NEWSLETTER OF THE TENNESSEE NATIVE PLANT SOCIETY Volume 40, Number 3 September 2016 Are computer models replacing field scientists? ~ Sunny Fleming R ecently, Discover Life in America (DLiA) was involved with a huge undertaking to model the distribution of all catalogued species in Great Smoky Mountains National Park. The results are visualized online and one can search for in- dividual species or entire suites to compare their ranges. (Try it yourself! https://science.nature.nps. gov/p arks/grsm/ species/) This visual product is smooth and easy, but the behind-the-scenes is actually a massive undertak- ing. However, before we dive into the details be- hind it, we first have to understand what exactly IS species distribution modeling (SDMs)? I discuss distribution modeling as a form of paint- ing. When someone creates, say, a painting or a drawing, typically a substrate is first chosen - the paper, canvas or wood, for example. Then, the medium or mediums is chosen - the paint (oil or acrylic?), watercolor, pencil, charcoal? These items are used in conjunction with another, layering on from general to specific until these things trick our mind into seeing something we recognize or intend to communicate. SDMs essentially do the same thing, but with layers of data and the end product is a prediction of where a species may occur. There are different ways to do this, but the most common way is with what’s called “presence data.” That is - understanding where a species already occurs and then asking the question “where else do these layers of data line up in the same way where this species already occurs?” Apios priceana is just one species Todd Crabtree has developed a model for more efficient search efforts. It’s just like picking out a good orange at the market. From experience, we know what a good orange looks and feels like so when we go to the market we look for the same criteria. While we know this is a pretty good way to measure the risk involved in purchasing a fruit that we can’t know is good until we open it up, we also know it’s not ioo% accurate and we can still pick out a bad fruit. The same risk can also be calculated in a model. . . essentially the computer can measure for itself how well it’s doing at predicting a species distribution. Now that we have a basic understanding of how species distribution models work - let’s put it in perspective: I work on modeling a single species; the Streamside Salamander ( Ambystoma barbouri.) On a produc- tive day, I’ll work on the geoprocessing component (preparing the layers of data) during the day and before I go to bed I tell my computer to run the model. . . it takes about ten hours for the model to complete. When I pour my coffee the next morning, I head downstairs to see what the model did overnight. Continued onpg. 7... TNPS Newsletter This newsletter is a publication of the Tennessee Native Plant Society and is published four times a year, generally in February, June, August, and November. The Tennessee Native Plant Society (TNPS) was founded in 1978. Its pu- poses are to assist in the exchange of information and encourage fellowship among Tennessee’s botanists, both amateur and professional; to promote public education about Tennessee flora and wild plants in general; to provide, through publication of a newsletter or journal, a formal means of document- ing information on Tennessee flora and of informing the public about wild plants; and to promote the protection and enhancement of Tennessee’s wild plant communities. Dues for each calendar year are: Regular: $20 Student: Complimentary Institution: $50 Life: $250 Dues may be sent to: Tennessee Native Plant Society P.O. Box 159274 Nashville, TN 37215 Officers Susan Sweetser, President Todd Crabtree, Vice-President Margie Hunter, Secretary Michelle Haynes, Treasurer Directors Sunny Fleming Louise Gregory Michelle Haynes Dennis Horn Bart Jones Larry Pounds Joey Shaw Sunny Fleming, editor Please send comments and material for the TNPS newletter to sannfleming@gmail.com A 'Letter from the President ello everyone, Hope everybody has had a wonderful summer. You will be getting this newsletter after registration was due for the annual meeting. If you want to come, you are more than welcome. Just contact Fall Creek Falls for you own reservations. You can pay the meeting fee when you come for the weekend. It is going to be a really good meeting with two very good speakers and a GREAT field trip. Just being at Fall Creek Falls is a joy in itself. Come join us and share the fun. Another reminder, the annual meeting next year will be in June. We will be going back to Bersheba Springs. We used to have our meetings in June but switched in recent years to September. This will give us a chance to see different plants and may work better with folks vacation schedules. Hope to see you there, 865-938-7627, Thanks to all, 2016 Annual Meeting Schedule: Fall Creek Falls State Park FRIDAY, SEPTEMBER 16 4:00-6:00pm 6:00-7:30pm 7:30pm Mix and Mingle Dinner Welcome Talk by Brandon Matheny UTK “ Ecology and Biology of Fungi” SATURDAY, SEPTEMBER 17 6:30am-8:00am Breakfast 8:30am - 1 1 :30am Field trips 12:30-3:30PM Field trips 1. -Tom Kimmerer 2. Larry Pounds 3. Tom Kimmerer 4. Auto Tour on your own of Fall Creek Waterfalls 6:00-7:30pm Dinner 7:30 pm General Business Meeting Talk by Tom Kimmerer: “Ancient trees in the woodlands of the Bluegrass and the Nashville Basin” SUNDAY, SEPTEMBER 18 6:30am-8:00am Breakfast CHECKOUT 8:30am - 1 1 :00am BOARD Meeting 3 A summary from the 2016 legislative session: - New Scenic River - SB2520 / HB2566 establishes that portions of Soak Creek are to be protected as a scenic river, which will allow it to remain free flowing and un- polluted. Enacted as Public Chapter 713 effective April 6, 2016. This is the first new scenic river approved in many years! - Electric and Hybrid Vehicles - SB 1451 / HB1851 Elec- tric and hybrid vehicles - would have increased registra- tion fees for electric and hydro-electric vehicles. Failed. This slap in the face of promoting sustainable practices thankfully failed thanks to advocates for conservation. Okay, maybe Nissan s lobbyists helped a bit too! - TWRA Payments - SB 1743 / HB1769 would have re- quired TWRA to pay local governments in lieu of taxes for all lands it owns. Failed. This effort, while understandable for counties needing revenue, would have stopped and even rolled back efforts to preserve important wilderness areas and waterways. - Gas Tax Revenues - SB 1716 / HB1650 would have pro- hibited gas tax revenues from establishing and maintain- ing bike lanes. Failed. This effort would have been a strong step backwards in terms of sustainable transportation. - Water Pollution - SB 1830 / HB1892 removes the authority of TDEC to develop appropriate water run-off regulations. Enacted as Public Chap- ter 1007 effective May 2, 2016. This effort unfortunately passed, although opposed by conservation groups, the Department of Environ- ment and Conservation and Gover- nor Haslam. The Homebuilders As- sociation got their way and in doing so have almost assured increased in run-off water pollution! Thanks to the Tennessee Native Plant Society and its members for support- ing our work! ~ Stewart Clifton Registered lobbyist for Tennessee Conservation Voters When Are Dues Due? Unless you are an email subscriber, check your mailing label for your membership date. You are paid through the year listed just above your name. You can pay TNPS dues at any time, and now you can pay online at the TNPS Website. Just go to www.tnps.org, click “Membership,” and follow directions there. If your address has changed, you can email the new address to info@tnps.org. We cannot print the newsletter in full color, but you may be pleased to find all the color in email copies and at the website, www.tnps.org 4 2016 TNPS Field Trips Arnold Air Force Base & May Prairie SNA, TN, June 4th, 2016 T his was a joint field trip with the Middle TN Outsiders and the TNPS. A total of about thir- ty people participated including eight from the Middle TN Outsiders. The threat of rain accounted for the relatively moderate temperature. Our first stop was at the AEDC power lines. All of the or- chids advertised were in flower, including Rose Pogonia (. Pogonia ), Grass Pink ( Calopogon ), Spreading Pogonia ( Cleistes ), and Ragged- Fringed Orchid ( Platanthera lacera). Also in flower were two mountain mints ( Pycnanthemum mutilum and R tenuifolium ), Hyssop-leaf Skullcap ( Scutel- laria integrifolia), Whorled Loosestrife ( Lysimachia quadri- folia), Goafs Rue ( Tephrosia ), and Death Camas ( Zigadenus leimanthoides). Slender Blue Flag ( Iris Prismatica ) was in fruit. After an hour in the field the rain came, so we headed back to Manchester for lunch. Our 2nd stop was May Prairie. The prescribed burn in late March was evident, this time the burn extended through the wooded area all of the way to Asbury Rd. The prairie was Lysimachia quadrifolia (whorled loosestrife) not as colorful this year, but most of the flowers we expect- ed were found. These included Eryngo ( Eryngium integrifolium ), Swamp Candles ( Lysimachia terrestris ), Marsh Pea (Lathrus palustris), Horned Bladderpod (17. cornuta ), and Dwarf Sundew (Drosera brevifolia) not in flower, all five plants are rare in Tennessee. Among the many other species found were Smooth Phlox ( P. glaberrima) and Indian Paintbrush ( Castilleja coccinea). The paintbrush is partly parasitic on the roots of grass. Then the rain came again, this time heavy. Only those that departed early escaped getting wet. ~ Dennis Horn TNPS and Middle TN Outsiders photographing the barrens. 5 Selaginella rupestris * in the Adirondacks ~ Andrew Gibson T his small patch of vegetation was one of the most exciting of life plants for me while out in the Ad- irondacks. This is rock spike-moss ( Selaginella rupestris ), and despite the name isn’t a moss but a fern ally related to quillworts ( Isoetes ) and lycopods (Ly cop o diop sida). It’s an ancient and fascinating little plant with its sporangia tucked singly at the base of fertile leaves known as sporophylls; somewhat reminiscent of the lycopod genus Huperzia. Like a moss it’s poikilo- hydric and can withstand severe drying out/drought and bounce right back to green, lush life with some rain. Rock spike-moss is widespread throughout North America but extirpated from Ohio, and one I’ve looked for for years in its historic haunts. It was a real treat to finally see it in the flesh while out in New York growing on bedrock bluffs overlooking the Hudson River. Andrew Lane Gibson is afield botanist with the Ohio Di- vision of Natural Areas & Preserves. He often travels across the United States botanizing and writes a blog dedicated to the natural world. You can read other works and see his photos at www.floraofohio.blogspot.com Selaginella rupestris above the Hudson River. * Editors note - Selaginella rupestris is also present in Tennessee, although not often seen. It is currently known from six counties including Bledsoe, Polk, Cocke, Unicoi, Carter and Washington. The record in Polk Co. was first docu- mented by Augustin Gattinger in the late 1800 s, and later docu- mented again during a rare plant survey of the Ocoee River Gorge conducted by Drs. Joey Shaw, Dwayne Estes, Eugene Wofford and students. Don’t forget to bring your field guides to the annual meeting at Fall Creek Falls State Park! All books available through TNPS. LONE A HiNE WILDFLOWERS TENNESSEE Hit OHIO VALLEY 'and the SOUTHERN WOODY PLANTS OF KENTUCKY AND TENNESSEE The Complete Winter Guide to Their Identification and U3C Ronald L Jones ft G, Eugene Wofford 6 Volunteer at the TN Tech Herbarium T he Tennessee Technological University Herbarium is located in Cookeville, and housed in Pennebak- er Hall. The herbarium was established by Dr. S. K. Ballal, and was named in honor of Dr. Paul L. Hollister, Associate Professor of Biology at TTU from 1950 to 1971. Dr. Hollister was a talented botanist with a great interest in the Asteraceae, especially the genus Chrysopsis. He participated in the Smoky Mountain Wildflower Pilgrim- age for many years, and was an active member of the Tennessee Academy of Science for almost 29 years. The collection has grown since its creation in 1965, with more than 40,000 specimens now estimated in its holdings. Curators of the collection since its inception include S. K. Ballal (1965-1970), Gordon Hunter (1970-1990), Felix Coe (1996-2006), Christy Carter (2007-2012), and Shawn Krosnick (2013 to present). The TTU Herbarium is part of the newly forming Ten- nessee Consortium of Herbaria and is working collabo- ratively to share scientific resources with other Tennessee institutions. The TTU Herbarium is also affiliated with SERNEC (Southeast Regional Network of Expertise and Collections), whose goal is to join together ca. 150 re- gional herbaria in the Southeast US to produce an online database of all collections. In order to develop, maintain, and restore the specimens housed at HTTU, we are actively seeking volunteers to become involved in curation. Members of the TNPS would be ideal as volunteers because you already love plants, care about Tennessee’s flora, and are likely detail oriented. Activities would include mounting specimens, filing, databasing, and repairing specimens as needed. The herbarium staff include faculty in the Biology Dept, and many undergraduate and graduate students. Several local botanists are also affiliated with the collections and come by to visit and use the herbarium regularly. Volunteering in the collections would provide an opportunity to meet new friends and colleagues, gain new skills, and contribute to an important collection of specimens. We welcome volunteers to assist in any aspect of herbari- um management. No prior skills are needed, but an artistic flare and attention to detail are beneficial qualities for this type of work. Volunteer schedules also completely flexible, though the collection is closed on weekends. Please feel free to visit our webpage, https://www.tntech.edu/ cas. /biology /herbarium/, and contact Dr. Shawn Krosnick (skrosnick@tntech.edu) for more information. Don’t miss the annual meeting at Fall Creek Falls State Park! September 16 - 18, 2016 Do you have photos from trips you’d like to share? Are you interested in other TNPS news? Stay up to date and fol- low us on facebook! Post your photos, write your own field trip reports, let us know about botanical goings-on in your neck of the woods! facebook.com/tennesseenativeplantsociety 7 Contcl from front page... The species mapper app developed by DLiA in partnership with U.T. Knoxville Department of Electrical Engineer- ing and Computer Science and the GSMNP Inventory and Monitoring Branch maps over 1,800 species. If it takes my computer 10 hours to map one species, then how are they mapping over 1,800?! (It would take over 2 years of non-stop running for my computer to do that, but it would fry itself from the heat and run out of memory in just a couple days.) DLiA solved this issue by running all the data through the Oak Ridge super computer, but its still no easy task. (Hello? Department of Energy? Please call me, thanks.) With all new things, though, comes the inevitable pushback. In discussing SDMs with other botanists and ecologists the response hasn’t always been a rosy one. There is some concern that this technology is threatening to take the place of the classic field scientist. The fear is not totally unfounded. Just like any conclusion in science, if we don’t understand the limitations of the design and the question being asked and the methods that produced results, then we can’t understand the limitations of those results. Simply put - the data out is only as good as the data in. Let’s talk about this in terms of paint, again. If I want to produce the world’s next Rembrandt maT acrylic paints I found hiding in the closet - the likelihood of it standing the test of time or being any good is slim. Likewise- I’m NOT a trained painter studying with the world’s greats. Yet, if I posted my painting on social media I would have some nice (and horribly untruthful friends) who tell me how great it looks, and in their lack of understanding techniques of the fine arts, they may genuinely like my cardboard mess! In a similar way, results of science can often be misinterpreted and taken for absolute when the ultimate audience doesn’t clearly under- stand the limitations of the process. Where does the role of on-the-ground trained professionals come in? First of all, if a model works by using known presence data then it requires that someone has gone out and identified where these species occur. Furthermore, the computer can’t Shown: Eggs of streamside salamander. New populations REALLY know if it’s done its job. Lrom a statistical standpoint were found this winter using predictive modeling. it can give us a value that tells us something about how well it’s done, but until we go out there and test a model and verify its usefulness, it’s simply just a bunch of numbers and some pretty visuals. You can’t just throw any data into the model, either. One must understand something about the habitat in the first place in order to decide what kinds of data to include or exclude. For example, on an early permutation of my salamander model, I used a layer of data that tells me the type of vegetation at any given point with a resolution of 30m2. A lot of my presence points had been collected within 30 meters of where a road crossed a creek because that’s the easiest place to jump into a stream. The resulting model mistook roads and roadside vegetation for excellent habitat and kicked back a model that looked more like a fiery atlas of the central basin. The value it spit out told me “Hey look! This model is superb!” but common sense and an understanding of ecology and the concept of “sample bias” let me know that the initial model was actually trash (but excellent for navigation!) So, even though vegetation is an important component to the real-life ecology of this salamander, the data simply was too coarse to be useful and it was hurting the model. Ok - so when we finally get a model that we are satisfied with, are we going to find species where the model is hitting? No. SDMs don’t tell us where we can go find the species in question. What they tell us is the likelihood that the proper habitat exists. Todd Crabtree, who also works with species models, notes a weakness in this scenario as well: “Some rare species have broad habitat requirements. This may seem counterintuitive, but these species are limited in numbers by some factor other than habitat specificity.” 8 In these cases, models may not be successful at all. However, there is still a great utility in being able to predict where these species may occur. I paraphrase Todd Witcher with Discover Life in America: “Some of this data has been collected over hundreds of years. How can managers and decision makers use that data if we don’t have a tool that brings it all together?” SDMs are more than just a fun academic exercise. They are proving to be an extremely efficient and effective way to help decision makers, developers, park managers, and others make informed decisions. Whether it’s planning trails, improv- ing roads, or targeting land acquisition for conservation, models help visualize the expertise for the non-experts. It can help predict species’ responses to climate change scenarios, and helps experts more efficiently search for elusive and rare species/habitats. This technology is not the nemesis of field science. These models require experts to gather the data in the first place, and then to confirm the accuracy of the model afterwards. If anything, it emphasizes the importance of classic field methods and the need to continue teaching classic natural sciences. For more information about the species mapper app, visit: https://www.dlia.org/park-species-mapper SI£Z£ NX ^IITAilSBN fZZ6£l xog *0‘d J9JJ9ISM3N SdNX