Advances in Computational and Cyberinfrastructure in the Earth Sciences III Posters - Special Focus: Advances in Data Acquisition, Management, Analysis and Display [SF]

Special Focus: Advances in Data Acquisition, Management, Analysis and Display [SF]

Advances in Computational and Cyberinfrastructure in the Earth Sciences III Posters

Presiding:G Steinle-Neumann, Bayerisches Geoinstitut, Universitat Bayreuth; A A Koppers, Scripps Institution of Oceanography; D Seber, San Diego Supercomputer Center; V J Salters, National High Magnetic Field Laboratory

An Ontology for Seismic Sources and its Application in the SCEC Community Modeling Environment

* Zechar, J D (zechar@usc.edu) , Southern California Earthquake Center, University of Southern California, 3651 Trousdale Parkway, Suite 169, Los Angeles, CA 90089-0742 United States
* Zechar, J D (zechar@usc.edu) , Department of Earth Sciences, University of Southern California, Zumberge Hall of Science (ZHS) 3651 Trousdale Pkwy, Los Angeles, CA 90089-0740 United States
Jordan, T H (tjordan@usc.edu) , Southern California Earthquake Center, University of Southern California, 3651 Trousdale Parkway, Suite 169, Los Angeles, CA 90089-0742 United States
Jordan, T H (tjordan@usc.edu) , Department of Earth Sciences, University of Southern California, Zumberge Hall of Science (ZHS) 3651 Trousdale Pkwy, Los Angeles, CA 90089-0740 United States

Seismologists have devised many ways to represent earthquake sources: double-couples, centroid moment tensors (CMTs), elementary dislocations, and other point sources; finite moment tensors (FMTs) and other higher-moment expansions, fault-slip distributions (FSDs), and volumetric sources. The mathematical and physical diversity of these representations is further expanded by the host of coordinate systems, physical units, topological assumptions, and geometrical conventions in common use. In this presentation, we outline a machine-readable ontology containing conceptual relationships among a hierarchy of source representations. Our purpose is to create a more flexible set of tools for constructing and manipulating source representations, describing sources in numerical simulations of earthquakes, translating from one representation to another, and mapping high-order representations into lower-order objects (e.g., calculating a CMT or FMT from an FSD). We have built the earthquake source ontology using the PowerLoom Knowledge Representation System (http://www.isi.edu/isd/loom/powerloom/) and leveraged standardized ontologies that deal with physical units and coordinate systems. An important issue is the assimilation of mathematical relationships among source representations into the ontology. The process we have used includes developing a set of Java objects which specify canonical source representations and their varieties. We illustrate an application of the ontology to a user interface for ground-motion simulations (Pathway 2) of the Southern California Earthquake Center (SCEC) Community Modeling Environment.

* Muench, J (joanna@iris.washington.edu) , IRIS Data Management Center, 1408 NE 45th Street, Suite 201, Seattle, WA 98105 United States
Maechling, P (maechlin@usc.edu) , Southern California Earthquake Center, University of Southern California, Los Angeles, CA 90089
Kamb, L (linus@iris.washington.edu) , IRIS Data Management Center, 1408 NE 45th Street, Suite 201, Seattle, WA 98105 United States
Ahern, T (tim@iris.washington.edu) , IRIS Data Management Center, 1408 NE 45th Street, Suite 201, Seattle, WA 98105 United States

Currently, analysis of synthetic seismograms and comparison to observed seismograms requires multiple tools to gather disparate sources of information and run various analysis routines. The SOSA application (Synthetic and Observed Seismogram Analysis) seeks to combine common tasks into a single software application. The tool enables scientists to retrieve observed seismograms accessible through a DHI (Data Handling Interface) and seismograms saved locally or accessible through an SRB (Storage Resource Broker). DHI implementations currently provide access to the IRIS DMC, the University of South Carolina, the Northern California Earthquake Data Center at Berkeley and the Southern California Earthquake Data Center at Caltech. Retrieved seismograms include the instrument signal and station information with instrument response. Seismograms retrieved from any source can be processed together, allowing easy comparison between synthetics and observed seismograms. Existing processing abilities include re-sampling, filtering, rotation, convolution/deconvolution with instrument response and correlation between seismograms. Scientists may save seismograms into SAC or mSEED format. The utility of the application will be demonstrated and future directions discussed.

* Hansen, T (tshansen@nlanr.net) , Cecil H. and Ida M. Green Institute for Geophysics and Planetary Physics, University of California, San Diego 9500 Gilman Dr. MC 0225, La Jolla, CA 92093-0225 United States
Vernon, F (fvernon@ucsd.edu) , Cecil H. and Ida M. Green Institute for Geophysics and Planetary Physics, University of California, San Diego 9500 Gilman Dr. MC 0225, La Jolla, CA 92093-0225 United States
Lindquist, K (kent@lindquistconsulting.com) , Lindquist Consulting, 59 College Rd. Suite #7, Fairbanks, AK 99701 United States
Orcutt, J (jorcutt@ucsd.edu) , Cecil H. and Ida M. Green Institute for Geophysics and Planetary Physics, University of California, San Diego 9500 Gilman Dr. MC 0225, La Jolla, CA 92093-0225 United States

For anyone who has managed a moderately complex buffered real-time data transport system, the need for reliable adaptive data transport is clear. The ROADNet VORBrouter system, an extension to the ROADNet data catalog system [AGU-2003, Dynamic Dataflow Topology Monitoring for Real-time Seismic Networks], allows dynamic routing of real-time seismic data from sensor to end-user. Traditional networks consist of a series of data buffer computers with data transport interconnections configured by hand. This allows for arbitrarily complex data networks, which can often exceed full comprehension by network administrators, sometimes resulting in data loops or accidental data cutoff. In order to manage data transport systems in the event of a network failure, a network administrator must be called upon to change the data transport paths and to recover the missing data. Using VORBrouter, administrators can sleep at night while still providing 7/24 uninterupted data streams at realistic cost. This software package uses information from the ROADNet data catalog system to route packets around failed link outages and to new consumers in real-time. Dynamic data routing protocols operating on top of the Antelope Data buffering layer allow authorized users to request data sets from their local buffer and to have them delivered from anywhere within the network of buffers. The VORBrouter software also allows for dynamic routing around network outages, and the elimination of duplicate data paths within the network, while maintaining the nearly lossless data transport features exhibited by the underlying Antelope system. We present the design of the VORBrouter system, its features, limitations and some future research directions.

* Grant, L B (lgrant@uci.edu) , University of California, Dept. Environmental Health, Science, and Policy, Irvine, CA 92697-7070 United States
Donnellan, A (donnellan@jpl.nasa.gov) , Jet Propulsion Laboratory, Mail Stop 183-335 4800 Oak Grove Drive, Pasadena, CA 91109-8099 United States
McLeod, D (mcleod@usc.edu) , University of Southern California, Computer Science Department, Los Angeles, CA 90089-0781 United States
Pierce, M (marpierc@indiana.edu) , Indiana University, Community Grid Computing Lab 501 N. Morton, Suite 224, Bloomington, IN 47404-3730 United States
Chen, A Y (yunanche@usc.edu) , University of Southern California, Computer Science Department, Los Angeles, CA 90089-0781 United States
Gould, M (miryha@uci.edu) , University of California, Dept. Environmental Health, Science, and Policy, Irvine, CA 92697-7070 United States
Noriega-Carlos, G (gnoriega@uci.edu) , University of California, Dept. Environmental Health, Science, and Policy, Irvine, CA 92697-7070 United States
Paul, R (prundle@cse.ucdavis.edu) , University of California, Center for Computational Science and Engineering, Davis, CA 95616 United States
Sung, S (sangsung@usc.edu) , University of Southern California, Computer Science Department, Los Angeles, CA 90089-0781 United States
Ta, M V (mvta@uci.edu) , University of California, Dept. Environmental Health, Science, and Policy, Irvine, CA 92697-7070 United States

The QuakeSim project will provide the first web-services based, interoperable environment for doing large-scale forward models for earthquake processes. Through a web-services based portal, QuakeSim provides global access to geologic reference models of faults and fault data, simple analysis tools, new parallel forward models, and visualization support. The database system for this project must manage a variety of types of earthquake science data and information, including real and simulated data, and pre-existing structured collections containing "validated" data from official sources such as U.S. and California Geological Surveys, and "non-validated" data sets such as Virtual California. The fault database component of QuakeSim is called QuakeTables. QuakeTables was developed using a basic public domain database management system (DBMS), MySQL, to be ported to a more fully functional relational DBMS. These systems support the definition, storage, access, and control of collections of structured data. To provide for the access and manipulation of heterogeneous data sources, the integration of information from such sources, and the structural organization and data mining of this data, QuakeTables employs techniques for wrapper-based information fusion to support data source access and integration. QuakeTables is searchable with annotated fault records from original sources. The QuakeSim team designed the fields that constitute the database records and provided a web-based interface that enables the submitting and accessing of those records. The fields include primary geologic and paleoseismic fault parameters (fault location/geometry, slip rate at measured location, measurements of co-seismic displacement, dates and locations of previous ruptures) as well as non-primary fault parameters such as names, segments and characteristic recurrence interval.

* Egbert, G D (egbert@coas.oregonstate.edu) , Oregon State University, College of Oceanic and Atmospheric Sciences , Corvallis, OR 97331-5503 United States

An important application of computational modeling in the geosciences is to problems in parameter estimation, inversion, and data assimilation. These applications can require complex software systems that combine simulations of Earth processes and systems with data processing and mathematical optimization algorithms. A modular approach to software design for these systems offers many obvious advantages: for example, optimization and data handling components can be developed once, and applied across a range of problems or numerical models. Developing numerical modeling modules that will be useful for inversion and assimilation applications also presents some challenges. For example, for many optimization strategies tangent linear and/or adjoint models are required. These modules are comparable in complexity to the original code, but tools and procedures have been developed (e.g., adjoint compilers) to facilitate development of these modules given a working forward code. Here we discuss recent experiences with two modular software systems being developed for inversion and data assimilation. The first is a small modular inversion system for electromagnetic induction data that we are developing ourselves, primarily for prototyping of efficient inversion algorithms. The second is a much larger project, the Inverse Ocean Modeling (IOM) system being developed in the physical oceanographic research community for variational data assimilation in a range of ocean modeling applications. One important, if unsurprising, lesson from these projects is that even a bit of foresight in initial development of modular forward modeling codes and interfaces can greatly simplify incorporation of these modules into data assimilation and inversion schemes. Since inversion, data assimilation, and parameter estimation are likely to remain very important in solid Earth studies, this issue deserves some consideration as the geoscience community prepares to invest in new software infrastructure for future research needs.

GeoFramework: Coupling multiple models of mantle convection within a computational framework

* Tan, E (tan2@gps.caltech.edu) , Seismological Laboratory, California Institute of Technology, Pasadena, CA 91125 United States
Choi, E (ces74@gps.caltech.edu) , Seismological Laboratory, California Institute of Technology, Pasadena, CA 91125 United States
Thoutireddy, P (puru@cacr.caltech.edu) , Center for Advanced Computing Research, California Institute of Technology, Pasadena, CA 91125 United States
Gurnis, M (gurnis@gps.caltech.edu) , Seismological Laboratory, California Institute of Technology, Pasadena, CA 91125 United States
Aivazis, M (aivazis@cacr.caltech.edu) , Center for Advanced Computing Research, California Institute of Technology, Pasadena, CA 91125 United States

Geological processes usually encompass a broad spectrum of length and time scales. Traditionally, a modeling code (solver) is developed for a problem of specific length and time scales, but the utility of the solver beyond the designated purpose is usually limited. As we have come to recognize that geological processes often result from the dynamic coupling of deformation across a wide range of time and spatial scales, more robust methods are needed. One means to address this need is through the integration of complementary modeling codes, while attempting to reuse existing software as much as possible. The GeoFramework project addresses this by developing a suite of reusable and combinable tools for the Earth science community. GeoFramework is based on and extends Pyre, a Python-based modeling framework, developed to link solid (Lagrangian) and fluid (Eulerian) solvers, as well as mesh generators, visualization packages, and databases, with one another for engineering applications. Under the framework, a solver is aware of the presence of other solvers and can interact with each other via exchanging information across adjacent mesh boundary. We will show an example of linking two instances of the CitcomS finite element solver within GeoFramework. A high-resolution regional mantle convection model is linked with a global mantle convection model. The global solver has a resolution of $\sim180$ km horizontally and 35-100 km (with mesh refinement) vertically. The fine mesh has a resolution of $\sim40$ km horizontally and vertically. The fine mesh is center on the Hawaii hotspot. A vertical plume is used as an initial condition. Time-varying plate velocity models are imposed since 80 Ma and we have investigated how the plume conduit is deflected by the global circulation patterns as a function of mantle viscosity, plume flux, and plate motion.

Application of high order compact finite difference algorithms on non-uniform grids to numerical geodynamo simulation

* Liu, D (dliu@bowie.gsfc.nasa.gov) , JCET, UMBC, Baltimore, MD 21250 United States
Kuang, W (Weijia.Kuang-1@nasa.gov) , Space Geodesy Branch, NASA GSFC, Greenbelt, MD 20771 United States

The Earth possesses a magnetic field of internal origin (core geomagnetic field) in much of its history. This field is generated and maintained by convective flow in the Earth's liquid outer core (geodynamo). Currently several numerical models have been successfully developed with various algorithms to model the dynamo process. Among them include the mixed approach of applying spectral algorithms on spherical surfaces and finite difference algorithms in radius. In our study, we focus on examining the applicability of high (6th and 8th) order, compact finite difference algorithms with non-uniform radial grids to numerical geodynamo modeling. The non-uniform radial grids are used to resolve thin boundary layers at the core-mantle boundary (CMB) and at the inner core boundary (ICB). We are in particular interested in the effect of the algorithms on nonlinear couplings among the magnetic field, the velocity field and the density perturbations. The results are used to compare those from our MoSST core dynamics model which employs a 4th order finite difference algorithm in radius.

The SCEC Community Modeling Environment (SCEC/CME) - An Overview of its Architecture and Current Capabilities

* Maechling, P J (maechlin@usc.edu) , Southern California Earthquake Center, 3651 Trousdale Parkway, Los Angeles, CA 90089 United States
Jordan, T H (tjordan@usc.edu) , University of Southern California, 3651 Trousdale Parkway, Los Angeles, CA 90089 United States
Minster, B (jbminster@ucsd.edu) , Scripps Institution of Oceanography, 8602 La Jolla Shores Drive, La Jolla, CA 92037 United States
Moore, R (moore@sdsc.edu) , San Diego Supercomputer Center, 10100 Hopkins Drive, La Jolla, CA 92093 United States
Kesselman, C (carl@isi.edu) , USC Information Sciences Institute, 4676 Admiralty Way, Marina del Rey, CA 90292 United States
SCEC ITR Collaboration, T (scecinfo@usc.edu) , Southern California Earthquake Center, 3651 Trousdale Parkway, Los Angeles, CA 90089 United States

The Southern California Earthquake Center (SCEC), in collaboration with the San Diego Supercomputer Center, the USC Information Sciences Institute, the Incorporated Research Institutions for Seismology, and the U.S. Geological Survey, is developing the Southern California Earthquake Center Community Modeling Environment (CME) under a five-year grant from the National Science Foundation's Information Technology Research (ITR) Program jointly funded by the Geosciences and Computer and Information Science & Engineering Directorates. The CME system is an integrated geophysical simulation modeling framework that automates the process of selecting, configuring, and executing models of earthquake systems. During the Project's first three years, we have performed fundamental geophysical and information technology research and have also developed substantial system capabilities, software tools, and data collections that can help scientist perform systems-level earthquake science. The CME system provides collaborative tools to facilitate distributed research and development. These collaborative tools are primarily communication tools, providing researchers with access to information in ways that are convenient and useful. The CME system provides collaborators with access to significant computing and storage resources. The computing resources of the Project include in-house servers, Project allocations on USC High Performance Computing Linux Cluster, as well as allocations on NPACI Supercomputers and the TeraGrid. The CME system provides access to SCEC community geophysical models such as the Community Velocity Model, Community Fault Model, Community Crustal Motion Model, and the Community Block Model. The organizations that develop these models often provide access to them so it is not necessary to use the CME system to access these models. However, in some cases, the CME system supplements the SCEC community models with utility codes that make it easier to use or access these models. In some cases, the CME system also provides alternatives to the SCEC community models. The CME system hosts a collection of community geophysical software codes. These codes include seismic hazard analysis (SHA) programs developed by the SCEC/USGS OpenSHA group. Also, the CME system hosts anelastic wave propagation codes including Kim Olsen's Finite Difference code and Carnegie Mellon's Hercules Finite Element tool chain. The CME system can execute a workflow, that is, a series of geophysical computations using the output of one processing step as the input to a subsequent step. Our workflow capability utilizes grid-based computing software that can submit calculations to a pool of computing resources as well as data management tools that help us maintain an association between data files and metadata descriptions of those files. The CME system maintains, and provides access to, a collection of valuable geophysical data sets. The current CME Digital Library holdings include a collection of 60 ground motion simulation results calculated by a SCEC/PEER working group and a collection of Greens Functions calculated for 33 TriNet broadband receiver sites in the Los Angeles area.

* Lindquist, K G (kent@lindquistconsulting.com) , Lindquist Consulting, 59 College Rd. #7, Fairbanks, AK 99701 United States
Hansen, T S (tshansen@nlanr.net) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
Newman, R L (rlnewman@ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
Vernon, F L (flvernon@ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
Nayak, A (atul@epicenter.ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
Foley, S (sfoley@ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
Fricke, T (tobin@splorg.org) , U. of Rochester, Dept. of Physics and Astronomy U. of Rochester, Rochester, NY 14627-0171 United States
Orcutt, J (jorcutt@ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, MC 0225, La Jolla, CA 92093-0225 United States
Rajasekar, A (sekar@sdsc.edu) , San Diego Supercomputer Center, U. of California, San Diego 9500 Gilman Drive, MC 0505, La Jolla, CA 92093-0505 United States

The ROADNet real-time monitoring infrastructure has allowed researchers to integrate geophysical monitoring data from a wide variety of signal domains. Antelope-based data transport, relational-database buffering and archiving, backup/replication/archiving through the Storage Resource Broker, and a variety of web-based distribution tools create a powerful monitoring platform. In this work we discuss our use of the ROADNet system for the collection and processing of digital image data. Remote cameras have been deployed at approximately 32 locations as of September 2004, including the SDSU Santa Margarita Ecological Reserve, the Imperial Beach pier, and the Pinon Flats geophysical observatory. Fire monitoring imagery has been obtained through a connection to the HPWREN project. Near-real-time images obtained from the R/V Roger Revelle include records of seafloor operations by the JASON submersible, as part of a maintenance mission for the H2O underwater seismic observatory. We discuss acquisition mechanisms and the packet architecture for image transport via Antelope orbservers, including multi-packet support for arbitrarily large images. Relational database storage supports archiving of timestamped images, image-processing operations, grouping of related images and cameras, support for motion-detect triggers, thumbnail images, pre-computed video frames, support for time-lapse movie generation and storage of time-lapse movies. Available ROADNet monitoring tools include both orbserver-based display of incoming real-time images and web-accessible searching and distribution of images and movies driven by the relational database (http://mercali.ucsd.edu/rtapps/rtimbank.php). An extension to the Kepler Scientific Workflow System also allows real-time image display via the Ptolemy project. Custom time-lapse movies may be made from the ROADNet web pages.

Earth Applications Group in the EGEE Europan Project : First experiments and objectives

* Petitdidier, M (monique.petitdidier@ipsl.jussieu.fr) , CETP Institut Pierre et Simon Laplace, 10-12 Avenue de l'Europe, Velizy, 78140 France
Fusco, L (luigi.fusco@esa.int) , ESA/ESRIN, Via Galileo Galilei, Frascati, RM, 00044 Italy
Nercessian, A (nerces@ipgp.jussieu.fr) , D\'epartement de Sismologie Institut de Physique du Globe de Paris, 4 Place Jussieu, Paris, 75252 France
Vilotte, J (vilotte@ipgp.jussieu.fr) , D\'epartement de Sismologie Institut de Physique du Globe de Paris, 4 Place Jussieu, Paris, 75252 France

The aim of the EGEE (Enabling Grids E-science in Europe) project, funded by the EC is to build on recent advances in Grid Technology and to develop a service grid infrastructure, built on the EU Research Network GEANT, available to scientific and industrial applications. The Earth Sciences group is concern in the deployment of selected applications on the EGEE production testbed. Today applications include Earth Observation, Climate, Hydrology, Solid Earth and applied Geophysics. Recent developments made in the domains of Earth Observation and Solid Earth geophysics will be presented. In Earth observation, the goal is to drive the satellite operational and research community to use the Grid to process, validate and exchange large sets of data. This objective has begun during the European DataGrid project. First results have been obtained for Ozone data processing and validation. Large numbers and large volume of files from different satellite instruments (Gome and Gomos) and ground measurements have be handled on the Grid. More than 70,000,000 of Ozone profiles have been produced and stored on the Grid by orbit. The data has been processed with complex algorithms (e.g. Neural approach, inversion methods) using IDL-runtime. Creation and secure access to metadata and satellite data has been made possible on the Grid. Moreover, ESA built a secure portal to access to its Archive Mass Storage to transfer satellite data on the Grid and to process the selected data. This Grid interface has provided a very progressive way for end-users to access and process satellite data. In EGEE an attempt to access other satellite archives from the Grid and process scientific jobs requiring access to several databases for quasi-real time prediction, like ozone, is in progress. In The Solid Earth, a first example of the use of the GRID has been done in the case of a seismic tomography study by IPG Paris. Secure and restricted access to database has been demonstrated as well as complex processing of seismic data (least-square inversion with parallel solver). The implementation on the GRID will be detailed. Ongoing extensions concern GPS data access and processing using the GAMMIT tool. This includes access to external data from the GRID. The goal is to provide a complete workflow of data storage, processing, analysis and visualisation with a secure portal built on the work done in Earth Observation. Another ongoing work concerns processing and sharing large data sets of synthetic seismograms produced by large 3D numerical simulations in complex Earth models. One of the key issues here is the data exchange format for synthetic seismograms. Progress in that direction is part of the objectives of the European SPICE project, and has to be done in closed collaboration with the FDSN/IRIS group. Grid technology allow not only the exchange of data but also the access in the same job to several distributed databases in the emerging large scale European projects like GMES and SPICE. The Earth Applications group involves today several institutes : IPSL, IPGP, ESA/ESRIN, CGG, CRS4, DKRZ, KNMI, SRON, UNINE, UTV

The SPICE Project: Towards a Digital Library With Codes, Training Material, and Simulation data

* Igel, H (heiner.igel@lmu.de) , Department of Earth and Environmental Sciences, Theresienstrasse 41, Munich, 80333 Germany
Wassermann, J (jowa@geophysik.uni-muenchen.de) , Department of Earth and Environmental Sciences, Theresienstrasse 41, Munich, 80333 Germany
Stupazzini, M (stupa@stru.polimi.it) , Department of Earth and Environmental Sciences, Theresienstrasse 41, Munich, 80333 Germany
Vilotte, J (vilotte@ipgp.jussieu.fr) , Institut de Physique du Globe, 4, place Jussieu, Paris, 75252 France

SPICE (\bf{S}eismic wave \bf{P}ropagation and \bf{I}maging in \bf{C}omplex media: a \bf{E}uropean network) is a Marie Curie Research Training Network funded within the 6th EU-framework programme. The network aims at integrating 14 European institutions with specialisations in physical, mathematical, geological, and computational aspects of wave propagation. The goal is to develop, verify and apply computational tools for wave propagation and imaging problems on all scales. Specific research areas include the determination of global Earth structure; the quantitative estimation of strong ground motion; the characterization and monitoring of reservoirs; understanding the structure and processes inside volcanoes; and simulating the physical processes of earthquake rupture. While the overall goal of the research training network is to train young scientists in the particular research areas, the specific technical goal of the SPICE network is to develop infrastructure for an open digital library with (1) computational algorithms ready for use by the scientific community; (2) accessible simulation results (e.g. ground motions from earthquake scenarios, global seismograms, dynamic rupture modelling); and (3) training material. One of the most fundamental questions to address therefore is the specific format with which synthetic data should be stored in the long term. The development of such standards and programs to access and visualize synthetic data is being carried out in close collaboration with the national and international seismic data centres (e.g., ORFEUS, IRIS). In our poster we classify the synthetic data types, discuss interfaces to observational data and present preliminary results based on SEED type formats.

Representation of Structural Geology Knowledge with Ontology: Design and Implementation Issues

* Babaie, H A (mylonite_7@yahoo.com) , Department of Geology, Georgia State University, Atlanta, GA 30303 United States

Ontology of a specific geological discipline is a domain model providing a vocabulary about key geological concepts and their relationships, and processes, theories, and principles in that field. Geologists apply existing domain knowledge when they study a geologic object such as the San Andreas Fault. The knowledge is based, explicitly and/or implicitly, on geologists' conceptualization or view of the world of geological objects, concepts, and their meaning and relationships. Ontology represents the domain knowledge by explicitly and formally specifying such conceptualizations shared by a community of geoscientists in a specific domain (e.g., stratigraphy). Ontology is used to more efficiently and formally share, reuse, and analyze domain knowledge (not just data), and to identify its implicit assumptions. Building an ontology begins by defining the purpose and scope of the ontology, and involves capturing the domain knowledge by identifying, evaluating, and documenting the concepts and their relations. These activities lead to the definition of classes representing the key domain concepts, in a taxonomic hierarchy, and their properties and values. This paper discusses the design principles, development process, life cycle, and methods required to build domain ontologies in geosciences. The Protägä 2000 ontology-building tool and OWL markup language are applied to build an ontology for a part of the structural geology domain.

Addressing Multi-Scale Problems in Geophysics With Coupled Modeling Tools: Current Development Status of the Lithomop Deformation Modeling Code Within the Pyre Framework

* Williams, C A (willic3@rpi.edu) , Dept. of Earth and Environmental Sciences, Rensselaer Polytechnic Institute, Science Center, 2C01B, Troy, NY 12180 United States
Aagaard, B (baagaard@usgs.gov) , U. S. Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025 United States
Aivazis, M (aivazis@caltech.edu) , Center for Advanced Computing Research, California Institute of Technology, M/C 158-79, Pasadena, CA 91125 United States
Simons, M (simons@caltech.edu) , Seismo Lab, California Institute of Technology, M/C 252-21, Pasadena, CA 91125 United States

Many fundamental problems in geophysics may be viewed as the interaction of phenomena that occur on different temporal and spatial scales. As an example, the occurrence of multiple earthquakes in a tectonically active region could be characterized in terms of at least three distinct scales: 1) The contribution to the regional stress field in the lithosphere from mantle convection (dynamic computations on large temporal and spatial scales). 2) Interseismic stress buildup and postseismic stress relaxation in the lithosphere (quasi-static computations on intermediate temporal and spatial scales). 3) Coseismic stress changes (dynamic computations on short time scales and small to intermediate spatial scales). In the past, the tendency has been to focus on one of the scales pertinent to the problem, and approximate the effects due to the other scales. With increasingly powerful computational abilities, however, it is now becoming more feasible to address the entire scope of these complex problems. We describe here the current development status of Lithomop -- a quasi-static finite element code for modeling deformation in the solid earth. This code, which was developed from a version of the finite element code TECTON, is being integrated into the Pyre framework as one member of a suite of geophysical modeling tools being assembled by the GeoFramework project. One important facility provided by the Pyre framework is the ability to couple different codes, which will allow the simulation of complex multi-scale problems. Recent improvements to Lithomop include the addition of 8 new element types, better integration within the Pyre framework, a new modular approach to defining material models, and a number of efficiency improvements. When completed, Lithomop will be fully parallelized, will have several flexible mechanisms for importing and exporting data, will include real-time visualization capabilities, and will be able to couple with several other codes in the framework.

* Bensen, G D (gbensen@cires.colorado.edu) , University of Colorado, 2200 Colorado Ave, Boulder, CO 80309 United States
Meertens, C M (meertens@unavco.org) , UNAVCO, Inc., 6350 Nautilus Dr, Boulder, CO 80301 United States
Sheehan, A F (afs@cires.colorado.edu) , University of Colorado, 2200 Colorado Ave, Boulder, CO 80309 United States

Some recent research at UNAVCO and the University of Colorado has been focused on Rocky Mountain tectonics, and Information Technology (IT) in the areas of data visualization and distributed data serving. At UNAVCO, we are participating in the geodynamics work in the Rocky Mountain Testbed of the GEON NSF funded (IT) Research project (www.geongrid.org). As part of this work, a variety of seismic tomography models, GPS velocity vector data, strain rate models and other data have been recompiled into a standard format. These data and models are being incorporated into our OPeNDAP server and the Integrated Data Viewer (IDV). OPeNDAP servers are platform independent, self-describing distributed data servers allowing easy access to a wide audience. The IDV is a freely distributed visualization and analysis tool developed by UCAR that has several exciting capabilities such as online collaboration, and a variety of 1-d, 2-d and 3-d viewing options. Necessary solid earth viewing capabilities (earthquakes, focal mechanisms, faults, etc.) are currently being added to the IDV. Both our OPeNDAP server and visualization tool are being integrated into the GEON portal, a website for data searching, analysis, and visualization. Designing and implementing such systems now allows us to be more prepared for the volumes of data anticipated from various EarthScope projects. As part of the scientific research for GEON, we have also begun investigations of Colorado seismicity. The 1992 Rocky Mountain Front IRIS/PASSCAL seismic experiment recorded many local earthquakes. We have begun to locate these events and are working to create focal mechanisms and calculations of stress drop for this region. These will aid in improving seismic hazard and risk assessments for the rapidly growing Rocky Mountain population. New IT capabilities will help augment the quality of this work through sharing the data with a larger audience, providing a means to view and analyze integrated data, and quickly providing a computationally intensive baseline by which results can be checked.

* Kuang, W (Weijia.Kuang-1@nasa.gov) , Space Geodesy Branch, NASA GSFC, Greenbelt, MD 20771 United States
Gary, P (James.P.Gary@gsfc.nasa.gov) , Earth and Space Data Computing Division, NASA GSFC, Greenbelt, MD 20771 United States
Seablom, M (Michael.S.Seablom@nasa.gov) , Earth and Space Data Computing Division, NASA GSFC, Greenbelt, MD 20771 United States
Truszkowski, W (Walt.Truszkowski@nasa.gov) , Advanced Architectures and Automating Branch, NASA GSFC, Greenbelt, MD 20771 United States
Odubiyi, J (jodubiyi@att.net) , Dept. Comp. Sci., Bowie State University, Bowie, MD 20715 United States
Jiang, W (jiang@bowie.gsfc.nasa.gov) , JCET, UMBC, Baltimore, MD 21250 United States
Liu, D (dliu@bowie.gsfc.nasa.gov) , JCET, UMBC, Baltimore, MD 21250 United States

Large-scale numerical simulation has been one of the most important approaches for understanding global geodynamical processes. In this approach, peta-scale floating point operations (pflops) are often required to carry out a single physically-meaningful numerical experiment. For example, to model convective flow in the Earth's core and generation of the geomagnetic field (geodynamo), simulation for one magnetic free-decay time (approximately 15000 years) with a modest resolution of 150 in three spatial dimensions would require approximately 0.2 pflops. If such a numerical model is used to predict geomagnetic secular variation over decades and longer, with e.g. an ensemble Kalman filter assimilation approach, approximately 30 (and perhaps more) independent simulations of similar scales would be needed for one data assimilation analysis. Obviously, such a simulation would require an enormous computing resource that exceeds the capacity of a single facility currently available at our disposal. One solution is to utilize a very fast network (e.g. 10Gb optical networks) and available middleware (e.g. Globus Toolkit) to allocate available but often heterogeneous resources for such large-scale computing efforts. At NASA GSFC, we are experimenting with such an approach by networking several clusters for geomagnetic data assimilation research. We shall present our initial testing results in the meeting.

Integration of Kepler with ROADNet: Visual Dataflow Design with Real-time Geophysical Data

* Fricke, T T (tobin@pas.rochester.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92-93-0225 United States
Ludaescher, B (ludaesch@sdsc.edu) , San Diego Supercomputer Center, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92093-0505 United States
Altintas, I (altintas@sdsc.edu) , San Diego Supercomputer Center, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92093-0505 United States
Lindquist, K G (kent@lindquistconsulting.com) , Lindquist Consulting, 59 College Rd. #7, Fairbanks, AK 99701 United States
Hansen, T S (tshansen@nlanr.net) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92-93-0225 United States
Rajasekar, A (sekar@sdsc.edu) , San Diego Supercomputer Center, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92093-0505 United States
Vernon, F L (flvernon@ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92-93-0225 United States
Orcutt, J (jorcutt@ucsd.edu) , Institute of Geophysics and Planetary Physics, U. of California, San Diego 9500 Gilman Drive, La Jolla, CA 92-93-0225 United States

The ROADNet project concentrates real-time data from a wide variety of signal domains, providing a reliable platform to store and transport these data. Ptolemy is a general purpose visual programming environment in which work flows on data streams can be constructed by connecting general purpose components. The Kepler scientific workflow system extends Ptolemy to approach design and automation of scientific data analysis tasks. In this work we discuss our integration of ROADNet (and the Antelope platform on which ROADNet is based in part) with the Ptolemy environment. We have produced interface components that allow someone using the Kepler scientific workflow system to readily use ROADNet data resources. Presently we have working components to gather real-time waveform and image data from ROADNet object ring buffers, and we are working to provide the ability to perform Datascope database queries from Kepler. The Kepler project, including the Antelope interface, is entirely free and open-source, and will run on any platform where Java is available. We discuss existing applications in addition to possible future directions, such as coherent array processing, event detection, and online stream processing. A major advantage of the Ptolemy environment is the ease with which it may be used for rapid prototyping of analyses by even inexperienced users. For instance, a user can drag-and-drop an Orb Waveform Source component and several general purpose analysis and display components, connect them visually, and immediately perform an analysis on real-time data.

* Saez, E (saez@esscc.uq.edu.au) , QUAKES, Earth Systems Science Computational Centre, The university of Queensland, St Lucia, QLD 4072 Australia
* Saez, E (saez@esscc.uq.edu.au) , Australian Computational Earth Systems Simulator, The university of Queensland, St Lucia, QLD 4072 Australia
Mora, P (morap@esscc.uq.edu.au) , QUAKES, Earth Systems Science Computational Centre, The university of Queensland, St Lucia, QLD 4072 Australia
Mora, P (morap@esscc.uq.edu.au) , Australian Computational Earth Systems Simulator, The university of Queensland, St Lucia, QLD 4072 Australia
Gross, L (gross@esscc.uq.edu.au) , QUAKES, Earth Systems Science Computational Centre, The university of Queensland, St Lucia, QLD 4072 Australia
Gross, L (gross@esscc.uq.edu.au) , Australian Computational Earth Systems Simulator, The university of Queensland, St Lucia, QLD 4072 Australia
Weatherley, D (dion@esscc.uq.edu.au) , QUAKES, Earth Systems Science Computational Centre, The university of Queensland, St Lucia, QLD 4072 Australia
Weatherley, D (dion@esscc.uq.edu.au) , Australian Computational Earth Systems Simulator, The university of Queensland, St Lucia, QLD 4072 Australia

We introduce a dynamic Finite Element method using a novel high level scripting language to describe the physical equations, boundary conditions and time integration scheme. The library we use is the parallel Finley library: a finite element kernel library, designed for solving large-scale problems. It is incorporated as a differential equation solver into a more general library called escript, based on the scripting language Python. This library has been developed to facilitate the rapid development of 3D parallel codes, and is optimised for the Australian Computational Earth Systems Simulator Major National Research Facility (ACcESS MNRF) supercomputer, a 208 processor SGI Altix with a peak performance of 1.1 TFlops. Using the scripting approach we obtain a parallel FE code able to take advantage of the computational efficiency of the Altix 3700. We consider faults as material discontinuities (the displacement, velocity, and acceleration fields are discontinuous at the fault), with elastic behavior. The stress continuity at the fault is achieved naturally through the expression of the fault interactions in the weak formulation. The elasticity problem is solved explicitly in time, using the Saint Verlat scheme. Finally, we specify a suitable frictional constitutive relation and numerical scheme to simulate fault behaviour. Our model is based on previous work on modelling fault friction and multi-fault systems using lattice solid-like models. We adapt the 2D model for simulating the dynamics of parallel fault systems described to the Finite-Element method. The approach uses a frictional relation along faults that is slip and slip-rate dependent, and the numerical integration approach introduced by Mora and Place in the lattice solid model. In order to illustrate the new Finite Element model, single and multi-fault simulation examples are presented.

High Tech High School Interns Develop a Mid-Ocean Ridge Database for Research and Education

Staudigel, D (dstaudig@igpp.ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92117 United States
Delaney, R (rdelaney@hightechhigh.org) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92117 United States
Staudigel, H (hstaudig@igpp.ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92117 United States
* Koppers, A A (dstaudig@igpp.ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92117 United States
Miller, S P (spmiller@ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92117 United States

Mid-ocean ridges (MOR) represent one of the most important geographical and geological features on planet Earth. MORs are the locations where plates spread apart, they are the locations of the majority of the Earths' volcanoes that harbor some of the most extreme life forms. These concepts attract much research, but mid-ocean ridges are still effectively underrepresented in the Earth science class rooms. As two High Tech High School students, we began an internship at Scripps to develop a database for mid-ocean ridges as a resource for science and education. This Ridge Catalog will be accessible via http://earthref.org/databases/RC/ and applies a similar structure, design and data archival principle as the Seamount Catalog under EarthRef.org. Major research goals of this project include the development of (1) an archival structure for multibeam and sidescan data, standard bathymetric maps (including ODP-DSDP drill site and dredge locations) or any other arbitrary digital objects relating to MORs, and (2) to compile a global data set for some of the most defining characteristics of every ridge segment including ridge segment length, depth and azimuth and half spreading rates. One of the challenges included the need of making MOR data useful to the scientist as well as the teacher in the class room. Since the basic structure follows the design of the Seamount Catalog closely, we could move our attention to the basic data population of the database. We have pulled together multibeam data for the MOR segments from various public archives (SIOExplorer, SIO-GDC, NGDC, Lamont), and pre-processed it for public use. In particular, we have created individual bathymetric maps for each ridge segment, while merging the multibeam data with global satellite bathymetry data from Smith & Sandwell (1997). The global scale of this database will give it the ability to be used for any number of applications, from cruise planning to data

* Gotberg, N K (ngotberg@ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92093 United States
Koppers, A A (akoppers@ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92093 United States
Staudigel, H (hstaudigel@ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92093 United States
Perez, J (jmperez@ucsd.edu) , Scripps Institution of Oceanography, 8800 Biological Grade, La Jolla, CA 92093 United States

Seamounts are important to research and education in many scientific fields, providing a wide range of data on physical, chemical, biological and geological processes. In order to make a diverse set of seamount data accessible we have developed the Seamount Catalog in EarthRef.org, available through the http://earthref.org/databases/SC/. The primary goal of the Seamount Catalog is to provide access to digital data files on a large assortment of interdisciplinary seamount research. The catalog can be searched at a variety of ability or expert levels allowing it to be used from basic education to advanced research. Each seamount is described in terms of its location, height, volume, elongation, azimuth, irregularity, rifts, morphological classification and relation to other features. GEBCO (General Bathymetric Chart of the Ocean) gazetteer data (2002; 2003) is included in the database in order to provide information on the history, discovery and names of the seamounts. Screen-optimized bathymetry maps, grid files and the original multibeam data files are available for online viewing with higher resolution downloadable versions (AI, PS, PDF) also offered. The data files for each seamount include a map made from the multibeam data only, a map made from Smith and Sandwell's (1996) predicted bathymetry, a merged map incorporating both data sets, and a map showing the differences between the two data sets. We are working towards expanding the Seamount Catalog by integrating bathymetry data from various sources, developing and linking disciplinary reference models, and integrating information from multiple disciplines and from the literature. We hope to create a data integrative environment that provides access to seamount data and the tools needed for working with that data.

* Sarbas, B (sarbas@mpch-mainz.mpg.de) , Max-Planck Institut fuer Chemie, Postfach 3060, Mainz, 55020 Germany
Nohl, U (nohl@mpch-mainz.mpg.de) , Max-Planck Institut fuer Chemie, Postfach 3060, Mainz, 55020 Germany

Since its introduction about five years ago, the geochemical database GEOROC of the Max-Planck-Institut für Chemie in Mainz established itself as a major resource available to the scientific community. In spring 2003, together with the databases PetDB and NAVDAT it has initiated the EarthChem consortium, with the aim to increase the synergy between the three geochemical database efforts and to build up a broad cyberinfrastructure for the Solid Earth Geochemistry. GEOROC provides geochemical data published for volcanic whole rocks, minerals and inclusions from ocean islands, large igneous provinces and convergent margins. Currently, the database is being expanded to samples from Archean cratons. Data for representative greenstone belts from Canada, South Africa, Russia, and Australia have been included with the latest update. The database now provides about 170,000 analyses published in 4,100 papers. The web interface allows the selection of samples by bibliographic, tectonic, geographic, petrological as well as chemical criteria. Recently, GEOROC has been implemented to an SQL server. This improved the performance of the web interface considerably, compared to the former MS Access-based version. The speed of queries as well as the possible size of data downloads increased. In addition, it removed restrictions in the possible number of simultaneous users of the database, an improvement that is essential when considering the constant increase in number of users to currently more than 5000 per month. In cooperation with the other members of EarthChem, the quality of the petrological metadata will be improved. At the moment, the rock name is adopted from the original publications without modifications. As a result, GEOROC contains more than 2000 different rock names. For many samples, instead of a rock name only terms like "pillow" or "lava" are available. Thus, further classification tools, besides the already available (Na$_{2}$O+K$_{2}$O) vs. SiO$_{2}$ TAS diagram, are necessary to help the user to query for rocks based on chemical composition, irrespective of the rock name given in the original publication. The revision of metadata will also include additions of geographic information beyond that given in the original publications. A consistent structuring of locations into given sequences from major to minor location subunits is necessary to utilize the full potential of the GEOROC web interface. To further improve the capabilities of the databases GEOROC, PetDB and NAVDAT, the development of a common web interface is in progress. This interface will contain basic query criteria based on bibliography, geography and rock names and will allow simultaneous access to all three datasets. More specific researches have to be done in the individual web pages of the databases. A further extension of GEOROC and also part of the common web interface will be spatial data queries and analyses. They will allow the selection of samples by drawing polygons on maps as well as the plotting and interpretation of chemical data against geography.

Building the EarthChem System for Advanced Data Management in Igneous Geochemistry

* Lehnert, K (lehnert@ldeo.columbia.edu) , Lamont-Doherty Earth Observatory, 61 Route 9W, Palisades, NY 10964 United States
Walker, J D (jdwalker@ku.edu) , Department of Geology, University of Kansas, Lawrence, KS 66045 United States
Carlson, R W (carlson@dtm.ciw.edu) , Carnegie Institution of Washington, 5241 Broad Branch Road, NW, Washington, DC 20015 United States
Hofmann, A W (hofmann@mpch-mainz.mpg.de) , Max Planck Institut fuer Chemie, Postfach 3060, Mainz, 55020 Germany
Sarbas, B (sarbas@mpch-mainz.mpg.de) , Max Planck Institut fuer Chemie, Postfach 3060, Mainz, 55020 Germany

Several mature databases of geochemical analyses for igneous rocks are now available over the Internet. The existence of these databases has revolutionized access to data for researchers and students allowing them to extract data sets customized to their specific problem from global data compilations with their desktop computer within a few minutes. Three of the database efforts - PetDB, GEOROC, and NAVDAT - have initiated a collaborative effort called EarthChem to create better and more advanced and integrated data management for igneous geochemistry. The EarthChem web site (http://www.earthchem.org/) serves as a portal to the three databases and information related to EarthChem activities. EarthChem participants agreed to establish a dialog to minimize duplication of effort and share useful tools and approaches. To initiate this dialog, a workshop was run by EarthChem in October, 2003 to discuss cyberinfrastructure needs in igneous geochemistry (workshop report available at the EarthChem site). EarthChem ran an information booth with database and visualization demonstrations at the Fall 2003 AGU meeting (and will have one in 2004) and participated in the May 2003 GERM meeting in Lyon, France where we provided the newly established Publishers' Round Table a list of minimum standards of data reporting to ease the assimilation of data into the databases. Aspects of these suggestions already have been incorporated into new data policies at Geochimica et Cosmochimica Acta and Chemical Geology (Goldstein et al. 2004), and are under study by the Geological Society of America. EarthChem presented its objectives and activities to the Solid Earth Sciences community at the Annual GSA Meeting 2003 (Lehnert et al, 2003). Future plans for EarthChem include expanding the types and amounts of data available from a single portal, giving researchers, faculty, students, and the general public the ability to search, visualize, and download geochemical and geochronological data for a wide variety of rock types of global distribution. We also hope to implement more mapping and query functions to open the application of these data across the spectrum of geoscientists. Lastly, we want to facilitate data submission and data entry in anticipation of incorporating more data directly from publishers and from a broader cross section of researchers. The building of interfaces for the transfer of raw and reduced data will allow data to be archived, and most importantly accessed, in a manner that will facilitate the use of geochemical data to address a wide variety of problems in the solid earth sciences.

Making Water Chemistry Data From Volcano-Hydrothermal Systems Accessible Using Open Source Tools

* Venezky, D Y (dvenezky@usgs.gov) , U.S. Geological Survey, 345 Middlefield Road, MS 910, Menlo Park, CA 94025 United States
Mariner, R H (rmariner@usgs.gov) , U.S. Geological Survey, 345 Middlefield Road, MS 434, Menlo Park, CA 94025 United States
Hurwitz, S (shaulh@usgs.gov) , U.S. Geological Survey, 345 Middlefield Road, MS 439, Menlo Park, CA 94025 United States
Evans, W C (wcevans.usgs.gov) , U.S. Geological Survey, 345 Middlefield Road, MS 434, Menlo Park, CA 94025 United States

Chemical and isotopic data collected over several decades by the U.S. Geological Survey from volcano-hydrothermal systems were recently organized into a web-accessible database for public use. The data were collected by members of the Barnes and/or Mariner projects and were supplemented with data from samples submitted for analysis by other researchers with similar interests. The data are primarily chemical and isotopic analyses of waters (thermal, mineral, or fresh) and associated gas (free and/or dissolved) collected from hot springs, mineral springs, cold springs, geothermal wells, fumaroles, and gas seeps. Additional data for a few streams, lakes, and oil wells are included. The web site follows a multi-stage design, first allowing for basic access to the MySQL database, then a user-friendly GIS (Geographic Information System) interface, and finally access to additional documentation and searching features. The initial web pages allow the user to choose the type of data (site, physical parameters, major and minor dissolved constituents, dissolved and free gas composition, water isotopes, and other isotopes) and the sample location. The data are then shown in a table that can be downloaded in several formats. The second stage of the project added an open-source GIS package called WorldKit, which gives easy-to-code and easy-to-use clickable icons on a base map using XML (Extensible Markup Language). WorldKit is also adding a zoom interface (zoomify) that uses new technology to reduce the display time. The final stage of the project involves more complex queries, alternative data presentation, and integrated background information. The more complex queries allow users to select multiple types of data from multiple sites. The data can be found at http://hotspringchem.wr.usgs.gov/.

Author(s) (2004), Title, Eos Trans. AGU, 85(47), Fall Meet. Suppl., Abstract #####-##.