Printer Friendly

Wind and prey nest sites as foraging constraints on an avian predator, the glaucous gull.


Predation may strongly influence population dynamics, community structure, and prey behavior (Zaret 1980, Taylor 1984, Sih 1989). The influence of predators on their prey often changes with the environment. For example, predators may switch diet in response to changes in prey availability (Sinclair 1989, Newton 1993). Consequently, it is of interest to quantify how predators respond to changing foraging constraints, and how these dynamics influence prey populations (Werner et al. 1981, Cooper 1990, Mills and Shenk 1992, Suhonen et al. 1994).

The most appropriate foraging strategy predicted for a predator is often to maximize the trade-off between energy reward and risk of injury while foraging, rather than to maximize net energy gain alone (Stein 1977, Pettifor 1990). Hence, strategies that provide a low return in energy may be favored over energy-maximizing behavior if they involve a lower risk of injury for the predator (Helfman 1990). The appropriate foraging strategy may vary with environmental and behavioral changes that influence the risk of injury for the predator.

The trade-off between energy gain and risk of injury may be particularly acute among avian predators that forage within seabird colonies and face group defenses by prey. Risk of injury while foraging, and the ability of avian predators to avoid it, may be influenced by colony topography, changes in prey nesting densities, and weather conditions (Siegel-Causey and Hunt 1981, Spear 1993, Young 1994).

Glaucous Gulls, Larus hyperboreus, are common predators of eggs and chicks of the Thick-billed Murre, Uria lomvia, in the Arctic. Thick-billed Murres breed in dense colonies on exposed cliff ledges, and they defend themselves collectively against gull attack (Gaston and Nettleship 1981). Previous experimental studies have shown that Glaucous Gulls, because of their large size, find it difficult to attack narrow cliff ledges, and that they are constrained by murre defense when they attack exposed murre eggs (Gilchrist and Gaston 1997). In these experiments, windy conditions appeared to enhance the ability of gulls to reach exposed eggs on narrow ledges and avoid murre defenses.

In this study, we use multivariate statistical models to examine how Glaucous Gulls respond to changes in their foraging constraints. We also examine whether wind influences the choice of attack tactics by gulls, their ability to avoid contact with murres, and their foraging success.


Study area and species

This study was conducted at a Thick-billed Murre colony on the northeastern tip of Coats Island, Northwest Territories, Canada (62 [degrees] 30 [minutes] N, 83 [degrees] 00 [minutes] W) during 1990-1992. The colony consisted of [approximately]32 000 pairs of Thick-billed Murres (Gaston et al. 1993) and 18-25 pairs of Glaucous Gulls. Thick-billed Murres bred on a vertical cliff up to 65 m above the sea. Glaucous Gulls nested on ledges within the murre colony or occasionally on turf immediately above the murres. Glaucous Gulls were the primary predators of murres at this colony, as they are at many other colonies of the Thickbilled Murre (Gaston and Nettleship 1981), and they foraged almost exclusively ([greater than]85% of gull diet) on murre eggs and chicks after murres began to lay. Mean laying dates of murres were 26 June ([+ or -]3.2 d) and 25 June ([+ or -]4.8 d) in 1990 and 1991, respectively (deForest and Gaston 1996). Prior to the onset of laying by murres, gulls fed mainly on the carcasses of murre, ringed seal, Phoca hispida, and caribou, Rangifer tarundus, present on the sea ice below the colony.

Considerable variation existed in the types of cliff nesting-ledges and the density of breeding murres within the colony (deForest 1993). We recognized five murre nest site characteristics: (1) broad cliff ledge sites with high murre nesting densities; (2) broad ledge sites at low densities; (3) narrow ledge sites with high densities; (4) narrow ledge sites in low densities; and (5) crevice sites. We defined the characteristics of individual murre nest sites following Gaston and Nettleship (1981): (1) broad ledges could support two or more rows of breeding murres, whereas narrow ledges supported only a single row; (2) high-density sites had at least three neighbors on broad ledges and two neighbors on narrow ledges. Crevice sites were located under rock overhangs, in crevices, or within rock piles. The proportions of the types were determined by photographing study plots from the cliff at a distance of [approximately]20-30 m. We counted murres on 15 x 25 cm black and white prints. The number of breeding pairs was estimated by multiplying the total number of birds present on the cliff by a correction factor of 0.75 to account for birds absent from the colony or not included in the photographs (Nettleship 1976).

Behavioral observations

Gull foraging behavior was monitored from blinds on the cliff face that overlooked study plots. Study plots contained murre nest sites of all types (i.e., broad and narrow cliff ledges, as well as high and low nesting densities), interspersed at high densities. Study plots extended [approximately]40 m from blinds, and from the sea to the cliff top. Gulls appeared to be unaffected by our presence once we were inside blinds. Aerial search activity of gulls was monitored by recording the number of flight patrols made by gulls over the study plots during 30-min time intervals. Gulls flying close to the cliff face with their heads directed toward brooding/incubating murres were considered to be patrolling. This distinct head posture made foraging patrols easily distinguishable from other flight behaviors (e.g., travel to and from the nest).

The number of attacks made during each 30-min observation period was monitored. We considered that an attack had occurred if a gull made an aggressive advance toward brooding murres either on foot or from the air. For each gull attack, we recorded the attack behavior (aerial or on foot), nest site characteristics of the murres attacked (cliff ledge width and nesting density), murre defensive response (do nothing, flush from nest ledge to avoid gull, orient beak toward gull, lunge at gull, strike gull), and outcome (egg or chick taken or not). We also recorded whether gulls were struck by murres during the attack.

Attack and patrol rates were monitored in relation to date, time of day, wind conditions, visibility, and the number of murres present on the study plots. Murre numbers on study plots included both breeding and nonbreeding birds. Wind was monitored with a Digitar anemometer (Davis Instruments, Diablo Avenue, Hayword, California, USA) mounted 1.5 m from the cliff. Visibility was recorded as clear (no fog), light fog (40 m [less than] visibility [less than] 80 m), and dense fog (visibility [less than] 40 m). Because visibility, wind conditions, and murre attendance varied on a fine time scale, we chose 30 min as the duration of behavioral observations.

We examined the vulnerability of gulls to injury in two ways: first, we videotaped gull attacks; second, we used a model gull made of Styrofoam and plastered paper to mimic natural attacks. The model was introduced into groups of brooding murres from the edge of nesting groups, using an extendible 4 cm diameter pole. We recorded the location where murres struck the heads of live or model gulls, and whether defensive strikes were by the individual being attacked or by its immediate neighbors.

The influence of wind on aerial maneuverability was examined by timing the duration of patrols and hovering next to murre nest sites in relation to wind conditions on the cliff face, Hovering was defined as a gull trying to maintain its position next to a murre nest site during aerial attack. Patrols were timed with a stopwatch as birds passed through a 50 x 40 m area of cliff (termed study plot). Patrols in which gulls made attacks were not included in the analysis of patrol duration.

Statistical analysis

We used GLM, general linear modeling (Wilkinson 1989) to explore factors affecting aerial search activity, gull attack rates, and predation rates. We applied arcsine square-root transformations to rate data on search activity, attack frequency, and predation prior to analyses (Sokal and Rolf 1981). We used logistic regression analysis to examine factors influencing gull attack success, and used the likelihood ratio statistic to test the significance of the independent variable (Kleinbaum et al. 1988).

Some behavioral data were collected during consecutive 30-min observation periods. Time series analyses (Wilkinson 1989) revealed that aerial search activity was autocorrelated for up to 2.5 h. Therefore, we analyzed only one 30-min period extracted randomly from each 2.5-h period.

Within the murre colony, nest sites of all types (broad and narrow cliff ledges, as well as high- and low-density nests) were interspersed at high densities. Thus, gulls made their foraging decisions (e.g., which murre nests sites to attack) on fine temporal and spatial scales as they patrolled the cliff. This is in contrast to many predator-prey systems, in which foraging decisions are governed by the constraint that foraging habitats are distinct and mutually exclusive (Olive 1982, Caraco and Gillespie 1986). This is a subtle but important distinction, because it determines the appropriate statistical approach. For example, because gulls can select from a variety of nest sites occurring together within the colony habitat, our multivariate analyses examining factors that affect this selection (e.g., changes in wind and visibility) included all nest site characteristics. This approach is consistent with other studies that examine foraging decisions made within a single habitat (Spear 1993, Goss-Custard et al. 1995). The single exception to this approach was our analysis of murre behavior in response to gull attack. In this case, we analyzed broad and narrow ledges separately, because murres were constrained by the characteristics of their nest site, and ledge width influenced their response to attack.

Model selection

For cases in which several independent variables were examined together, we did not conduct full factorial analyses (James and McCulloch 1990). Instead, we included only a subset of possible interactions. These were selected based upon our previous knowledge of the system (Gilchrist and Gaston 1997) and the results of other studies (Birkhead et al. 1985, Spear 1993, Young 1994). We will explain how each test addressed specific hypotheses, and identify why specific variables and interactions were included in each test.

Search and attack activity. - Foraging activity of predators may vary diurnally or seasonally in response to temporal changes in prey (e.g., energy value, availability) or the energy demands of predators and their offspring (Bell 1991, Young 1994). Foraging activity may also vary with environmental conditions that influence the energy costs of search and/or the ability of predators to locate and subdue prey (Norberg 1977, Bell 1991). We examined this using two models in which patrol rate and attack rate were dependent variables. In each model, we included the following independent variables: date, time of day, wind speed, visibility, and murre attendance (number of murres present in the study plots). Previous studies of Glaucous Gulls showed that wind enhanced their aerial maneuverability and possibly decreased their energy demands while searching (Gilchrist and Gaston 1997). Consequently, we predicted that search and attack activity would be positively related to high winds. Given that Glaucous Gulls hunt by sight, foraging activity should be negatively influenced by poor visibility due to fog, and by high murre attendance, which could conceal eggs and chicks. In the model examining attack activity, we included search activity as an independent variable, because high search effort should increase foraging opportunities. We also included the interactions: (1) wind x search activity, which we predicted would be positively correlated; and (2) murre attendance x wind conditions, which should be negatively correlated (deForest and Gaston 1996).

Attack rates relative to nest site characteristics. - Because predators often respond to variation in habitat characteristics that determine the accessibility of prey, we predicted that gulls should attack nest sites selectively. We examined this in a model in which attack rate (number of attacks/30 min) was the dependent variable. We included the following independent variables: wind speed (kilometers per hour), ledge width (narrow or broad), and nesting density (high or low). We predicted that ledge width and/or nesting density would significantly influence attack frequency. We also included the following interactions: (1) ledge width x wind, because wind increased the ability of gulls to reach narrow ledges (Gilchrist and Gaston 1997); and (2) broad ledge width x nesting density, because broad ledges typically support more murres.

Predation rates and attack success. - The accessibility of prey to predators and the ability of prey to defend themselves may vary in relation to topography, changing environmental conditions, or over time (Johnson 1938, Young 1994), and this variation should influence the attack rate and the attack success of predators. We examined these issues in two models. In the first model, gull predation rate (number of eggs or chicks taken/30 min) was the dependent variable. We included the following independent variables: wind speed (kilometers per hour), search rate (number of patrols/30 min), attack rate on narrow ledges (number of attacks/30 min), attack rate on broad ledges, murre nesting density (high or low), and gull attack technique (on foot or flying). We predicted that ledge width and/or nesting density would significantly influence attack success. We also included the interaction terms wind x narrow ledge width and wind x attack technique, because wind increased the aerial maneuverability of gulls (Gilchrist and Gaston 1997). We also included ledge width x nesting density, because broad ledges typically support more murres, which could negatively influence attack success.

In the second model, we used logistic regression with attack success (number of eggs or chicks taken per attack) as the dependent variable. We included the following independent variables: wind speed, gull attack rate, attack technique (on foot or from the air), ledge width (broad or narrow), murre nesting density (high or low), murre attendance (number of murres present in the study plot), and murre response (do nothing, orient beak toward gull, strike gull). High wind speeds and attack rates were predicted to have positive influences on attack success because they increase aerial maneuverability and foraging opportunities (Gilchrist and Gaston 1997). Murre attendance and strong defensive response by murres were both predicted to have negative influences on gull attack success. We included attack technique x wind and attack technique x ledge width, because we predicted that attack technique would be influenced by both wind and ledge characteristics.

Murre response to attack. - Several factors may influence the ability of seabirds to defend themselves, including colony topography, weather conditions, and the attack technique of the predator. We examined this in a linear model in which murre response was the dependent variable (do nothing, flush, orient beak toward gull, lunge at gull, strike gull). We conducted separate analyses for broad ledges (BL model) and narrow ledges (NL model). In both models, we included the following independent variables: wind speed (kilometers per hour), gull attack technique (on foot or from the air), murre nesting density on a ledge (high or low), and murre attendance (number of murres present in study plots). We predicted that nesting density would positively influence defense in the BL model, because broad ledges support more birds and provide space to maneuver during attack; both factors may enhance group defense. We included technique x wind and technique x ledge as interactions, because attack technique should be influenced by wind and ledge characteristics.


Seabird colony structure and environment

There was considerable variation in the width of cliff ledges and the density of nesting murres within the Coats Island colony. All data are reported as mean [+ or -] I SE About half of the murres nested on broad ledges in high-density groups (47 [+ or -] 2.3%; n = 5); 22.0 [+ or -] 1.1% and 8.1 [+ or -] 0.78% of pairs nested on narrow ledges under high and low nesting densities, respectively. A further 17.7 [+ or -] 1.5% nested on broad ledges under low-density nesting conditions, including birds nesting alone or on the edge of large groups. A further 5 [+ or -] 1.2% of murres nested in crevices or caves, a figure which may be an underestimate because these sites were difficult to census.

Winds were frequently [greater than]40 km/h in 1990, but rarely exceeded 30 km/h in 1991 and 1992. Significantly stronger winds occurred in 1990 than in 1990-1991, but wind speeds in 1991 and 1992 were similar (Gilchrist 1995). Most of our behavioral observations were conducted under calm ([less than]10 km/h) or moderate (10-40 km/h) wind conditions [ILLUSTRATION FOR FIGURE 1 OMITTED]. Extended calm periods ([greater than]3 d at [less than]10 km/h) occurred each summer and were typically accompanied by warm temperatures (15 [degrees] -25 [degrees] C).

Gull search activity

Aerial search activity was significantly reduced under calm wind conditions in all years ([ILLUSTRATION FOR FIGURE 2 OMITTED], Table 1). Date had a significant influence on search activity in all years, and there was a statistically significant interaction between date and wind conditions in 1990 (Table 1). This reflected daily fluctuations in wind speed over the season.

Gull attack activity

In all years, attack rate was positively correlated with both aerial search activity and wind speed (Table 2a, b). Wind conditions and aerial search activity contributed significantly to the models when included independently or together. Changes in wind conditions between observation periods had a significant influence on search activity in the calm years of 1991 and 1992, but not in the generally windy year of 1990, indicating that gulls responded more strongly to increases in wind during calm years. Year, date, time of day, and visibility did not affect attack rate.

Gull attacks in relation to murre nest site characteristics

Rates of gull attacks were influenced by wind speed, nesting density, and ledge width (Table 3). Wind conditions had the strongest influence on rates of attack (Table 3). As wind increased, gulls attacked nest sites on narrow ledges more often, apparently because wind enhanced their ability to reach these sites in flight [ILLUSTRATION FOR FIGURE 3 OMITTED]. Consequently, there was a strong interaction between wind conditions and the width of cliff ledges attacked (Table 3). Cliff ledge width alone had only a weakly significant influence on attack rates (Table 3). Murre nesting density had a strong and negative effect on attack rates (Table 3). Thus, gulls selectively attacked areas of low murre nesting densities [ILLUSTRATION FOR FIGURE 3 OMITTED]. Wind did not interfere with the ability of gulls to attack broad ledges. The high rates of attack on narrow ledges under windy conditions indicate that gulls preferred to attack these sites even when all nest sites were accessible to them [ILLUSTRATION FOR FIGURE 3 OMITTED].

In summary, attack rates and the types of nest sites attacked by gulls were determined largely by moderate to high winds, which enabled gulls to reach sites of low nesting density, on narrow ledges, from the air.
TABLE 1. General linear models of factors affecting rates of aerial
search by gulls (no. patrols/30 min) in (a) 1990, (b) 1991, and (c)
1992. Murre attendance was not monitored in 1990.

Term                 cient     1 SE        t             P

Model a: [F.sub.5,165] = 44.7; 63.7% of variance explained

Date                 0.033     0.011     3.028    [less than] 0.003
Time                 0.000     0.000     0.814                0.417
Wind                 0.643     0.206     3.117                0.002
Wind change          0.017     0.026     0.643                0.521
Wind x date         -0.001     0.001    -2.533                0.020

Model b: [F.sub.5,395] = 28.0; 27.6% of variance explained

Date                -0.031     0.009    -3.571    [less than] 0.001
Time                -0.001     0.001    -2.774                0.006
Wind                 1.009     0.089    11.338    [less than] 0.001
Wind change         -0.041     0.020    -2.007                0.045
Murre attendance    -0.002     0.002    -1.022                0.308

Model c: [F.sub.5,638] = 30.1; 19.7% of variance explained

Date                -0.010     0.010    -1.019    [less than] 0.001
Time                -0.000     0.000    -2.490                0.013
Wind                 0.458     0.051     9.017    [less than] 0.001
Wind change          0.016     0.017     0.940                0.348
Murre attendance    -0.004     0.001    -3.030                0.003

Note: Terms with more than one component indicate interaction terms.
See Methods for transformations conducted prior to the analysis and
further rationale.

Gull predation rates

Year (1991-1992), date, time of day, and murre attendance had no detectable influence on predation rates. There were also no seasonal trends in daily predation rates. Instead, predation rates varied sharply over a few days, largely in response to changing wind conditions. Wind alone had a positive influence in 1990, but not in the generally calm years of 1991 and 1992 (Table 4a, b). Predation within the colony was determined by murre nest site characteristics and by wind speed. Nest sites on narrow ledges experienced higher rates of predation when winds were [greater than]10 km/h [ILLUSTRATION FOR FIGURE 4 OMITTED]. This was expected, because narrow ledges experienced higher rates of attack under windy conditions. Consequently, there was a significant interaction between narrow cliff ledge width and wind conditions in all years (Table 4a,b). Therefore, predation rates closely reflected the influence of wind conditions on the selection of murre nest sites that were attacked.
TABLE 2. General linear models of factors affecting rates of gull
attack activity (no. attacks/30 min) in (a) 1990 and (b) 1991-1992
combined. Murre attendance was not monitored in 1990.

Term                   cient     1 SE        t            P

Model a: [F.sub.4,165] = 100.7; 75.0% of variance explained

Wind                   0.178     0.034     5.303   [less than] 0.001
Wind change            0.001     0.010     0.030               0.976
Aerial search rate     0.363     0.037     9.791   [less than] 0.001

Model b: [F.sub.8,1033] = 80.7; 38.7% of variance explained

Wind                   0.103     0.050     2.076               0.038
Wind change            0.014     0.004     3.392               0.001
Murre attendance      -0.001     0.000    -2.695               0.007
Aerial search rate     0.196     0.019    10.203   [less than] 0.001

Note: Year (b), date, time, and interactions were not significant
and are not reported. Search and attack rates were
arcsine-transformed prior to analysis.

Gull attack success

In [approximately]21% of 2407 gull attacks, an egg or chick was taken. Attack success depended strongly on attack technique (Table 5). Attacks made on foot succeeded more often (308 of 671, or 46%) than aerial attacks (191 of 1736, or 11%). Attack success was negatively related to attack rate and wind speed (Table 5), because high attack rates occurred during windy conditions when the less successful aerial attacks predominated. Attack success was positively correlated with low murre nesting densities and with weak defensive responses by murres (Table 5). Attack success was correlated negatively with attack rates and did not increase with wind speed. Cliff ledge width did not influence attack success and, contrary to prediction, there was no detectable interaction between ledge width and murre defensive response. Fog and murre attendance also had no influence on attack success.

In summary, the high predation rates observed under windy conditions resulted from high attack rates and not from an increase in attack success.

Wind and gull maneuverability in flight

We examined the duration of search patrols and attack hovers under various wind conditions. Aerial searching by gulls consisted of two distinct components: the patrol (in which gulls flew into the wind along the cliff searching for prey), and the return (in which gulls circled back over the ocean, flying with the wind). Search patrol duration increased significantly with wind speed ([r.sup.2] = 0.61, P [less than] 0.001; log-transformed data) because stronger head winds reduced gliding speeds [ILLUSTRATION FOR FIGURE 5A OMITTED]. This probably increased the ability of gulls to locate exposed or poorly guarded murre eggs and chicks.

Hover duration during attacks also increased with wind speed ([r.sup.2] = 0.14, P [less than] 0.002; linear regression with data weighted by their variance, [ILLUSTRATION FOR FIGURE 5B OMITTED]). Wind conditions [greater than]10 km/h enabled gulls to maintain their position next to nesting ledges without flapping their wings (static glide).

Murre response to gull attack

Murres have the potential to kill adult-sized gulls. Several fully grown gull chicks were blinded and killed by murres when they fell onto murre nesting ledges. Adult gulls took flight from ledges when attacked, but several were observed with bloodied faces, backs, and legs following interactions with murres. Gulls attacked brooding murres headfirst, and this exposed their head and eyes to contact with defending murres. Video analysis [TABULAR DATA FOR TABLE 3 OMITTED] of gull attacks and experiments with gull models indicated that the greatest danger to gulls came from the neighbors of intended victims. This occurred because the murres being attacked typically struck the beak of the gull, whereas the neighbors to either side could strike the head and eyes of the attacker. This could explain why gulls avoided large groups of murres and preferentially attacked murres nesting alone or on the periphery of nesting groups.

Gulls that attacked on foot were struck three times more often (269 of 421 attacks, or 64%) than gulls that attacked from the air (193 of 921 attacks, or 21%). Murres on broad ledges were more likely to strike gulls than murres on narrow ledges because (1) they could turn to face attacking gulls without dislodging their eggs, and (2) they had time to respond to gulls walking on their nesting ledges. This is in contrast to the sudden attacks made by gulls on the wing, which often occurred during windy conditions. Consequently, wind had a negative influence on murre response to attack on broad ledges (Table 6a). Wind had no influence on murre response on narrow ledges (Table 6b), because narrow ledges were usually attacked under windy conditions [ILLUSTRATION FOR FIGURE 3 OMITTED]. Attack technique had a significant influence on murre response on both broad and narrow ledges. In summary, gulls that attacked narrow ledges from the air under windy conditions were least likely to be struck by defending murres.


Wind: a foraging constraint for Glaucous Gulls preying on Thick-billed Murres

Glaucous Gull foraging activity and predation rates were higher under windy conditions. During windy conditions, gulls selectively attacked narrow ledges, although broad ledges continued to be accessible to them. Typically, gulls patrolled the murre colony on the wing when wind conditions exceeded 10 km/h, and there are at least three possible explanations.

First, gulls flew into the wind when searching, and higher winds slowed their rate of search. Foraging patrols over nesting murres lasted twice as long under windy conditions as under calm conditions. This probably increased their ability to locate exposed or poorly guarded murre eggs, because slower search speeds typically increase the likelihood that a forager will detect concealed food items (Norberg 1977, Gendron and Staddon 1983, Bell 1991).

Second, wind also increased the ability of gulls to attack vulnerable eggs and chicks once they had been detected. Under low winds ([less than]10 km/h), gulls had difficulty hovering next to murre nest sites [ILLUSTRATION FOR FIGURE 5B OMITTED], and they typically had to circle out over the ocean to maintain [TABULAR DATA FOR TABLE 4 OMITTED] their flight speed before returning to make an attack. Under these circumstances, gulls often could not relocate the vulnerable sites after circling. In contrast, gulls could maintain their position next to murre nest sites during attack under windy conditions [ILLUSTRATION FOR FIGURE 5B OMITTED]. Indeed, we occasionally observed gulls gliding backwards under windy conditions to take eggs that they had initially passed over. Thus, windy conditions enabled gulls to make rapid and effective attacks almost immediately after prey were detected. This is important because, in most cases, eggs and chicks were only exposed to gulls briefly while an adult murre preened or an incubation changeover took place between brooding murres.

Third, and perhaps most important, windy conditions allowed gulls to reach nest sites that were poorly defended, but difficult to land on. Murre defense was weak on narrow ledges because most murres on narrow ledges faced toward the cliff and could not turn to face gulls without dislodging their eggs and chicks. Consequently, murres on narrow ledges could only defend themselves and their immediate neighbors. In addition, most gull attacks at narrow ledges were swift and from the air. In contrast, murres on broad ledges could turn to face gulls that were attacking on foot, and could act collectively to defend their neighbors (Birkhead 1977, Parish 1995). We conclude that windy conditions enabled gulls to overcome constraints imposed by both colony topography and prey defense, and thus reach narrow ledges where murre defense was ineffective.

Other avian predators use wind to enhance their foraging success: American Kestrels Falco sparverius (Rudolph 1981), Common Ravens Corvus corax (Spear and Anderson 1989), gulls at sea (Haney and Lee 1993), Herring Gulls Larus argentatus and Great Black-backed Gulls Larus marinus (Theil and Sommer 1994), South Polar Skuas Catharacta maccormicki (Young 1994), Northern Fulmars Fulmarus glacialis (Furness and Bryant 1996). However, not all aerial foragers are efficient in high winds, because prey detection and pursuit may be compromised in these conditions. For example, the foraging ability of Ospreys Pandion haleaetus and some terns, Sterna spp., falls during windy conditions despite increased maneuverability, longer hovering bouts, and more time spent gliding. The surface turbulence on water created by wind interferes with their ability to locate prey (Dunn 1973, Taylor 1984, Machmer and Ydenberg 1990). Wind may also increase the maneuverability of avian prey and their ability to escape predation or kleptoparasitism (Amat and Aguillera 1990). Wind and poor weather may also decrease the activity of prey and, in turn, the probability that predators will detect them (Mearns and Newton 1988). based upon these considerations and our findings, we predict that wind should enhance the foraging efficiency of avian predators when: (1) aerial maneuverability increases the accessibility of prey and/or the likelihood of successful attack; (2) the energy costs of search and attack during flight dramatically influence the net profitability of prey (Furness and Bryant 1996); and (3) wind does not enhance the ability of prey to escape or avoid detection.

The currency of Glaucous Gull foraging decisions

A basic theoretical premise of behavioral ecology is that animals select behavioral strategies that maximize fitness. In practice, indirect "currencies" for estimating fitness are derived from the natural histories of foraging animals and the constraints facing them (Sibly and McCleery 1985, McNamara and Houston 1990). For example, classical foraging theory predicts that animals should select foraging strategies that maximize their net energy gain while foraging (reviewed in Stephens and Krebs 1986). This could occur either through maximizing the rate of energy gain or the efficiency with which it is obtained, and there is considerable evidence that both strategies occur in the wild (Ydenberg et al. 1993). In this system, however, Glaucous Gulls appeared to do neither of these things.

Most gulls were inactive for extended periods of time [TABULAR DATA FOR TABLE 5 OMITTED] during calm ([less than]10 km/h) wind conditions at Coats Island [ILLUSTRATION FOR FIGURE 2-4 OMITTED]. Once windy conditions returned, however, gulls that had been loitering at the colony began to forage immediately, which suggests that they had been waiting for wind conditions to increase. This is surprising, because adult gulls were surrounded by their prey, and attacks made on foot under calm conditions were the most successful.

The foraging inactivity of gulls could simply have been a response to changes in their energy requirements under different weather conditions. For example, during calm conditions, the thermal maintenance requirements for Glaucous Gulls would be low (Gabrielsen and Mehlum 1984). Under windy conditions, gulls might need to increase their foraging activity to meet their higher energy requirements. However, three factors suggest that this does not explain the foraging inactivity of Glaucous Gulls under calm wind conditions. First, gulls occasionally left the colony for extended periods to forage at sea under calm conditions. This strategy yields less energy than that available from a single murre egg obtained by foraging on foot near the nest (Spear 1993, Gilchrist 1995). Second, under windy conditions gulls foraged primarily on the wing, although this probably incurred higher energy costs (Wiens 1984) and achieved lower attack success (11%) than when gulls foraged on foot (43%). Finally, gull chicks actively begged and harassed adults during calm conditions, and this activity appeared to intensify as calm conditions persisted and adults remained idle. Collectively, these results suggest that gulls were not satiated during calm conditions, and that they selected foraging modes that did not yield the highest net energy gains even when foraging conditions improved. We therefore suggest that energy considerations alone are insufficient to explain Glaucous Gull foraging behavior at Coats Island.
TABLE 6. General linear model for factors influencing the level of
murre response to gull attack on (a) broad and (b) narrow ledges.

Term                       Estimate    1 SE       t        P

a) Model [F.sub.4, 882] = 18.2: 13.5% of variance explained:
P [less than] 0.004

Wind speed                  -0.022     0.119    -2.53    0.011
Attack technique            -0.033     0.015    -2.14    0.032
Murre nesting density       -0.045     0.161    -0.57    0.567
Attack technique x wind      0.001     0.001     1.08    0.277

b) Model [F.sub.4, 998] = 21.3; 17.5% of variance explained;
P [less than] 0.001

Wind speed                  -0.010     0.007    -1.41    0.158
Attack technique             0.040     0.014     2.78    0.006
Murre nesting density       -0.031     0.053    -0.58    0.560
Attack technique x wind      0.001     0.001    -0.07    0.941

Foraging theory recognizes that environmental constraints often force foragers away from the strategy that simply maximizes net energetic gain (Mangel and Clark 1986, McNamara and Houston 1990, Stephens 1990, Ward 1990). For example, foragers may minimize their exposure to mortality risks while foraging (e.g., predation), and this may lower their foraging efficiency (Lima and Dill 1990, Suhonen et al. 1994).

For predators, mortality risks while foraging often result from prey fighting back (Curio 1976). Considering the longevity of many avian predators and the frequency of their attacks, factors that even subtly influence the risk of injury or death should strongly affect predator foraging decisions (Clark 1994). However, evidence is lacking. It is difficult to quantify the risk of injury for predators foraging in the wild (e.g., frequency of injury per attack) because (1) predators typically avoid injury during attack, (2) attacks and their outcome are rarely observed, and (3) injured or dead predators may be difficult to detect in the wild (Mills and Shenk 1992).

In this system, murres have the potential to blind and/or kill gulls by striking them repeatedly with their beaks, although we observed that gulls typically abandoned their attacks prior to being seriously injured. However, strikes from murres occurred most often when gulls attacked on foot. Consequently, the risks of injury associated with foraging on foot could reduce the fitness value of foraging on foot under calm conditions, We propose that a trade-off between energy gain and risk of injury while foraging explains the reluctance of gulls to forage on foot and, consequently, the reduced foraging activity observed under calm conditions. Sutherland and Moss (1988) predict such a foraging strategy in systems where (1) poor foraging conditions are short-lived relative to the energy reserves of foragers and their dependent young (e.g., the duration of calm wind conditions); (2) the forager is highly successful once conditions improve; (3) the energy value of a food item is large (here, 714 kJ per murre egg; Spear 1993); and (4) the mortality risks associated with improved conditions are low (e.g., the ability to reach poorly defended narrow ledges on the wing). Therefore, waiting for improved foraging conditions could be an appropriate foraging strategy for Glaucous Gulls at Coats Island, particularly considering their probable long life-span and, consequently, the limited contribution of a single season's reproduction to lifetime reproductive success.

In summary, we suggest that Glaucous Gull foraging behavior at Coats Island reflects a trade-off between the dangers of injury while foraging on foot (generated by murre defense) and the energy gain of capturing a murre egg or chick. This trade-off is mediated by wind conditions, which alter the reward/danger ratio of alternative foraging decisions.


We thank Thomas Alogut, Leah deforest, Garry Donaldson, David Noble, Marco Passeri, and Paul Prior for assistance with the fieldwork. We also thank Arnon Lotem, David Ward, Ray Pierotti, Adrian Farmer. George Hunt, David Jones, Ron Ydenberg, Kathy Martin, and an anonymous reviewer who provided helpful comments on earlier drafts of the manuscript. The research was funded by the Orville Erickson Memorial Scholarship, the Canadian Wildlife Service (operating funds and University Support Grant), the Natural Sciences and Engineering Research Council of Canada, and the Northern Science Training Program. Logistical support was provided by the Polar Continental Shelf Program and the Science Institute of the Northwest Territories Iqaluit Research Center.


Amat, J. A., and E. Aguillera. 1990. Tactics of black-headed gulls robbing egrets and waders. Animal Behaviour 39:70-77.

Bell, W. J. 1991. Searching behaviour: the behavioral ecology of finding resources. Chapman and Hall, London, UK.

Birkhead, T. R. 1977. The effect of habitat and density on breeding success in the Common Guillemot Uria aalge. Journal of Animal Ecology 46:751-764.

Birkhead, T. R., E. Greene, J. D. Biggins, and D. N. Nettleship. 1985. Breeding site characteristics and reproductive success in thick-billed murres. Canadian Journal of Zoology 63:1880-1884.

Birkhead, T. R., and D. N. Nettleship. 1986. Ecological relationships between common murres, Uria aalge, and thickbilled murres, Uria lomvia, at the Canner Islands, Labrador. II: breeding success and site characteristics. Canadian Journal of Zoology 65:1630-1637.

Caraco, T., and T. G. Gillespie. 1986. Risk-sensitivity: foraging mode in an ambush predator. Ecology 67:1180-1185.

Clark, C. W. 1994. Antipredator behavior and the asset-protection principle. Behavioral Ecology 5:159-170.

Cooper, S. M, 1990. The hunting behaviour of spotted hyaenas, Crocuta crocuta, in a region containing both sedentary and migratory populations of herbivores. African Journal of Zoology 28:131-141.

Curio, E. 1976. The ethology of predation. Springer-Verlag, New York, New York, USA.

deforest, L. 1993. The effects of age, timing of breeding, and breeding site characteristics on the reproductive success of the thick-billed murre, Uria lomvia. Thesis. University of Ottawa, Ottawa, Canada.

deforest, L., and A. J. Gaston. 1996. The effect of age on timing of breeding and reproductive success in the Thickbilled Murre. Ecology 77:1501-1511.

Dunn, E. K. 1973. Changes in the fishing ability of terns associated with windspeed and sea surface conditions. Nature 244:520-521.

Furness, R. W., and D. M. Bryant. 1996. Effect of wind on field metabolic rates of breeding Northern Fulmars. Ecology 77:1181-1188.

Gabrielsen, G. W., and F. Mehlum. 1984. Thermoregulation and energetics of arctic seabirds. Pages 137-145 in G. C. Whittow and H. Rahn, editors. Seabird energetics. Plenum Press, New York, New York, USA.

Gaston, A. J., L. deforest, H. G. Gilchrist, and D. N. Nettleship. 1993. Monitoring thick-billed murre populations at colonies in northern Hudson Bay, 1972-1992. Canadian Wildlife Service, Occasional Paper 80.

Gaston, A. J., and D. N. Nettleship. 1981, The Thick-billed Murres of Prince Leopold Island. Canadian Wildlife Service Report 46.

Gendron, R. P., and J. R. Staddon. 1983. Searching for cryptic prey: the effect of search rate. American Naturalist 121:172-186.

Gilchrist, H. G. 1995. The foraging ecology of Glaucous Gulls preying on the eggs and chicks of Thick-billed Murres. Dissertation. University of British Columbia, Vancouver, British Columbia, Canada.

Gilchrist, H. G., and A. J. Gaston. 1997. Effects of murre nest site characteristics and wind conditions on predation by glaucous gulls: an experimental study. Canadian Journal of Zoology 75:518-524.

Goss-Custard, J. D., R. W. G. Caldow, R. T. Clarke, and A. D. West, 1995. Deriving population parameters from individual variations in foraging behaviour. II. Model tests and population parameters. Journal of Animal Ecology 64:277-289.

Haney, J. C., and D. S. Lee. 1993. Air-sea heat flux, ocean wind fields, and offshore dispersal by gulls. Auk 111:427-440.

Helfman, G. 1990. Mode selection and mode switching in foraging animals. Advances in the Study of Behavior 19:249-298.

James, F. C., and C. E. McCulloch. 1990. Multivariate analysis in ecology and systematics: pancea or pandora's box? Annual Review of Ecology and Systematics 21:129-166.

Johnson, R. A. 1938. Predation by gulls in murre colonies. Wilson Bulletin 50:161-170.

Kleinbaum, L. L., L. Kupper, and K. E. Muller. 1988. Applied regression analysis and other multivariate methods. Duxbury Press, Scituate, Massachusetts, USA.

Lima, S. L., and L. M. Dill. 1990. Behavioral decisions made under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68:619-640.

Machmer, M. M., and R. C. Ydenberg. 1990. Weather and osprey foraging energetics. Canadian Journal of Zoology 68:40-43.

Mangel, M., and C. W. Clark. 1986. Towards a unified foraging theory. Ecology 67:1127-1138.

McNamara, J. M., and A. I. Houston. 1990. The value of fat reserves and the trade-off between starvation and predation. Acta Biotheoretica 38:37-61.

Mearns, R., and I. Newton. 1988. Factors affecting breeding success of peregrines in south Scotland. Journal of Animal Ecology 57:903-916.

Mills, M. G. L., and T. M. Shenk. 1992. Predator-prey relationships: the impact of lion predation on wildebeest and zebra populations. Journal of Animal Ecology 61:693-702.

Nettleship, D. N. 1976. Census techniques for seabirds of arctic and eastern Canada, Canadian Wildlife Service Occasional Paper 25:1-33.

Newton, I. 1993. Predation and limitation of bird numbers. Current Ornithology 11:143-197.

Norberg, R. A. 1977. An ecological theory on foraging time and energetics and choice of optimal food-searching method. Journal of Animal Ecology 46:511-529.

Olive, C. W. 1982. Behavioral response of a sit-and-wait predator to spatial variation in foraging gain. Ecology 63:912-920.

Parish, J. K. 1995. Influence of group size and habitat type on reproductive success in Common Murres, Uria aalge. Auk 112:390-401.

Pettifor, R. A. 1990. The effects of avian mobbing on a potential predator, the European kestrel, Falco tinnunculus. Animal Behaviour 39:821-827.

Rudolph, S. G. 1981. Foraging strategies of American Kestrels during breeding. Ecology 63:1268-1276.

Sibly, R., and R. McCleery, 1985. Optimal decision rules for herring gulls. Animal Behaviour 22:506-515.

Siegel-Causey, D., and G. L. Hunt. 1981. Colonial defense behavior in Double-crested and Pelagic Cormorants. Auk 98:522-531.

Sih, A. 1989. Predator and prey lifestyles: an evolutionary and ecological overview. Pages 203-224 in W. C. Kerfoot and A. Sih, editors. Predation: direct and indirect impacts on aquatic comunities. University Press of New England, London, U.K.

Sinclair, A. R. E. 1989. Population regulation in animals. Pages 364-369 in J. M. Cherret, editor, Ecological concepts: the contribution of ecology to an understanding of the natural world. Blackwell, Boston, Massachusetts, USA.

Sokal, R. R., and F. J. Rohlf. 1981. Biometrics: the principles and practice of statistics in biological research. Second edition. Freeman, San Francisco, California, USA.

Spear, L. 1993. Dynamics and effect of western gulls feeding in a colony of guillemots and Brandt's cormorants. Journal of Animal Ecology 62:399-414.

Spear, L. B., and D. W. Anderson. 1989. Nest site selection of Yellow-footed Gulls. Condor 91:91-99.

Stein, R. A. 1977. Selective predation, optimal foraging, and the predator-prey interaction between fish and crayfish. Ecology 58:1237-1253.

Stephens, D. W. 1990. Foraging theory: up, down, and sideways. Studies in Avian Biology 13:444-454.

Stephens, D. W., and J. R. Krebs. 1986. Foraging theory. Princeton University Press, Princeton, New Jersey, USA.

Suhonen, J., K. Norrdahl, and E. Korpimaki. 1994. Avian predation risk modifies breeding bird community on a farmland area. Ecology 75:1626-1634.

Sutherland, W. J., and D. Moss. 1988. The inactivity of animals: influence of stochasticity and prey size. Behavior 92:1-8.

Taylor, R. J. 1984. Predation. Chapman and Hall, New York, New York, USA.

Theil, M., and T. Sommer. 1994. Wind-dependent impact of gulls (Larus sp.) on the breeding success of common terns (Sterna hirundo). Ophelia Supplement 6:239-251.

Ward, D. 1990. Putting ecology back into optimal foraging theory. Theoretical Biology 1:387-410.

Werner, E. E., G. G. Mittelbach, and D. J. Hall. 1981. The role of foraging profitability and experience in habitat use by the bluegill sunfish. Ecology 62:116-125.

Wiens, J. 1984. Modelling the energy requirements of seabird populations. In G. C. Whittow and H. Rahn, editors. Seabird energetics. Plenum Press, New York, New York, USA.

Wilkinson, L. 1989. Systat: the system for statistics. Systat, Evanston, Illinois, USA.

Ydenberg, R. C., C. V. J. Welham, R. Schmid-Hempel, P Schmid-Hempel, and G. Beauchamp. 1993. Time and energy constraints and the relationships between currencies in foraging theory. Behavioral Ecology 5:28-34.

Young, E. 1994. Skua and penguin: predator and prey. Cambridge University Press, Cambridge, UK.

Zaret, T. M. 1980. Predation and freshwater communities. Yale University Press, New Haven, Connecticut, USA.
COPYRIGHT 1998 Ecological Society of America
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1998 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Gilchrist, H. Grant; Gaston, Anthony J.; Smith, James N.M.
Date:Oct 1, 1998
Previous Article:Nest predation and avian species diversity in northwestern forest understory.
Next Article:Conspecific reproductive success and breeding habitat selection: implications for the study of coloniality.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters