Wednesday, August 4, 2010: 9:20 AM
321, David L Lawrence Convention Center
Bradley A. Pickens, School of Renewable Natural Resources, Louisiana State University AgCenter, Baton Rouge, LA and Sammy L. King, USGS Louisiana Cooperative Fish & Wildlife Research Unit, Baton Rouge, LA
Background/Question/Methods The southwest Louisiana rice agricultural region is critical for millions of wintering, migratory, and breeding waterbirds. Wetland birds of concern include a variety of secretive marsh birds, wading birds, shorebirds, wintering and breeding ducks, and cranes. Land use conversions and invasive species threaten wetland birds by reducing available rice habitat in the region. Yet, priority areas for wetland bird conservation have not been identified, and the spatial pattern of habitat variables on the landscape was also unknown. Our objective was to develop a spatially explicit habitat suitability model for the king rail, Rallus elegans, a secretive marsh bird of high conservation concern. We used geographic information systems (GIS) and readily available GIS data to build a habitat model to identify priority king rail habitat and to identify potential threats to the bird in southwest Louisiana. The habitat model was derived from GIS data layers of ditches/streams, rice density, and canopy cover. Canopy cover and ditches were previously found to be associated with king rails at a localized spatial scale, however our broad-scale spatial analysis included these variables measured within a 1-km radius of each cell.
Results/Conclusions
We developed an empirical model based on presence/absence data obtained from call-back bird surveys in rice fields from 2007-2009. We used 50% of the data as training data for a logistic regression habitat model and used the other 50% of the data to validate the model. The analysis was conducted by using a receiver operating characteristic (ROC) statistic, and the results showed the model successfully distinguished the presence and absence of king rails on the landscape. The combination of canopy cover and ditches measured at a 1-km scale were important in predicting king rail presence/absence. The results also emphasized the importance of the southern parishes in southwest Louisiana for supporting a large king rail population. Recently, these parishes have been rapidly losing rice fields due to salt water intrusion and invasive woody species, posing a substantial threat to king rails in the region. This regional landscape approach can inform land use decisions and conservation programs in the region.
