IN13C-1169
Open Access to Data - Central Role for Geoinformatics
The open access to scientific information has become a contentious issue. In the United States there are calls to make all published literature available for free within 6 months of publication, the notion being that this will promote better science and policy decisions based on science. Here, I argue that this is the incorrect approach to the issue of open access to scientific information. A fundamental problem raised by the call for open access to government-supported research results is the viability of our not-for-profit professional scientific societies. These societies provide the base level framework for the exchange of scientific ideas, and hence the very core of how we do science and how scientific knowledge is advanced. Why should a scientist subscribe to a journal if they can read the article for free in six months? A large portion of a society's operational costs come from these subscriptions and the sale of specialized books, all of which contain the results of federally-funded research. Without revenue from journal subscriptions and book sales, not only will these publications disappear, but many of societies may as well. Without a broad venue to publish and in which to interact, our science suffers - many subdisciplines may fade or even die - those that don't "sell well." Very popular publications such as "Nature", "Science", "Tectonics", and "Geology" will continue to thrive, but what about the more specialized journals such as "Journal of Paleontology"? They are costly to publish yet fill a very critical niche for our science. Many will still pay for reading the Nature/Science/Tectonics/Geology article, but where do we publish the mainstream science paper? We have to guard against becoming a "Hollywood Science" - where only the glitzy gets published because those are the articles that sell. We must have peer-reviewed, independent publications and viable professional societies, or our science will severely suffer. We can better approach the need for open access to scientific information by concentrating on the data versus the written word. The written word can and should be copyrighted. This protects the viability of the journals of many societies - large and small - and therefore the viability of the societies and the science they support. Furthermore, the written word is largely interpretation - some of which flows directly from the funded research, but much of which reflects the accumulated knowledge of the scientists; knowledge that has been derived from sources that cannot be tracked. Interpretations and conclusions differ among scientists - that is what drives the progress of science, and that is the written word. The notion that all journal articles must be free to all effectively says that ideas and interpretations cannot be protected by copyright. It should also be noted that all articles are available in libraries as they are published, so it is hard to argue that they are not "freely available." What hinders science and public policy decision making, is the lack of complete access to all relevant data and metadata. To require that all relevant data and metadata be publically available six months after publication is a viable solution. The government dollars clearly pay for the data, but the source of support for scientific interpretation and discussion is impossible to determine. Once the data are public, they are free for all to reinterpret and to use as the basis for additional studies that reflect the unsolved issues of the previous study. Many of the government agencies that supply research funds already have data policies in place. What may be required is the funding base to allow them to implement these policies in consultation with the academic community that they serve.
IN13C-1170 INVITED
Digital Earth: Development of advanced software and databases to integrate plate motions with other datasets
Plate reconstructions enable analysis and interpretation of biologic, geologic, palaeoclimatic and palaeogeographic data throughout time and space. For this task, peer- reviewed digital databases and complementary software is urgently needed. No current plate reconstruction software can take into account plate deformation, link plate kinematics with variations in lithospheric rheology, predict �dynamic' topography, or incorporate as boundary conditions the convective processes of the mantle. The international GPlates initiative aims to accomplish these tasks within one code. In this contribution we discuss a prototype software package for GPlates (dubbed SPlates) that has been tuned for close interaction with non-commercial and commercial GIS software. We focus on software functionality and database connectivity and notably address how new databases of geological features such as terrane boundaries, sedimentary basins etc. must be combined with absolute (palaeomagnetic and hotspot) and relative (magnetic anomaly, fracture zones) reference frame databases in order to construct high-quality plate reconstructions. We also show examples on how appropriate reference frames linked to up-to-date time-scales, by providing kinematic velocity fields through time, provide critical input to mantle convection simulations oriented towards an improved understanding of subduction dynamics, the generation of large igneous provinces, and dynamic topography.
IN13C-1171
A 4D Framework for Ocean Basin Paleodepths and Eustatic Sea Level Change
A digital framework for paleobathymetry of the ocean basins requires the complete reconstruction of ocean floor through time, including the main ocean basins, back-arc basins, and now subducted ocean crust. We reconstruct paleo-oceans by creating "synthetic plates", the locations and geometry of which is established on the basis of preserved ocean crust (magnetic lineations and fracture zones), geological data, and the rules of plate tectonics. We reconstruct the spreading histories of the Pacific, Phoenix, Izanagi, Farallon and Kula plates, the plates involved in the Indian, Atlantic, Caribbean, Arctic, Tethys and Arctic oceanic domains and all plates involved in preserved backarc basins. Based mainly on the GML-standards compliant GPlates software and the Generic Mapping Tools, we have created a set of global oceanic paleo- isochrons and paleoceanic age and depth grids. We show that the late-Cretaceous sea level highstand and the subsequent long-term drop in sea level was primarily caused by the changing age-area distribution of Pacific ocean floor through time. The emplacement of oceanic plateaus has resulted in a 40 m sealevel rise between 125 and 110 Ma, and a further 60 m rise after 110 Ma, whereas the oceanic age and latitude dependence of marine sediments has resulted in a 40m sealevel rise since about 120Ma, offsetting the gradual post-80Ma drop in sealevel due to the ageing and deepening mainly of the Pacific ocean basin, with the net effect being an about 200m drop after 80 Ma. Between 140 Ma and the present, oceanic crustal production dropped by over 40% in the Pacific, but stayed roughly constant in the remaining ocean basins. Our results suggest that the overall magnitude of 1st order sealevel change implied by Haq's sea level curve is correct.
IN13C-1172 INVITED
The Five Dimensions of MagIC
The Magnetics Information Consortium (MagIC) provides information technology infrastructure for the international paleomagnetic, geomagnetic, and rock magnetic community. Currently MagIC provides the data model, web portal ( http://earthref.org/MAGIC), and related cyber-infrastructure for the interoperable RMAG (rock magnetic) and PMAG (paleomagnetic) databases. Paleomagnetic data are intrinsically metadata rich and cannot be fully exploited without information about both their geological and temporal context. Within MagIC, connections to geological information come initially from tables in the data model that are shared with EarthRef, and deal with information about publications, sampling expeditions, rock formations, sites, rock types and formations, and age determinations. It is thus possible to design searches for related geological, geochemical, or geophysical information across all databases residing under the EarthRef.org umbrella. Version 2.1 of the MagIC data model will make allowance for International Geo Sample Numbers (IGSNs) provided by SESAR the System for Earth SAmple Registration, providing the possibility of tracking related sample information across other databases. A challenge that remains to be resolved is how to integrate information from PMAG with independent new age controls and up-to-date time-scales, while respecting the specific age information available from a given paleomagnetic study. The related problem of determining paleo-locations requires both age information and a framework for plate reconstruction, along with an appropriate understanding of what they are based on. Although such information can be available from independent cyber-infrastructure efforts like Chronos and G-Plates it may not be straightforward to acquire in a readily useable form. Individual users may also have specific model needs. We will discuss strategies for implementing standard paleo-location and time scale models within MagIC, and initial ideas about how to access and assess the most current relevant information.
http://earthref.org/MAGIC
IN13C-1173
Experiences with the Application of Services Oriented Approaches to the Federation of Heterogeneous Geologic Data Resources
The federation of databases is not a new endeavor. Great strides have been made e.g. in the health and astrophysics communities. Reviews of those successes indicate that they have been able to leverage off key cross-community core concepts. In its simplest implementation, a federation of databases with identical base schemas that can be extended to address individual efforts, is relatively easy to accomplish. Efforts of groups like the Open Geospatial Consortium have shown methods to geospatially relate data between different sources. We present here a summary of CHRONOS's (http://www.chronos.org) experience with highly heterogeneous data. Our experience with the federation of very diverse databases shows that the wide variety of encoding options for items like locality, time scale, taxon ID, and other key parameters makes it difficult to effectively join data across them. However, the response to this is not to develop one large, monolithic database, which will suffer growth pains due to social, national, and operational issues, but rather to systematically develop the architecture that will enable cross-resource (database, repository, tool, interface) interaction. CHRONOS has accomplished the major hurdle of federating small IT database efforts with service-oriented and XML-based approaches. The application of easy-to-use procedures that allow groups of all sizes to implement and experiment with searches across various databases and to use externally created tools is vital. We are sharing with the geoinformatics community the difficulties with application frameworks, user authentication, standards compliance, and data storage encountered in setting up web sites and portals for various science initiatives (e.g., ANDRILL, EARTHTIME). The ability to incorporate CHRONOS data, services, and tools into the existing framework of a group is crucial to the development of a model that supports and extends the vitality of the small- to medium-sized research effort that is essential for a vibrant scientific community. This presentation will directly address issues of portal development related to JSR-168 and other portal API's as well as issues related to both federated and local directory-based authentication. The application of service- oriented architecture in connection with ReST-based approaches is vital to facilitate service use by experienced and less experienced information technology groups. Application of these services with XML-based schemas allows for the connection to third party tools such a GIS-based tools and software designed to perform a specific scientific analysis. The connection of all these capabilities into a combined framework based on the standard XHTML Document object model and CSS 2.0 standards used in traditional web development will be demonstrated. CHRONOS also utilizes newer client techniques such as AJAX and cross-domain scripting along with traditional server-side database, application, and web servers. The combination of the various components of this architecture creates an environment based on open and free standards that allows for the discovery, retrieval, and integration of tools and data.
IN13C-1174
An Update on CoreWall Tool Suite Development: An Integrated Approach for Data Discovery, Description, Visualization, and Correlation
The CoreWall Project is an NSF funded initiative (March, 2006) aimed at providing new geoinformatics tools to meet the broadly similar needs of various coring communities. A community wide workshop, sponsored by JOI- USSAC, was held in Washington DC in May 2006 to help establish the requirements for the Corewall Suite of software tools. The Corewall Suite consists of four main components: Corelyzer, Nclip, CoreNavigator, and the Workflow Database. This software is cross-platform compatible, requires normal computers, utilizes GeoWall technology, and is available over the web at www.corewall.org. Corelyzer is a wide-screen high-resolution computing facility for logging ice, lake, sediment and hard rock cores. It is described as a Collaborative Interactive Core Analysis Environment, allowing dispersed (even international and ship-side or in-the-field) logging and interpretation of core representing earth history via the internet. In addition, Corelyzer can retrieve images and data files from remote databases or web portals/services, such as CHRONOS. Corelyzer is designed to increase the efficiency of research, and the rate of new idea development and validation. NClip is newly upgraded version of the old IODP Splicer/Sagan software that is used for core-core or core-log integration. NClip provides graphic, interactive data analysis products for depth-merging and integrating core and downhole log data. CoreNavigator is a data discovery/ data browser tool that takes advantage of Google Earth capabilities to both find and explore existing data sets in spatial queries. The Workflow Database is a built in database that allows the user to work both locally or remotely from other databases and allow data to be both accessed and moved on the fly. Details are available: http://sqlcore.geo.umn.edu/CoreVault/docs/WorkflowDbPoster.pdf Through 2006, the Corewall group has made substantial advances: (1) Deployed Corelyzer for the first ANDRILL Antarctic field coring program along with a plug-in support for their sediment description software, PSICAT. (2) NClip version 1.0 was released in August, 2006 that provided upgraded and enhanced core-core integration. (3) CoreNavigator is improved deployed with Ocean Drilling and dbSEABED datasets supported with a Google Earth interface to index globally extensive data collections. (4) Ability to launch Corelyzer from the internet. (5) Interactive windows located to points on the core within Corelyzer can accept data, commentary, and data files, with outputs logged to the point. CoreWall software will be demonstrated at the meeting using a range of datasets including lake (LACCORE), ocean bottom (IODP), marine piston cores (TAMU Gulf Mexico hydrates).
http://www.corewall.org
IN13C-1175
Spatial Indexing of Datasets for CoreWall: CoreNavigator
Corewall is a community computing facility for logging ice, lake, sediment and hard rock cores. It is described as a Collaborative Interactive Core Analysis Environment, allowing dispersed (even international and ship-side / in- the-field) logging and interpretation on sections of core representing Earth history. Many institutions support Corewall, including NSF. CoreNavigator is a 3D Visual Indexer of core and stratigraphic datasets, necessary because GIS systems available to researchers do not adequately display the vertical stratigraphic structure, or let users browse at will through the stratigraphy. CoreNavigator including its Google Earth extension, is likely to be a primary point of entry for the community into CoreWall. By a combination of 3D VRML and KML visualization technologies CoreNavigator indexes thousands of cores for user selection leading to a variety of actions. By clicking on visual 3D elements of CoreNavigator, users can obtain tables of integrated ready-to-use data (e.g., from dbSEABED, see web). They can also drill down into the original field notes, core photos, equipment types, lab analysis files, calibrations, etc. They can launch applications including the Corelyzer part of Corewall. [CoreNavigator 3D VRML displays are also editable and publishable, and can have seismic, oceanography, culture objects inserted. In the Geowall environment they are a resource for education.] CoreNavigator will be demonstrated as part of Corewall. By adopting a single spatial - global approach in this way to all types of cored stratigraphic data - ice, sediment, rock sea and lake - researchers will be able to transfer their enquiries and validation exercises in questions of environmental change, across the whole Earth surface.
http://www.corewall.org