Motivations for breeding adults to guard nests change throughout the nesting season. Investments in nest guarding are expected to reflect fitness tradeoffs between risks to territory, eggs, or brood and other duties required for successful brood-rearing (e.g., Markman et al. 1995). Guarding in the pre-laying stage may be intended to deter extra-pair copulation and ensure paternity (Korpimäki et al. 1996, Mougeot 2000, Mougeot et al. 2002). In territorial species, guarding prior to egg-laying may also advertise that the territory is occupied and help to ward off competition (Nice 1941, Mougeot 2000, Margalida and Bertran 2005). In species subject to brood parasitism, nest guarding during the incubation stage prevents parasitic species from laying eggs in the nest (Møller 1987, Gowaty et al. 1989). Guarding also deters potential predators of eggs or chicks (Slack 1976, Woodard and Murphy 1999). In the event of a predation attempt, guarding adults are better prepared to defend the nest by chasing or attacking intruders (Turrin and Watts 2014). Nest guarding likely fulfills multiple purposes simultaneously.

Nest sites are a limiting resource for Bald Eagles (Haliaeetus leucocephalus) in the Chesapeake Bay area in the eastern U.S.A. (Watts et al. 2008). As the population approaches regional carrying capacity, the number of floaters is growing disproportionally and competition for nesting sites is increasing. An analysis of annual recruitment rates of new breeders showed that nearly 100% of reproductively mature eagles were assimilated into the breeding portion of the population in the early 1990s. By 2013, that proportion had fallen to 17%, indicating that most young adult birds were excluded from breeding locally (Turrin 2013). These birds may disperse out of the Chesapeake Bay or may become floaters until they attain a breeding opportunity.

As a surplus of adult birds competes for a limited number of nesting territories, territorial interactions may become increasingly frequent (Newton 1979, Hunt 1998, López-Sepulcre and Kokko 2005). Adult mortality resulting from intraspecific conflict can be important in regulating raptor populations near carrying capacity (Newton 1979, 1998) and has been documented in adult Bald Eagles (Jenkins and Jackman 1993, Elliott et al. 2011). A recent investigation of territorial interactions in Bald Eagles in the Chesapeake Bay found that intraspecific intrusions around active nests are common during the nesting season (Turrin and Watts 2014). Though documented accounts are rare, there is anecdotal evidence in Bald Eagles of conspecific intruders killing nestlings (Markham and Watts 2007). More frequently, intrusion indirectly affects reproductive success by creating additional stresses on breeders (Kauffman et al. 2004, Bretagnolle et al. 2008). These stresses include the allocation of time and energy to chasing off intruders (Sunde and Bølstad 2004); the constraint of the territory to a more defensible size (Norton et al. 1982, Mougeot et al. 2003, Ridley et al. 2004); conflicts resulting in the wounding or killing of a breeding adult, which leaves the mate to care for the young alone (Newton 1979); and may include the allocation of more time to nest guarding at the expense of other parental care activities.

Nest-guarding behavior in Bald Eagles is poorly described. Our objective was to examine parental attendance and nest guarding by breeding Bald Eagles in the lower Chesapeake Bay. We attempted to determine whether nest-guarding coverage changed over the course of the nesting period and whether male and female breeders contributed equally to nest attendance and guarding effort. We also described the types of locations that breeders chose when nest guarding. We predicted that (1) guarding coverage would change throughout the stages of the nesting season. We expect adults to adjust nest guarding in response to shifts in time demands of the reproductive stages and changes in the risks associated with intrusions in different stages of the nesting season. We hypothesized that (2) there would be a sex bias in total nest attendance and (3) in nest-guarding effort during the nestling period. Sex biases in parental roles have been demonstrated for a variety of parental care behaviors in Bald Eagles (e.g., Gerrard and Bortolotti 1988, Cain 1998, Bryan et al. 2005). Because female breeders' time budgets are dominated by incubation and brooding, particularly in the early post-hatching period, we expected females to spend more time at the nest overall but to have less time available to dedicate specifically to guarding. Though males are responsible for the majority of provisioning, males' time budgets seem to have more room for flexibility; thus, we expected males to guard nests more than females. Finally, we hypothesized that (4) breeding adults would use some types of nest-guarding locations more often than others.

Methods

Study Sites.

We studied Bald Eagles at active nest sites in Virginia along the James River (n  =  21), Pamunkey River (n  =  2), Pagan River (n  =  2), Lynnhaven River (n  =  2), Nansemond River (n  =  2), Elizabeth River (n  =  1), and the Southwest Branch Back River (n  =  1; Fig. 1). We chose nests located along the salinity gradient characteristic of the major tributaries of the Chesapeake Bay in order to examine a sample that was representative of the population. Nests were selected based on accessibility and visibility of the site, our observation of breeding activity such as nest maintenance and presence of the breeding pair early in the breeding season, permission to access the property, and proximity to another suitable study nest to allow for two observation sessions to be conducted in succession each day.

Figure 1.

Locations of study nests in the lower Chesapeake Bay where nest-guarding behavior of breeding Bald Eagles was observed (2012–2013).

Results

During the 2012 and 2013 nesting seasons, we monitored 31 Bald Eagle nests located along tributaries of the lower Chesapeake Bay. Total observation time was 540.8 hr. In 2012, nests were observed for 4.6 ± 0.7 hr in the pre-laying period, 10.4 ± 0.7 hr during incubation, and 10.7 ± 0.6 hr in the nestling period, on average. In 2013, observation time averaged 12.7 ± 0.4 hr per nest during the nestling period.

Nest-guarding effort varied with nesting stage. On average, nests were guarded during 13.7 ± 4.2% of the pre-laying period, 6.8 ± 2.2% of the incubation stage, and 21.9 ± 5.2% of the nestling period during the 2012 breeding season (one-way ANOVA: F_2,31  =  3.611, P  =  0.039). Pairwise post-hoc comparisons indicated that guarding coverage was significantly higher in the nestling period than during incubation (P  =  0.030). All other pairwise tests were nonsignificant (all P > 0.364). There was no effect of observation effort within each nesting stage on observed guarding coverage (r_s  =  0.060, n  =  34, P  =  0.736). The study-wide (2012 and 2013) average nest-guarding coverage during the nestling period was 26.3 ± 3.2%.

During the 2013 nestling period, females attended the nest area more often (t  =  −5.344, P < 0.001, df  =  30.9; Fig. 2) and for longer average bouts of time than males (t  =  −6.271, P < 0.001, df  =  25.2). Observations for which breeder sex was unknown made up 0.8% of total attendance events, and a sensitivity analysis indicated that these observations had no effect on the results (all t > −5.111, P < 0.001). We found no effect of differences in observation time across nests on average attendance bout length of male (r_s  =  −0.081, n  =  19, P  =  0.741) and female (r_s  =  0.034, n  =  19, P  =  0.889) breeders at individual study nests. The proportion of time that females attended nests was positively correlated with observation time (r_s  =  0.562, n  =  19, P  =  0.012), while male attendance was not (r_s  =  0.353, n  =  19, P  =  0.137). On average, nests were left unattended during 5.0 ± 1.4% of observation time in 2013. Two nests were never observed unattended. At other study nests, the average bout length of nonattendance was 6.9 ± 1.2 min (n  =  17).

Figure 2.

Mean nest attendance ±SE by male (M) and female (F) breeding Bald Eagles at 19 nests in the lower Chesapeake Bay during the first 3 wk of the nestling period (2013). Nest attendance was defined as the proportion of time a breeding adult was present within 500 m of the nest.

Males guarded nests more than females during the nestling stage (t  =  4.281, P < 0.001, df  =  35.8; Fig. 3A). The difference between male and female guarding bout lengths was not significant (t  =  1.733, P  =  0.092, df  =  36.0; Fig. 3B). There was no effect of observation effort on the proportion of time nests were guarded by male and female breeders (r_s  =  0.172, n  =  19, P  =  0.481; r_s  =  0.268, n  =  19, P  =  0.268, respectively). There was no correlation between observation effort and average male guarding bout length (r_s  =  0.313, n  =  19, P  =  0.192), but female guarding bout length was positively correlated with observation time (r_s  =  0.568, n  =  19, P  =  0.011).

Figure 3.

Total coverage (A) and average bout length (B) ±SE of nest-guarding behavior exhibited by male (M) and female (F) Bald Eagles at 19 nests in the lower Chesapeake Bay during the first 3 wk post-hatch (2013). In this study, a breeding adult was considered to be nest guarding when it was within 500 m of the nest and not engaged in brooding, incubation, or feeding, and the mate was also present in the nest area, regardless of the mate's activity.

In 2013, adults guarded broods most often from perches within 25 m of the nest but not in the nest itself (one-way ANOVA: F_6,126  =  7.676, P < 0.001; Fig. 4). Pairwise tests showed that guarding in the 0–25 m distance category was significantly higher than in all other distance categories (all P < 0.004) except for in the nest, though that relationship approached significance (P  =  0.052). All other pairwise comparisons were nonsignificant (all P > 0.064). Adults engaged in nest-guarding behavior guarded most often from an adjacent tree (43.6 ± 5.7% of guarding events; one-way ANOVA: F_3,72  =  9.054, P < 0.001). Of total guarding events, 24.2 ± 5.0% were conducted from a perch in the nest tree but not in the nest itself. Nest guarding from a position in the nest and aerial guarding while circling or soaring over the nest site made up 16.3 ± 2.5% and 15.9 ± 3.3% of guarding events, respectively. Pairwise comparisons indicated significantly more guarding from perches in adjacent trees than all other locations (all P < 0.011). No other pairwise comparisons were significant (all P > 0.530). There was no correlation between observation time and total guarding events observed per nest (r_s  =  −0.070, n  =  19, P  =  0.776).

Figure 4.

Distance to the nest during nest-guarding bouts of 19 Bald Eagle pairs nesting in the lower Chesapeake Bay (2013). Proportion of guarding events per breeding pair in each distance category was averaged (±SE) across study nests. A breeding adult was considered to be nest guarding when perched within 500 m of the nest and not engaged in brooding, incubation, or feeding, and the mate was also present in the nest area, regardless of the mate's activity. Distances presented refer to the distance between the nest-guarding perch and the nest structure.

Discussion

Nest-guarding coverage changed over the nesting season, supporting our first hypothesis. In 2012, nests were guarded in the nestling period nearly twice as often as in the pre-laying period and three times more often than during the incubation period. However, the difference between guarding coverage during the pre-laying and nestling stages was not significant. Our remaining hypotheses were supported in that there were sex biases in (2) nest attendance and (3) nest-guarding frequency, and (4) breeders used some nest-guarding locations more frequently than others. Although nest attendance was higher in female breeders, males guarded nests more often than females. Adults typically guarded within 25 m of the nest from a perch in an adjacent tree or the nest tree, excluding in the nest itself, possibly to maximize response time and visibility for effective surveillance and nest defense.

Differences in observation effort among nests did not have significant effects on the results, with two possible exceptions. Female attendance coverage and nest-guarding bout length showed moderate positive correlations with observation effort (r_s  =  0.562 and 0.568, respectively). Despite the effect of observation time on female attendance coverage, our general finding that females attended nests more than males during the nestling period was consistent with observations of eagles within other populations. Bryan et al. (2005) observed a South Carolina Bald Eagle nest from hatching through fledging in 1997 and 1999, reporting attendance of 31.6–47.3% and 8.8–21.1% by females and males, respectively. Cain (1998) found attendance of approximately 60–70% and 20–30% by female and male breeders, respectively, during the first 3 wk of the nestling period at three nests in Alaska. A possible cause of the positive correlation between female guarding bout length and observation time is an observation boundary problem, as some birds were already guarding when a session began or were still guarding when the session ended. In this case, we expect that female bout length was underestimated at some nests. Interestingly, male behavioral bout lengths showed no effect of differences in observation time among nests. Taking into account the guarding bout lengths of males and females (16.5 ± 1.7 min and 12.4 ± 1.6 min, respectively) and the moderate positive correlation between female bout length and observation time, we expect that the average female bout length is greater than our reported estimate, but we still expect no difference between the averages for males and females. Because there was no effect of observation time on the number of guarding events observed per nest, we do not expect that analyses of nest-guarding locations were affected by variation in observation time. No other study parameters were correlated with observation effort. We conclude that differences in observation time among nests did not influence the results concerning shifts in guarding effort during the nesting season, sex biases in total guarding effort, or variation in guarding distances and locations used.

We defined nest guarding as the presence of a member of the breeding pair within 500 m of the nest while its mate was also present, excluding individuals engaged in brooding, incubation, or feeding. Though this definition does not discriminate between guarding and loafing, the results support our assumption that the behaviors included under our definition were guarding and not incidental loafing near the nest. We would expect breeding pairs to loaf most often in the pre-laying and incubation periods, as there are relatively few demands on their time. In the pre-laying period, males may guard more intensively to prevent EPC, but females have more flexibility than they do in any other stage. In the incubation period, one member of the pair must be in attendance almost constantly to incubate the eggs, but the demands on the non-incubating adult are only to feed and possibly to guard the nest. During the nestling period, the need to guard the nest is still present, the consistent demand of incubating is replaced by that of brooding, and a new demand on the pair is introduced, as provisioning becomes essential to nestling-rearing. Despite the relatively higher demands on the pair during the nestling period, nest-guarding coverage was nearly double that of the pre-laying period and triple that of the incubation period, although only the latter comparison was statistically significantly different. This suggests that loafing did not constitute a large portion of the behaviors we defined as nest guarding. Furthermore, loafing need not occur within close proximity to the nest; thus, if adults are idling near the nest rather than near a food source or some other location, their choice of loafing site may indicate a motive of nest supervision. An individual loafing near the nest may switch to guarding upon the appearance of an intruder. For many reasons, nest guarding and loafing are difficult to distinguish, and thus our inclusive definition was used.

Because the motivation for nest guarding differs in each stage of the reproductive period, our results may offer insight into the environmental pressures that force pairs to nest guard. Guarding in the pre-laying period is often intended to guard the mate to deter extra-pair copulation (EPC; Korpimäki et al. 1996, Mougeot et al. 2002) or to ward off potential attempts to usurp the territory (Nice 1941, Mougeot 2000). Bald Eagles form persistent pair bonds and are considered monogamous (Stalmaster 1987, Jenkins and Jackman 1993); however, in some species social monogamy masks relatively high rates of extra-pair fertilization. Mougeot (2004) examined copulation behavior in 19 socially monogamous raptor species and found that EPC occurred in 68% and that incidence of EPC increased with breeding density. Although there are no data on EPC prevalence in our study population, the prevalence of floaters suggests a high risk of both EPC and attempts to usurp the territory early in the breeding season when non-territory holders compete for breeding opportunities. Breeders guarded nests during 13.7 ± 4.2% of pre-laying observations. Studies in other raptor species indicate that males spend 40% of the pre-laying period mate guarding or attending the territory, presumably to guard mates from EPC attempts (Korpimäki et al. 1996, Mougeot et al. 2002). These studies defined mate guarding as the attendance of the male in the vicinity of the nest with his mate (Korpimäki et al. 1996) and territory attendance as the percentage of time spent within the breeding territory, either at the nest or perched within 300 m of the nest (Mougeot et al. 2002). The relatively low rate of guarding at our study nests suggests that guarding by Bald Eagles in the pre-laying period may not be motivated by EPC risk. However, there may be considerable variation in pre-laying guarding effort due to differences in local breeding density, site characteristics, location relative to flyways and roosts, breeder experience, and other factors. In addition, our definition of nest guarding was fairly conservative and may not be directly comparable to other studies.

Once eggs are laid, new risks are introduced and antipredator behaviors become more important in ensuring nest success. Though some passerine species nest guard for 60% of the incubation period to deter nest parasitism and predators (Slack 1976, Woodard and Murphy 1999), nest-guarding coverage by Bald Eagles during incubation was 6.8 ± 2.2%. This low value may be an artifact of our relatively conservative definition of guarding; the aforementioned studies defined nest guarding to include male attendance in the nest area in the absence of the female (Slack 1976, Woodard and Murphy 1999), while our definition was restricted to guarding by an adult in the presence of the mate. It seems likely that an incubating female could simultaneously scan for predators or intruders and would be more able to do so than an adult engaged in feeding young, but her activity would not be considered guarding by our definition. During our observations, both adults were rarely in attendance at the same time during incubation. This is likely in part because eagles do not face the same nest parasitism threats as passerines. Furthermore, adult eagles have no true predators in this ecosystem. When the incubating adult is in attendance, their nests likely do not face the same magnitude of predation risk as passerine nests. In addition, adults are expected to invest time in guarding based on the potential for renesting if failure should occur and the probability of offspring survival (Montgomerie and Weatherhead 1988). Wood and Collopy (1993) found that 78% of pairs in northern Florida laid a second clutch within 29 d of experimental removal of the first clutch. The Chesapeake Bay population has a shorter potential breeding season than the Florida population as a result of the latitudinal gradient (Newton 1979). Reproductive surveys of the Chesapeake Bay population indicate that most pairs lay a replacement clutch when nests fail during incubation, but failures in the nestling stage generally occur too late in the season for pairs to attempt to renest (B.D. Watts and M.A. Byrd unpubl. data). Furthermore, the probability of offspring survival is higher for nestlings than for unhatched eggs, and nestlings are therefore a more valuable investment than eggs (Ricklefs 1973, Barash 1975). Guarding in the nestling period is critical to protecting altricial young from predators. In this population, nest cameras documented predation of nestlings up to 8 wk old (A.C. Markham and B.D. Watts unpubl. data). Finally, the observed peak in guarding may also relate to behavioral assessments of offspring and mates. Breeders may station themselves near the nest to gauge offspring feeding demands (e.g., Quillfeldt and Masello 2004, Hamer et al. 2006) or to indicate to the mate readiness to take over brooding (e.g., Debus et al. 2007). Though the relative importance of these motivators is unknown, nest guarding likely fulfills many roles simultaneously.

The timing of peak nest-guarding activity necessitates the presence of both breeders at the nest, with one adult brooding young and the other perched nearby to guard the nest. Past nest defense studies have focused on the nest-guarding role performed by males in the absence of females (Ricklefs 1969, Slack 1976, Greig-Smith 1980, Hayes and Robertson 1989, Martin 1992, Markman et al. 1995, Komdeur and Kats 1999) or on nest guarding by sentinels in cooperative breeding systems (Skutch 1935, McGowan and Woolfenden 1989, Hailman et al. 1994, Burton and Yasukawa 2001, Wright et al. 2001). Our observations indicate that in eagles the second adult, generally the male, plays an important role in nest defense by acting as a sentinel to guard against potential threats while the first adult broods young. This may suggest (1) that the female is vulnerable during brooding (Burton and Yasukawa 2001), (2) that both adults are essential for warding off intruders without leaving nestlings unprotected (Regelmann and Curio 1986), or (3) that both adults are at least twice as effective as one adult in deterring predators (Montgomerie and Weatherhead 1988). There is evidence supporting the second and third explanations. Simultaneous intrusions were observed during the study in multiple instances, with as many as four birds intruding at once (Turrin 2013). Breeders responded more often and more aggressively, usually with one breeder chasing or attacking the intruder, when both members of the breeding pair were present at the time of the intrusion event (Turrin and Watts 2014). The absence of the second adult at nests with older young suggests that reproductive tradeoffs shift for adults with older broods, possibly because older young are less vulnerable to potential predators.

As the Bald Eagle population in the Chesapeake Bay reaches saturation, breeders may reinforce nest-guarding efforts in response to the increasing risk of intraspecific intrusion. Time allocated to nest guarding during the nestling period may detract from the time available for other fundamental nestling-rearing behaviors. Males perform more provisioning than females, and our results indicate that they also nest guard more than females. If the quantity and quality of food provided to young is compromised by increased guarding effort, nestling rates of growth and mass gain may be negatively affected (Bortolotti 1989, Markham and Watts 2008). Alternatively, if adults do not adjust time budgets to increase nest-guarding coverage, pairs may face greater risk of nest failure resulting from an intrusion event. Infanticide and cannibalism by an intraspecific intruder has been documented in this population (Markham and Watts 2007).

Further research is needed before drawing any broad conclusions about the potential reproductive implications of increased guarding rates and the tradeoffs between different parental care strategies. Future studies should investigate the relationships between nest-guarding effort and reproductive success as well as guarding effort and provisioning rate, as males seem to be principally responsible for both of these behaviors. In addition, further research into the causes of nest failure and the relative effect of factors that contribute to nest failure, though challenging, would be a valuable addition to our understanding of Bald Eagle breeding ecology.