Preprints of the Metadiversity Conference Proceedings

  Session 2: The Challenge in Species Discovery and Taxonomic Information

Discover Life in America & the Database Needs of the All-Taxa Biodiversity Inventory (ATBI) of Great Smoky Mountains National Park

JOHN PICKERING, Associate Professor at the Institute of Ecology, University of Georgia; Chair of Discover Life in America

Globally we are faced with a colossal challenge. In the next several decades, we must document and understand the natural history, environmental requirements, distribution, and ecological function of millions of species. If we succeed, we will be able to manage natural communities wisely and continue to benefit from our planet's natural bounty. If we fail, many species will be lost forever, ecosystem functions and services will decline, and our environmental and economic well-being will be threatened. Better scientific understanding and successful education programs could reverse the ongoing, widespread environmental decline. However, the world's scientific and educational communities must make fundamental changes in their modi operandi if we are to succeed. The task at hand is too large for those currently involved in science and education to accomplish. So much information needs to be collected, analyzed, and distributed that we cannot rely on traditional means. We must change our ways, involve the public, and embrace informational technology–most notably, we need to build a trained army of individuals who collect and share scientific information via the Internet.

The number of species on the planet is unknown. Current estimates generally range by an order of magnitude, from 3 to 30 million, possibly more. Yet fewer than 1.5 million species have been named scientifically, and we know little about the biology of most of these. If the current rate of habitat destruction continues, we are likely to lose many potentially beneficial species before they are known to science. Without changing our methods and organizational structure to meet the challenge, there are simply too few taxonomists and ecologists to get the basic information on each species in time to assure that human pressures will not be responsible for mass extinction. We need to develop better ways to allow non-specialists, particularly students, teachers, and amateur naturalists, to supplement the efforts of professional scientists.

Here I consider how we might organize and coordinate such an army of individuals, describing a goal of a new organization, Discover Life in America, to coordinate a large, comprehensive study of all the species in Great Smoky Mountains National Park (GSMNP) in eastern North America. I will address some of the difficulties in mustering students and others to help professional scientists collect quality data and posit possible technical solutions that we may wish to develop to help in our global endeavor.

All-Taxa Biodiversity Inventory of the Great Smokies

Discover Life in America has started an All-Taxa Biodiversity Inventory (ATBI) of the Great Smoky Mountains National Park (GSMNP). The ATBI will collect information on the distribution, abundance, and natural history of the estimated 100,000 species that inhabit the Park. Our goal is to disseminate information that is useful in resource management, science, education, and recreation. In particular, we wish to make detailed information on the natural history and ecology of all species available to the wider, non-specialist audience. We plan to develop interactive identification guides and Web pages for each species. We will make these available through our site's Flora & Fauna section. While completing the ATBI, we will develop methods, train personnel, and form partnerships that will facilitate inventories of other parks and conservation areas.

Our Park offers several advantages for such a comprehensive study. Its geography, geology, and climate have made the Smokies one of the areas of highest species richness in the temperate zone, possibly rivaled in diversity only by parts of eastern China. It is geographically located near many universities, museums, and other research centers, and it has a large user constituency–an estimated 9 million visitors annually. The Park contains over 500,000 acres, including the largest remaining stands of eastern old-growth forest, which total over 100,000 acres. Its terrain can be rugged and boasts 16 mountains over 6,000 feet.

Of the estimated 100,000 species within the Park, less than 10 percent have been documented to occur within the park's boundaries and many–on the order of 10 percent–are likely to be new species to science. Because of the large number of species that need to be studied and because of the logistic difficulty of sampling such an extensive, rugged area, Discover Life in America hopes to team volunteers with scientists to more effectively complete the study. If one assumes that each scientist on average will oversee the study of 100 species, then it is evident that each scientist is likely to need considerable help in sampling the park to understand the distribution, habitat requirements, and natural history of their organisms. Furthermore, they will need help in photographing specimens, building geo-referenced databases, Web pages, and interactive identification guides for each taxon.

By developing local, national, and international partnerships among educators, researchers, resource managers, and other concerned citizens, the ATBI gives individuals from all walks of life an opportunity to study nature. In helping with the research, school and community participants will get hands-on experience with scientific methods and state-of-the-art technology. Students will do science, use technology, and learn valuable skills, such as how to collect, process, and present information in a meaningful way. They will learn to discover the diversity of life and uncover its wonders. We invite individuals who are interested in contributing to go to our Website at http://www.discoverlife.org, register under "Get Involved," and then contact us. In addition to research professionals, we encourage nature lovers from all walks of life–teachers, naturalists, photographers, writers, and others, young or old–to become involved.

Obstacles to Overcome

To enable an individual who is not a taxonomic expert to contribute useful data to the ATBI, we must be vigilant about data quality. We must build the appropriate checks and balances to allow us to understand the source and approximate the level of accuracy of each datum. For an army of volunteers to collect data and present information on a global scale, we are presented with the additional challenge of coordinating, assembling, and presenting their efforts. The World Wide Web is a wonderful tool to assemble and present information–but how reliable is what it yields? How do we oversee the quality of information that we collect and present? Extensive sampling by volunteers could contribute significantly to our knowledge of the distribution of species and of their habitat requirements, for example. Conversely, the inclusion of misidentified specimens into geo-referenced databases could cause considerable confusion about ranges and detract from the value of the ATBI.

The data that we need to collect have a fairly simple structure. For the most part, they pertain to documenting events in time and space. Events such as observations, collections, photographs, sound recording, sensor reading, and molecular and chemical analyses will need to be databased. For each event, critical data will include (1) WHO (or what) collected the information; (2) WHEN it was collected; (3) WHERE it was collected; (4) WHAT it is or pertains to; (5) HOW it was collected; and (6) RELATIONSHIP or link to other events in the data structure. We anticipate using barcodes and other methods to assign a unique identifier to each event. These unique identifiers will allow us to track event handling, such as the distribution of barcoded vials and specimens to taxonomists. They will also facilitate linking parent-child relationships, such as the processing of parasites from a host, and analyzing individual events as members of a set, such as generating species maps from the WHERE of individual specimens.

Although the exact fields and structure of the database have yet to be chosen, we need to address certain issues and select standards to meet our needs. For example, the WHO variable will be critical in determining the level of confidence that we can place in an event's data records. We need to make sure that each WHO participating in data collection and analysis is uniquely identified from other participants. We should also specify WHO's level of expertise and update this level as WHO is trained. Thus, we should be able to put confidence weights on the level of data reliability over time for each participant.

Similarly, how should we standardize WHEN and WHERE across the database? Will date and time be Y2K compatible? Should WHERE be specified in latitude and longitude so that the data are globally accessible, or in UTM coordinates so that they can integrate into the Park's Geographic Information Systems (GIS), or both? How should we document and use confidence intervals for WHERE? Such records will include everything from very accurate GPS or surveyors' measurements to historical information from museum specimens that simply state location as GSMNP. These are questions that need to be resolved. We seek help and advice and do not wish to reinvent the wheel.

We have divided up taxonomic responsibilities into approximately 20 taxonomy teams that will coordinate the activity of their members and oversee and correct the work of the volunteers working with them. These teams are listed under "Who's Involved" on our Website. For each species within a taxon, team members will oversee the answering of questions such as: How do we identify it? What is its name? Or who will name it, if it is new to science? Are we going use the taxonomy in ITIS or the Species 2000 Project as authority files? Where can the species be found in terms of both latitude-longitude and habitat type? How is the species observed or captured? How does it make a living? What is its importance in the ecosystem? What is the effect of rainfall, deforestation, and fragmentation on the species?

For each species we intend to build a homepage with text, images, and, when appropriate, video and sound recordings. Because of the labor involved with building these Web pages, we anticipate that much of this work will be done by students under the supervision of teachers. To date, over 300 pages have been started by students at the University of Georgia and by elementary and secondary students who are participating in our program to enable teachers of minority students to study biodiversity and put their findings on the Web. Ultimately, these pages will be refined and edited under the supervision of taxonomic experts. We also intend to link them to our database, in part so that distribution maps can be dynamically generated.

The primary scientific impediment to our success in studying the planet's multitude of life is our inability to identify it. If we cannot tie event records together with correct names, we will have a jumble of disjunct facts. The participation and training of taxonomists who can put names on species and build identification guides is the first essential step in the process of collecting the information that we need about each species. Because non-experts cannot contribute to the wealth of knowledge about a species if they cannot identify it, we intend to build interactive keys on the Web that will allow students and others to identify species easily, using images and non-technical characters as much as possible. We are developing a software package called "20 Questions." When finished, the goal of this software is to create a tool to identify all species. The computer will ask the user up to 20 questions about a specimen, tell you what you have, and then link to the species homepage. We have developed a demonstration of this technology for species of Rogadinae, a subfamily of parasitic wasps (see "Identification Guides" under "Flora & Fauna").

Our database planning is being coordinated in conjunction with USGS-BRD. Our initial plans include using commercially available Biota software to manage data entry and specimen-based event records. We intent to integrate machines running Biota into a network that includes a Unix-based SQL database, so that records can be uploaded and downloaded. Eventually, as we empower numerous volunteers to collect and contribute data, I envision that we will develop the ability to submit data directly through forms on the Web.

Coda

The ATBI is a long-term project that we hope can be accomplished in 15 years. If successful, we will achieve a comprehensive understanding of the species and community interactions within the Great Smokies. In terms of general principles, we will gain scientific understanding on a scale and resolution never attained before from any ecological study. With regard to the Smokies specifically, Park managers will be able to better manage their natural resource using broad-based knowledge of numerous taxa.

If we can successfully develop and implement an information management system that allows non experts to collect useful data on the distribution and natural history of species, we will have a very powerful tool to extend the technology beyond the Park. Ultimately, I envision having a palm-size device that will combine existing technologies so that novices can contribute to inventories as they walk along trails. Such a device might include an interactive key to allow them to identify plants and caterpillars, a global positioning system to record their exact location, a digital camera to take voucher photographs, and the ability to communicate with the Internet from remote locations. With a press of a few buttons, a photograph is automatically sent documenting a WHO, WHAT, WHEN, WHERE, HOW, RELATIONSHIP between a plant and insect.

Yes, the challenge of inventorying life is colossal. But we could complete it in time.

Previous | Next

Copyright © 2003 NFAIS. All rights reserved. No part of this product or service may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior written consent.

Privacy Policy