How does a specimen collected in the field winds up on the maps you see here? The collected specimen is sent to Bob Glotzhober. If it is already identified, Bob confirms the ID and records the collection info on a data sheet. If the specimen is not identified, Bob identifies it, or sends it to one of the other members of the Ohio Odonata Survey Steering committee (such as Bob Restifio or Carmen Trisler) for identification. Sometimes, unusual specimens are sent to experts in particular taxa for further study.
Once the information is recorded, it is entered into the main database, which is in the Microsoft Access format. This is either done by Bob Glotzhober or by one of several people Bob has contracted to do data entry. The database - which is about 18 megabytes in size - is then placed on a zip disk and sent to Dave McShaffrey at Marietta College.
At this point, a special program "scans" the database and looks for any unusually formatted entries which might cause problems. These are then fixed up.
The "clean" Access file is then opened with the HAGENIUS program (Hagenius is a genus of dragonflies that hunts down other dragonflies; we've adopted the dragonhunter as the symbol of our dragonfly data-retrieval software). The program has a number of options, one of which is a mapping module. Among the options in the mapping module is the ability to automatically generate web pages for each of the species maps. At this point, a single click of the mouse will set the computer to work. A 200Mhz Pentium with 32mb of RAM can print the 150 or so web pages in about 20-40 minutes. Each web page has three components: a GIF file of the species distribution map, a GIF file of the species time-of-flight line, and an HTML page with the species name, disclaimers, and links to the previous and next species. With all the species maps printed, the computer then generates a species list HTML page with links to each of the species pages.
The Visual Basic Program, HAGENIUS, cannot produce GIF images directly. It produces BMP files, which must then be converted to GIF files. This can be done automatically, and takes about 10 minutes for the 300 or so files (150 species maps and 150 flightlines). Once all the GIF files are completed, an FTP program is used to transfer them from the Pentium PC to the MCNET Unix server. MCNET is connected to the Internet and serves as the college's WWW server.
The maps are updated every few months, however anyone looking at the maps should be aware of several things. First, the data are not checked for accuracy, and we know there is some bad data in there. For instance, several latitude/longitude values are plotting outside the boundaries of the state, but the rest of the label information describes a site within the state. We don't have the correct latitude/longitude data, however, to replace the faulty data, and we don't know at this point if the data was recorded improperly in the field or if the problem lies in data entry. There are probably a number of records like this which are not as obvious because the data point gets plotted within the state's boundaries. Second, even immediately after an update on the WWW, we probably have several hundred records which are not represented on the maps. These represent specimens still in the collector's possession, specimens for which ID's are being made or verified, specimens whose data has not been transcribed onto a data sheet, and specimens for which the data on the data sheet has not been entered into the computer.
What's next? We would like to make more data available over the web - possibilities include setting up a download site for the database and the software. We would also like to make up informative pages for each species. These pages would include photographs, descriptions, and Ohio and national distribution maps.