Font Size: a A A

Predicting threatened orchid (Isotria medeoloides [Pursh] Raf.) habitat in the southern Appalachian region using Maxent model

Posted on:2015-04-29Degree:M.SType:Thesis
University:Western Carolina UniversityCandidate:Montgomery, Ashley DallasFull Text:PDF
GTID:2470390017994202Subject:Conservation biology
Abstract/Summary:
Isotria medeoloides (Pursh) Raf. or small whorled pogonia is one of the rarest orchids in the eastern U.S. and is currently threatened by habitat loss in the southern Appalachians. The purpose of this study was to improve our understanding of where I. medeoloides is found and to develop a habitat suitability model so that monitoring and conservation efforts can be prioritized in the southern Appalachian region of the orchid's range. Habitat characteristics, including topographic, soil, and vegetation variables were measured at 15 extant locations during the initial habitat characterization. The results of the habitat characterization were used to guide development of the habitat suitability model. The maximum entropy modeling method (Maxent) was used to predict potential suitable habitat for Isotria medeoloides in the southern Appalachian region of NC, SC, TN, GA, and AL.;In general, Isotria medeoloides occurred in mid-successional, mixed acidic cove forests with a sparse herbaceous layer, high canopy cover (81-98%), moderately-sloped terrain (5-40%), low soil pH (4.0-4.9), and soils with a hardpan layer. Additionally, one-way ANOVA tests between small and large Isotria medeoloides populations revealed that overstory snag density was significantly higher (0.029 ;The Maxent jackknife test determined that annual average precipitation was the most important environmental predictor of suitable habitat for Isotria medeoloides, suggesting that moisture may be one of the most critical factors controlling survival of the orchid species. The Maxent model had high predictive performance with a statistically significant Area Under the Curve (AUC) score of 0.954. It predicted that 0.74% of the total study area was suitable habitat for Isotria medeoloides. Field validation to evaluate the predictive performance of the model revealed differences in habitat characteristics at predicted high suitability sites compared to predicted low suitability sites, and these differences were similar to habitat characteristics that differed between large and small extant Isotria medeoloides sites. Additionally, a new population of 19 plants was found in an area predicted as highly suitable by the model indicating that the Maximum Entropy modeling method is a valuable tool for predicting suitable habitat for Isotria medeoloides in its southern Appalachian range thus contributing to conservation of the species.
Keywords/Search Tags:Isotria medeoloides, Habitat, Southern appalachian, Model, Maxent
Related items