Stratigraphic Workshop: Building a Comprehensive Suite of Integrated Databases, Study notes of Chemistry

Download Stratigraphic Workshop: Building a Comprehensive Suite of Integrated Databases and more Study notes Chemistry in PDF only on Docsity! 1 INTEGRATION OF CHRONOSTRATIGRAPHIC DATABASES FOR THE 21ST CENTURY SUMMARIES OF NSF-SPONSORED WORKSHOP (AMHERST, MA, NOV. 8-10, 2001) AND CHRONOS STEERING COMMITTEE RECOMMENDATIONS FOR 5-YEAR PROGRAM Contents I. Executive Summary II. Mission Statement, Chronos Goals, and Applications III. Summary of Chronos Database System IV. Chronostratigraphic Data Issues– Types, Standards, Interpretation, Compatibility, Data Usage V. Database Design Issues – Public Interfaces, Toolboxes, and Information Analysis VI. 5-Year Plan and Budget VII. Appendices -- Participants, Workshop Issues, Agenda, and Selected Current Databases 2 I. Executive Summary Workshop Goals and Themes of Workshop A group of thirty quantitative stratigraphers and database specialists from diverse research teams met in a three-day workshop (Nov, 2001) to discuss methods and strategies for: • Assembling, integrating and distributing data relevant to geologic time • Producing a dynamic global timescale to frame Earth history events and processes for social benefit. Three major working sessions, each consisting of focus teams and full-group discussions, considered scientific and information technology aspects of database development, including: • Coordination of database standards and compatibility • Designing strategies and tools for information retrieval and analysis of all types of global and regional stratigraphic data • Discussing future directions for database integration and centralization of currently distributed depositories. A subset of workshop participants, representing information technology and various stratigraphic shareholders, prepared a detailed 5-year plan to accomplish the Workshop Chronos concepts. Major Decisions and Recommendations (1) A chronostratigraphic database system is vital for future earth science studies. Establishing an integrated chronostratigraphic database, provisionally to be called Chronos, will facilitate greater efficiency in stratigraphic studies. By providing both greater ease of data gathering and allowing for metadata synergies, the Chronos system would be of fundamental importance in a wide variety of research, including time scale construction, paleoenvironmental analysis, paleoclimatology, paleoceanography, and paleotectonics. Beyond scientific research, Chronos would provide educational and societal benefits by providing a readily accessible source of information on topics of general interest (e.g., evolution, frequency of catastrophic events in earth history) and facilitating studies of issues of modern concern (e.g., atmospheric change, biodiversity). 5 Applications and Components Unified suite of thematic hubs There is an intrinsic value in developing a unified suite of chronostratigraphy databases that that seamlessly unites several distributed thematic hubs. The chronostratigraphic data to be included or linked within this system include: • Life (paleontological assemblages, evolution, biodiversity, extinction, productivity proxies from carbon isotopes and other criteria, terrestrial and marine trends, etc.) • Climate (orbital forcing, glaciations, temperature records from oxygen isotopes and other proxies, ecosystem changes, dust accumulation, etc.) • Surface processes (carbon and other biogeochemical cycle monitors, weathering balances from strontium isotopes and other data, sediment accumulation rates, etc.) • Core-Mantle dynamics (magnetic reversals and intensity trends, rates and directions of plate motions, hydrothermal fluxes of elements, volcanic ash frequency, etc.) • Catastrophic episodes (iridium and other impact-related anomalies, oceanic anoxia episodes, climatic excursions, etc.) • Time (absolute ages derived from -- radiometric decay, astronomical cycles, annual varves and other methods; and relative ages derived from correlations to global and regional geological stage boundaries) Purposes and major products of Chronos database network The purpose of the chronostratigraphic database system is improved access to chronostratigraphic information in order to provide dynamic temporal resolution for earth science data and interpretation. • Standard Geological Time Scale is the centerpiece of the database, both as a specific deliverable as well as a format for underlying data retrieval. • This Standard Time Scale is linked to geochronologic and biostratigraphic time scales for dynamic handling. • The Standard Time Scale is based on the highest quality data. An important function of the associated database is to continuously improve the precision and accuracy of this Standard Time Scale. • The Standard Time Scale is agreed upon, updated and maintained by the international community. • The Standard Time Scale is transparent – the foundation data and interpretation methods are given. Indeed, a user can determine the degree that each datum is critical to the Standard. • Database readily ties the chronostratigraphic control to the broadest range of correlative biostratigraphic, geological and physical data. 6 • Database provides the capability to go to an outcrop and link it to the Standard Time Scale. The database would enable error bars to be placed on the precision of these correlations. • Chronostratigraphic data should be tiered to focus accuracy and precision of available datasets. • Database structure must be dynamic as a platform for continuous improvement and revisions. • Database should be a catalyst for defining new research directions and problems. • Database would be linked to other Earth System and geoscience resources throughout the world, especially those which have data that bear on the time scale. Benefits of Chronos database to Earth System research Major research areas that would benefit from this integrated chronostratigraphic database include: • Evolution and controls on biodiversity • Catastrophes and abrupt climatic change (e.g., the end-Cretaceous impact, the end- Paleocene methane release, the Oligocene transition, ecologic change and recovery following critical events) • Climate history and climate oscillations (e.g., greenhouse and icehouse trends, orbital forcing, glaciations) • Basin modeling (sediment and elemental accumulation rates, subsidence and thermal maturity, etc.) • Rate studies (e.g., magnetic intensity, geochemical fluxes, responses at human timescales) • High-resolution and high-precession geochronometry (calibrated sets of decay constants, relation of tectonic and volcanic episodes to biologic events, etc.) • Linkage of systems (e.g., terrestrial vs. marine biogeochemistry, influence of mantle processes on evolution) • Forward projection (e.g., sea level, climate, atmosphere, and evolution trends) All of these research areas require a time frame. In addition, the development of the Chronos integrated database system would advance the field of information management. 7 III. Summary of Chronos Database System Database Infrastructure DISTRIBUTED DISCIPLINE NODES CENTRAL CHRONO- STRATIGRAPHIC NODE Bio- strat Magneto- strat Chemo- strat Radio- metrics Cyclo- strat SUB-NODES etc.HYDRO SEISMO GEO- INFOR- MATICS PETRO etc. Central Hub and Distributed Nodes The Chronos database infrastructure (schematic above) will have a central hub linked to a suite of topic-oriented chronostratigraphic databases forming Distributed Nodes (e.g., Bio, Mag, Chem, Cycles, Radiometrics, Stable Isotopes). Each of these Distributed Nodes, in turn, will coordinate the efforts of many distributed database “sub-nodes”, which are the real basis of the 10 Chronos Steering Committee Tasks: 1. Information compilation and distribution: A. Brochure and Web version (new CHRONOS website at www.eas.purdue.edu/chronos/) B. White Paper (and associated detailed “proceedings” of the Workshop) –. This will be circulated to all participants, and the final version will also be a formal report to NSF. C. Publicity of the Workshop and Chronos initiative. 2. Prepare Strategic Plan 3. Acquire seed funding for tactical needs 4. Establish widespread collaboration and support 5. Establish Chronos development structure and funding coordination Initial Membership of Chronos Steering Committee • Academic – Charles Marshall, Jim Ogg • Government – Bruce Wardlaw • ICS – Felix Gradstein and Jim Ogg • Industry – Paul Sikora, Jeff Stein, Tony Gary • Public institutions – Brian Huber (Smithsonian) • Education – Cinzia Cervato • Adviser (NSF) – Rich Lane There is a 1-year “sunset” on each member’s term, so that new blood can be continuously involved in the planning, development and coordination. 11 IV. Chronostratigraphic Data Issues – Types, Standards, Interpretation, Compatibility, Data Usage Scope of Chronostratigraphic Database What data? • Data useful for correlation (all potential correlation tools). Also assemblages. • Use all available data, including gray literature, government reports, theses, etc. • No time-filter –from annual varves to multi-million-year sequences • No scale constrictions – use global, regional, and local • Need parallel databases (geochronology, isotopes, “biology”, etc.) that are inter- linked • Encourage funding to institutions that contribute databases for community use Data types • Two basic types: • Points (e.g., geochronology) • Sequential or stratigraphic (e.g., biostrat, chemostrat) • Two basic forms: • Quantitative (e.g., physical values) • Qualitative (e.g., taxa assignments, facies) • Marine versus Terrestrial • Core versus Outcrop • Biostratigraphy (and assemblages for paleoecology) • Stable isotopes (C, O, S, Sr) • Radiogenic (U-Pb, K-Ar, Sr, etc.) • Magnetostratigraphy • Cyclostratigraphy • Event stratigraphy (tephra, etc.) • Sequence stratigraphy • Dendrochronology • Varves • Ice cores Data Issues Data characteristics • Must include access to RAW data. • But, what is the “Raw” data? All published “observations” are interpreted to some degree. 12 • Compiling Raw data: • Retrospective acquisition (data behind previous interpretations) • Forward acquisition (promote standards for reporting and publishing results) from now on • Acceptance of qualitative observations – This is also important. • Digital image atlas for species, and where these samples are stored. Links. • Language not a limitation: Include a program (software) of translation from other languages for inclusion in database. • Sequencing (e.g., Sadler techniques) is important. Data quality • Difficult to assign quality, but minimum requirements established by Central Hub. • Full attribution of data and source • Data collection • Where, when, how collected, by who, and why. • Need to state what the reason was for sampling, and what are the implications and how this might limit its use for other purposes. • Data storage – Where and how • Details of lithology and geologic context of samples • Processing of data • Methods, instruments, sample preparation • Uncertainties, measurement precision, potential errors • Reference to published data • Published versus un-published • Everything would have “dates of entry” • Quality marking • By originator (who best knows the limitations) – why done, nature of database • By later users (e.g., reviews, as in some on-line book sellers). • However, even though such critiques would be very useful, these might cause legal/political problems. • After debating pros and cons (including legal aspects), it was decided not to include this feature of outside ratings, even if it would be useful for future investigators. Instead, links or pointers could be included to published re-evaluations or discussions of the data. Data interpretation • One applies Reference Standards and Models to the basic observations (databases) to derive Interpreted Datasets • Taxonomic concepts, taxonomic dictionaries • Radiometric decay constants • Magnetic stripes on ocean floor • The methods that raw data is processed to get interpreted datasets should be transparent. • Filters – user can select subsets based on different criteria. • Interpretations can be peeled off 15 • User Personal Workspace • Work with downloaded Chronos or uploaded personal data in temporary "workspace" • Have control of it until research is complete • Input data to workspace for using tools on central server • Example is ODP -- data is withheld until public release Ways to retrieve data (Search Engines) Search engines will be both comprehensive and user friendly (i.e., accessible to the general public). Likely possibilities include geographic retrieval through GIS map capability and graphic retrieval via mouse clicks on various chronostratigraphic displays such as time scales and paleomagnetic columns. Index and keyword searches will also be included, with user options for binary or fuzzy search parameters. Distinction will also be made between searches for descriptive or qualitative data (i.e., for basic education and general public inquiries) and raw data for scientific research. • Drilling starting with metadata • Example -- Logic of the 200 Ma age assignment for base-Jurassic • General "drill-down" to specific data • Specific "drill out" across nodes and disciplines • "Drill up" to implications of data item • Index or keyword searches • Meta-database to contain keywords, indices • Taxonomy -- synonymies • Images or photographs • Key index fossils • Cores • Queries • Stored or canned queries • Capture frequently used queries • Options to edit parameters of saved queries • User can create canned queries • Cookies • Forms to provide selections • Concept search tools • San Diego Supercomputer Center has been working on such tools • Knowledge-based system Centralized and Distributed Data • Central server • Contains metadata for routing to other nodes • Contains some "Core" data • Time scales • Essential data behind "Standard Time Scale" • Hosts some "orphaned" databases 16 • Distributed thematic nodes • Contain or link to raw data from individual databases • Contain or link to data processed through interpretation models • One issue is the archiving of data Results of Searches Search engines will enable data integration and capability for downloading. Downloads of the search results will be provided in formats (e.g., Excel, ASCII, comma-delimited files) most applicable for importation to a wide variety of applications. • Three main pathways for users: • User is curious to know an answer => Return "answer" to query with supporting data • User wants the data for own purpose => Return basic or raw data for screen listing or output • User wants to examine or analyze data on-line => Return data into personal workspace, which has set of tools to do "complete job" • Scope of data availability • Temporary display of holdings and relevant expanded sets for further selection by user • Simple visualization is coupled with data retrieval to aid user in more sophisticated selections • Fuzzy query or fuzzy logic • Bring back data within bounds • Fuzzy logic applied (from here to here with some uncertainty applied, score applied, and retrievals are graded) • Data compiler • Retrieves raw data from distributed databases • Aggregates data • Sorts data by different categories • Data interpreter • Distinguishes between raw data and interpreted data • Ability to apply simple interpretation models (e.g., a time scale) • Manipulates and displays (e.g., chronogram, seafloor spreading rates) • Limited ability to "do the science" • Data is exported in common "portable" forms • Low-level tables (ASCII, Tab-delimited) • Plots in PDR • Options for integration into GIS -- Spatial information analysis • For time-event data, one can choose which time scale is applied (e.g., the current Chronos "standard", a previous "standard", or other common published scales) 17 Toolboxes, Workspace, and Other On-Line Capabilities Chronos will have a smooth, sophisticated suite toolbox of applications for all users. The visualization and research tools will be flexible, and designed so that a high-school student can apply them without requiring advanced training. Beyond basic visualization tools, Chronos will also have more sophisticated capabilities for data manipulation and metadata transformations. Providing licensing issues can be resolved, these will include statistical applications (e.g., SAS plots) and programs for ranking and scaling and graphic correlation. Most tools can be smaller versions of larger commercial packages. Some tools will be available for download into the user's host computers, or they can apply them within a temporary personal Workspace. Tool development would be in public domain. In addition, output options will include format conversions to other common tools. On-Line Data Visualization and Manipulation Output from searches can be put into other tools (on-line & commercial) for visualization and simple analyses. The ability to "do the science" can be accomplished in a personal temporary workspace. • Visualization tools will be available to user • Two options for users: • Work with tools on data in central server workspace • Download selected tools into one's personal computer • Use what's available ("Buy versus Build") • E.g., SAS/SAS Plot • Export in format for uploading into Excel or other personal software • License issues • Make available unique tools • RASC, GraphCor • Develop tools that are not already available • Public domain • Examples of visualization and analysis tools • Graphic presentation of integrated stratigraphies • Fossil ranges • Creation of zonations using available algorithms (Unitary Associations, RASC, Graphic correlation, General operations) Other Database Issues Dataset Submission (Direct or Indirect) • Encourage contributions • Prioritization of what data should be captured • Standard forms • Thematic nodes should capture data • Distributed nodes handle their own databases • Responsibilities of nodes include security 20 Provides a template for future rich client development Central Hub Standards All open source / free software – Tomcat, Java (JDK), Apache, etc. All open standards and protocols – SOAP, XML (Xforms, XSLT, etc.), UDDI, WSDL No locking of data on certain platforms or operating systems Database (Could be any OS . and any JDEC/ODBC reachable DRAFT concept for creation of database) Database Service Publisher with Central Hub to coordinate services and interface methods along with initial reference client for users. Some Tarms: NA UDDI (Universal Description Discovery and Integration) WSDL (Web Services Description Language) SOAP (Simple Object Access Pratacal} XML (eXtensible Markup Language) JDBC ODEG Forms (an XML UI application} ee S 5 S Serviet After locating appropriate service a client talks se ‘Chaining directly to the the SOAP publishing servlet. The ae 2 structure of this communication will be defined by 8 gE ISOAP publishing servlet their WSDL for the service 2a8 Bs e Serviet Engine (Tomcat) ___—__ Boar) id eta 8 6s Java Virtual Machine _ B28 a a5 oes Host OS = |] Uses UDDI to register 9 service with repository Uses UDDI to register service with repository ‘of hosted or mirrored Central . databases Hub ‘Client uses UDDI to Thematic SL find appropriate service (SOAP/XML) Service Index Repository Thematic Hub - provides guidelines and support to locat database creators - replication or hosting of databases as needed or requested White Pages: Provider names ~ feedback mechanism to and information Central Hub on development and usage Yellow Pages: Providers “The clients used to access. issues, arranged by location, type, may be either the reference and protocols in use client or a custom client developed by other Green Pages: Specific API groups and interface information and specific information about individual services 21 22 Time line for Central Hub (Information technology functions) Year 2003: - Staffing and equipment integration - Creation of initial Database Service Publisher Software Development Kit (SDK) with Java Database Connectivity (JDBC) interface to allow for publishing of data services on the net - Draft initial XML scheme for Database Service Publisher Simple Object Access Protocol (SOAP) publisher and initial Extensible Stylesheet Language transformation (XSLT) document - Draft of the Web Services Description Language (WSDL) spec for initial Database Service Provider SDK. This allows for a means for the automatic user interface creation on the client. - Simple test client for Database Service Publisher SOAP publisher - Assess the integration servlets the Database Service Publisher will need in order to interface to the databases and pass the information along via the SOAP servlet Year 2004: - Add Universal Description Discovery and Integration (UDDI) ability to the initial Database Service Provider SDK. UDDI will allow the databases to register their services with the central hub for any to search and access a service. - Bring Universal Description Discovery and Integration (UDDI) repository on-line to allow the initial databases to begin publishing their services to the repository - Creation of initial draft reference client with XForms integration - Creation of addition Database Service Publisher database interface servlets to access a greater range of database configurations - Draft WSDL, XForms and UDDI specs published allowing other groups to begin creating clients. In addition the source code for the initial client would be opened for all to use as a baseline. Year 2005: - Final WSDL, XFORMS and UDDI specs published allowing based on feedback from users of the year 2 draft - White/Yellow/Green Pages are created and published to allow for more detailed information about the groups, organizations, databases and published services - Continue interface servlet development as needed for Database Service Publisher - Aggressive deployment of Database Service Publisher to the various Databases Year 2006: - Full deployment mode - Rich client development (more advanced client interface mechanisms) Year 2007: - Full network systems support - Publish final API specs for Database Service Publisher, Central Node systems, schemes and both the reference and rich clients 25 F. Budget Estimate (see attached Excel table for details by project and institution) F. Initial Chronos key institutions (preliminary suggestions) Central Hub (tasks and portals are possibly shared among more than one institutions) Iowa State University (Cinzia Cervato) Biostratigraphy Node EGI at University of Utah (Paul Sikora, Tony Gary) – Database integration and tool development Harvard University and Smithsonian (Charles Marshall, Brian Huber) – Node & Paelobiology database USGS (Bruce Wardlaw) – Paleodata database Iowa State University (Cinzia Cervato) – Neptune database American Assoc. Strat. Palynology (Martin Farley) – Palynodata database Ocean Drilling Program – Janus database Universities of Bremen and Kiel – Ocean Drilling Stratigraphic Network (ODSN) databases Other Stratigraphic Nodes MIT or Berkeley (initial workshop) – Radiometric thematic node Purdue University (Jim Ogg) and Scripps Institution of Oceanography (Cathy Constable) – Magnetostratigraphy node Johns Hopkins (Linda Hinnov, initial workshop) – Cycle stratigraphy node University College London (John McArthur, initial workshop) – Geochemiscal stratigraphy node Other institutions for eustatic, ice core or other thematic node workshops are yet to be suggested $ 1.2 M100K300K300K300K200KTime-slice $ 8.8 M $ 2.3 M $ 2.7 M $ 0.5 M $ 2.2 M Total 350K350K400K800K800KBiostrat node $ 1.3 M$ 1.7 M$ 1.9 M$ 2.1 M$ 1.6 MTOTAL 300K550K700K500K200KOther nodes 100K100K100K100K50KOutreach 450K450K450K450K400KCentral hub 20072006200520042003Task 26 Time-Slice Projects Boise State University (Permian Research Institute) and Nanjing University (China) for initial workshop – Permian-Triassic database assembly and trial Rutgers University (?) for initial workshop – Mid-Miocene database assembly and trial University of Utah (Paul Sikora) for initial workshop – Mid-Cretaceous database assembly and trial Oxford University (?) – Cambrian database assembly and trial ************************************************************************ Acknowledgements The Conference Center at the University of Amherst arranged meeting facilities, audio- visual and Internet systems, hotel accommodations, coffee breaks, and luncheon and dinner banquets. Funding for conference travel, web site logistics, and facilities at University of Amherst was provided by a workshop grant from the National Science Foundation. Purdue University provided secretarial and business office support and website hosting at no cost. 27 VII. Appendices A. Workshop Participants Name, e-mail, web Mail address, Phone Expertise, Role BARNOSKY, Anthony (Tony) barnosky @socrates.berkeley.edu http:// ib.berkeley.edu/labs/barnosky /adbprofile.htm Museum of Paleontology University of California 1101 Valley Life Sciences Building Berkeley, CA 94720-4780 Tel: (510)- 643-6275 FAX: (510)-642-1822 PALEO, coordinating on-line availability of many paleo databases (Paleontology Database Network) BOBE, Rene Bobe.Rene @nmnh.si.edu Evolution of Terrestrial Ecosystems Program Dept. of Paleobiology, NHB MRC 121 Smithsonian Institution, Washington, DC 20560 (202) 357-3033 office (messages with Paleoecology lab) Geraldine E. McBrinn ETE database administrator): (202)-357-1974) (703) 750-3443 home (Bobe) Fax: (202)-786-2832 PALEO, Evolution of Terrestrial Ecosystems project CERVATO, Cinzia cinzia_spencer @hotmail.com Dept. Geological and Atmospheric Sciences 253 Science Hall I, Iowa State University Ames, Iowa 50011 Tel: (515) 294-4477 FAX: (515) 294-6049 PALEO, ISOTOPE, strat databases (NEPTUNE and NORGES) with global correlation CONSTABLE, Cathy cconstable @ucsd.edu http:// mahi.ucsd.edu/tauxe/cathy/ Institute of Geophysics and Planetary Physics, Scripps Institution of Oceanography, University of California San Diego, La Jolla, CA 92093-0225, USA Tel: (858) 534-3183 Fax: (858) 534-8090 MAGNETICS, databases and geomagnetic analysis DIVER, Patrick diverpl @bp.com BP Exploration 501 Westlake Park Blvd., P.O. Box 3092 Houston TX 77253-3092 SYSTEMS, Stratigraphic database design DAVYDOV, Vladimir C/O wsnyder @boisestate.edu http:// pri.boisestate.edu/ Permian Research Institute Boise State University MG205, 1910 University Dr. Boise, ID 83725 Tel: 208-426-???? Fax: 208-385-4061 PALEO, PaleoStrat system (specialty is Permian) FIRTH, John Ocean Drilling Program, Texas A&M U i it INTEGRATED, SYSTEMS, O D illi d t b l b l 30 vspiess @uni-bremen.de Klagenfurter Strasse, P.O.Box 330440 D-28359 Bremen, Germany Tel: (49) 421-218 3387 Fax: (49) 421-218 7179 Home: (49) 421-382 751 stratigraphy network), on-line access and tools STEIN, Jeff steinja @bp.com BP Exploration 501 Westlake Park Blvd. P.O. Box 3092 Houston TX 77253-3092 SYSTEMS, chronostratigraphic integration TAUXE, Lisa ltauxe @ucsd.edu http:// sorcerer.ucsd.edu/tauxe/tauxe.html Scripps Institution of Oceanography Univ. Calif. San Diego, La Jolla, CA 92093- 0220 Tel: (858) 534-6084 Fax: (858) 534-0784 MAGNETICS, stratigraphic and other applications VEIZER, Jan veizer @science.uottawa.ca http:// www.science.uottawa.ca/ ~users/jveizer/ http:// www.science.uottawa.ca/geology/ isotope_data/ Department of Earth Sciences, University of Ottawa 365 Nicholas St., P.O. Box 450, Stn. A Ottawa, Canada K1N 6N5 Tel.: (613) 562-5800 ext.6461 Fax: (613) 562-5192 and Institut fuer Geologie, Mineralogie und Geophysik, Ruhr Universitaet 44780 Bochum, Germany Tel.: +49-(234) 32-23250 Fax: (234) 32-14571 ISOTOPE, carbon-oxygen array through Phanerozoic WARDLAW, Bruce bwardlaw @usgs.gov U.S.Geological Survey, MS 926A National Center 12201 Sunrise Valley Drive Reston, VA 20192 Tel: (703) 648-5288 FAX: (703) 648-6953 INTEGRATED, statistics applied to stratigraphy Plus, logistical help by: Chunfu ZHANG (Purdue University) – Web site for Workshop czhang6@purdue.edu, Tel: 1-765-494-0257 Steve NATHAN (Univ. Massachusetts) – Logistics and computer assistance at Workshop snathan@geo.umass.edu, Tel: 1-413-545-2593 Sent Regrets: Kay Berensmeyer (Smithsonian), Samuel Bowring (Univ. Harvard), Lucy Edwards (USGS), Jason Hicks (Denver Museum), Steve Holland (Univ. Georgia), Susan Kidwell (Univ. Chicago), Ken Miller (Rutgers Univ.), A. Krishna Sinha (Virginia Tech.), Walter Snyder (Boise State Univ.), and Hermann Zimmerman (NSF) 31 B. Workshop -- Selected Objectives and Issues (Initial document distributed to participants at the Workshop) Introduction -- Selected Examples of Chronostratigraphic Databases and Approaches, and Future Needs. Representatives of selected academic and industrial database groups will present brief overviews and demonstrations of compilations, software capabilities, application tools, and planned objectives. These include outcrop and well records, stratigraphic and paleogeographic overlays, and global temporal compilations of isotopic and biological data. Considering these current and ongoing efforts, which major types of stratigraphic data (quantitative, qualitative, temporal, and spatial) is “falling between the cracks” or requires special emphasis in the future? What components of the distributed analytical toolbox are absent? How can future growth of present databases be most efficiently handled to prevent duplication of effort? These presentations will set the stage for the working groups on Friday and Saturday. Knowledge Management Strategy. For each type of database and for the entire distributed suite, a knowledge management strategy can be proposed based on desired input and output by stakeholders and an evaluation of current and future technology feasibility. The relevant task models are a key component. It is important that database contributors and users share their ‘perfect world’ desires with the developers, and that developers share their visions of future technological and analytical opportunities to encourage more creative utilization. Theme #1 Chronostratigraphic Data -- Types, Standards and Compatibility Stratigraphic Database Applications and Future Directions. What would be the ideal stratigraphic database system (contents and tools) for various types of Earth system history problems? Who are the current and future users? Which problems require focused and specialized databases, and which require a broad integrated system? How can we mine the stratigraphic information of the past century and the forthcoming information of the next decades? This major topic is part of all themes and will be a major goal of the Workshop. Chronostratigraphic Data Types and Standards. Chronostratigraphic data comes in many flavors that are intimately associated and inter-calibrated with each other. How can the values and inter-relations of different data types be included into a database system? Each type of chronostratigraphic data (fossils, radiometric ages, sedimentary records of sea level or climate oscillations, paleomagnetic polarity, stable isotopic ratios, etc.) has different methods of interpretation, reliability control, and associated quality. What background information should be included, how are obsolete values removed, etc.? Industry – Data Sharing, Database Usage. Research teams in the petroleum industry and government surveys have been leaders in assembling extensive and comprehensive stratigraphic databases. Examples include the former Amoco well and outcrop database (now a repository at the University of Utah maintained by the Energy & Geoscience Institute), the ‘BIOLOG’ program at ELF Exploration-Production, the Late Paleozoic biostratigraphy database for Eurasia at BP Amoco (under John R. Groves), and the Mesozoic-Cenozoic sequence stratigraphy and chronostratigraphy compilation coordinated by Exxon under Jan Hardenbol and Martin Farley (published as a chart series in SEPM Special Volume 60 in 1998). Providing public access to portions of these confidential databases and stratigraphic tools in the industrial and government domains would be beneficial to all geoscience, and industry may gain substantially from integration of their subsets into a larger global compilation. It is only fair that industry and government surveys, which are potential major users of the stratigraphic database network, provide a share of the funding for the continued development and accessibility. Scientific Focus Group Suggested Discussion Issues IT Focus Group Suggested Discussion Issues Data Quality What to include? What type (e.g., raw or interpreted)? How, who decides? What disciplines? Compatibility Hardware Central server vs. distributed Best archival and retrieval methods Software How data stored (e.g., standard vs. multiple formats)? If standardized – best format?; import/ export issues 32 Paleontology – e.g., taxonomic consistency Radiometric – e.g., standardization of methodology Stable Isotope, Cycles, Paleomag – e.g., interpretation, methods Administrative Who makes these decisions?; For how long? If multiple – integration, linkage issues, HTML vs. XML Data Entry and Type Quality control Binary vs. fuzzy Compatibility issues Theme #2 Database Design – Data Retrieval and Analysis Data Exchange and Format Compatibility. Some systems, such as Amoco’s former in-house stratigraphic database and ODP’s site-core archive, were designed for flexible inclusion and manipulation of all types of lithologic, paleontological, geochemical, isotopic and chronostratigraphic data. Other systems have specialized tools for particular data types. Regardless of the internal database formats of each site, it would be ideal to establish a common format for efficient interchange among systems and applications. Public Interfaces and Toolbox. Ease of public access, including downloading, by all geoscience researchers must be a major objective of any public-funded database program. Currently, very few systems offer this capability. Possible types of queries would include global and regional event lists, range charts and population dynamics data and paleoenvironmental data for both modern or paleogeographic reconstructions. Other possible interfaces include the abilities for researchers to upload their own data either for inclusion in the main database (after quality controls) or for trial comparisons and to manipulate stratigraphic data from different repositories. A common Web directory with pointers to all databases and tools is essential. This Web page requires a dedicated web page administrator and communication of updates from participants. Placement or mirroring of this central Web site, selected databases and tools within the National Geophysical Data Center (NGDC) or World Data Center System is an option. Scientific Focus Group Suggested Discussion Issues IT Focus Group Suggested Discussion Issues Archival vs. Application What do we expect of our database? Data Table Structure Integrated vs. discipline-segregated Quantitative, qualitative or both? Output Type Who are expected users? What are their expectations? Raw data vs. interpreted? User Issues User interface – functions? Direct or indirect? Format – how friendly? Search Engines, Cross-indexes Data manipulation/retrieval Gathering data - distributed databases (e.g., NASA) vs. central database Output retrieval – data compiler vs. data interpreter User Issues User interface – type? Direct or indirect? Format – how friendly? Search Engines Cross-indexes Theme #3 Integrated Chronostratigraphic Databases – Future Directions, Needs, Frameworks Database Administration Mechanisms and Issues. How are the quality and reliability of database entries determined? Some paleomagnetic databases contain subjective ‘quality’ factors, and perhaps similar cautions are required for certain types of geochemical data and radiometric ages. A taxonomic dictionary and equivalence table is important for paleontological databases, and it is probable that other standardizations are required for chronostratigraphic data (e.g., monitor standards and decay constants) and other stratigraphic information. Temporary confidentiality may be a requirement for industrial well data or ongoing individual research efforts. Another aspect is intellectual property issues for data and tools. Visions, Strategic Goals and Timetables. Where do we wish to be with stratigraphic databases at the end of this decade and the next decade? What are the required innovations, data acquisition, user interfaces, manipulative tools, infrastructure, and other components for these information systems? Should stratigraphic data be available via a