Night Roosts of Bald Eagles (Haliaeetus Leucocephalus) Wintering in Northern Arizona PDF Download

Author: Prabin K. Joshi
Publisher:
ISBN:
Category : Bald eagle
Languages : en
Pages : 156

Get Book

Book Description
Bald eagles (Haliaeetus leucocephalus) were delisted from threatened or endangered status in 2007 in the conterminous states because of their encouraging comeback throughout most of North America. However the recent court decision on 1 May 2008 forced USFWS to issue a rule to amend the regulations for the Federal List of Endangered and Threatened Wildlife by designating bald eagles in the Sonoran Desert of central Arizona (also called the ‘southwest population’) as threatened under the authority of Endangered Species Act of 1973. The southwest population is believed to exist in an ecological setting that is unusual and unique among bald eagle populations. Although it is unknown if the southwest population differs markedly from other bald eagle populations in its genetic characteristics, morphological and behavioral differences suggest that genetic differences may exist. Bald eagles are still protected under the Migratory Bird Treaty and the Bald and Golden Eagle Protection Acts. Bald eagles are habitat generalists and opportunistic feeders (they typically take the easiest and most abundant prey regardless of class or whether live or dead), and roost in a single or group of trees in winter. A roost is an individual or group of trees where ≥1 bald eagle(s) perch overnight. We hypothesized that bald eagles selected night roosts during winter that provided physical safety (e.g., hiding cover), thermal protection, and close proximity to food sources. Our objective was to describe habitat characteristics of night roosts used by migratory bald eagles wintering in ponderosa pine (Pinus ponderosa) forests of northern Arizona. Bald eagles are not usually thought to be a migratory species but some populations migrate while other eagles stay on or near their breeding territories yearround. We considered habitat use at two scales: microhabitat (habitat characteristics within a 30-m diameter area centered at the roost tree measured on ground-based plots) and macrohabitat (habitat characteristics within a 90-m diameter plot centered at the roost tree based on spatial data layers from a geographic information system). Arizona Department of Emergency and Military Affairs and Arizona Game and Fish Department captured and fitted 10 bald eagles (≥3.70 kg, mean and standard error [SE]: 4.75 ± 0.22 kg; 7 adults, 1 subadult, and 2 juveniles) with 70-g, solar-charged satellite Platform Transmitter Terminals at Camp Navajo, Flagstaff, Arizona in 2005 and 2007. We measured microhabitat characteristics of 54 night roosts for 7 bald eagles (6 adults, 1 subadult; mean number of locations per eagle and SE: 8 ± 2) in ponderosa pine forests and ≥500 m apart (for spatial independence). We paired each roost site with a randomlyselected plot ≥500 m away. Random plots had to be in a ponderosa pine stand with at least one tree ≥20 cm diameter at breast height (dbh; large enough for eagles to perch overnight). To measure macrohabitat characteristics, we selected 200 night roosts (mean and SE: 25 ± 11 roosts per eagle; range: 7 to 43) for 8 bald eagles (6 adults, 1 subadult, 1 juvenile) that were located within ponderosa pine forests of northern Arizona. Spatial data layers we used were mapped by the Forest Ecosystem Restoration Analysis (ERA) project with a 90-m resolution raster dataset. We generated 200 random points within ponderosa pine stands ≥500-m apart for comparison points and measured the same characteristics as for night roosts. We developed 13 and 11 a priori models for microhabitat and macrohabitat use, respectively. We used Akaike’s Information Criteria or Akaike’s Information Criteria adjusted for small sample sizes to evaluate models describing night roosts. Models with ∆AICc ≤2 were considered the best approximating models. Microhabitat analyses showed that trees used as night roosts were larger dbh (roost: 75.2 ± 2.2 cm, random: 58.2 ± 1.5), on steeper slopes (roost: 12.4 ± 2.2 percent, random: 6.6 ± 1.1) and surrounded by greater basal area (roost: 19.7 ± 8.0 m2 /ha, random: 17.7 ± 7.8), higher densities of large trees (roost: 61.9 ± 4.8 trees/ha, random: 39.6 ± 4.5) and lower densities of small trees (roost: 137.3 ± 17.4 trees/ha, random: 158 ± 24.0) than comparable characteristics of randomly-selected sites. Mean dbh for trees in the plot was larger (61.9 ± 4.8 cm) than for random plots (39.6 ± 4.5 cm). Roost plots were more likely to face east (67%) than west (33%). The global model best predicted use of a roost by bald eagles (wi = 0.998); other models performed poorly in comparison (∆AICc ≥12.34). Most (94%) of the large trees in roost plots were co-dominant or dominant and had ≥50% live crown ratio (91%). We used a post-hoc analysis to find a more parsimonious model to describe night roost characteristics. The top 3 models (∆AICc ≤3.94) that best predicted use of a roost by eagles included dbh of the largest tree in each plot, density of large trees, density of small trees, slope, eastness (Sin [aspect in degrees] transformed aspect using trigonometric function), tree size variability, and number of trees with live crown ratios ≥75%. At the macrohabitat scale, eagles used roosts that were closer to highways and interstates and farther from lakes than randomly-selected points. Since highways and interstates are believed to provide major sources of food (e.g., road-killed mammals) for bald eagles during winter, eagle roosts appeared to be located in close proximity to food sources. Eagles selected north-facing roosts which may offer protection from southwest winds. Twelve hotspots for bald eagles (used by ≥1 eagle ≥5 times) were identified. Hotspots used by multiple eagles were on generally northeastfacing, on greater slopes and closer to lakes, highways, and interstates. Location and structure of night roosts seemed to provide physical safety, better thermal cover, and sources of food to bald eagles wintering in northern Arizona.