Making Good Use of SEINet By L. R. Landrum, E. E. Gilbert, and E. Makings Arizona State University The Southwest Environmental Information Network or SEINet was created by a National Science Foundation Grant (BDI 9983132) to Peter McCartney, Nancy Grimm, Charles Redman, Timothy Craig, and Corinna Gries at the Global Institute of Sustainability (GIOS) at Arizona State University starting in 1999. It is a series of linked mainly specimen databases and accompanying web software that helps one use those databases. Subsequent databasing grants have expanded SEINet (e.g., BRC 0237418) and continue to do so (e.g., DBI 0847966) and other databases have joined SEINet without any special funding. SEINet has become a cooperative venture with about 20 contributing member institutions and now hosts ca. 1.1 million specimen records, principally from the Southwestern states of Arizona, New Mexico, Nevada, Colorado, Utah, and California, and Sonora, Mexico. An additional programming grant (BDI 0743827) has allowed for the enhancement of software (Symbiota, http://symbiota.org/tiki/tikiindex.php) associated with SEINet. The herbaria included in SEINet have been organized into the Southwest Biodiversity Consortium and had their first meeting of providers and users in February 2010 (http://swbiodiversity.org/tiki/tiki-index.php?page=Meetings). We explain below some of the interesting things you can do with SEINet. Making checklists: Suppose you would like to have a checklist of plants from a particular area or county. First go to SEINet (http://swbiodiversity.org/seinet/index.php) and click on "Search Collections." This will take you to another page where about 20 different collections are listed by their names and acronyms. Choose among these for the appropriate collections to search. Suppose you want a checklist from central Arizona, perhaps Tortilla Flat in Maricopa County, you would check the collections of Arizona State University (ASU) and the Desert Botanical Garden (DES) because they are located near by and are likely to have many collections from the area and perhaps also the University of Arizona (ARIZ) and Northern Arizona University (ASC). The Navajo Nation herbarium (NAVA) and Grand Canyon National Park herbarium (GCNP) are much less likely to have collections from Tortilla Flat. (Alternatively you can select all collections by clicking on the appropriate box at the top of the list. You may find this method faster.) After checking the box for each of these herbaria you would click on "Next." A new screen appears entitled "Search Parameters," with categories under "taxonomic criteria, locality criteria," and "collector criteria." Under "Locality Criteria," you could type "Tortilla Flat" in the Locality field, Arizona in the State field, and Maricopa in the County field and click on "search." After a short wait, during which all the designated collections are searched for specimens that have "Tortilla Flat" in the locality field, and are from Maricopa County, Arizona, a new page appears. It is June, 2010 and I have searched ASU and DES only and get a list of 332 specimens. They are listed by herbarium and then alphabetically by name. There is a tab at the top labeled "checklist" and clicking on this tab allows one to convert the list of specimens into a list of taxa. There are options for making checklists: you can use the raw data (names as they appear on the labels) or a taxonomic thesaurus that account for synonyms based on a taxonomic opinion. Raw data defines a list of 211 taxa and when I use the ASU Taxonomic Thesaurus the number of taxa is reduced to 205. Thus, one is able to create a checklist, based on specimens in the databases today, which might be useful for a class field trip, land managers, conservation efforts, a family outing or help in identifying new specimens from the Tortilla Flat area. There are numerous images of living plants and herbarium specimens linked to the names in the checklist, so by clicking on a name, you can view images. Clicking on an image allows you to zoom in a take a closer look.

Sometimes localities have unique names (e.g., Tortilla Flat) but other names are used various times in a state (e.g., Sycamore Creek is found many times). Including the county for these widely used names is important. If you do not also designate a county, then you may get specimens from more than one "Sycamore Creek." If you wish to search multiple counties or multiple keywords, be sure to separate the criteria with a semicolon. For example, if you want to see all of the plants from Mohave and Yavapai counties, type in "Mohave; Yavapai." If you wish to see all of the specimens from Schnebly Hill and Sedona, type in "Schnebly Hill; Sedona." A large portion of the specimens at ASU, ARIZ and ASC have been georeferenced and these can be queried by designating limits based on coordinates. Thus we might want to know what plants have been collected from 32.5 degrees N to 32.7 degrees N and 113.5 degrees west to 113.7 degrees west. In doing that query for ASU, ARIZ, ASC, and DES we find that there have been only 85 specimens collected and there are only about 64 species known. It looks like more collecting needs to be done in that square that we picked at random of ca. 100 square miles near the Mohawk Mountains. Note that searching by coordinates will only query specimens that have been georeferenced and exclude all specimens that have not been. Another way to build a checklist is by choosing a point-radius search. This time I have chosen all the collections and start with a point at 33 degrees N and 112 degrees W (a few miles south of Maricopa), and a radius of 10 miles. I find 146 specimens that account for 98 taxa using the ASU Taxonomic Thesaurus. You may insert a name in the Taxonomic Keyword Criteria using the genus name. This allows you to limit your search, for example, to Brickellia (a group of shrubs in the aster family) of Schnebly Hill. You may also create a family checklist for a particular locality. For example, if you would like to see all of the Brassicaceae from Yuma County, Arizona enter "Brassicaceae" in the taxonomic criteria box, and "Yuma" in the county box and Arizona in the state box. I found 146 specimens and 39 taxa using the ASU Taxonomic Thesaurus. There is also a Yuma county in Colorado, so including the sate is important. Plant images as an aid to identification: One of the most valuable aspects of SEINet is the image library. This database includes over 17,000 records of scanned herbarium sheets, field photos, live scans, and many close-up photos of important structures such as grass spikelets. Many taxa have associated images linked to them so that when one makes a checklist, clicking on the taxon will bring up all images that have been linked to that plant. In a collection database search, you can also click on "more taxon info and images" to see photos or herbarium sheets of the plants on your list. If you can make an educated guess as to the family or genus of a plant, then the image library can be a great help, because you can compare several species quickly. Suppose you had an unknown Lotus from Tortilla Flat. With the checklist maker you have discovered that only three species have been found there before. You might quickly view images of the three species and compare them to the new collection. This might be enough to give you a good idea of what you had found. By checking the keys in Arizona Flora or consulting herbarium specimens you could confirm the identification. We have found that this process makes plant identification much more efficient. Recently someone sent Landrum an image of Petrophyton caespitosum from the North Rim of the Grand Canyon. He had never seen the plant before, but was able to guess its family, Rosaceae, after a couple of mistakes. He made a checklist of possible plants for the North Rim, and was able to identify this very distinctive plant in a few minutes. Mapping: There are two ways of making maps with SEINet. One can search the collections for a particular family, genus or species as was outlined before. Suppose you are interested in Arizona only, then putting Arizona in the State category would be a good idea. When I search all the collections for Quercus gambelii in Arizona I get 511 collections. Next I click on the Maps tab. Two options appear:

Google Maps and Google Earth (which requires that you have Google Earth on your computer). You can search on more that one species by separating the names with a semicolon. Here I have made a Google map of Q. gambelii (red) and Q. turbinella (blue). It is interesting to compare their overlapping ranges.

Using Maps to Help Identify an Unknown: Suppose you could recognize an unknown plant to family or genus and you knew where it came from. The mapping tool can be very helpful for identifying. Suppose I have an unknown Camissonia from Yuma County. I could make a map of Camissonia for all of Arizona to start. I get 511 collections. I make a map with nearly all the 511 dots displayed. My unknown comes from the Copper Mountains at about 114 degrees W and 32.5 degrees W. At the bottom of the map I fill in those coordinates, being careful to put a minus sign before 114 (it will warn me if I don’t). A blue X appears on the map at that point and I can zoom in and see which dots appear close to it.

By clicking on the dots I can see that C. californica, C. claviformis, C. arenaria, C. chamaererioides, and C. boothii grow in the vicinity of my unknown, so I can start by looking at these species, either in the image library or as specimens in an herbarium.

Cool Things Students and Others can do with SEINet: In SEINet there are many choices of what to do. Say you want to study plants for your Sonoran Desert Field Botany class. You can go to home page http://swbiodiversity.org/seinet/index.php and there is a list of options below. One is for “Species Lists.” Click on it and a long list of checklists appears. Most of these are documented local floras. About 2/3 of the way to end there are three teaching lists under “Teaching Checklists.” One is “Sonoran Desert Field Botany 2010.” Click on that checklist and you get a list of species by family that 2010 class. Click on a name, e.g., Trixis californica and you see the species page with a description and some images. By clicking on “more photos,” more images will appear if there are any. Click on an image and it expands. Click on the “back button” and you go back to the species page.

You can convert your checklist to an illustrated one by checking “Display as Images” box and then click on “Rebuild Checklist.” Go back to the unillustrated checklist and click on Cylindropuntia acanthocarpa. On the species page right after the name you see a green curved arrow. Click on that and you can compare all the species of Cylindropuntia with one image each and a map. Click on the image and it expands. Click on the name and you go to the species page with other images and a description. Go back to the Cylindropuntia page and click on the green arrow and you see all the Cactaceae, the family description and an image of all species.

Go back to the page with the checklist and you will see a little yellow key and the word “quiz” in blue after the checklist name. Click on “quiz” and an image guessing quiz begins. It should be a great way to study. Click on the yellow key and you come to a multiple entry key and you can start trying to key out a plant based on characters you remember it has. Say it was Olneya tesota, you could click on “tree.” That narrows the list to a few species. Principal flower color is purple and that gives you “Olneya.” You can use the multiple entry key for all static checklists.

The variable radius Dynamic Key: Go to http://swbiodiversity.org/seinet/ident/dynamickeymap.php and you will see a Google Map of the Southwest. You can pan and zoom in to find the point you want. I have zoomed into Young, Arizona at 34.09896 N, 110.96191 W. I click on “submit coordinates” and in a few moments an interactive key appears for that area. Symbiota has searched the specimen collections in increments of 6 miles until a good checklist can be made; in this case it is 24 miles in radius and the

number of species 1118. Then Symbiota drew upon a database of characters to make a key for this particular area. So suppose you want to know what that funny looking tree is growing near Young.

First click on the tree box to the left under habit. That reduces the number or species down quite a bit. The leaves are opposite so click on that under arrangement. That reduces the number again and gives you more characters to choose. Click on compound leaves and toothed margins. Then on white flowers and you are down to 2 species. You could just compare images of those two species by clicking on the names or if you recognize the inflorescence as a panicle you can click on that and you are down to a single species, Sambucus nigra. Click on the name to see if you have got it right, and if you do congratulations. If not, go backwards and try again. This is a great way to learn about plants and will work for areas where we have many georeferenced specimens and character data for the species.

Something Very cool indeed. You can make a checklist based on coordinates or place names, just as described above. I am choosing number of my home address for this one, as I have collected many plants in my yard over the years. I get a checklist of 67 species. In the upper right corner you will see the yellow key. Click on that and you will be able to make an interactive key to your dynamic checklist. It may take a few moments because the character data have to be harvested from the character database. If some of the plants in my yard are not in the character database the key won’t well for them; but in any case it works remarkably well, especially for native plants. I was able to key out Justicia californica and Encelia farinosa. Who knows what the potential uses of SEINet might be? You may be able to invent some of your own. In the future we will be developing new web tools that will make SEINet even more useful.