Preprints of the Metadiversity Conference Proceedings

  Session 2: The Challenge in Species Discovery and Taxonomic Information

ITIS, The Integrated Taxonomic Information System

BRUCE B. COLLETTE, Senior Scientist, National Marine Fisheries Service Systematics Laboratory

First I would like to say a few words about the Integrated Taxonomic Information System, or ITIS. ITIS is a unique government organization because it is a non-bureaucratic. It has no director and no direct funding. It exists only because the people involved had a vision: that we need a standard, yes a standard, name–not as an expression of taxonomy–as a way to get from one database to another database. So, for the purpose of communicating between databases, we have to agree on a name. It would be nice if it were the "right" name, but the important thing is to have the same name so that someone can get data about a species from all your databases. And that is simply what ITIS is all about–a way of doing this.
It is not the same way that Frank Bisby spoke of. We are coming from different directions and hopefully we will meet in the middle. One of the reasons that ours is different is because we started with an existing database of hundreds of thousands of names that had already been entered. This was the National Oceanographic Data Center (NODC) database, which we loaded and now need to modify.

An Introduction to ITIS

What I want to do this morning is tell you a little about the ITIS project, including the key components of interrelationships with other projects, where we are now, and where we are going. ITIS is an online database (see http://www.itis.usda.gov/itis) built through partnerships with the world taxonomic community, sponsoring agencies, and organizations. Our goal is to provide on the Web quality taxonomic information about all organisms from both aquatic and terrestrial habitats. (There are, according to various accounts, 300,000 to 400,000 names available now.) Our original focus is North America because the governmental agencies involved are North American. But where databases are available, like for mammals of the world, we go global immediately.

Creating a Standardized List

We believe that informed decisions for managing our biological heritage can best be made with easy access to the wealth of information already existing about plants and animals. The problem is communicating between databases. As I said before, we have to agree on the same name so that we can get into each other's databases. In fact, we started as an organization because several federal agencies responsible for managing the nation's biological heritage found that they had information stored under different names about the same organisms, so they could not communicate with each other. This was on both the federal and the state level. We needed interoperability.

As a result, our goal is to standardize credible lists of species names, which have unique identifiers. This falls within the recommendation of the National Research Council and other agencies that have said that taxonomy is important if we are going to manage biodiversity. I once spent 10 minutes trying to explain to Jim Baker, the head of the National Oceanic and Atmospheric Administration (NOAA), why there is not a list. There are, of course, lists of lots of things–but we don't even have a list of commercial species within the United States. Thousands of taxonomists have been working on this for hundreds of years, and we still have a long way to go. Depending on the source, there are from 6 million to 40 million entries we need to make. The names are in Latin. The original descriptions are dispersed among thousands of biological journals, from obscure societies and little museums, and in all languages. The only thing you can read in some of them is the Latin.

Rules to Follow

We have rules of nomenclature. We also have separate rules for zoology, biology, and bacteria in constructing the infrastructure of ITIS. It was necessary that the business rules for each of the codes were integrated into the structure so that the names would not violate those codes. Yes, classification is constantly changing, and it has to be constantly updated. But that is the beauty of the Internet–it can be updated.

The History of ITIS

ITIS actually began in 1972 in the Chesapeake Bay region, when the Virginia Institute of Marine Sciences started one of a lot of local codes. This was then taken on by the National Oceanographic Data Center. NODC is responsible for archiving physical chemical data, including biological data.

This was in the 1970s, and computer fields were limited, so it was not a good option to put in the entire species name. A code was better. So, a flat file, a so-called intelligent number system, was devised. The first number is the phylum, the second number is the class, and the third number is the order. This works okay up to a certain point, but then it collapses because you have too much data. Further, the emphasis was on quickly putting the names in the database because the data were available. Employees were charged with getting names in quickly, and this meant that many of the names entered were unreviewed.

In 1985, EPA entered into a partnership with NOAA. This then broadened to include what is now the United States Geological Survey (USGS). In 1992 there was formal commitment to replace the code. As a result, in 1993, the real effort started to develop a relational database based on a system of classification, so that, for example, if you wanted data on a given species and there were taxonomic problems with identification, you might be able to go up a level in the relational database structure and get the data on a generic level.

Where We Are Now

We went online in 1996, and it became necessary to migrate the old National Oceanographic Data Center database into ITIS. This means we inherited some good data and some bad data. This also means we now have a big cleanup job. So when you go to our Web site and you see bad data, it is because we have not yet cleaned it up. The reason is some people were using those names and those codes, and they can continue to use them. We will create links so you can type in the old name, the scientific name, or the common name and the system will tell you the currently accepted scientific name. There is also a "change link" so that you can get back to the original taxonomic serial number. So, where we are right now is getting new data and cleaning up our old data. In this sense we need help from the systematics community.

Partnerships

Our partners include the Department of Agriculture, NOAA from the Department of Commerce, the USGS, EPA, and the National Museum of Natural History. We are directly linked to the KDI project in the University of Kansas, and recently we have added Canadian participation. These are active partners that are participating as we develop our program.

ITIS as a Relational Database

The system is built around a relational database. The database includes scientific names, the authors, the dates in a single classification, a unique identifier, and a taxonomic serial number (TSN). In addition, there are associated data, some of which are obligatory and some of which are "nice-to-know" data that we put in when available. There is an online system that can be queried. Reports can be asked for. There is a system that allows you to compare two lists of names if they are in the proper format. In addition, you can download the data. In fact, if you don't like our particular system of classification, download it and change it. It is there for people to use.

Taxonomic Workbench

We also have a taxonomic workbench, which is designed to enter data. We are still working out the best procedures to make this more interoperable, to make it more accessible to people in the field, and to make corrections. Right now all updates and changes funnel through us.

This is a simplified database, consisting of a scientific name with a number for computer purposes–the taxonomic serial number–behind it. Synonyms are linked to it. There is a record of change-tracking from the time a name is originally entered. Every time there is a change there is a reason for the change. There are publications sited. There is a series of vernacular names linked, and it is indicated if these are approved by some organization or other authority. There is a lookup table of authors of publications and of species. There is a comment field for anything else that is not required but about which we have information (for example, is it an endangered species?).

Our Homepage

Visit our homepage to see what we have to offer. You’ll find you can query the database by typing in a common name or a scientific name. You can generate reports. You can extract the scientific name and other data from ITIS. You can download anything you want and modify it any way you want. You can compare different databases because when you have local lists from different places, there are going to be a lot of matches (and some non-matches–you actually might want to focus on those non-matches and figure out why they are not matching).

You also will see on the homepage places to pull up publications, experts, and names. Credibility can be found in the right-hand corner. It also will alert you regarding whether the data have been reviewed. If they have not, be cautious–we haven’t gotten to that record yet. Remember, we don't have millions of dollars and hundreds of people, so we have not yet achieved the goal of updating all the data. But we will get there. In fact, you can help us by visiting our site, then telling us how you think it could be improved.

Reaching Our Goal

One way we’re trying to reach our goal is to take some of the small amount of money that is available and contract with systematists to produce lists for groups that are important. For example, we have a contractor working on a beetle list for North America. And we have an algae contract.

In addition, we are particularly interested in finding old-time taxonomists who still have lists of species on 3x5 cards. We want to make sure that we can get that information captured electronically while these people are still alive. Taxonomy is itself an endangered discipline. In many groups there is only one expert in the entire world. If we don't get that information now, we will have to do it all over again, which is not very cost effective.

We also have people to evaluate the data. Many of these are systematists based at the Natural History Museum of the Smithsonian. Many others are, and will increasingly be, in other parts of the country and from other parts of the world. Experts will review the data and make a rational decision on which name to use.

We have a data-development team that is trying to obtain new sources of data, new lists, just as Frank Bisby does for his distributive system. We acquire the data from whatever source we can get, but then it has to be developed. It has to be formatted. It then must undergo peer review. Hopefully, it is certified. Then it is loaded and managed online.

All of this is accessible on the Internet. We realize we cannot make changes as fast as we would like. There are some glitches and problems in development, as there are in any large system. But we still have lots of names up there, and names for many important groups are in very good condition.

We also plan to interact with different groups. Information presumably will flow out from ITIS to groups like the public and science writers, people who want the correct name for an organism for a high-school report or a story they are writing. This is the place to get that correct name.

We are trying to be representative of good systematics, but we also have to be practical and results have to be immediate. We have to manage our resources now–not with tomorrow's taxonomy, but with the information we have right now. We cannot wait for perfect taxonomy. We have to make some decisions to give you some standard names to enable you to move back and forth between the databases right now.

Interaction with the systematics community is essential. The data stewards come from the systematics community. The data sources come from the systematics community. And peer review is provided by systematists.

We are in cooperation and coordination with a long suite of organizations including Species 2000, CONABIO in Mexico, and FAO, which produces aquatic species catalogues on taxa of importance. FAO covers mostly food fishes, but they include turtles and other organisms. All the names of all the organisms of the FAO catalogues are in, updated, and correct in the list. Bit by bit we will move through and get the rest of the names updated.

In addition, we have been endorsed by the National Performance Review and Access America. We have space at the National Museum, where we are in direct contact with the largest group of systematists in this country. We have a data-standardization process. We have hundreds of thousands of names in there. We have been recognized by Vice President Gore with a Hammer Award for Interagency Cooperation for making information available at a really cheap cost. There is very little direct funding money. As I said before, it is a volunteer effort.

Where We’re Going

So, where do we go from here? We have to finish cleaning up the data. We have to expand the geographic and taxonomic coverage. We have to redo the Web page to make it easier to retrieve information. We have to become a bureaucracy, because at some stage the buck has to stop someplace and somebody has to be in charge.

We also need somebody to encourage the sponsoring cooperating agencies to contribute more money so we can get the job done. We need to expand partnerships with various organizations. We have to expand our relationships with the systematics community. When I say these things, some of my taxonomic colleagues won't talk to me because they think I am trying to make the ITIS system a standard. But it is not a standard–it is just a means for helping people move among and use various databases. It is just a method of communication.

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