Imagine a fisheries resource manager collecting samples of fish in various streams throughout a watershed area. Upon returning to the office, he accesses an online fish-identification program, and quickly determines that his catch includes several alien species. Another Internet database tells him the native ranges of these species, their reproductive requirements, ecological preferences, natural predators, and parasites. Now the manager deduces that only one of the alien species is likely to survive and spread throughout the watershed.

Concurrently, he combines environmental data specific to each fish's native range (measured by satellite passes for the past 20 years) with the habitat conditions of the other areas where they have been introduced - and thereby ascertains which management strategy is most likely to succeed in the current scenario. In a mere few hours, the resource manager has used a rich network of online databases to monitor the populations of invading fish species before they can grow out of control.

Does this kind of online database management tool sound far-fetched or impossible? It's not. In fact, scientists worldwide will soon be able to use this type of electronic network "yellow pages" that highlights all of the world's plants, animals, and microorganisms. Called the Global Biodiversity Information Facility (GBIF), the linked databases will be the first to include a definitive list of the world's 1.5 million named species. This initiative, sponsored by the Working Group on Biological Informatics of the Megascience Forum of the Organisation for Economic Co-Operation and Development (OECD), is the largest of several that plan to make collections-based information accessible to both general and scientific communities by digitizing it and storing it on networked databases.

A Global Resource

"The Global Biodiversity Information Facility will be a fantastic resource for everybody," says James Edwards, an assistant director at the National Science Foundation in Washington, D.C., and one of several scientists involved in the project from among the OECD's 29 industrialized nations.

For those having trouble envisioning the massive scope of the GBIF, Steve Young, a computer specialist working on the project at the Environmental Protection Agency, advises, "Bear in mind that the GBIF is not a monolithic facility like a new giant telescope or particle accelerator. It will be an organic, distributed, growing, evolving facility, much like the World Wide Web itself."

GBIF's Arrival: Step-by-Step

Edwards explains that the GBIF will become functional in stages. In March, an Internet site will open. Its ultimate goal: to provide a Web page for every species on the planet. Links to existing data will then be created, and information from other online projects that maintain inventories of the world's species will be integrated into the site. Edwards expects some of these shared database links to be operational by January 2001, although more in-depth integrated searches may not be possible until 2002 or later.

The databases that operate jointly within the GBIF will remain the property of the institutions that created them. "This seemed more appropriate and more practical than starting a single central database for biological information from scratch," said Michael Oborne, deputy director for the OECD in Paris.

Thomas Lovejoy, chief biodiversity adviser to the World Bank and one of the planners of the GBIF project, suggests its greatest appeal lies in updating and validating the efforts of natural history scientists worldwide: "Natural history began hundreds of years ago and has stayed behind many younger core science fields - such as molecular biology and genome research - that have always used network databases to share their research."

Rapid Transit

Faced with the immediate need to implement conservation plans in an effort to save dwindling populations of flora and fauna worldwide, scientists will be able to use the GBIF to obtain information rapidly. Scientists, for example, will greatly simplify their searches and conclusions by identifying data sets for a species or family with the click of a mouse.

"Right now, if I wanted to know all the specimens of a particular bird,'' Edwards says, "I would have to write to all the museums that may have a specimen, then go to the museums myself and look at labels and cards. Our initial focus is to get the data from the tags and the drawers into the computer, and - where possible - include images."

And the Poor Get Richer

In addition to providing untold benefits to scientists, professors, researchers, biology students, and natural science students, the massive and comprehensive nature of the GBIF system promises to assist poorer countries that are rich in diversity of species but have no species cataloging system of their own. These developing countries currently host most of the world's living plant and animal species, Edwards points out, but information and specimens on these species are mostly housed in the databases and natural history collections of northern European and North American universities, zoos, and museums. Strange as it may sound, a field ornithologist from South America seeking information about the native tanagers he is observing daily may have to travel overseas to access an archival collection. For this reason, Lovejoy emphasizes, "the idea [of creating GBIF] is to pull all of this together and make it available to everybody."

Challenging Times

Initially, Young says, the GBIF is expected to have a staff of around 10. Developed countries will pay more to join than developing countries. But once all parties begin sharing their biodiversity databases with the world, more complex challenges are expected.

One of the most formidable roadblocks recalls the recent cataloging and search engine maladies that plagued the World Wide Web when the general population first began browsing for information. Dozens of search engine companies - among them Yahoo, AltaVista, Lycos, and Excite - were challenged to their performance limits when troubling technical questions sometimes resulted from linking independent databases.

Edwards is not trembling with fear at the similarly complex technical barrier facing the GBIF in its quest to link, organize, and track the world's huge volumes of biological data. "Until a few years ago, the technical hurdles in linking disparate datasets were formidable," he says. "In recent years, Web-based search engines have begun to proliferate, and it is now feasible to do distributed searches across a wide array of different kinds of data and database formats."

Common Bonds and Links

Crucial to overcoming the challenge of linking databases, Edwards explains, is identifying a common element when a scientist begins a search within GBIF. "For most biodiversity data, a valid species name will provide that common element; this is why GBIF has placed such an emphasis on completing a catalog of the names of known organisms. For other kinds of data, other common elements must be used; for example, biogeographic data use place names of latitude-longitude coordinates.

"Part of what we'll have to do is to come up not only with the named species but also all their synonyms," Edwards says. "Now that we're trying to computerize things, we'll have to weed out any ambiguity that exists. This will be of value not only to scientists but also to policy makers. If people prepare a new law, say to protect a species, there can be no ambiguity."

Shared Knowledge

Is there opposition to the GBIF? Especially from developing countries that may both be wary of foreign intrusion upon their ample supply of natural resources and mindful of past invasions by multinational companies looking to obtain virgin resources? Not yet, say the project's developers, and Young points out: "One of the key aspirations of many of the GBIF organizers is to support the repatriation of information back to source countries; in other words, to share with the countries of origin of biological specimens information about those specimens, so that the knowledge is common property. We believe this will be well received by the developing countries. If they are able to access the knowledge without incurring all the costs of curating large collections over a long time, then this should be very beneficial to them."

The project's organizers are also quick to point out that great pains have been taken to establish constructive relationships with existing governmental and nongovernmental agencies in the informatics and conservation fields. To date, these strategic and diplomatic efforts have included soliciting the involvement of the Convention on Biological Diversity, and its subsidiary bodies, the Clearinghouse Mechanisms, and the Subsidiary Body on Scientific, Technical, and Technological Advice. The Convention will have a free, nonvoting seat on GBIF's Governing Board.

Indeed, scientists worldwide are expected to use the GBIF regularly as they "find and use biodiversity data from other scientists and organizations and use the GBIF as a way of 'publishing' and exchanging their own scientific findings with others around the world," according to Glady Cotter, associate chief biologist of the Biological Information Resources Division of the United States Geological Survey.

Of greater concern to Cotter may be the incompatibility of GBIF systems with existing biodiversity and ecosystem databases. Citing her work for the United States' current version of a biodiversity database, the National Biological Information Infrastructure (NBII), she emphasizes that the key to successful exchange of GBIF data among scientists will be maintaining consistent infrastructure standards when listing regional, national, and global biological information. These catalogs currently exist, for example, as part of online Natural Heritage Programs that most states already operate, and in other regional biodiversity listings.

The NBII, in fact, will be the first building block of the GBIF database, along with similar worldwide projects. Some of these online programs, including current ones in Australia, Britain, and Canada, offer biological information on their major domestic species, but these countries' biodiversity inventories are also not yet complete. In addition, many of the world's historical plant and animal collections, such as those in Britain, France, Germany, and Russia, often date back to the 18th century, and have not been entered into computer records.

"Creating a GBIF is connecting those [worldwide] pieces into a larger whole; it's not preempting [them] in the least," Lovejoy says.

Integrated Virtues and Extended Features

Other GBIF developers are also enthusiastic about extended features that will eventually make the integrated biological network popular among professionals, like economic planners, in fields other than the environment, such as pharmaceutical companies that do biological prospecting. These features - called "vertical links" - will facilitate management decisions based on linking the species databases with others dealing with soil, climate, human populations, animal distribution, vegetation patterns, watersheds, and others.

A simplified "vertical link," Cotter suggests, might challenge scientists to use the GBIF to locate, retrieve and synthesize data from multiple sources as they investigate the effect, say, of a particular pesticide on an amphibian known to be declining in population size in certain areas of North America. Using the GBIF network, the results of the study could then help resource managers identify the climates where the species lives, its competitors in an ecosystem, and how all of them might be affected by the pesticide.

Edwards declares his own growing optimism for the value of the GBIF is partially based on a current "real world" success - the integration of environmental information in Sweden. "There," he said, "use of environmental information is contributing to the development of environmental policies, assessment of environmental impact, and plans for sustainability. For example, information about plant and animal species in Sweden has been integrated with climate models, to predict the distribution of a range of species under a number of climate change scenarios developed from global climate models. This information has then been integrated with other information about vegetation for use in the development of species management plans."