Habitat Ecology & Restoration II

Contributed Oral Presentations

Contributed paper sessions will be available on-demand for the duration of the conference, then again at the conclusion of the conference.


Characterizing Methylmercury Bioaccumulation in Larval Dragonflies
Cailin Mackenzie; Tiffany Sacra Garcia; Collin Eagles-Smith
Monitoring aquatic organisms is critical to inform management of mercury contamination within food webs and across landscapes. Predatory invertebrates are increasingly being used to gauge biotic mercury exposure, given their import as both prey and predator, as well as their site fidelity, widespread distribution, and biphasic linkage of aquatic and terrestrial habitats. Quantitative understanding of mercury bioaccumulation, biomagnification, and toxicity to predatory invertebrates remains limited, requiring further research in order to estimate population and community level impacts. To quantify dietary bioaccumulation over time, we collected dragonfly larvae (Libellula pulchella, L. forensis, Plathemis lydia, Pachydiplax longipennis) from a single pond and reared them in the laboratory on a clean diet to foster depuration of background mercury for seven weeks. After this period, larvae were fed California blackworms (Lumbriculus variegatus) dosed at four ecologically relevant concentrations of methylmercury. Dosing continued for eight weeks with regular larval subsampling to quantify total- and methyl-mercury accumulation over time, with assays performed to quantify feeding rate, predator avoidance behavior, and immune response. Preliminary data suggest that dietary doses of 1, 20, 100, and 500 ppb methylmercury led to dragonfly total mercury tissue concentrations of 28.6±4.1, 41.1±3.8, 115.2±6.6, and 462.7±53.7 ppb respectively. Low, medium, and high treatment tissue concentrations were estimated to be 1.4, 4.0, and 16.2 times higher respectively than controls (p<0.025, 0.00001, 0.00001 respectively). Trophic transfer factors decreased with increasing mercury dose - 21.7 (control), 2.05 (low), 1.16 (medium), and 0.93 (high) - indicating higher bioaccumulation efficiency at lower doses. These trophic transfer rates inform how different environmentally relevant levels of mercury contamination influence assimilation to secondary consumers, and how this accumulation changes over time. Additionally, these findings augment the utility of the US National Park Service’s comprehensive Dragonfly Mercury Project dataset by empowering inference of contamination at lower trophic levels from dragonfly tissue.
Environmental Factors Influencing Heavy Metal Bioaccumulation in Freshwater Turtles
Darien N. Lozon; Donald J. Brown; Jason A. Hubbart; James T. Anderson
West Virginia has historic heavy metal contamination due to acid mine drainage from coal extraction (iron sulfide, selenium, manganese), coal-fired power generation (mercury), and abandoned mines (aluminum). Agricultural runoff can simultaneously contribute zinc, cadmium, chromium, and lead accumulation in wetland systems over time. Heavy metals can pose a threat to the health and safety of humans who interact with contaminated water or consume animals with accumulated levels higher than the EPA regulation. To ensure safe levels in the environment, freshwater turtles can be used as a biological indicator for quantifying environmental health because of their long life and high trophic status. Within the Upper Deckers Creek watershed in Preston County, WV, we quantified heavy metal levels (cadmium, chromium, total mercury, lead, selenium, and zinc) in 62 painted turtles (Chrysemys picta) and 32 snapping turtles (Chelydra serpentina) through non-destructive tissue (blood and nail) sampling. Heavy metals in turtles were compared between species, age-sex classes, tissue types, soil heavy metals, and surrounding land use practices. All metals were consistently higher in nails than blood. Model selection using Akaike’s Information Criterion indicated an interaction between carapace length and sex was a strong predictor for cadmium and lead, which is to be expected as bioaccumulation increases with growth over time and females transfer heavy metals to their offspring. Selenium was the only metal that included soil or species as predictors. Percent agriculture (within 250 or 500 m) was a predictor for all top models—except for cadmium—with positive (chromium, total mercury, and lead) and negative (selenium and zinc) relationships found. Further research should be conducted to investigate agricultural effects on heavy metals found in wetland soils (a proxy for bioavailability) as land use variation such as seasonal vegetation removal (e.g., crop harvest) could affect the amount of metals in the environment.
Predictive Mapping of Rare Plants in Texas
Jordan P. Craven; Marissa Pensirikul; Hemanta Kafley; Darrel Murray; Heather Mathewson; Kim Taylor
Rare plants are responsible for some of the most vulnerable ecosystem functions and their loss could affect not only surrounding flora and insects, but larger vertebrates as well. Species distribution models allow ecologists to reliably predict habitat and develop maps that can be used in subsequent species management plan. Using historical collection data, herbaria databases, and citizen science records, we will use presence-only data of occurrence for 17 species of rare plants to predict potential habitat throughout the state of Texas. The main objective was to identify potential locations of present-day populations of each species. We created predictive suitability models using MaxEnt and several environmental variables including bioclimatic data (WorldClim), solar radiation, soil, geological features (USGS), and land cover. We determined the contribution of variables to model output using jackknife tests. We assessed predictive ability of the models using diagnostic ROC curves and AUC values. Each model indicated high probabilities of finding existing populations in the expected regions of Texas for each species. These species have diverse life histories and are found in 8 of the 10 ecoregions. Botanical Research Institute of Texas will perform field surveys for us to use to validate four species models. These models will be integral in the future management of these 17 species, especially if they were to become listed as threatened or endangered in the state of Texas.
Clear as Mud: Using Occupancy Modeling to Better Understand the Distribution of Bog Turtles in Virginia
Joseph C. Barron II; Michael T. Holden; Carola A. Haas; Emmanuel A. Frimpong; Jeffery Feaga
Comparing factors that affect a species distribution at the inter-patch scale can prove challenging, especially when the breadth of current sampling is narrow and the majority of suitable habitat is on private land. In this study, we conducted 184 surveys on 49 discrete wetlands identified using a habitat suitability model. Using an occupancy modeling framework, we compared factors that affect the presence/absence of a rare habitat specialist, Glyptemys muhlenbergii. We evaluated several different factors at the patch, inter-patch and landscape scales, as well as covariates that could possibly affect detection during our surveys. Factors from all three scales affected presence of Glyptemys muhlenbergii in identified patches. Additionally, we found that detection was not constant across surveys, but was affected by both environmental and site factors. Our study shows how occupancy modeling can build upon habitat suitability modeling to better assess factors related to presence/absence and provides insights to surveyors looking for Glyptemys muhlenbergii so they may allocate their resources more efficiently.


Contributed Oral Presentations
Location: Virtual Date: Time: -