• Introduction
• Objective
• Study Area
• Methods
• Analysis
• Results
• Discussion

Models for Managing Habitat of a Swainson’s Warbler Breeding Population, Bond Swamp National Wildlife Refuge, Georgia    

J. Michael Meyers1,2,3 and Elizabeth A. Wright2 1USGS Patuxent Wildlife Research Center-Athens, GA, USA and  2 D. B.Warnell School of Forest Resources, The University of Georgia, Athens, GA, USA

Swainson's Warbler

Nest-Atypical for BSNWR-on dead palmetto leaf.  Pair fledged young on 20th July 2002.

INTRODUCTION

The Swainson’s Warbler (Limnothlypis swainsonii) is a neotropical migratory songbird which breeds predominantly in bottomland hardwood forests of the southeastern United States (Meanley 1971, Brown and Dickson 1994).  Its population status and trends are unclear (Brown and Dickson 1994, Sauer et al. 2001).  It appears, however, that populations have declined throughout the species’s breeding range as the result of agricultural, forestry, and watershed management practices.   These practices may have substantially altered southeastern floodplains dense thickets of a native bamboo (Arundinaria gigantea) or cane, which may be important habitat of breeding territories in some areas (Meanley 1971, Brown and Dickson 1994).  Little quantitative information is available about the characteristics of habitats used by Swainson’s Warblers in core areas of the breeding range (Meanley 1945, 1966, 1971).  In Virginia, Graves (2001) created a 5-variable habitat model which correctly classified 90% of the 74 habitat plots used to create the model in terms of presence versus absence of Swainson’s Warblers.  Interestingly, cane stem density was negatively correlated with Swainson’s Warbler presence in the Virginia data set, and median number of cane culms was about five times greater in random plots than territory plots.  Total number of trees with diameter breast height (dbh) <5 cm and greenbrier (Smilax sp.) stem density were the variables most positively correlated with Swainson’s Warbler presence.  Standing or pooled water was recorded in 61% of random plots, but not in any territory plots, and presence of standing water was positively correlated with cane stem density at the Virginia site.  The latter result seems to contradict information assembled by Platt and Brantley (1997), which suggests that cane may benefit from seasonal flooding (perhaps due to nutrient transport into cane patches), but may be killed by prolonged inundation and thus prefers relatively dry bottomland soils.

This research was made possible only with funding and support from our partners: U.S. Fish and Wildlife Service - Region 4, Warnell School of Forest Resources Alumni Scholarship Fund, and the Georgia Ornithological Society.

[ back to the top]

STUDY AREA

            Our study area was located within Bond Swamp National Wildlife Refuge (BSNWR) administered by the U.S. Fish and Wildlife Service (USFWS).  BSNWR comprises 2,630 ha of land located along the upper Ocmulgee River (pictured below) in Bibb and Twiggs counties, Georgia, ca. 10 km south of the city of Macon.  Habitat types include bottomland hardwood forest, tupelo gum (Nyssa aquatica) swamp forest, and mixed pine/hardwood ridges, along with beaver swamps, oxbow lakes, and tributary creeks.  Some areas of the bottomland hardwood forest have characteristics of old growth forests (e.g., very large trees and openings).  The refuge is typically inundated by seasonal floodwaters in late winter through spring, and some swamp forest areas remain inundated year-round  (USFWS 2000).  We focused our research within the main body of BSNWR, in an area of ca. 1,740 ha. 

Ocmulgee River near Macon, Georgia

 [ back to the top]

METHODS 

            Field data collection—We established 21 transects totaling 18.4 km in April-June 2001 for Swainson’s Warblers searches at BSNWR .  Swainson’s Warbler territories were located also by searching the entire refuge in addition to conducting distance sampling and playback surveys for territorial Swainson’s Warblers along the transects. 

            From late April through mid-June 2001, we collected data by following singing males as they moved through their territories, maintaining a distance of at least 10 m from these individuals.  During our intensive territory observations, we recorded GPS coordinates for each location at which the individual bird was seen or heard.  Each individual was followed for at least 1-2 hr.  We obtained an average of 20 locations per individual.    

Counting dense cane stems at BSNWR (3.5 m x 97.5 m transect) in a 50 x 50 m plot of a Swainson's Warbler territory.

            We produced a simple map (10 x 10 m grid) from GPS locations of each SWWA territory.  These maps identified a 50 x 50 m (0.25 ha) area used most intensively by each individual during our territory observations.  We then established the 50 x 50 m plot above for measurement of habitat variables in the corresponding area of each territory.

            In addition, we established 50 x 50 m (0.25 ha) random unoccupied plots for habitat measurements along our transects for the Swainson’s Warbler surveys.  We measured the following habitat variables in all Swainson’s Warbler territory plots and random plots between 20 June and 1 August 2001: average cane height (m) in plot, to nearest 0.25 m, based on mean of 2 observers; total cane area (m2) in plot, mean from 2 observers; number of cane stems on 2 perpendicular 3.5- x 50-m belt transects (stem density); number of shrub, vine, and small tree stems (≤2.5 cm dbh) on same transects above (stem density); number of samples with dead ground cover (100 point samples of dead ground cover, every m on transects); number of samples with no dead ground cover (ibid.);  number of samples with water ground cover (ibid.); number of samples with broad-leaved forb ground cover (100 point samples for live ground cover, every m on transects); number of samples with grass ground cover (ibid.); mean leaf litter depth from 20 samples, taken every 5 m along transects; % canopy cover (number of point samples with canopy cover present, out of 100 samples; diameter at breast height (dbh in cm) of tree in plot with maximum dbh; distance from center of plot (m) largest tree (dbh); number of tree falls in plot; number of canopy tree openings in plot; basal area (x 10 = sq. ft./acre); pH water of plot soil sample; pH salt (CaCl2) of plot soil sample; % silt in plot soil sample; % sand in plot soil sample; % clay in plot soil sample; and electrical conductivity (deciSiemens/m [dS/m]) of plot soil sample. 

[ back to the top]

ANALYSIS

             We used SAS Version 8 for statistical analysis of habitat plot data.  We analyzed data for 56 Swainson’s Warbler territory plots (SWWA) (50 x 50 m, n = 56) and 110 randomly selected and unoccupied (RAND) plots.  We developed a set of alternative multivariate habitat models for Swainson’s Warbler territories at BSNWR using stepwise binary logistic regression.  We avoided potential multi-collinearity by preparing a Pearson correlation matrix prior to logistic regression analysis and eliminated 4 variables that were highly correlated (r² > 0.49) with others in our data set.  We then ran several different logistic regression analyses with the remaining 18 variables, using varying alpha-levels for entry and retention of variables in each model (alpha = 0.05, 0.10, and 0.15, respectively). 

[ back to the top]

RESULTS 

          We developed 3 habitat models using logistic regression, with increasingly permissive alpha-levels for entry and retention of variables in each model.  The most parsimonious model included 4 variables: cane stem density, shrub+vine+sapling stem density, mean litter depth, and ground coverage by water (alpha = 0.05 for both entry and retention; AIC = 80.6).  Other models contained more variables (6-11) and described the habitat in more detail, both biologically and intuitively.  Not surprisingly, predictive capability improved with the addition of more variables to the model (i.e., with increasingly "-levels), as estimated by several rank correlation indices calculated for the three models (SAS Institute Inc. 1990).  Rank correlations for assessing predictive ability of the models by Goodman-Kruskal Gamma values were 0.941, 0.950, and 0.981 (out of 1.00) for the 4-, 6-, and 11-variable models, respectively.  Percent concordance of the predicted probabilities and observed responses were high for all models (97.0, 97.5, and 99.0 respectively).   

[ back to the top]

DISCUSSION 

          Our analysis confirms that cane is an important component of Swainson’s Warbler breeding habitat at BSNWR.  Not only were mean cane area, cane height, and cane stem density much greater in territories than in random plots, but cane stem density was the most significant variable entered into each of our 3 logistic regression models.   The greater shrub stem density we found in territories, compared with unoccupied random plots, is of interest with regard to Graves’s (2001) work in Virginia.  Shrub stem density was the second variable entered in our logistic regression models, demonstrating that density of non-cane shrub-level vegetation may be nearly as important as the cane-related variables in separating Swainson’s Warbler territory plots from unoccupied plots at BSNWR.  While Swainson’s Warblers may preferentially establish breeding territories within cane patches in areas where cane is abundant, overall structure of the shrub layer (including both cane, if present, and other shrub-layer species) may be a more important factor in determining suitable breeding habitat.  The shrub stem count in our study area also included vines such as greenbriar.  In Graves’s (2001) logistic regression model, Greenbriar abundance was one of the most important variables for predicting Swainson’s Warbler presence.

            Leaf litter depth was the third most important variable in our models and was greater in territories than in random plots.  Conversely, unoccupied plots had a greater percentage of areas with no leaf litter and/or the presence of standing water; the latter variable was the fourth most important variable in the habitat models, and was negatively correlated with Swainson’s Warbler territories. As in our study, Graves (2001) also found the presence of standing water to be negatively correlated with Swainson’s Warbler presence in Virginia.   

            The mean difference in litter depth between our Swainson’s Warbler territory plots and unoccupied plots was 50% greater in territories.  This difference in litter depth may be important, in terms of food availability and foraging efficiency for the Swainson’s Warbler.  

³ E-mail:  joe_meyers@usgs.gov             

[ back to the top]