OOS 50-8 - Detection and occupancy of anuran adults and tadpoles in wetland restorations

Friday, August 12, 2011: 10:30 AM
14, Austin Convention Center
Michelle L. Hellman , School of Natural Resources, Nebraska Fish and Wildlife Cooperative Research Unit, University of Nebraska-Lincoln, Lincoln, NE
Craig R. Allen , Nebraska Cooperative Fish and Wildlife Research Unit, University of Nebraska-Lincoln, Lincoln, NE
Martin P. Simon , Department of Biology, Benedictine College, Atchison, KS
Background/Question/Methods

Amphibians are declining globally due to habitat loss, climate change and the spread of pathogens Wetland restorations are a potential solution to protect and bolster amphibian populations. In order to assess the efficacy of restoration efforts, monitoring is needed to determine whether amphibians are utilizing restored wetlands.  Occupancy modeling is a new method of modeling presence that accounts for variability in detection rates and utilizes covariates.  For anurans, occupancy is primarily based on call surveys, which allow rapid survey of many wetlands.  However, detection and occupancy of anurans may vary depending on the life stage sampled.  To assess the difference between detection rates and occupancy estimates of calling adults versus netted tadpoles we selected 40 wetlands in three restored Missouri River bends in southeast Nebraska.  Each wetland was visited four times during May, 2010, twice to conduct call surveys, and twice more to conduct tadpole sampling.  We used habitat covariates and detection histories for Woodhouse’s Toad (Anaxyrus woodhousii), the Boreal Chorus Frog (Pseudacris maculata) and the Plains Leopard Frog (Lithobates blairi) to estimate occupancy and assess factors that affect detectability and occupancy.

Results/Conclusions

Our site-specific covariates were: wetland slope, vegetation cover, presence of fish and presence of emergent vegetation.  With multi-model inference we assessed six call survey models and seven tadpole survey models.  For Pseudacris maculata, a relative habitat generalist and opportunist, only weak inference differentiating models was possible, and all covariates were plausible in our model set (evidence ratios between top and bottom models were 5.13 and 6.27 respectively).  For Anaxyrus woodhousii and Lithobates blairi occupancy estimates derived from call surveys indicate that occupancy is negatively associated with wetland slope.  Slope is also negatively correlated with detection rates of Anaxyrus woodhousii tadpoles.  Both slope and woody vegetation were negatively correlated to occupancy of Lithobates blairi, although the presence of fish was positively correlated.  Naïve estimates of occupancy varied between adults and tadpoles for Pseudacris maculata (0.225 and 0.525 respectively).  This could be reflective of adults calling from, or breeding in, habitat that is unsuccessful for larvae.  Amphibians are utilizing restored wetlands in multiple life stages, and in the case of generalist species such as the Pseudacris maculata, tadpoles may serve as a better indicator of true occupancy.

Copyright © . All rights reserved.
Banner photo by Flickr user greg westfall.