COS 69-5 - Using historic survey data to determine whooping crane (Grus americana) habitat use patterns at Aransas National Wildlife Refuge

Thursday, August 11, 2016: 9:20 AM
207/208, Ft Lauderdale Convention Center
Katherine Golden1, Amy Frazier2, Ben Hemingway2, Craig Davis1 and Samuel Fuhlendorf1, (1)Natural Resource Ecology and Management, Oklahoma State University, Stillwater, OK, (2)Geography, Oklahoma State University, Stillwater, OK
Background/Question/Methods

The recovery of the only remaining wild population of whooping cranes (Grus americana) from fourteen individuals in 1941 to 310 cranes in 2015 has been a success story for the Endangered Species Act.  The wild population overwinters at the Aransas National Wildlife Refuge (ANWR) in Austwell, Texas, which was declared critical habitat for the whooping crane in 1978.  The refuge began conducting prescribed burns during the mid-1980s to control the encroachment of Texas live oaks (Quercus fusiformis) on the refuge.  Anecdotal evidence and periodic observations have noted that whooping cranes use recently burned habitat, but few studies have examined the role of fire in habitat use by whooping cranes.  Using a historical survey dataset of winter whooping crane locations, we examined their habitat use patterns relative to frequency of fires, size of fires, and a variety of other landscape variables on the refuge.  The normalized burn ratio method was used with Landsat data to map the documented burns on the refuge.  Crane locations were then intersected with burn polygons and related to the fires and other land cover data from the National Land Cover Database (NLCD) at various time intervals. Binary logistic regression was used to analyze crane occurrence in habitat patches. 

Results/Conclusions

Preliminary results indicate cranes occurred in burned habitat, but occurrence likelihood decreased per week as time since burn increased.  Cranes were less likely to occur in grasslands and developed open spaces, while the likelihood of crane occurrence in or near water was eight times greater than other land cover type.  We will continue to analyze additional occupancy models related to fire, time since fire, vegetation, and other habitat variables that influence whooping crane occurrence at Aransas National Wildlife Refuge.  Results from this study will be informative in guiding management objectives and future research as the whooping crane population continues to grow and expand beyond their protected winter range on the refuge.