since the actual difference in processing speed has been found to be within a factor of 2, which can be offset by the parallel processing capability. Certain legacy models are incorporated directly using wrappers rather than being rewritten. The new software is embedded in the Enterprise Java CORBA Component Model, an implementation of the CCM that was developed at CPI. The new modeling software offers improved maintainability, more extensive code reuse where similar algorithms are used in different models, and greater flexibility for studying terrestrial and planetary atmospheres. It is being added to a more general suite of open source software called Aeronomy Modeling Phenomenology Tool (APMT), which also includes a photochemistry model developed under a previous AISR project.
| |
| Climatological Database of Auroral Images for Solar Cycle 23: An Online Synoptic Search and Metadata (PI: Germany, Glynn A.,
University of Alabama, Hunstville
) | |
| We propose to provide an online search and visualization tool of auroral and geophysical metadata covering the ascending phase of the current solar cycle. This tool will address a demonstrated
problem in using auroral image collections, namely the extreme difficulty of performing non-event studies and surveys of large (~4M) collections of auroral images. The tool will be based on the
collection of POLAR UVI images available from March 1996 to present, but will allow ingestion of metadata from the other POLAR imagers (VIS and PIXIE) as well as other imaging missions such as
the imager suite on IMAGE.
The proposed work directly addresses a major problem associated with all the POLAR imagers (and spaced-based imagers in general), namely that data is typically available only on an event-driven
basis. There is currently no tool available for synoptic searches of auroral features or for visualization of derived parameters such as boundaries, energy input, or presence of given auroral morphologies.
We propose to provide that tool.
The methodology is feasible and appropriate to the proposal goals, and has been explored in sufficient detail to insure its success. The methodology builds on previous success with image classification techniques and development of user interfaces to mission databases essentially identical to that proposed here.
In summary, this work builds on previous experience with analysis of auroral image morphologies and with online access to scientific databases, uses a team intimately familiar with the data sets to be
enhanced, addresses a significant problem within the magnetospheric community that is directly relevant to the AISR goals and areas of interest, promises significant results early in the funding
period, and has a high probability of successfully achieving the proposed goals.
| |
| Automated Event Detection and Classification in High Volume Space Physics Time Series Data using Machine Learning (PI: Perkins, Simon,
Los Alamos National Laboratory
) | |
| Modern satellite-based space physics instruments generate vast streams of temporal multi-band data. A trained expert can often examine a suitably displayed portion of the data stream and identify such events by hand, but for high-volume data streams this approach is entirely impractical. Writing robust feature detectors is very time-consuming and requires a high degree of programming skill and
familiarity with pattern recognition techniques -- certainly not skills that every scientist has. We propose to investigate the use of supervised machine learning techniques to make this job much
easier. We intend to develop a general purpose user-friendly software tool for automatically generating event classifiers that operate on multi-band time-series data using advanced machine learning techniques. We intend to look at using the relatively recently developed technique of Support Vector Machines. SVMs have a strong mathematical foundation and have been empirically shown to
be one of the most powerful and robust machine learning technologies available today. We will also consider the use of Boosting and Gaussian Processes, two other recently developed and
mathematically justifiable machine learning paradigms. Using whichever technique works best, we will develop a general software tool that can be applied to a variety of different space physics data
streams. The primary target of our initial research will be magnetospheric data collected by satellites such as the Los Alamos geosynchronous satellites (Magnetospheric Plasma Analyzer, MPA;
energetic particles, SOPA and ESP), The GPS energetic particle detectors, with further application to other data sets existing in-house (POLAR CEPPAD/CAMMICE/HYDRA) as well as upcoming data
from the recently launched Cluster II sets of satellites.
| |
| Development, Visualization, Retrieval, and Analysis of Multispacecraft Data Sets (PI: Roberts, D. Aaron,
Goddard Space Flight Center
) | |
| We propose to produce and disseminate new methods for the retrieval, visualization, and analysis of multi-spacecraft data. Data set browsing will initially be made possible through the use of uniform,
averaged data sets, but the software will be compatible with multi-platform, distributed data retrieval. The work builds on previous visualization and data analysis tools from AISRP-supported and other
work, and will be applicable to nearly all NASA space physics missions, including ISTP, Magnetospheric Constellation and related missions, and the "COHOWeb" missions to the heliosphere. The results will be incorporated into the National Space Science Data Center (NSSDC) data dissemination tools, allowing scientists and the public to visualize spacecraft data in the 3-D context that it was measured (tied to the spacecraft trajectories) and to download data intervals of interest in directly useable form (interpolated, despiked, etc.).
The software will be developed using common industry tools (e.g., Java3D, OpenGL, C, C++, Fortran, and X) and will be linked to very
extensively used commercial packages such as IDL. Server-based versions of the software will be used for browsing and data set preparation, and stand-alone routines will be available to analyze the data after download. A variety of file formats will be available for downloaded data, and the software will be as platform-independent as possible. To add to the utility of the project, we will develop a
parallel processing version of the retrieval software, so that more and larger data requests can be handled efficiently.
| |
| Statistical Object Identification, Tracking and Analysis (PI: Turmon, Michael J.,
Jet Propulsion Laboratory
) | |
| Scientists are overwhelmed by the vast amounts of data available from instruments, experiments, and simulations; all information technology trends point to a continuing increase in the size of these
datasets. Unfortunately, the ability to collect, store, and manipulate ever-greater volumes of data will not translate into greater scientific understanding without development of suitable software. One way to conceptualize the process of discovering patterns in these large spatiotemporal datasets is to
identify objects, which we take to be spatially coherent regions of interest that evolve through time. This provides one way for scientists to rise above a pixel-level understanding of their data to a more natural and physically relevant description in terms of salient features.
To analyze objects, they must be identified in individual images, tracked through image sequences, and their temporal behaviors modeled quantitatively. We may solve a large class of object-identification problems with modern, general-purpose spatial statistics methods which employ unsupervised learning or scientist-provided labelings to define local assessments of activity. These pixel-level cues are progressively linked into a coherent, object-level scene description. Similarly general statistical time series models are used for object tracking and analysis. Expressing the models in a neutral, mathematical/statistical language allows repeatability, model refinement over time, and model exchange among investigators. Our use of open standards for structured data (e.g., object models and datasets in XML with suitable schemata) ensures portability across applications and future extensibility. The system is field-tested on a multiwavelength series of solar images in conjunction with solar physicists. Strong collaborations with domain scientists ensure successful deployment into other Sun-Earth connection missions, and OSS application areas like planetary science and geophysics.
| |
| Package for Interactive Analysis of Line Emission (Analysis of UV--X-Ray High-Resolution Emission Spectra) (PI: Kashyap, Vinay L.,
Harvard/Smithsonian Center for Astrophysics
) | |
| PINTofALE is an IDL based package to analyze high-resolution grating spectra. The first version was made available to the public on 3 February 2001. Since then we have carried out numerous changes and subsidiary releases. The current release is version 2.0 (released 6 Apr 2004), and we are preparing to release v2.1 within the next month. The changes include bug fixes, upgrades to handle higher versions of IDL and the CHIANTI database, enhancements in user-friendliness, handling of instrument response matrices, and the release of a Markov Chain Monte Carlo based DEM fitting routines. A detailed description of the package, together with fairly detailed documentation, example walks-throughs, and downloadable tar files, are available on-line from http://hea-www.harvard.edu/PINTofALE/. The website also lists papers that have used PINTofALE in their analysis.
| |
| Tomographic analysis and visualization of remotely-sensed heliospheric data (PI: Jackson, Bernard V.,
University of California San Diego
) | |
| The
proposed project consists of two intimately related tasks. The first is
the development of a generic tomographic deconvolution tool, capable of
reconstructing the global, 3D density and velocity structure of the
solar wind in the inner heliosphere using remote sensing data. The
technique takes a novel approach by including outward motion of
structures in the solar wind to obtain the different perspectives
required for a tomographic analysis. Observations from a single vantage
point are needed only, making the technique applicable to e.g.
interplanetary scintillation (IPS) observations, Thomson scattering
data from the HELIOS photometers, coronagraphs (e.g. SOHO/LASCO), and
future all-sky imagers such as the Solar Mass Ejection Imager (SMEI)
and imagers on missions such as STEREO, Solar Probe and the Solar Polar
Sail Mission. Proper understanding of tomographic results requires
advanced visualization tools designed to deal with a 3D plasma
environment in a heliospheric context. The second task is the
development of tools for displaying and manipulating the 3D tomography
results. This will target a system capable of simultaneously displaying
density, velocity and magnetic field, and will allow the user to
interactively explore the heliospheric model from any desired
perspective. We will also experiment with advanced techniques, such as
stereoscopy, drawing on the experience and hardware at the VisLab of
the San Diego Supercomputer Center (SDSC). The proposed tools have the
potential to become crucially important to researchers active in
heliospheric remote sensing. The CASS scientists on this proposal have
worked with remote sensing data for many years, including HELIOS
photometer and IPS observations, and, more recently, the SOHO/LASCO
coronagraph data. They are involved in the construction of SMEI, and
have contributed to the design of prospective all-sky imagers for Solar
Probe and the Solar Polar Sail Mission as co-investigators in
NASA-funded concept studies.
| |
|
|
|
|
|
