Behavioral Ecology Advance Access originally published online on July 22, 2008
Behavioral Ecology 2008 19(6):1314-1325; doi:10.1093/beheco/arn073
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Alternative strategies of space use and response to resource change in a wintering migrant songbird
Department of Ecology and Evolutionary Biology, Tulane University, New Orleans, LA 70118, USA
Address correspondence to D.R. Brown, who is now at Department of Biological Sciences, Eastern Kentucky University, Richmond, KY 40475, USA. E-mail: david.brown{at}eku.edu.
Received 25 September 2007; revised 24 May 2008; accepted 27 May 2008.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
The causes and consequences of nonbreeding space-use strategies are poorly understood. We studied 2 alternatives, sedentary and floating behaviors, in a wintering population of the Ovenbird (Seiurus aurocapilla), a Neotropical–Nearctic migrant, in response to manipulated and natural variation in food availability over 4 years in Jamaica, West Indies. Using radio transmitters, we documented in sedentary individuals use of a fixed home range, greatly overlapping those of neighbors, as well as core areas that overlapped little, suggesting that Ovenbirds defend the core of their home range as a territory. Floaters included individuals with multiple disjoined home ranges and individuals that undertook frequent excursions, but floaters always occupied relatively large feeding areas. Floaters comprised 8–17% of the population, and in some individuals, the behavior persisted in multiple winters, but the behaviors were not sex or age restricted. Sedentary birds were attracted to artificial feeding stations located within their home range. However, reduction of food availability did not induce sedentary individuals to expand or shift their home range or to adopt floater behaviors. By contrast, floaters appeared better able to exploit seasonal or experimentally induced variation in food availability by matching their space use to the resources. The physical consequences of these alternative wintering strategies were situation dependent: Whereas body mass of territorial birds was positively correlated with food availability, floaters showed the opposite response, with higher mass in food-reduced situations. These results suggest that alternative behaviors represent a trade-off in response to resource availability.
Key words: alternative strategies, floater, food supply, home range, Jamaica, migrant songbird, nonbreeding, Ovenbird, Seiurus aurocapilla, trade-off, winter.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
Vertebrates use space in a variety of ways, best studied in the context of competition for mates (Smith 1978
; Smith and Arcese 1989; Solomon and Jacquot 2002; Shuster and Wade 2003; Stapley and Keogh 2004). Alternative strategies of space use, frequently called alternative territorial strategies, also occur during the nonbreeding season (hereafter "winter") and have been observed in a number of species of migrant songbirds (Schwartz 1964; Rappole and Warner 1980; Rappole et al. 1989; Winker 1998; reviewed by Brown and Long 2007).
Alternative space-use behaviors may include strategies such as a fixed home range, flocking, and solitary floating. The term "floater" generally applies to any nonsedentary, nonflocking individual. Among wintering migrant songbird species whose individuals primarily occupy winter territories, many appear to include a substantial proportion of floaters (Rappole et al. 1989; Brown and Long 2007). Whereas the ecological context of flocking and territoriality is well studied, we know surprisingly little about floating compared with territorial behavior, let alone the context, consequences, or trade-offs.
In migratory birds, the principle resource requirements during the winter season are food and safe areas with respect to predators. At multiple spatial and temporal scales, the distribution of wintering birds in Jamaica matches food supply (Johnson and Sherry 2001), supporting the hypothesis that food availability affects space use (Sherry et al. 2005). Winter space-use patterns of migrant songbirds are also influenced by predation pressure (Cuadrado 1997). A third factor, intraspecific dominance interactions, also affects these birds’ space use. In dimorphic migrant songbird species with small winter home ranges, males and females often occupy different habitats (Lynch et al. 1985; Wunderle 1995; Marra 2000). In some cases, such differential occupancy is driven by choice (Morton 1990), but at least in the American redstart (Setophaga ruticilla), dominant males exclude females from higher quality habitats (Marra 2000). Sexual habitat segregation did not occur in at least one sexually dimorphic, winter-territorial species, the Kentucky warbler (Oporonis formosus; Maybe and Morton 1992). Age is also an important factor affecting dominance (Stutchbury 1994; Wunderle 1995; Marra 2000). By contrast, little evidence involving monomorphic species exists that age- or sex-based dominance is operating despite subtle average body size differences between the sexes (Bates 1992; Brown et al. 2000; Koronkiewicz et al. 2006; Sogge et al. 2007). This does not preclude variation in space-use strategies within and between sexes or age classes of monomorphic birds, and body size may be an important factor. Given that food supply, predator abundance, and dominance status all appear to contribute to structuring winter local-scale distributions, one would expect these factors also to correspond with alternative space-use strategies within a population.
In this study, we radio tracked wintering Ovenbirds (Seiurus aurocapilla) to characterize space-use patterns and to investigate how space use varies in relation to food supply and dominance status, as determined by sex and age class. Risk of predation is minimal for migratory songbirds in Jamaica, the mongoose (Herpestes javanicus) being their only occasional predator (Wilson B, personal communication). With predation thus controlled, we focused this study on the role of food and dominance in structuring local space-use behavior.
Ovenbirds are long-distance migrants that spend the stationary portion of the winter throughout the Caribbean and Central America (Van Horn and Donovan 1994). Ovenbirds are site tenacious through the winter; persistence from early- to late-winter ranges from 35% to 80% (Strong and Sherry 2000). Ovenbirds also frequently return from migration to occupy the same home range between years (Van Horn and Donovan 1994). Wintering Ovenbirds occupy a variety of habitats but forage almost exclusively on the ground. Their diet is primarily ants and other ground surface-active arthropods, fruit and seeds are uncommon (Strong 2000). Male and female Ovenbirds have indistinguishable plumage, but males average slightly larger body size (Van Horn and Donovan 1994; Pyle 1997). Immature and adult birds also look similar and do not differ in body size.
Although songbirds display considerable variation in space-use behaviors (Rappole and Warner 1980; Staicer 1992; Greenberg and Salewski 2005; Brown and Long 2007), in general, male and female individuals of a large proportion of species display tenacity to small territories throughout the winter (Greenberg and Salewski 2005). In Ovenbirds, individuals are "territorial" only in a loose sense, in that they do not actively patrol and defend their entire home range but instead maintain spatiotemporal territories by defending their immediate space only during chance encounters (Strong 1999, 2000). In Jamaica, most Ovenbirds occupy solitary, but overlapping, home ranges throughout winter and are site faithful to a home range among years (Strong 1999). Strong (1999) found that the degree of home range overlap was related to variation in food supply among habitats. He also conducted a small-scale experimental reduction of food supply within the core home range of 4 Ovenbirds and found that home ranges appeared to expand relative to control birds. Because food manipulations have been a valuable tool for studying behavior (Boutin 1990; Newton 1998), we expanded on the experimental approach of Strong (1999) by using both food reductions and supplementations to further explore space-use behavior of Ovenbirds.
Based on the Ovenbird's natural history and our knowledge of floating behavior in other species, we made the following predictions about wintering space use: 1) Most Ovenbirds should occupy solitary feeding home ranges that are similar in size and overlap among neighbors and randomly dispersed with respect to sex and age in this monomorphic species. 2) Ovenbirds should respond to variation in food availability by changing the size of their home range and thus also changing the degree to which their home range overlaps with neighbors. Specifically, food supplementation should lead to home range contraction, especially when the resource is concentrated, and to increased overlap as birds shift to areas of increased food supply. Food reduction should force home ranges to expand and shift as birds seek areas of higher food supply. 3) We also expected that food reduction would induce some birds to adopt a floater strategy. Because we suspected that prior experience influences space-use strategy, we predicted that immature birds should be more likely than adults, regardless of sex, to adopt space-use behaviors alternative to a fixed home range when resource conditions decline (naturally or by manipulation) within a winter season (Brown and Long 2007). 4) Whereas body condition of sedentary birds should correspond to induced variation in food supply, floater body condition should not respond as individuals alter their space use to match food availability.
To test these predictions, we quantified space use of radio-telemetered birds in different demographic classes. We did not specifically investigate how body size influenced space use; instead, we considered sex as a classification that includes body size and other characteristics that may have influenced dominance but were undetectable to observers. If body size was an important factor in determining dominance, we expected a statistical signal to be detected in comparisons between sexes because males average slightly larger than females (mean mass ± standard error [SE]: females 18.05 ± 0.09, males 19.21 ± 0.12; mean wing chord ± SE: females 71.20 ± 0.14, males 75.97 ± 0.18; females: N = 147; males: N = 108; all from study population). We then tested how birds adjusted space use in response to food supply by separately increasing and decreasing food availability in independent plots over multiple years. Because the study spanned years that varied naturally in climate and food availability, we also examined bird responses to this natural food variability.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
Food supply treatments
We studied Ovenbirds and their food supply from January to April of 2002–2005. Ovenbirds arrive in Jamaica in August and September, reside throughout the winter, and depart in late April and May. Thus, our study spanned the middle of the winter season of each year until the migration period. This period coincides with the late-winter Caribbean dry season. We reasoned that naturally dry conditions would make supplemental food more attractive to birds and thus produce stronger patterns of behavioral response. The study sites became increasingly desiccated through each study period, with more extreme drought conditions in each successive year (Table 1). We studied Ovenbirds in 2 habitats, shade coffee and second-growth scrub, but mostly in scrub habitat. We do not compare patterns between these habitats.
|
We manipulated food supply over 4 years in Jamaica. Food manipulations were begun in February of each year after an initial premanipulation sampling period. In 2002, we conducted a food reduction experiment in a single 5-ha plot and maintained a separate 5-ha control plot separated by 0.5 km at Baronhall shade coffee plantation (18°13'N, 77°22'W). In the food reduction plot, we found individual ant colonies by visual searching and applied AMDRO ant-bait pesticide directly to the colony entrance (Strong 1999
). AMDRO is a fast-acting pesticide: Ants carry it into the colony and feed the queen and larvae, effectively destroying the entire colony in several days. In 2003—2005, we separately reduced and augmented food availability and maintained controls in replicate 1-ha plots in second-growth scrub habitat at Luana Point Nature Reserve (18°02'N, 77°55'W). Although seemingly small, 1-ha plots were large enough to include several Ovenbirds each (see Results: Figure 3) and small enough to allow manipulation of food supply. Each year from 2003 to 2005, we maintained two 1-ha control plots in the same locations in all years. In 2003, we manipulated food in 4 plots, 2 each of food reduction and supplementation. In 2004 and 2005, we manipulated 6 plots, 3 of each experimental treatment. Plots were in the same location from year to year, but in most cases, plot treatment was switched in successive years to avoid any site effects. Nearest-neighbor distance among plots ranged from 100 to 1000 m. To reduce food supply on scrub plots, we also used AMDRO, but with a single broadcast application in the evening. The food reduction lasted through the end of the study in each year. We supplemented Ovenbird food availability by uniformly distributing 21 piles of cut oranges across each 1-ha plot. Orange piles were replenished every 1–3 days to compensate for effects of consumption and desiccation. We also maintained food-supplementation plots through the end of the study each year, that is, until the spring migration period. In addition to serving as a direct source of nutrients and water, oranges increased prey availability by attracting large numbers of ants and other arthropods.
|
Arthropod and bird sampling
We visually surveyed arthropods in replicate quadrats of 0.25 m2 for 5 min each (range: 468–586 total arthropod surveys per year). Surveys were randomly located within each plot and repeated in identical locations in the early- and late-sampling periods. We also conducted surveys centered on supplementation plot feeding stations. We counted and estimated lengths of all arthropods and then converted these estimates to biomass using length–weight regression equations specific to Jamaica (Johnson and Strong 2000
). Because manipulations had the greatest effect on ants, and ants are numerically the principle food of wintering Ovenbirds (>60% of stomach items), we report biomass of ants and not other arthropods (Strong 2000). There were a variety of ant species present in the study area, and alates were uncommon. For additional details on study design, study sites, experimental manipulations, and bird and arthropod sampling, see Strong (2000) and Brown and Sherry (2006).
We used mist nets for capture–mark–recapture of birds. We sampled birds in both pre- and postmanipulation periods in each plot. Within each of these periods, we sampled each plot over 2 days. Each day, we erected 10–20 mist nets. After sampling all plots once, we returned to each plot a second time, placing nets in new locations. This sampling design effectively saturated each plot (176.21 ± 6.28 mean mist net h plot–1 period–1 year–1 ± SE) and produced capture probabilities of >60% (Brown 2006). All birds were marked with a unique combination of 1 aluminum numbered USGS plus 2 colored plastic bands. On each individual, we recorded total body mass (Ohaus balance with 0.1 g precision), visible fat score (6-class scale ranging from 0 = no visibly detectable fat to 5 = thick fat covering entire body), and breast muscle score (3-class scale: 1 = concave; 2 = flat; 3 = convex). We recorded several body size measurements (bill width, depth, and length at anterior margin of nares and wing chord length, tail length, and tarsus length) and combined these into an index of structural body size with principal components analysis (Brown and Sherry 2006). We aged each bird as immature or adult using subtle differences in flight feather shape, coloration, and wear (Pyle 1997) and collected a blood sample (70 µL drawn from brachial vein of wing) for molecular sexing. Most birds were sexed by either body size measurements and/or a polymerase chain reaction technique (Griffiths et al. 1996). We could not determine the sex of 27% of all captured individuals.
We attached 157 radio transmitters (Holohil Systems, Carp, ON, Canada) to 140 individuals using prefabricated leg-loop harnesses made of elastic thread. In general, harnesses lasted for the duration of the season and were absent on birds that returned the following year. Eighteen of these 140 individuals were tracked using one radio each in shade coffee; the remaining 139 radio tags were used in scrub habitat on 122 individuals over 3 years. Several of these latter individuals (N = 6) received more than one radio in 1 year to replace radios with dead batteries. Among the many birds that were captured after returning among years (N = 86 of 255), 11 different individuals received radios in more than 1 year and were selected as a random subset of all individuals captured each year. In 2002–2004, we used radio tags weighing 0.5–0.6 g (<3% body mass) with a battery life of 21 days. In 2005, we used heavier radio tags (0.9 g, <7% body mass) so that we could track birds longer (56 days). We noticed no adverse behavioral reaction such as difficulty flying, walking, or foraging because of the larger radios. Birds were located on a gridded trail system by single-observer triangulation of 2 bearings with a hand-held receiver (Wildlife Materials, Murphysboro, IL, RS-64) and 3-element Yagi antenna. The topography of the study areas is generally flat and the vegetation homogenous, allowing little, if any, bias in detectability among radio-tagged individuals. Most bearings (>95%) were taken within 50 m of a bird's location. Maximum detection distance generally ranged from 100 to 200 m, although individuals were occasionally detected at a greater distance. Locations were calculated by triangulation from bearings with program LOAS (2002). Location accuracy was estimated to be within 87 m2 using maximum likelihood–based area of confidence ellipses (Saltz and White 1989). We collected an average of 39 ± 1 (mean ± SE) locations per individual. We collected pre- and postmanipulation data on a subset of radio-tagged birds in 2002 (N = 11) and 2003 (N = 18). However, it was not practical to collect pre- and postmanipulation data on all birds because of the short battery life of the radio tags. Thus, for most birds, we have only postmanipulation telemetry data.
We calculated plot persistence as the proportion of birds recaptured in the postmanipulation period that had been originally captured prior to manipulation. We calculated relative abundance as the total number of individuals captured in each 1-ha plot by time period by year combination.
Designation of behavior and home range estimates
We classified the fate of radio-tagged birds as surviving through the battery life span, depredated, or disappeared. Surviving birds were further classified as sedentary versus floaters. Sedentary birds occupied relatively small, fixed locations throughout the battery life span of a transmitter. Floaters were a more heterogeneous group, displaying different behaviors, but they had at least one of the following characteristics: a relatively large home range with multiple core areas; multiple, separated home ranges; or frequent, relatively long junkets. Some birds disappeared before we were able to distinguish whether they were sedentary or floaters. Frequently, these birds were followed for a short time as they moved away from the study area before disappearing. When birds disappeared, we searched the surrounding landscape for a radio signal. Although some of these birds could have died or been depredated, radio tags often continue to transmit after birds are killed. Thus, we suspect that most moved beyond the limits of the search area. There is little chance that these movements were migratory because the study took place 1–3 months before the earliest onset of spring migration. However, because we cannot make a confident determination of the fate of these birds, we report estimates of the proportion of floaters and disappeared birds as ranges to account for such uncertainty.
We used ARCVIEW Animal Movement Extension to compute fixed-kernel utilization distributions (UD) from point location estimates (Hooge and Eichenlaub 1997). Kernel estimates are based on densities of observed locations across a grid of artificial locations and produce both 2-dimensional area contours at user-defined increments and a 3-dimensional volumetric estimate (White and Garrott 1990; Millspaugh and Marzluff 2001). We measured the area of overlap between neighbors at 95% and 30% of the total UD for all pairs of intersecting home ranges and calculated the average area of overlap for individuals with multiple overlapping ranges.
Statistical analyses
To test for differences in food supply among years and treatments, we used analysis of covariance (ANCOVA) with premanipulation ant counts as the covariate. This essentially acts as a repeated measures analysis of treatment effect. We used birds’ space-use behavior data from coffee only in the experiment to induce movement behavior by food reduction. For all other tests of home range characteristics, we used only birds from scrub habitat. Because we radio tracked some individuals in scrub more than once, we used only the sample that had more locations in analyses of home range characteristics. We used analysis of variance (ANOVA) to compare home range size (95% UD) and area of overlap (both 95% and 30% UD) among sexes, age classes, years, and food supply treatments. We assessed the dispersion of males and females by counting the number of overlapping neighbors at the 95% UD of each sex and the combined number for every radio-tracked bird and then comparing the sexes with t-tests. For this test, we expected a random dispersion of the sexes such that individuals of both sexes would have similar numbers of male, female, and total neighbors. We followed the same approach to determine if there was a nonrandom dispersion based on age class, but in this case, we expected adults, in contrast to immatures, to have fewer immature and total neighbors. We tested for changes in space use (home range size and overlap) by individual birds from pre- to postmanipulation using paired t-tests. For birds in supplementation plots, we investigated the proximity of birds to feeding stations by comparing observed locations to random locations within each home range generated with a uniform probability across the home range and not with respect to the activity and then compared means of each individual using a paired t-test. We compared home range size of sedentary birds and floaters with a t-test. For analyses of home range characteristics, we tested for differences in the frequency of sedentary and floater behavior among manipulation treatments, sexes, age classes, and years using chi-squared tests. We tested for year and treatment differences in plot-level persistence using ANOVA. To test for differences in plot-level relative abundance, we used ANCOVA with year and treatment factors and premanipulation values as a covariate. We compared body condition of floaters and sedentary birds using an ANCOVA model with time period (pre- and postmanipulation), behavioral strategy, and food treatment as factors. For these analyses, we included known floaters plus birds that disappeared but that we suspected were floaters. To control for structural body size, we included as a covariable the first principal component of all body size measurements. We used SYSTAT 10.0 for all statistical analyses (Wilkinson 2000) and report descriptive statistics as means ± 1 SE.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
Climate and food manipulation
In the single year of sampling coffee habitat, drought conditions did not occur (December to April precipitation = 437 mm; Table 1). Ant biomass in the coffee food reduction plot was reduced by 38% from premanipulation level (F1,231 = 24.37, P < 0.001). In the coffee control plot, by contrast, ant biomass did not change from pre- to postmanipulation.
In scrub habitat, annual variation in premanipulation ant biomass mirrored precipitation by declining in each successive year (F2,807 = 64.47, P < 0.001). Randomly located arthropod surveys revealed that across all years, manipulations caused a strong, linear response with lower ant biomass in reduction plots and higher in supplementation plots (F2,691 = 38.40, P < 0.001). In supplementation plots, feeding stations provided birds with arthropod and fruit resources in superabundant patches compared with surrounding areas. For example, ant biomass at feeding stations increased 5–10 times relative to premanipulation levels (F1,226 = 149.54, P < 0.001). In reduction plots, ant biomass was reduced relative to premanipulation levels by 97%, 88%, and 81%, respectively, in 2003–2005. The progression of the dry season each year also caused food supply to decline in control plots, but to a lesser extent (2003: 68%; 2004: 30%; 2005: 46%).
Dispersion of sedentary birds
The sexes did not differ in the number of males with overlapping 95% UD (males: 2.43 ± 0.23; females: 2.53 ± 0.20; t87 = 0.34, P = 0.73), the number of females with overlapping 95% UD (males: 2.52 ± 0.38; females: 2.53 ± 0.28; t87 = 0.01, P = 0.99), or the total number of overlapping neighbors (males: 5.32 ± 0.45; females: 5.47 ± 0.43; t97 = 0.23, P = 0.82). The age classes did not differ in the number of adults with overlapping 95% UD (adults: 3.21 ± 0.332; immatures: 2.93 ± 0.36; t97 = 0.56, P = 0.57) or the total number of overlapping neighbors (adults: 4.96 ± 0.40; immatures: 5.68 ± 0.40; t97 = 1.27, P = 0.21), but there was an almost significant trend for adults to have fewer immature birds overlapping their home range (adults: 2.01 ± 0.27; immatures: 2.80 ± 0.31; t97 = 1.92, P = 0.06). The degree of overlap among neighboring home ranges at the 95% level did not differ by age (F1,67 = 0.64, P = 0.42; Table 2), sex (F1,67 = 0.06, P = 0.80), treatment (F1,67 = 1.61, P = 0.20), or year (2003: 21.03 ± 2.38%; 2004: 23.29 ± 2.67%; 2005: 22.80 ± 3.38%; F1,67 = 0.34, P = 0.71). However, at the 30% UD level, overlap was higher in immature birds than adults (F1,67 = 4.64, P = 0.03) but did not differ by sex (F1,67 = 0.40, P = 0.52), treatment (F1,67 = 1.98, P = 0.16), or year (2003: 2.60 ± 0.072%; 2004: 3.22 ± 0.73%; 2005: 3.21 ± 1.43%; F1,67 = 0.72, P = 0.49).
|
For sedentary birds radio tracked only after manipulation, home range size at the 95% UD level did not differ between age classes (adults: 0.68 ± 0.06 ha; immatures: 0.81 ± 0.10 ha; F1,73 = 1.64, P = 0.20; Figure 1), but there was a nonsignificant trend for larger home ranges in males (0.79 ± 0.09 ha) than females (0.67 ± 0.07 ha; F1,73 = 2.67, P = 0.11). Despite large variation in food supply, home range size of sedentary birds did not differ among years (2003: 0.69 ± 0.08 ha; 2004: 0.75 ± 0.09 ha; 2005: 0.83 ± 0.14 ha; F2,73 = 0.20, P = 0.82) or treatments (F1,73 = 1.61, P = 0.80).
|
For birds radio tracked before and after manipulation, mid-season changes in food supply did not cause detectable changes in home range structure in either scrub or coffee habitat: home range area (95% UD and 30% UD), centroid location, and bird movement rates did not differ from before to after food reduction (all t < 1.0, all P > 0.33; Table 3). Food supplementation in scrub also had no detectable effect on home range size, location, or movement rate (all t < 1.0, all P > 0.33). However, among all individuals in supplementation plots, birds were on average located closer to feeding stations (13.7 ± 0.5 m) than to random locations within their home ranges (16.7 ± 0.6 m; t45 = 5.38, P < 0.001). We also observed these birds frequently at the feeding stations, eating both the ants attracted to the stations and the orange pulp.
|
Floater behavior
Of 120 birds radio tagged over 4 years in scrub habitat, 8–17% displayed a nonsedentary (i.e., floater) behavior based on the assumption that disappeared birds could have been floaters (Table 4). In scrub, the ratio of floaters to sedentary birds did not differ between food reduction and food-supplementation treatments (
2 = 2.54, P = 0.11). Among all floaters, the proportion of males and females were similar (male: 53%; female: 47%) and did not differ from the sex ratio of sedentary birds (male: 54%; female: 46%; 2 = 0.02, P = 0.96). Likewise, the proportion of all floaters in each age class was similar (immature: 53%; adult: 47%) and did not differ from the age class ratio of sedentary Ovenbirds (immature: 45%; adult: 55%; 2 = 0.48, P = 0.48). Several birds (N = 3) retained floating status in subsequent winters, but most floaters were never observed again. The proportion of radio-tracked birds that were known floaters or that disappeared did not differ between years (2003: 14%; 2004: 24%; 2005: 15%; 2 = 1.58, P = 0.45).
|
Based on 7 floaters for which we had sufficient locations (>10) to conduct home range analysis, floaters had considerably larger home ranges and greater variation in home range size (12.60 ± 5.21 ha; range: 3.00–42.09 ha) than sedentary birds (0.74 ± 0.06 ha; range: 0.08–2.67 ha, N = 85; t90 = 8.37, P < 0.001). Because of their use of large areas, we do not have, for most floaters, detailed information on their space use. However, we do know that behavior varied considerably among these individuals. Three examples illustrate this variation. One individual maintained a small home range (<0.25 ha) within a food reduction plot where it was located 38% of 70 occasions. On 15% of occasions, this bird could not be found at all and was likely several hundred meters away. For the remaining 53% of occasions, it was located on a nearby supplementation plot (150 m distant) where it was observed at almost every orange pile feeding station and it used a total area in excess of 4 ha. It would be tempting to suggest that this bird was responding directly to the addition of food, but we had radio tracked this bird the prior year and observed similar behavior when the food reduction plot containing its small home range was a food-supplementation plot and the nearby plot was not yet established.
The second individual floater shuttled back and forth between supplementation plots separated by more than 1 km but stayed on each plot for 13 and 24 days (Figure 2). Within each plot, this individual stayed within a relatively small area (0.13 and 0.85 ha) but occasionally (8% and 15%, respectively) disappeared or moved outside its normal home range.
|
The third individual moved 200 m within a day of our attaching the radio tag. Over the subsequent 3 days, it continued to move away from the area of capture and eventually into a different habitat (mangrove) before disappearing altogether.
Persistence, abundance, and body condition
For all birds captured in premanipulation sampling, plot-level persistence from pre- to postmanipulation did not differ between treatments (F1,12 = 0.06, P = 0.80; reduction: 0.50 ± 0.11; supplementation: 0.55 ± 0.08). Prior to manipulation, relative abundance of Ovenbirds per 1 ha plot also did not differ between treatments (F1,12 = 0.01, P = 0.92). However, after manipulation, supplementation plots had higher abundance of Ovenbirds than food reduction plots (F1,11 = 7.83, P = 0.01; Figure 3).
Prior to manipulation, body mass did not differ between floaters and sedentary birds (F1,29 = 0.06, P = 0.81). The physical body mass response to treatments differed between floaters and sedentary birds (treatment x behavior x time period interaction: F1,82 = 14.93, P < 0.001; Figure 4A,B). In response to food manipulation, body mass of sedentary birds increased over the winter in supplementation plots and decreased in reduction plots. Floater body mass responded in the opposite manner, increasing on the food reduction plots and decreasing on the food-supplementation plots. This total body mass pattern was paralleled by the lipid component of body mass as measured by visible fat deposits (treatment x behavior x time period: F1,82 = 13.78, P < 0.001; Figure 4C,D). The pattern for breast muscle score was also similar for all groups except floaters in supplementation plots (treatment x behavior x time period: F1,80 = 9.62, P = 0.003; Figure 4E,F).
|
|
DISUCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
Wintering Ovenbirds displayed a range of localized space-use behaviors that we partitioned into 2 broad categories: 1) Sedentary birds of both sexes and both age classes that were randomly dispersed with respect to these demographic classes and that occupied small, overlapping, solitary, fixed home ranges with a core area (30% UD) apparently defended as a territory and used for nighttime roosting (Brown and Sherry forthcoming
); 2) floaters used larger areas, but their behaviors were more diverse. The proportion of floaters was similar for each sex and age class and did not differ from ratios of these demographic classes among sedentary birds. Both plot-level manipulations and successive decreases in annual precipitation enabled us to compare movement behaviors across a gradient of Ovenbird food supply that spanned years and treatments. To our surprise, sedentary birds did not respond spatially to reductions in food supply by expanding or shifting home ranges or by switching to floater behavior. Below, we suggest possible constraints on extra–home range movement by sedentary birds. Although home range size and location did not change detectably after food supplementation, sedentary birds did frequent feeding stations and located themselves closer to feeding stations than to random positions within their home range. Thus, sedentary birds in supplementation plots responded to small-scale variation in food availability by adjusting within–home range activity but did not alter overall home range location, size, or degree of overlap with neighbors. In contrast, floaters responded to variation in food supply at larger spatial scales. Increased plot-level abundance of individuals in supplementation plots was likely caused by floaters because it could not be attributed to any significant movement by sedentary birds.
Perhaps, most interestingly, the body condition of individuals with sedentary and floater strategies responded in opposite manner to food manipulation: sedentary birds responded as expected, with additional food leading to improved physical condition over the winter. In contrast, and counter to expectation, physical condition of floaters declined in supplemental food plots and improved in food reduction plots (Figure 4). Although Ovenbirds did not shift strategies, at least in mid-winter, each strategy appears to have entailed different advantages for obtaining food such that there exists a trade-off in risk at different scales. This is only the second demonstration, of which we are aware, of a trade-off in risk of assuming alternative behavioral strategies in relation to spatially heterogeneous environments (see Dingemanse et al. 2004). Below, we discuss in detail the causes, constraints, and consequences of Ovenbird winter social strategies in relation to food supply and dominance.
Ovenbird winter social system
The majority (83–92%) of Ovenbirds in the study population were sedentary. These birds maintained solitary feeding and associated roosting home ranges (Brown and Sherry forthcoming) that were relatively stable throughout the winter. These home ranges had considerable overlap. Strong (1999) argued that wintering Ovenbirds maintain spatiotemporal territories, actively moving away from each other after chance encounters. The skulking behavior and infrequent vocalizations of Ovenbirds during the late winter result in home range defense occurring only after chance encounters. However, overlap of home range cores (30% kernels) was low (Table 3). Thus, these birds appear to maintain a subtle kind of territoriality through active defense of just their home range core.
The explanation for the substantial overlap of home ranges as represented by the 95% kernels is not clear but could be a response to the diet, comprised largely of ants (Strong 2000), which appears to represent a shifting and perhaps unpredictable resource (Levings 1983). Large and overlapping home ranges could thus constitute a trade-off between food availability and competition with neighbors. Alternatively, feeding on the ground may make difficult the kind of territory surveillance behaviors typical of other birds such as wintering American redstarts (Holmes et al. 1989). However, other species of wintering migrants forage on the ground while maintaining minimally overlapping territories including Hooded Warbler (Ogden and Stutchbury 1994), Kentucky Warbler (McDonald 1998), and Wood Thrush (Hylocichla mustelina; Roth et al. 1996). The Ovenbirds in this study system appear to have a more specialized diet, with a larger proportion of ants than these other ground foragers. These subtle differences in feeding behavior may help explain the high degree of home range overlap in Ovenbirds.
A subset of the local population displayed nonsedentary behaviors. Such behaviors included use of multiple home ranges and frequent extended junkets of distances measured up to 1 km. In addition, more than 8% of the birds we radio tracked disappeared soon after tagging (Table 4), and many of these were followed for one to several days as they moved progressively farther away from their initial capture location. Thus, we suspect most of the unexplained disappearances to be floaters.
Social dominance and space use
According to our results, birds of both sexes and age classes were equally likely to display alternative spatial strategies. Regardless, prior experience, measured here operationally as age, may be an important component of the social spacing system of wintering Ovenbirds. Because annual return rates are high in Neotropical migratory birds, we are confident that most adults benefit from priority effects. Although we found almost no difference in home range structure, dispersion, or behavioral response to food supply among demographic classes of sedentary birds, overlap of home range cores (30% UD) with neighbors was higher in immature than adult birds (Table 3), and adults tended to share home range area with fewer immature birds than did other immatures. This suggests that immatures were either not able to defend home range core areas and were thus at a behavioral disadvantage to adults or not as good at finding food. These findings are consistent with the subordinate status of immatures in other wintering migratory birds (Stutchbury 1994; Marra 2000). Strong and Sherry (2000) did not find differences in persistence rates for adult and immature Ovenbirds. However, their early-sampling period came well after fall arrival, when there is probably more home range abandonment and greater landscape-scale movement (Marra 2000).
Interestingly, in the present study, we found no evidence of sex-based differences in space-use patterns. This suggests that females are able to compete with males for space and, hence, food. This is consistent with the evidence that females are in general not strictly relegated to winter social subordination in monomorphic species (Brown et al. 2000; Koronkiewicz et al. 2006). Sexually monomorphic species like the Ovenbird thus seem to differ from the American redstart, a model (and sexually dimorphic) organism in population studies of winter season events (Runge and Marra 2005), by allowing simpler models of population dynamics. The apparent dominance by adults and absence of sex-based dominance in monomorphic species suggest that prior experience is more important than body size or other morphological characteristics in determining individual ability to dominate space.
Food supply and space use
Despite drastic induced reduction of food supply (38–97% of ant biomass), individual Ovenbirds monitored both before and after manipulation in 2 separate habitats did not expand their home range, one of the most surprising results of our study. Ovenbirds were tenacious on food-reduced plots even as their physical condition deteriorated (Brown and Sherry 2006). Although this finding is consistent with a food reduction experiment on wintering Hermit Thrush (Brown and Long 2006), it contrasts with a small-scale study by Strong (1999) in which he induced home range expansion of 4 Ovenbirds by reducing food within their core home range. It is possible that we did not adequately reduce nonant food, allowing Ovenbirds simply to switch diet to compensate for the decreased ants. Another alternative is that drought conditions naturally limited food supply outside the reduction treatment area so that the gradient of food availability from the reduction treatment to ambient (control) conditions was not strong enough to create a response. However, we strongly reduced ant biomass, and birds in reduction plots lost body mass at the same time as richer food sources were available, at least in the form of nearby supplementation plots, if not in control plots. We also found that floater frequency did not differ among years, despite large variation in the severity of annual drought and food supply. Therefore, we suspect that some mechanism or constraint caused sedentary birds to maintain their previously established home ranges. We hypothesize several nonexclusive alternative explanations for such a constraint: 1) Predation pressure favors restricted movement within the confines of a familiar home range. We discount this possibility as a proximate mechanism in this system simply because of the essential absence of Ovenbird predators, at least those causing much risk, in Jamaica. However, if these behaviors are largely genetically controlled and winter populations are mixing during the breeding season, predation could be the ultimate cause. 2) Social interaction with neighbors may incur a cost to home range expansion or shifting. Although vocalizations, chases, and agonistic displays were not frequently observed during this study, such interactions are common during the fall arrival period for Ovenbirds (Strong AM, personal communication) and for other species (Brown et al. 2000; Marra 2000). Thus, home range establishment and habitat sorting likely occur early in the winter, and the territory neighborhood is stable during the mid- and late-winter periods. 3) Space-use behavior may be a genetically polymorphic behavior (Winker et al. 1990) or constrained by the expression of other genetically correlated behaviors (Sih et al. 2004).
On food-supplementation plots, Ovenbirds with established home ranges foraged regularly in close proximity to feeding stations. In fact, many Ovenbirds found feeding stations within minutes of our initiating supplementation treatments. Feeding stations also attracted floaters. One individual floater made round-trip excursions between distant supplementation plots. Its floating behavior likely facilitated discovery of the supplementation plots that it frequented. Mark–recapture data also revealed that floaters found and used supplementation plots. Whereas before manipulation, the density of Ovenbirds did not differ between treatments, food supplementation increased bird density, and food reduction decreased density. The decreased density in reduction plots cannot be attributed to abandonment by sedentary birds because pre-to-post manipulation persistence rates did not differ between treatments, and we were not able to induce floating. Thus, changes in density must largely be driven by movements of floaters, as they tended to abandon low-food areas and then located and used high-food areas. Our findings support the contention of Johnson and Sherry (2001) that the apparent match between food supply and bird distribution is driven by movements of a subset of the population.
Because mid-season food supply did not control home range size in sedentary birds or causes shifts in space-use strategy, we suspect that food supply during the Fall arrival period is relatively important in determining patterns of home range selection. After home ranges are established in the Fall, it appears that the sedentary class of birds is unlikely to move beyond their familiar home range. In contrast, floaters appear to be better able to find patches of high-quality food throughout the winter, perhaps making the floater space-use strategy advantageous when landscape-scale resources are low and heterogeneous.
Causes and consequences of floater behavior
We have previously demonstrated that body condition of sedentary individual Ovenbirds is tightly linked to climate through its effect on food availability (Brown and Sherry 2006). Sedentary birds given supplemental food maintained or improved body condition, whereas birds in reduction and control plots experienced deteriorating physical condition through the dry season. Here we report that the physical body condition response to variation in food availability differed for floaters. Although the sample size of floaters is low and we do not know the specific space-use behaviors of many of these individuals, we found that body condition was affected by a strong 3-way interaction among time period, food treatment, and behavioral class of space use (Figure 4). Whereas floaters captured posttreatment in food reduction plots had relatively high mass, protein, and fat reserves compared with premanipulation, those in food-supplementation plots had lower mass and fat stores, suggesting that they were in relatively poor condition. These patterns conflict with our prediction that physical condition of floaters would not be affected by food manipulation. We reasoned that floaters could adjust their space use to match food supply. Instead, our results for floaters sampled after food reduction suggest that they may be better than sedentary individuals at finding food in an environment of unpredictable and heterogeneous resources. Floaters that moved into supplementation plots likely encountered increased competition from previously established, dominant territory holders for access to a limited number of high-quality feeding sites. In such situations, subordinate birds are known to experience elevated corticosterone, a stress hormone that causes protein catabolism and degraded physical condition when chronically elevated above baseline levels (Marra and Holberton 1998). Regardless, it remains difficult to understand how a floating strategy led to increases in mass, fat, and protein stores, when improved food searching should at best help floaters to maintain body condition in a naturally dry, low-food environment.
Increasing evidence supports the contention that wintering birds, including Ovenbirds, maintain relatively high fat reserves in unpredictable habitats (Strong and Sherry 2000), as predicted by the adaptive body mass hypothesis (reviewed by Rogers 2005). It might be argued that this hypothesis could be invoked as an alternate explanation for the response of floaters in this study because their body mass, including lipid deposits, increased, but only in the food reduction plots (Figure 4B,D). Yet, in extreme food shortages, even sedentary Ovenbirds invariably lose mass, fat, and muscle (protein) stores (Brown and Sherry 2006). This latter observation indicates that overwintering migrant birds have complex behavioral and physiological responses to resource variation, and we recognize that the mechanisms underlying patterns of body condition in this study are not well understood.
The literature generally assumes that floater behavior corresponds with subordinate social status. It is further assumed, because of their increased mobility and relative unfamiliarity with their area of use, that floaters are more vulnerable to predators and to starvation (Winker 1998). Thus, floaters should have reduced fitness overall relative to sedentary individuals. However, perhaps because of the difficulties in studying floaters, only limited evidence exists to suggest that floaters actually experience lower survival in the winter (Rappole et al. 1989; Winker et al. 1990). Our results suggest the possibility that floaters may not be socially subordinate. Four lines of evidence are consistent with this hypothesis that at least some individual Ovenbirds displaying nonsedentary strategies are not simply subordinates wandering the landscape in search of vacancies, that is, making the best of a bad situation: First, adults and males were floaters in equal proportion to immatures and females, respectively. Given that males and adults are typically dominant demographic groups, the assumption that floating is a socially subordinate strategy predicts a higher proportion of immature and female floaters, a prediction not supported by our results. Second, the behavioral class of floaters included a variety of space-use behaviors that frequently differ from the traditional view of individuals wandering the landscape in search of territory vacancies. Third, several birds displayed floater behavior in multiple years. Fourth, floaters maintained higher physical body condition in situations of low resource availability. Taken together, these observations lead to new, testable hypotheses about winter alternative behaviors: 1) at least some floaters are not social subordinates and 2) floating may offer situational advantages.
The present study is one of the first detailed descriptions of alternative space-use strategies by a wintering migratory bird, providing novel descriptions of space-use behavior related to feeding home ranges, food abundance, and floating. Our study also suggests that many individuals are constrained in their movement response to variation in food resources and thus during periods of low food supply face increased risk relative to individuals with alternative, often more flexible, space-use behaviors such as floating. Based on these results, we hypothesize that an individual's winter space-use strategy is largely determined during the life-cycle transitions after fall arrival, and thus, an increased research focus on this period is warranted.
|
FUNDING |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
National Science Foundation (DEB-008941, DEB-0408117).
|
ACKNOWLEDGEMENTS |
|---|
We thank Lee Dyer and 2 anonymous reviewers for helpful comments on the manuscript. We are extremely grateful to A. Anderson, L. Duda, A. Flowers, L. Forester, G. Levandoski, and C. Studds for assistance in the field. We thank M. Brown for laboratory assistance and M. White for use of her Southeastern Louisiana University laboratory for bird sexing. Bird banding, radio tagging, tissue collection, and tissue transport were permitted by the US Geological Survey, the Jamaican National Environment and Planning Agency Jamaica, and approved by Tulane University's Institutional Animal Care and Use Committee.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISUCUSSIONFUNDINGREFERENCES |
|---|
Bates J. Winter territorial behavior of Gray Vireos. Wilson Bull (1992) 104:425–433.
Boutin S. Food supplementation experiments with terrestrial vertebrates: patterns, problems, and the future. Can J Zool (1990) 68:203–220.[CrossRef]
Brown DR. Food supply and the dry season ecology of a tropical resident bird community and an over-wintering migrant bird species. (2006) [New Orleans (LA)]: Tulane University. 186. [dissertation].
Brown DR, Long JA. Experimental fruit removal does not affect territory structure of wintering hermit thrushes. J Field Ornithol (2006) 77:404–408.[CrossRef]
Brown DR, Long JA. What is a winter floater? causes, consequences, and implications for habitat selection. Condor (2007) 109:548–565.
Brown DR, Sherry TW. Food supply controls the body condition of a migrant bird wintering in the tropics. Oecologia (2006) 149:22–32.[CrossRef][Web of Science][Medline]
Brown DR, Sherry TW. Solitary winter roosting of ovenbird in core foraging area. Wilson J Ornithol (2008) 120:455–459.[CrossRef]
Brown DR, Stouffer PC, Strong CM. Movement and territoriality of wintering hermit thrushes in southeastern Louisiana. Wilson Bull (2000) 112:347–353.[CrossRef]
Cuadrado M. Why are migrant robins (Erithacus rubecula) territorial in winter?: the importance of the anti-predatory behavior. Ethol Ecol Evol (1997) 9:77–88.
Dingemanse N, Both C, Drent P, Tinbergen J. Fitness consequences of avian personalities in a fluctuating environment. Proc R Soc Lond B Biol Sci (2004) 271:847–852.[Medline]
Greenberg R, Salewski V. Ecological correlates of wintering social systems in new world and old world migratory passerines. In: Birds of two worlds: the ecology and evolution of migration—Greenberg R, Marra PP, eds. (2005) Baltimore (MD): Johns Hopkins University Press. 336–358.
Griffiths R, Daan S, Dijkstra C. Sex identification in birds using two CHD genes. Proc R Soc Lond B Biol Sci (1996) 263:1251–1256.[Medline]
Holmes RT, Sherry TW, Reitsma L. Population structure, territoriality and overwinter survival of two migrant warbler species in Jamaica. Condor (1989) 91:545–561.
Hooge P, Eichenlaub B. Animal movement extension 2.0. (1997) Anchorage (AK): Alaska Science Center—Biological Science Office, US Geological Survey.
Johnson MD, Sherry TW. Effects of food availability on the distribution of migratory warblers among habitats in Jamaica. J Anim Ecol (2001) 70:546–560.[CrossRef]
Johnson MD, Strong AM. Length-weight relationships of Jamaican arthropods. Entomol News (2000) 111:270–281.[Web of Science]
Koronkiewicz TJ, Sogge MK, Van Riper C, Paxton EH. Territoriality, site fidelity, and survivorship of willow flycatchers wintering in Costa Rica. Condor (2006) 108:558–570.
Levings SC. Seasonal, annual, and among-site variation in the ground ant community of a deciduous tropical forest: some causes of patchy species distributions. Ecol Monogr (1983) 53:435–455.[CrossRef][Web of Science]
LOAS [software]. Urnasch (Switzerland): Ecological Software Solutions. (2002).
Lynch JF, Morton ES, van Der Voort ME. Habitat segregation between the sexes of wintering hooded warblers (Wilsonia citrina). Auk (1985) 102:714–721.[Web of Science]
Marra PP. The role of behavioral dominance in structuring patterns of habitat occupancy in a migrant bird during the nonbreeding season. Behav Ecol (2000) 11:299–308.
Marra PP, Holberton RL. Corticosterone levels as indicators of habitat quality: effects of habitat segregation in a migratory bird during the nonbreeding season. Oecologia (1998) 116:284–292.[CrossRef][Web of Science]
Maybe SE, Morton ES. Demography and territorial behavior of wintering kentucky warblers in Panama. In: Ecology and conservation of neotropical migrant landbirds—Hagan JM, Johnston DW, eds. (1992) Washington (DC): Smithsonian Institution Press. 329–336.
McDonald MV. Kentucky warbler (Oporornis formosus). In: The birds of North America, No. 324—Poole A, Gill F, eds. (1998) Philadelphia: The Academy of Natural Sciences. 1–20.
Millspaugh JJ, Marzluff J. Radio tracking and animal populations (2001) San Diego (CA): Academic Press.
Morton ES. Habitat segregation by sex in the hooded warbler: experiments on proximate causation and discussion of its evolution. Am Nat (1990) 135:319–333.[CrossRef][Web of Science]
Newton I. Population limitation in birds (1998) New York: Academic Press.
Ogden LJ, Stutchbury BJ. Hooded Warbler (Wilsonia citrina). In: The birds of North America, No. 110—Poole A, Gill F, eds. (1994) Philadelphia: The Academy of Natural Sciences. 1–20.
Pyle P. Identification guide to North American birds (1997) Bolinas (CA): Slate Creek Press.
Rappole J, Warner D. Ecological aspects of migrant bird behavior in Veracruz, Mexico. In: Migrant birds in the neotropics: ecology, behavior, distribution, and conservation—Keast A, Morton E, eds. (1980) Washington: Smithsonian Institution Press. 353–393.
Rappole JH, Ramos MA, Winker K. Wintering wood thrush movements and mortality in southern Veracruz. Auk (1989) 106:402–410.[Web of Science]
Rogers CM. Food limitation among wintering birds: a view from adaptive body mass models. In: Birds of two worlds: the ecology and evolution of migration—Greenberg R, Marra PP, eds. (2005) Baltimore (MD): John's Hopkin's University Press. 106–113.
Roth RR, Johnson MS, Underwood TJ. Wood Thrush (Hylocichla mustelina). In: The birds of North America, No. 246—Poole A, Gill F, eds. (1996) Philadelphia: The Academy of Natural Sciences. 1–28.
Runge MC, Marra PP. Modeling seasonal interactions in the population dynamics of migratory birds. In: Birds of two worlds: the ecology and evolution of migration—Greenberg R, Marra PP, eds. (2005) Baltimore (MD): John's Hopkin's University Press. 375–389.
Saltz D, White GC. Reporting error measures in radio location by triangulation: a review. J Wildl Manage (1989) 58:181–184.
Schwartz P. The northern waterthrush in Venezuela. Living Bird (1964) 3:169–184.
Sherry TW, Johnson MD, Strong A. Does winter food limit populations of migratory birds? In: Birds of two worlds: the ecology and evolution of migration—Greenberg R, Marra PP, eds. (2005) Baltimore (MD): Johns Hopkins University Press. 414–425.
Shuster S, Wade MJ. Mating systems and strategies (2003) Princeton: Princeton University Press.
Sih A, Bell A, Johnson J. Behavioral syndromes: an ecological and evolutionary overview. Trends Ecol Evol (2004) 19:372–378.[CrossRef][Medline]
Smith JNM, Arcese P. How fit are floaters? consequences of alternative territorial behaviors in a nonmigratory sparrow. Am Nat (1989) 133:830–845.[CrossRef][Web of Science]
Smith SM. The "underworld" in a territorial sparrow: adaptive strategy for floaters. Am Nat (1978) 112:571–582.[CrossRef][Web of Science]
Sogge MK, Koronkiewicz TJ, Van Riper C 3rd, Durst SL. Willow flycatcher nonbreeding territory defense behavior in Costa Rica. Condor (2007) 109:475–480.
Solomon N, Jacquot J. Characteristics of resident and wandering prairie voles Microus ochrogaster. Can J Zool (2002) 80:951–955.[CrossRef]
Staicer C. Social behavior of the Northern Parula, Cape May Warbler, and Prairie Warbler wintering in second-growth forest in southwestern Puerto Rico. In: Ecology and conservation of neotropical migrant landbirds—Hagan JM, Johnston DW, eds. (1992) Washington (DC): Smithsonian Institution Press. 308–320.
Stapley J, Keogh J. Exploratory and antipredator behaviors differ between territorial and nonterritorial male lizards. Anim Behav (2004) 68:841–846.[CrossRef][Web of Science]
Strong AM. Effects of food abundance on non-breeding habitat quality for two species of ground-foraging neotropical migrant warblers (1999) New Orleans: Tulane University. 200. PhD dissertation.
Strong AM. Divergent foraging strategies of two neotropical migrant warblers: implications for winter habitat use. Auk (2000) 117:381–392.[CrossRef][Web of Science]
Strong AM, Sherry TW. Habitat-specific effects of food abundance on the condition of ovenbirds wintering in Jamaica. J Anim Ecol (2000) 69:883–895.
Stutchbury BJ. Competition for winter territories in a neotropical migrant: the role of age, sex, and color. Auk (1994) 111:63–69.[Web of Science]
Van Horn MA, Donovan TM. Ovenbird (Seiurus aurocapilla). In: The birds of North America, No. 88—Poole A, Gill F, eds. (1994) Philadelphia: The Academy of Natural Sciences. 1–24.
White GC, Garrott RA. Analysis of wildlife radio-tracking data (1990) San Diego (CA): Academic Press.
Wilkinson L. Systat 10.0 for Windows [software] (2000) Chicago: SPSS.
Winker K. The concept of floater. Ornitol Neotrop (1998) 9:111–119.
Winker K, Rappole JH, Ramos MA. Population dynamics of the wood thrush in southern Veracruz, Mexico. Condor (1990) 92:444–460.
Wunderle JM. Population characteristics of black-throated blue warblers wintering in three sites on Puerto Rico. Auk (1995) 112:931–946.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||