Risk-Taking by Incubating Common Goldeneyes and Hooded Mergansers ’

The Condor 100:694-701 0 The Cooper Ornithological Society 1998 RISK-TAKING BY INCUBATING COMMON GOLDENEYES AND HOODED MERGANSERS ’ MARK L. MALLORY, DONALD K. MCNICOL AND RUSSELLA. WALTON~ Canadian Wildlife Service (Ontario Region), 49 Camelot Drive, Nepean, ON, Canada, KIA 0H3, e-mail: [email protected] MARK WAYLAND Canadian Wildlife Service (Prairie and Northern Region), 115 Perimeter Road, Saskatoon, SK, Canada. S7N OX4 Abstract. We studied nest defense by Common Goldeneyes (Bucephala clangula) and Hooded Mergansers (Lophodytes cucullatus) nesting near Sudbury, Canada between 1989- 1995. As incubationproceeded, female Common Goldeneyestook greaterrisks by allowing the observer to approachthe nest more closely before flushing, landing closer to the nest after flushing, vocalizing more commonly when flushed, and giving more broken wing or distraction displays. Hooded Merganser females exhibited relatively strong defense at all stages,including giving distractiondisplays much earlier in incubation.An index of all four behaviorsincreased for each speciesas incubationproceeded. Nest defense by these cavity- nesting ducks differed from patterns known for ground-nestingwaterfowl, perhapsdue to effects of nest location, predationtype and intensity, and nest parasitism.Future nest defense studies should consider the overall suite of behaviors that comprise parental strategiesfor defending their nests and/or offspring. Key words: Bucephalaclangula, cavity-nesting, Common Goldeneye, Hooded Mergan- ser, Lophodytes cucullatus,nest defense. INTRODUCTION as “weapons” (Montgomerie and Weatherhead Egg-laying and incubation expose female birds 1988) in active defense against most predators, to a variety of risks, perhaps the greatest being so their nest defensebehaviors may rely on other predation on the female or her clutch (Montgo- tactics. One tactic is to remain on the nest in the merie and Weatherhead 1988). When females presence of a predator. This protects the clutch defend their nests against predators, they face a by not drawing attention to it, but puts the fe- tradeoff: decreasedrisk of clutch predation (An- male at greater risk because it reduces her time dersson et al. 1980) vs. increased risk of injury to escape if the predator gets too close (Forbes or death which affects future prospects for et al. 1994). A second defense tactic is to per- breeding (Curio and Regelmann 1985). The form a distraction display. A series of behaviors costs and benefits associated with this tradeoff can be grouped under this tactic, ranging from will be determined in part by the current value vocalizations when flushed, to landing nearby on of the clutch, which generally increases with the water, to the familiar “broken wing” dis- clutch size and as incubation proceeds (Mont- plays exhibited by many species (Montgomerie gomerie and Weatherhead 1988). Females are and Weatherhead 1988). The common element therefore expected to defend larger clutches to all of these behaviors is that the female risks more rigorously and to increase their efforts as being capturedby the predator.Nest defense the- incubation proceeds (Montgomerie and Weath- ory predicts that she gauges the level of accept- erhead 1988). able risk by the current value of her clutch. One group of precocial birds that has received In this paper, we quantify such risk-taking by little attention in nest defense researchis water- incubating Common Goldeneyes (Bucephala fowl, particularly cavity-nesting ducks. They clangula) and Hooded Mergansers (Lophodytes lack adequate physical features that could serve cucullatus), both cavity-nesting waterfowl. We monitored female behaviors during nest visits which simulated approaches by predators. In ‘Received 16 March 1998. Accepted 20 June 1998. ZCurrent address: Ministrv of Forests. Kamlooos preliminary studies (Mallory et al. 1993a, Mal- Region, 515 Columbia St., Kamloops, RC, Canada, lory and Weatherhead 1993), both species ex- v2c 2Tl. hibited the risk-taking behaviors described 16941 NEST DEFENSE IN CAVITY-NESTING WATERFOWL 695 above (remaining on the nest, vocalizing, land- at least once during incubation (late May) each ing nearby, and broken wing displays) during year between 1992 and 1995. Because of their incubation. Our objectives in this study were to isolation and the lack of sportfishin the majority refine earlier analysesby examining each behav- of our study lakes, nesting females probably ex- ior independently in relation to incubation stage perienced no other human disturbance at the nest and clutch size, and to test several specific pre- box during incubation. We did not measure dis- dictions regarding nest defense and nesting hab- turbance by predators at the nest site, but suc- its of these species. First, because female gol- cessful predation was generally low (McNicol et deneyes and mergansers could use any combi- al. 1997). Also, female age (i.e., residual repro- nation of the above four behaviors to defend ductive value and breeding experience, which their nest, we predicted that individual nest de- can influence nest defense) was not known for fense behaviors would exhibit similar changesin most females in this study, and thus its potential intensity through incubation, and thus that over- effects were not examined. all nest defense intensity would increase as in- When we visited a nest, we observed the be- cubation proceeded. Second, we examined havior of incubating females on our approach. whether nesting females appeared to alter their We recorded the distance of the observer from behavior in response to repeated visits to the the nest when the female flushed (estimated to nest by observers, because patterns of defense 1 m), the estimated distance she flew before related to incubation stage might instead be ex- landing (estimated to 10 m, to 100 m if she land- plained by effects of previous visits on female ed > 100 m away), any vocalizations given in behavior (Knight and Temple 1986). Finally, be- flight, and whether she performed any distrac- cause nest parasitism is known to influence nest tion displays. Because the nests of these species defense in some altricial species (Weatherhead are located above the ground, we had to treat et al. 1994), we analyzed whether nest defense flushing distances somewhat differently than for was related to interspecific nest parasitism. ground-nesters(Forbes et al. 1994). When a fe- METHODS male flushed from a nest box, we used the actual distance of the observer from the nesting tree as We collected data between 1989 and 1995 on the score.However, if a female flushed when the risk-taking by female Common Goldeneyes and observer was at the base of the tree or on the Hooded Mergansers nesting in nest boxes erect- tree, we assigned scoresto represent this “finer ed on small study lakes (generally < 20 ha) scale” behavior: “0.5 m” if the observer was at northeastof Sudbury, Ontario, Canada (46”54’N, the base of the tree, “0.25 m” when the observ- 80”41 ‘W). Characteristics of the study site and er was climbing the tree, and “0.1 m” if the nest boxes are describedin McNicol et al. (1987, observer was opening the nest box. 1997). Physical dimensions and position of the nest box site on the lake were similar among To assesstotal nest defense, we ranked each boxes. One box was erected per lake, and in al- of the four measuredvariables (flushing distance most all casesthere was only one nesting female (range 1.5-71.5), landing distance (16-77), with per lake (Mallory et al. 1993b; known excep- closer distances ranked higher; vocalizations tions occurred on two lakes with occupied nat- (24.5-67), with higher rank given to birds that ural cavities). Both species have one annual gave calls; broken wing displays (39.5-82), with breeding attempt and incubation periods of higher rank to birds that gave displays) for all about 30 days, and typically lay 8-10 eggs first visit data among all females from both spe- (Dugger et al. 1994, Eadie et al. 1995). The cies within each behavior. We then divided ranks study site is at the southern boundary of the for each variable by the maximum rank for each Common Goldeneye’s breeding range but is in variable, and added the resultant indices for each the northern third of the Hooded Merganser’s variable together for each female. These cumu- breeding range (McNicol et al. 1995). lative indices were then divided by the maxi- Females were studied intensively in 1989 and mum cumulative index to generate a total com- 1990 (Mallory et al. 1994, Wayland and Mc- posite nest defense index for each female, with Nicol 1994), with nest visits occurring as often a maximum value of 1.O. Note that this approach as every 5 days during May and June. Nest box- assumesequal weighting for each behavior; we es were not checked in 1991, but were checked considered this the most conservative approach NEST DEFENSE IN CAVITY-NESTING WATERFOWL 697 Common Goldeneye Hooded Merganser E >lOO- 25 7 $ fJ g 25-100 - 17 18 i? 55 ~25 - 6 3 J 7 -I 0 10 20 30 0 10 20 30 Incubation Stage (days) FIGURE 1. First visit flushing distance (m) and landing distance(m) for incubatingfemale Common Golden- eyes and Hooded Mergansers in relation to incubation stage (days since the start of incubation). Numbers on bar charts are sample sizes. COMPOSITE NEST DEFENSE incubation (where data were well-represented Although the above analyses described patterns for both species). in particular behaviors, we also were interested To test whether larger clutches were defend- in the whole pattern of nest defense of each spe- ed more vigorously and whether interspecifi- cies (i.e., the combined effects of each of the tally parasitized clutches were defended less four measured behaviors), and thus we gener- vigorously, we used the residual composite ated a compositenest defense index. In a prelim- nest defense index (from above, after control- inary analysis, we included incubation stage and ling for incubation stage) as our estimate of Julian date as independent variables in stepwise defense intensity, because this incorporated multiple regressions,but Julian date was not in- possible combinations of behaviors for each cluded as a significant predictor of the defense female.

Load more