GeoInformatics

[Note: In preparation for the workshop, the organizers have compiled a summary of the key elements and attributes of GeoInformatics as it is evolving in a series of reports and other workshops. This summary is intended to serve as the basis for discussions. If the domain sciences are largely in agreement on the basic design of GeoInformatics, we can more readily focus our time in the workshop on implementation and organizational issues.]

System design

The community appears ready to emulate the successful origin of the Internet and the WWW by designing and establishing an organic, innovative, and adaptive system that not only allows but indeed, requires dynamic interaction and development from the users of the system and voluntary contributors.

Define key attributes of the GeoInformatics system

The workshop organizers have tried to synthesize the major elements and attributes of a geoinformatics system as a strawman hypothesis to facilitate discussion at the workshop and to allow more time to be spent on developing the response to organizational and process needs.

We adopt a prototype in which geoinformatics is intended to be an open, free, persistent, user-driven system that is adaptive, flexible, innovative and organic (Allison, et al, 2002). It will be designed for research, but it is likely and in fact, desirable, that other applications and uses will spin off the initial system. Use of the system by government and industry would be an indication of its utility and influence.

Integration and interoperability are key elements, regardless of whether data are point, gridded, vector, or text. The most daunting challenges may be connecting existing databases into a distributed system and making this system accessible to all interested users. The system must honor the ownership and integrity of all data, and provide a user interface that facilitates widespread adoption. To be truly successful, the system must link data and tools not only from the academic sector, but also from all the other segments of the earth science community including from federal and state agencies, professional societies and groups, and industry.

Existing technologies need to be used, integrated, and widely applied to initiate the system. New technologies will certainly be spawned by the system, and others will be adapted by the system as they appear. Technology changes so quickly that the system must be dynamic. A rigid, static system will be obsolete before it is fully established.

The system needs to be accessed anywhere with ease (an informatics version of the worldwide web--infoweb). GeoInformatics will improve the democratization of the science. The system will carry capabilities to small and minority institutions and communities, worldwide.

The vision of the geoinformatics system that is quickly evolving includes these attributes:

0. Knowledge-based integration of multidisciplinary databases;
1. Persistence;
2. Recognizes an ongoing need to populate databases and make them available;
3. Advanced search and access tools for distributed databases, including user-friendly, smart, interdisciplinary search engines;
4. Online workspace, storage, software, and tutorials;
5. Grid computing technologies to provide access to high performance computational resources and large databases;
6. Flexible and extensible standards that are responsive to community needs, which also allow "personalization" of information views;
7. Providing a base set of capabilities to enable easy access and effective use of resources;
8. An array of distribution options - clearinghouses or central repositories (permanent), portals or nodes (long-term), websites (variable) and peer-to-peer (ephemeral) access;
9. Tools for data mining, visualization;
10. Middleware that serves as metadata engines between users and databases; and
11. Catalogues of databases.

Thus, a scientist might automatically accept, for example, new age dates posted by certain refereed journals that he or she has pre-selected as data sources. Other sources of data might generate a notification to the user of their availability, after which the user can decide whether to accept (or retrieve) them. A corollary is a local weatherman automatically incorporating meteorological data from a set of weather satellites and ground stations for routine forecast updates.

Eventually, it is expected that NSF and other funding agencies will require participation in the system and database (plus software, tools, etc.) dissemination as a condition of their funding. NIH has announced this requirement effective Oct. 2003 for recipients of large grants.

Organization

Service based: The research communities appear to be looking for someone to build and maintain the infrastructure so that scientists can use it but won't have to worry about it unless they want to, much as they use the telephone, electric lights, or highways now.

Funding for projects remains with NSF: The research communities are more likely to support a geoinformatics initiative that does not compete for funding with existing entities or serves as a "mini-NSF" by becoming another funding distribution center. Rather, we anticipate that NSF and other funding sources could set aside funds specifically for GeoInformatics-related projects, or preferentially select projects that contributed to or built upon GeoInformatics. In these ways, GeoInformatics would serve as a catalyst or focus for the traditional funding mechanisms.

Strategies to achieve goals

Federally funded research should require putting resulting data into the system. The system and data need to be in the public domain.

Operation

Creation of middleware (in essence a distributed hierarchical digital library) to provide protocols and a digital "card catalogue" system that will allow every database to be part of the system.

Discipline-oriented data coordination nodes or hubs will work with the scientific communities to ensure their particular needs and peculiarities are adequately and appropriately addressed.

A coordination center will compile, integrate, and create the tools, protocols, and standards necessary to establish and maintain the geoinformatics system.

Virtual/Digital Library System: a distributed, hierarchical system of permanent libraries/data repositories, databases, and ephemeral sources. A preliminary model is AGI's National Geoscience Data Repository System.

Other scientific disciplines

The oceanographic community is pursuing coordination efforts similar to those in the earth sciences. Others, including those in the fields of biodiversity and atmospheric and environmental sciences, who were represented in the Boulder workshop, will be doing so in the near future. While each community has its own needs and unique attributes, early coordination among all the sciences is crucial to the long-term successful implementation of GeoInformatics.

Proprietary industry systems

Many petroleum service companies and IT firms are putting together large databases, software tools, and support services for marketing to the petroleum industry. These systems tend to be proprietary and independent. Part of the goal is to create interoperability between these stand-alone systems and the larger geoinformatics system.