Nesting ecology of common Goldeneyes and Hooded Mergansers in a boreal river system.
Common Goldeneye (Bucephala clangula) and Hooded Merganser (Lophodytes cucullatus) are secondary cavity-nesting ducks that breed in the boreal forest (Dugger et al. 1994, Eadie et al. 1995). These species are relatively large and require large cavities, which represent only a small proportion of all cavities available in the boreal forest (Gauthier and Smith 1987). Intra- and interspecific competition for these large cavities is therefore severe (Erskine 1990, Semel and Sherman 2001). Selection of a suitable nest site may be especially critical in regions of intensive for est management because forestry practices can reduce the availability of suitable cavities. Many studies have shown that addition of nest boxes results in increases in breeding populations of secondary cavity-nesters (Savard 1988, Newton 1994, Poysa and Poysa 2002) suggesting that nest site availability was limiting these populations at least locally.
Common Goldeneyes and Hooded Mergansers commonly breed on ponds and lakes of the boreal forest. Fast-flowing rivers, although abundant throughout the boreal forest, are generally not considered an important breeding habitat for these species (Eadie et al. 1995). Nest site selection of Common Goldeneyes has been previously studied in lake systems (Lumsden et al. 1980, 1986), but the nesting biology of these species in river systems is little known. The overall objective of this study was to evaluate the potential of rivers as nesting habitat for Common Goldeneyes and Hooded Mergansers in the boreal forest. We studied the nesting ecology of both species using nest boxes in a fast-flowing river system in the Quebec boreal forest. We also assessed the potential for nest site competition between both species and investigated whether occupancy of nest boxes along the river could be associated with a scarcity of natural cavities in the area. Finally, we identified river characteristics and components of the riparian habitat influencing nest site selection and nesting success of these species in this habitat.
Study Area.--This study was conducted in the Ste. Marguerite River Valley (48[degrees] 23' N, 1 70[degrees] 12' W), 200 km northeast of Quebec City, Canada. This is a meandering but fast flowing river typical of the boreal forest in Quebec (mean annual discharge = 58 [m.sup.3]/sec). The river flows for ~100 km in a 0.5-1.0 km wide valley, often bordered by steep mountain slopes rising to 350 m above the valley bottom. About 53 ponds (including oxbows) occur in the valley of the studied section and can be used by breeding adults or broods after hatch. The Ste. Marguerite River is a prime Atlantic salmon (Salmo salar) and brook trout (Salvelinus fontinalis) fishery managed by a private fishing association.
Mature stands of balsam fir (Abies balsamea) and yellow birch (Betula alleghaniensis) dominate the forest of the valley. The forest in the upper half of the study area has remained almost untouched for at least 100 years and large yellow birch trees are abundant. However, clear-cutting occurred in the early 1980's on about 6% of the catchment area, mainly in the lower half of the valley; this area still has some 80-year-old trees.
Experimental Design.--Ninety-two nest boxes were erected in fall 1997 along a 50-km stretch of the river, encompassing a diversity of river habitats (rapids, fast current, slow current, meanders, etc.); 10 additional boxes were installed in fall 2000. The number of boxes varied annually from 82 to 93 because some nest boxes were damaged by falling trees or disappeared. Their distribution along the river followed fishing access paths in most cases but boxes were isolated from fishing pools to minimize human disturbance. Boxes were separated by at least 30 m and nailed to trees close to the water edge (range = 0 to 29 m) with the entrance hole facing the river, between 2 and 5 m above the ground. Boxes were made of plywood (inside dimensions: 23 x 26 X 58 cm; entrance: 10 X 12 cm; depth below the entrance: 45 cm) and a 10-cm layer of wood shavings was added as nest material. We removed branches in front of the boxes to ensure visibility from the river and ease of entry for waterfowl. Boxes were cleaned at the end of each breeding season (abandoned eggs removed) and wood shavings were changed.
Nest Box Monitoring and Measurement of Nesting Parameters.--Nest boxes were monitored from early May to early July weekly in 2001, three times in 2000 and 2002, and twice in 1998 and 1999. We recorded the number of eggs of each species and boxes containing eggs of more than one species were identified as interspecific nest parasitism. The species that first started the nest or, if unknown, the incubating species was considered the host. Eggs were numbered with a permanent marker, and measured (length and width, [+ or -] 0.1 mm) with a caliper and weighed ([+ or -] 1 g) with a spring scale. These measurements were used to calculate the Maximum Euclidian Distance (MED), a criterion used to identify intraspecific nest parasitism (Eadie 1989, Poysa et al. 2001). Nests were classified as parasitized by conspecifics when at least one of the following criteria was met: (1) the MED among the eggs was >2.5, (2) clutch size was >12 eggs, (3) more than one egg was laid per day, and (4) eggs were laid two or more days after the start of incubation. More than one of these criteria was met in most nests where parasitism was detected. These criteria should yield a minimum estimate of nest parasitism (Andersson and Ahlund 2001).
We calculated the date on which the first egg of the clutch was laid (i.e., laying date) for nests discovered during laying by backdating using the number of eggs of the host species at the time of discovery (maximum of 12 eggs; additional eggs were considered parasitic) and assuming a mean laying rate of one egg every 1.4 days. The laying date for nests discovered during incubation was backdated from the hatching date assuming an incubation period of 30 days. We calculated hatching date when it was unknown (for some nests in 2000 and 2002), using the relationship between incubation stage and egg density (Senechal 2003). We estimated the age of embryos (Caldwell and Snart 1974) for deserted nests with unknown laying dates and backdated the laying date. Clutch size was defined as the highest number of eggs recorded in an incubated nest.
Nesting success was the proportion of successful nests (i.e., those in which at least one egg hatched). Deserted nests were those abandoned during laying or incubation with no signs of predation and were included in the calculation of nesting success. Hatching success, defined as the proportion of eggs that hatched within a clutch, was calculated for successful nests.
Availability of Natural Cavities.--We estimated the availability of living trees and snags sufficiently large to support cavities and of cavities along the river using line transect sampling (Buckland et al. 1993). We positioned 17, 300-m transects every 3 km at right angle to the river axis. We recorded the position and perpendicular distance from the transect of all trees and snags [greater than or equal to] 30 cm in diameter at breast height (DBH) encountered. We ascertained species, height, DBH, and presence of cavities with entrance diameter [greater than or equal to] 8 cm (visual estimate from the ground) for these trees. We recorded the cavity type (hollow top of standing trees, woodpecker hole, broken tree limbs, or crack on the side of a tree), its height above ground, an index of its accessibility from the air (under the canopy, in the canopy, over the canopy, or open forest), and the approximate size of the entrance hole. Any signs of waterfowl use (e.g., presence of down at the entrance) were noted. Cavities with entrances [greater than or equal to] 8 cm, [greater than or equal to] 1.8 m above ground, and easily accessible (i.e., above the forest canopy or in an open stand) were considered suitable (Bergeron et al. 1997). Sampling was conducted in late May-early June and at the end of July 2001.
Nest Site Characterization.--We sampled the habitat surrounding each nest box in late July-early August 2001. We measured river depth at the deepest point with a graduated rod, river width with a measuring tape (at times estimated from the shore), and surface water velocity at the deepest point using a portable current meter (Marsh McBirney, model 201M; Frederick, MD, USA). We combined velocity measurements and the general aspect of the fiver to create a new variable that categorized river sections as rapids ([greater than or equal to] 60 cm/sec with riffles at the surface), fast-flowing ([greater than or equal to] 60 cm/sec, water surface still) or slow-flowing (<60 cm/sec).
We sampled 12 terrestrial habitat variables at each nest box. These variables included: distance to water edge, to nearest lake or pond, to nearest used nest box, and to nearest tree [greater than or equal to] 10 cm in diameter; entrance height above the river and above the ground; compass direction faced by the entrance (categorized as North, East, South, or West); and canopy height. Distances to nearest pond or nest box were obtained from a digital map (scale: 1:20,000) whereas other distance variables were measured in the field. We also sampled density of trees with a DBH of [greater than or equal to] 20 cm in a semi-circle centered in front of the nest box (11.3-m radius) and evaluated shrub height. Stems [greater than or equal to] 4 m in height were defined as trees and those <4 m as shrubs. We also counted the number of lakes and ponds in the vicinity on the digital map, i.e., those included in a 2 x 6 km ellipse (long axis parallel to the river) centered at the nest box, and calculated the total surface area of these water bodies.
Data Analysis.--Statistical analyses were completed with software SAS 8.0 (SAS Institute 1999). Interactions in analyses with more than one factor were examined and removed from the model when non significant (P > 0.05).
We examined the effects of species and year on laying date, hatching date, and clutch size with two-way ANOVAs. Dates were rank-transformed and we used the ART procedure to test the interaction as recommended by Salter and Fawcett (1993). We examined the effects of year, species, and parasitic egg laying on nesting success using a log-linear analysis. We analyzed the effect of egg status (host species in a parasitized nest, heterospecific parasite in a parasitized nest, or in a nest non-parasitized by a heterospecific parasite), year, and species on hatching success using a three-way ANOVA on rank-transformed data. We examined the effects of species and year on the rate of parasitism using logistic regressions, and on the number of heterospecific parasitic eggs using two-way ANOVAs. We compared the laying date of parasitized and non-parasitized nests of each species using ANOVA.
We estimated densities of living trees ([greater than or equal to] 30 cm DBH), snags, and cavities with DISTANCE 3.5 (Buckland et al. 1993, Research Unit for Wildlife Assessment 1999). We compared DBH and height between snags and living trees using Mann-Whitney tests, and the proportion of trees with cavities using [chi square]. We examined if snag and living tree densities, cavity abundance (because of the small number of suitable cavities found), and tree characteristics (DBH, height) were related to distance to the river (in 50-m classes) with linear regression.
[FIGURE 1 OMITTED]
We generated three independent data sets to analyze characteristics of nest sites used by each species from 1999 to 2002. Boxes were classified whether they were used: (1) at least once versus not used, (2) at least twice versus not used, and (3) at least three times versus not used. Separate analyses with each data set generally yielded similar results (Senechal 2003), and we present only the results based on boxes used at least twice because this avoided using data from boxes rarely used (i.e., only once) while retaining sufficient sample size. The effect of the 13 quantitative habitat variables on nest site use was first analyzed using a stepwise discriminant analysis, followed by a logistic regression including variables retained by the first analysis (P < 0.15) and qualitative variables. We used the same procedure to test the relationship between nesting success and habitat characteristics for goldeneyes (sample sizes were too small for mergansers). We combined data from 2000 to 2002 and, for nests used more than once, we assigned success or failure based on the occurrence that was most frequent, excluding those with the same number of successes and failures.
Nest Box Use and Nesting Parameters.--The number of nest boxes used by waterfowl was low the first year but increased in the second year (Fig. 1). Occupancy rate stabilized at 51-55% by the fourth year of the program. Goldeneye use of nest boxes outnumbered mergansers three to four times in all years but the first.
Both species started their clutches at about the same time from 2000 to 2002 ([F.sub.1.109] = 2.0, P = 0.17) with no differences between years ([F.sub.2,109] = 1.8, P = 0.18). Overall, mean ([+ or -] SE) laying date was 4 May [+ or -] 0.9 days (range = 14 Apr-24 May, n = 81) for goldeneyes and 7 May [+ or -] 1.9 days (21 Apr-5 Jun, n = 34) for mergansers. Goldeneye nests hatched slightly earlier than mergansers ([F.sub.1,90] = 5.8, P = 0.02). Overall, the mean hatching date was 15 Jun [+ or -] 0.9 days (31 May-2 Jul, n = 68) for goldeneyes and 20 Jun [+ or -] 1.9 (31 May-13 Jul, n = 28) for mergansers.
Mean ([+ or -] SE) clutch size in incubated goldeneye nests (9.8 [+ or -] 0.4 eggs, n = 78) was similar to that of mergansers (9.2 [+ or -] 0.5, n = 33; [F.sub.1.l07] = 0.8, P = 0.36) when eggs resulting from interspecific parasitism were excluded and did not vary among years ([F.sub.2,107] = 0.2, P = 0.83; Table 1). The total number of eggs (i.e., including parasitic eggs) was higher for merganser nests (11.6 [+ or -] 0.6) than for goldeneye nests (10.2 [+ or -] 0.4; [F.sub.1,107] = 3.7, P = 0.06).
Nesting success of both species decreased between 2000 and 2002 ([[chi square].sub.2] = 16.1, P < 0.01), and was lower for goldeneyes than for mergansers ([[chi square].sub.1] = 3.3, P = 0.07; Table 2). Non-parasitized nests were less successful than those parasitized (64 vs. 82%, [[chi square].sub.1] = 8.8, P < 0.01). Desertion was the main cause of nesting failures of goldeneyes (54%, n = 28). Other causes included desertion due to marking (18%), death of incubating females (7%), clutch predation (7%), other human disturbance (7%), and unknown (4%). The main cause of nest failure for mergansers was desertion due to human disturbance other than marking (33%, n = 6). Other causes included clutch predation (17%) and unknown (50%).
Overall, heterospecific parasitic eggs did not fail more often than eggs from host species or eggs in non-parasitized nests ([F.sub.2,107] = 0.2, P = 0.80; Table 3). Goldeneyes and mergansers had similar hatching success (90 vs. 88% overall, [F.sub.l,107] = 0.4, P = 0.56) and there were no differences between years (F.sub.2,107] = 2.3, P = 0.11).
Pattern of Nest Parasitism.--Interspecific nest parasitism started in the second year (1999) of the nest box program and appeared to increase with the proportion of nest boxes used (Fig. 2; year effect: [[chi square].sub.3] = 5.6, P = 0.13). A much greater proportion of merganser nests were parasitized by goldeneyes than the reverse ([[chi square].sub.1] = 8.6, P < 0.01). Parasitized merganser nests also contained on average more goldeneye eggs (4.7 [+ or -] 0.6 eggs, n = 17) than the reverse (2.8 [+ or -] 0.3, n = 16; [F.sub.1,29] = 7.2, P = 0.01).
Parasitism by conspecifics was also common in both species (species effect: [[chi square].sub.1] = 0.5, P = 0.49), ranging from 9 to 32% of goldeneye and 7 to 30% of merganser nests during 2000-2002 (year effect: [[chi square].sub.2] = 3.2, P = 0.20). Among the total number of nests initiated by each species (i.e., nests where each species laid at least 1 egg), mergansers tended to lay more often as a parasite (intra- and interspecific egg laying combined) than goldeneyes ([[chi square].sub.1] = 3.2, P = 0.07; Fig. 3). On average, 25 [+ or -] 5% (range = 18-35%) of all nests containing goldeneye eggs were associated with parasitic events by this species compared to 39 [+ or -] 8% (22-50%) for mergansers. Nests parasitized by goldeneyes (intra- and interspecifically) were initiated earlier (30 Apr [+ or -] 1.3 days, n = 31) than non-parasitized nests (6 May [+ or -] 1.1, n = 50; [F.sub.1,77] = 9.1, P < 0.01); this was not the case for mergansers ([F.sub.1,29] = 2.2, P = 0.15).
Availability and Suitability of Natural Cavities.--Of 688 trees ([greater than or equal to] 30 cm DBH) recorded along transects, 17% were snags and 83% were alive. Large snag density was estimated at 6.5 [+ or -] 0.8 snags/ha, almost six times lower than large living trees (36.6 [+ or -] 2.0 trees/ha). Snags were slightly larger (mean DBH: 42.4 [+ or -] 0.8 cm, n = 114) than living trees (40.4 [+ or -] 0.4 cm, n = 574; U = 5.9, P = 0.02), but shorter (7.8 [+ or -] 0.7, n = 45 vs. 18.1 [+ or -] 0.3 m, n = 204, respectively; U = 82.6, P < 0.01).
Suitable cavities were present in 6% of the snags (n = 114; 1 with 2 cavities) but were rarely found in living trees (1%, n = 574; [[chi square].sub.1] = 32.3, P < 0.01). We found 16 suitable cavities during sampling for an estimated density of 0.92 [+ or -] 0.31 cavity/ha, but none had signs of waterfowl use. Thirty-seven cavities sufficiently large for ducks were found (1.71 [+ or -] 0.37 cavities/ha) when the accessibility criterion was not considered. Pileated Woodpecker (Dryocopus pileatus) holes were the most abundant cavity type (50%), followed by chimney cavities (38%) and natural cracks (12%). Most suitable cavities were in snags (69%) at a mean height of 6.2 [+ or -] 1.0 m (n = 11).
The dominant species of large trees were yellow birch (57%), balsam fir (19%), eastern white cedar (Thuja occidentalis; 7%), and white spruce (Picea glauca; 6%). Black ash (Fraxinus nigra), eastern white pine (Pinus strobus), paper birch (Betula papyrifera), trembling aspen (Populus tremuloides), balsam poplar (P. balsamifera), and white elm (Ulmus americana L.) each accounted for [less than or equal to] 3% of the large trees. The 16 suitable cavities were mainly in yellow birch (81%); the rest were equally distributed among elm, ash, and fir (6% each).
Yellow birch was the only tree species for which density increased with distance to the river up to 300 m ([beta] = 3.73 [+ or -] 0.47 trees/ha/ 50 m, [r.sup.2] = 0.94, P < 0.01, n = 6). Density of large snags decreased away from the river ([beta] = -0.87 [+ or -] 0.37 snags/ha/50 m, [r.sup.2] = 0.58, P = 0.01). There was no relationship between the abundance of suitable cavities and distance to the river ([r.sup.2] = 0.17, P = 0.41), nor between tree height or DBH and distance to the river (P > 0.12).
Nest Site Characteristics.--Only 19 of all nest boxes available were not used. Boxes in the upper half of the river were generally used more often than those in the lower half. Few habitat variables explained nest box use by ducks. Boxes used by goldeneyes had a greater total surface area of ponds in the vicinity (40.7 [+ or -] 6.1 ha) than those not used (8.3 [+ or -] 4.0 ha) and goldeneyes used boxes predominantly oriented toward the south (Table 4). Boxes not used by this species were usually farther from water than those used (9.4 [+ or -] 1.2 vs. 6.2 [+ or -] 0.7 m, respectively). Used boxes were farther from trees [greater than or equal to] 10 cm (2.4 [+ or -] 0.3 m) than those not used (1.5 [+ or -] 0.2 m). Goldeneye nesting success was affected mainly by the total surface area of ponds in the vicinity ([[chi square].sub.1] = 13.2, P < 0.01); failed nests were surrounded by a larger surface of ponds than those that were successful (61.6 [+ or -] 15.0 ha, n = 14 vs. 35.5 [+ or -] 7.9, n = 36, respectively).
Boxes used by mergansers were also farther from the nearest tree [greater than to equal to] 10 cm in diameter than those not used (3.3 [+ or -] 0.7 vs. 2.0 [+ or -] 0.2 m), and tree canopy around used boxes (21.8 [+ or -] 2.4) was higher than around boxes not used (14.5 [+ or -] 0.7, Table 4).
Nesting Biology.--Common Goldeneye and Hooded Merganser nesting biology in river habitat appears similar to that in the more usual lacustrine habitats of northeastern North America. Nest initiation dates (16 Apr-27 May) reported for goldeneyes in Ontario (Mallory et al. 1994, Eadie et al. 1995) are similar to those in our study. Mean Clutch sizes reported in Ontario (7.4-9.4 eggs), New Brunswick (9.0 eggs), and Minnesota (10.2 eggs) (Eadie et al. 1995) are also similar. Nesting success in our study was higher than reported by Bouvier (1974) in southern Quebec (57%) but lower than observed in Maine (87%) (Allen et al. 1990). Clutch size of Hooded Mergansers usually ranges from nine to 13 eggs and nesting success ranges from 66 to 82% (Dugger et al. 1994, Mallory et al. 2002). This suggests that nesting in habitats adjacent to fast-flowing rivers does not impose severe constraints on nesting phenology or success for these species.
The similarity in the observed breeding phenology of Common Goldeneyes and Hooded Mergansers is not surprising given the similarities in life history traits of the two species (Dugger et al. 1994, Eadie et al. 1995). The few merganser nests initiated after 25 May were possibly renesting attempts or nests of first-time breeders that nest later than older, experienced females (Dow and Fredga 1984).
Desertion was a major cause of nesting failure for goldeneyes. This may be due to a high proportion of first-time breeders in the area, as inexperienced birds are more prone to desertion (Eriksson and Andersson 1982, Gauthier 1989). A high proportion of first-time breeders may result from a recent increase in the breeding population possibly due to installation of nest boxes. Human disturbance may also be a cause of nest desertion of goldeneyes. This species shows relatively weak nest attentiveness and defense relative to Hooded Mergansers, flushing more readily when disturbed (Mallory et al. 1993a, 1998). Greater sensitivity to disturbance may contribute to the high rate of nest abandonment of goldeneyes. It is also possible that more frequent visits to nest boxes in 2001, and more capture and handling of females, contributed to high nest desertion that year.
Nest parasitism often has a negative impact on nesting success of waterfowl due to disturbance by parasitic females (Semel et al. 1988, 1990). This was not the case in our study, as parasitized nests generally had greater nesting success than non-parasitized nests (also reported by Eadie 1989). Large clutches resulting from nest parasitism at times also have reduced hatching success (Mallory et al. 2002), but normal hatching success has also been reported (Eriksson 1979, Eriksson and Andersson 1982, McNicol et al. 1997). Large clutches (>20 eggs) were rare in our study, which may explain why parasitism had little effect on hatching success.
Nest Parasitism and Nest Site Competition.--Intra-and interspecific nest parasitism were both common in our study. The frequency of conspecific egg laying is likely a minimum value because Andersson and Ahlund (2001) showed with biochemical techniques that it is underestimated with traditional methods. Our results support the suggestions of Eadie (1989), Semel et al. (1988), and Haramis and Thompson (1985) that high breeding densities of cavity-nesting waterfowl and relative scarcity of suitable nest sites may contribute to nest parasitism. Almost half of our nest boxes were unused, apparently suggesting that nest sites were not limiting. However, this assumes that all nest boxes and sites selected for their placement were of equal quality, which is unlikely to be true. We found evidence that goldeneyes preferred some boxes over others. Moreover, boxes were distributed over 50 km of river that may vary greatly in quality for ducks. For example, boxes in the lower half of the river received relatively little use. Thus, even though only 55% of the boxes were used, there could be competition for high quality nest sites.
Merganser nests were more likely to be parasitized by goldeneyes and received a larger number of parasitic eggs than goldeneye nests, as reported in other studies (Bouvier 1974, Mallory et al. 1993a). The greater proportion of parasitized merganser nests may be related to the smaller size of the merganser population rather than a lower propensity for parasitic egg-laying. Mergansers tended to lay more often as a parasite than goldeneyes (also reported by Mallory et al. 1993a) when we compared the number of nests where one species laid parasitically in relation to the number of nests initiated by that species. Intraspecific nest parasitism in cavity-nesting ducks may be an adaptive strategy (Semel et al. 1988, Poysa 1999, Andersson 2001, Ahlund 2005), but it can also be the outcome of two individuals inadvertently choosing the same nest site to lay their eggs (Erskine 1990). Semel and Sherman (2001) supported this hypothesis for Wood Ducks (Aix sponsa) and interspecific nest parasitism is perhaps more likely to result from such nest site competition. Hooded Mergansers are probably dominated in competitive interactions by the larger, more aggressive goldeneye. Thus, they may be more easily evicted from a nest box, giving the impression that mergansers "parasitize" more often than goldeneyes.
Availability and Suitability of Natural Cavities.--No breeding pair surveys were conducted in the area before nest boxes were installed and we do not know with certainty if high use of boxes by goldeneyes and mergansers reflects a real increase of their local populations, as reported for other populations (Fredga and Dow 1984, Savard 1988, Poysa and Poysa 2002). One explanation for the increase in nest box use may be that local females moved from nearby natural cavities to the more conspicuous nest boxes near the river (Gauthier and Smith 1987). Our sampling showed that density of suitable cavities was low in our study area considering the high density of large snags and living trees present. Our density estimate for natural cavities may also be a maximum value as we were not able to inspect the interior of cavities. Similar cavity densities have been reported elsewhere (Senechal 2003), but the density was lower than expected for a mature forest (Prince 1968, Gilmer et al. 1978). None of the cavities found appeared to be used by waterfowl. Suitable cavities may have been beyond our 300-m transects, but steep slopes or cliffs were often encountered beyond this distance, which would reduce their attractiveness because of predation risks associated with long and difficult overland travel by broods.
Common Goldeneyes appear flexible in the type of cavity used. Carter (1958) mentioned they use mainly lateral openings in New Brunswick, whereas Prince (1968) identified chimney cavities as the main type used. Maisonneuve et al. (2002) reported that half of Hooded Merganser and Common Goldeneye nests found in Quebec boreal forest were in old Pileated Woodpecker cavities. Abandoned Pileated Woodpecker holes in our study were the main source of suitable cavities and this woodpecker may be a key species for cavity-nesting ducks (Bonar 2000).
Nest Box Use and Habitat Characteristics.--The preference of goldeneyes for nest boxes in areas with a high surface area of ponds suggests that many pair territories were on ponds. Mallory et al. (1993b) reported goldeneyes preferred nest boxes in wetlands without fish isolated from other water bodies in acid-stressed areas of Ontario. Many lakes and ponds in our study area supported fish and were little acid-stressed, but may provide more invertebrate food than the river where Atlantic salmon were present. An abundance of ponds in the vicinity may also provide many potential brood-rearing sites, as ponds are important brood habitats for females nesting along the river (Senechal 2003). Golden eyes preferred nest boxes closest to water, presumably because proximity to water minimizes overland travel for broods after hatch (Morse et al. 1969, Poysa et al. 1999) and decreases predation risks. Nest boxes near water are also more conspicuous than those further in the forest and are more easily found by ducks. The preference for nest boxes far from large trees is another indication that goldeneyes prefer nesting in open habitats. Mergansers also selected nest sites in open habitats with a high canopy. The preference of goldeneyes for boxes with a southward orientation may be related to a more favorable micro-climate (i.e., away from prevailing westerly wind and toward the sun). No characteristics of the river influenced nest box use by either species, but this may not be surprising if pair territories were mostly away from the river. Presence of a suitable cavity (nest box) along the river may be the main stimulus for nesting in this habitat.
Few characteristics of the habitat surrounding nest sites affected goldeneye nesting success. Surprisingly, goldeneye nests with a high surface area of ponds in their vicinity were less successful. Pond area positively influenced nest site use, and the attractiveness of nest sites in these areas may have increased disturbance from prospecting females (Eadie and Gauthier 1985) and may have decreased nesting success. Mallory et al. (1993b) reported that lake isolation was an important factor in nest site selection by goldeneyes in Ontario and suggested that high bird density may lead to interference in this highly territorial species (Savard 1984). Eadie (1989) found no differences in habitat characteristics between successful and failed Barrow's (Bucephala islandica) and Common goldeneye nests in British Columbia.
We acknowledge the Fondation de la faune du Quebec, Fonds Quebecois de Recherche sur la Nature et les Technologies, Natural Sciences and Engineering Research Council of Canada, Canadian Wildlife Service, Ducks Unlimited, and the Centre Interuniversitaire de Recherche sur le Saumon Atlantique for financial support. We also thank Nadia Aubin-Horth, Dominic Savard, Gerald Picard, Gregory Bourguelat, Jerome Leger, M.-C. Cadieux, Antoine Morrissette, J.-F. Savard, Karine Plante, Catherine Bonenfant, and Felix Ledoux for help in the field, and Jean Huot and Cyrille Barrette for comments on the manuscript.
Received 1 October 2007. Accepted 23 February 2008.
AHLUND, M. 2005. Behavioural tactics at nest visits differ between parasites and hosts in a brood-par asitic duck. Animal Behaviour 70:433-440.
ALLEN, R. B., P. O. CORR, AND J. A. Donso. 1990. Nesting success and efficiency of waterfowl using nest boxes in central Maine: a management perspective. Pages 291-296 in Proceedings of the 1988 North American Wood Duck Symposium (L. H. Fredrickson, G. V. Burger, S. P. Havera, D. A. Graber, R. E. Kirby, and T. S. Taylor, Editors). St. Louis, Missouri, USA.
ANDERSSON, M. 2001. Relatedness and the evolution of conspecific brood parasitism. American Naturalist 158:599-614.
ANDERSSON, M. AND M. AHLUND. 2001. Protein fingerprinting: a new technique reveals extensive conspecific brood parasitism. Ecology 82:1433- 1442.
BERGERON, D., M. DARVEAU, A. DESROCHERS, AND J.- P. L. SAVARD. 1997. Impact de l'abondance des chicots sur les communautes aviaires et la sauvagine des forets coniferiennes et feuillues du Quebec meridional. Technical Report Series Number 271F. Canadian Wildlife Service, Quebec City, Quebec, Canada.
BONAR, R. L. 2000. Availability of Pileated Wood pecker cavities and use by other species. Journal of Wildlife Management 64:52-59.
BOUVIER, J. M. 1974. Breeding biology of the Hooded Merganser in southwestern Quebec, including interactions with Common Goldeneyes and Wood Ducks. Canadian Field-Naturalist 88:323-330.
BUCKLAND, S. T., D. R. ANDERSON, K. P. BURNHAM, AND J. L. LAAKE. 1993. Distance sampling: estimating abundance of biological populations. Chapman and Hall, London, United Kingdom.
CALDWELL, P. J. AND A. E. SNART. 1974. A photographic index for aging Mallard embryos. Journal of Wildlife Management 38:298-301.
CARVER, B. C. 1958. The American Goldeneye in central New Brunswick. Wildlife Management Bulletin 2. Number 9. Canadian Wildlife Service, Ottawa, Ontario, Canada.
Dow, H. AND S. FREDGA. 1984. Factors affecting reproductive output of the goldeneye duck Bucephala clangula. Journal of Animal Ecology 53: 679-692.
DUGGER, B. D., K. M. DUGGER, AND t. H. FREDRICKSON. 1994. Hooded Merganser Lophodytes cucullatus. The birds of North America. Number 98.
EADIE, J. M. 1989. Alternative reproductive tactics in a precocial bird: the ecology and evolution of brood parasitism in goldeneyes. Thesis. University of British Columbia, Vancouver, Canada.
EADIE, J. M. AND G. GAUTHIER. 1985. Prospecting for nest sites by cavity-nesting ducks of the genus Bucephala. Condor 87:528-534.
EADIE, J. M., M. L. MALLORY, AND H. G. LUMSDEN. 1995. Common Goldeneye Bucephala clangula. The birds of North America. Number 170.
ERIKSSON, M. O. G. 1979. Clutch size and incubation efficiency in relation to nest-box size among goldeneyes Bucephala clangula. Ibis 121 : 107-109.
ERIKSSON, M. O. G. AND M. ANDERSSON. 1982. Nest parasitism and hatching success in a population of goldeneyes Bucephala clangula. Bird Study 29: 49-54.
ERSKINE, A. J. 1990. Joint laying in Bucephala ducks--"parasitism" or nest-site competition? Ornis Scandinavica 21:52-56.
FREDGA, S. AND H. Dow. 1984. Factors affecting the size of a local population of goldeneye Bucephala clangula (L.) breeding in Sweden. Viltrevy 13: 225-255.
GAUTHIER, G. 1989. The effect of experience and timing on reproductive performance in Buffleheads. Auk 106:568-576.
GAUTHIER, G. AND J. N. M. SMITH. 1987. Territorial behaviour, nest-site availability, and breeding density in Buffleheads. Journal of Animal Ecology 56:171-184.
GILMER, D. S., I. J. BALL, L. M. COWARDIN, J. E. MATHISEN, AND J. H. RIECHMANN. 1978. Natural cavities used by Wood Ducks in north-central Minnesota. Journal of Wildlife Management 42: 288-298.
HARAMIS, G. M. AND D. Q. THOMPSON. 1985. Density-production characteristics of box-nesting Wood Ducks in a northern greentree impoundment. Journal of Wildlife Management 49:429-436.
LUMSDEN, H. G., R. E. PAGE, AND M. GAUTHIER. 1980. Choice of nest boxes by Common Goldeneyes in Ontario. Wilson Bulletin 92:497-505.
LUMSDEN, H. G., J. ROBINSON, AND R. HARTFORD. 1986. Choice of nest boxes by cavity-nesting ducks. Wilson Bulletin 98:167-168.
MAISONNEUVE, C., R. McNICOLL, A. DESROSIERS, AND G. LUPIEN. 2002. Caracterisation de l'habitat de reproduction des canards arboricoles. Societe de la faune et des parcs du Quebec, Direction de la recherche sur la faune et Direction de l'amenagement de la faune du Saguenay-Lac Saint-Jean, Quebec City, Quebec, Canada.
MALLORY, M. L., H. G. LUMSDEN, AND R. A. WALTON. 1993a. Nesting habits of Hooded Mergansers Mergus cucullatus in northeastern Ontario. Wild fowl 44:101-107.
MALLORY, M. L., D. K. McNICOL, AND P. J. WEATHERHEAD. 1994. Habitat quality and reproductive effort of Common Goldeneyes nesting near Sudbury, Canada. Journal Wildlife Management 58: 552-560.
MALLORY, M. L., D. K. McNICOL, R. A. WALTON, AND M. WAYLAND. 1998. Risk-taking by incubating Common Goldeneyes and Hooded Mergansers. Condor 100:694-701.
MALLORY, M. L., A. TAVERNER, B. BOWER, AND D. CROOK. 2002. Wood Duck and Hooded Merganser breeding success in nest boxes in Ontario. Wildlife Society Bulletin 30:301-316.
MALLORY, M. L., P J. WEATHERHEAD, D. K. McNICOL, AND M. E. WAYLAND. 1993b. Nest site selection by Common Goldeneyes in response to habitat features influenced by acid precipitation. Ornis Scandinavica 24:59-64.
McNICOL, D. K., R. A. WALTON, AND M. L. MALLORY. 1997. Monitoring nest box use by cavity-nesting ducks on acid-stressed lakes in Ontario, Canada. Wildlife Biology 3:1-12.
MORSE, T. E., J. L. JAKABOSKY, AND V. P. MCCROW. 1969. Some aspects of the breeding biology of the Hooded Merganser. Journal of Wildlife Management 33:596-604.
NEWTON, I. 1994. The role of nest sites in limiting the numbers of hole-nesting birds: a review. Biological Conservation 70:265-276.
NILSSON, S. G. 1986. Evolution of hole-nesting in birds: on balancing selection pressures. Auk 103: 432-435.
POYSA, H. 1999. Conspecific nest parasitism is associated with inequality in nest predation risk in the Common Goldeneye (Bucephala clangula). Behavioural Ecology 10:533-540.
POYSA, H. AND S. POYSA. 2002. Nest-site limitation and density dependence of reproductive output in the Common Goldeneye Bucephala clangula: implications for the management of cavity-nesting birds. Journal of Applied Ecology 39:502-510.
POYSA, H., M. MILONOFF, V. RUUSlLA, AND J. VIRTANEN. 1999. Nest-site selection in relation to habitat edge: experiments in the Common Goldeneye. Journal of Avian Biology 30:79-84.
POYSA, H., P. RUNKO, V. RUUSILA, AND M. MILONOFF. 2001. Identification of parasitized nests by using egg morphology in the Common Goldeneye: an alternative to blood sampling. Journal of Avian Biology 32:79-82.
PRINCE, H. H. 1968. Nest sites used by Wood Ducks and Common Goldeneyes in New Brunswick. Journal of Wildlife Management 32:489-500.
RESEARCH UNIT FOR WILDLIFE ASSESSMENT. 1999. DISTANCE. Version 3.5. University of St. Andrews, Fife, Scotland.
SALTER, K. C. AND R. F. FAWCETT. 1993. The ART test of interaction: a robust and powerful rank test of interaction in factorial models. Communications in Statistics: Simulation and Computation 22:137- 153.
SAS INSTITUTE. 1999. SAS/STAT user's guide. Version 8.0. SAS Institute Inc., Cary, North Carolina, USA.
SAVARD, J.-P. L. 1984. Territorial behaviour of Common Goldeneye, Burrow's Goldeneye and Bufflehead in areas of sympatry. Ornis Scandinavica 15: 211-216.
SAVARD, J.-P. L. 1988. Use of nest boxes by Burrow's Goldeneyes: nesting success and effect on the breeding population. Wildlife Society Bulletin 16: 125-132.
SEMEL, B. AND P W. SHERMAN. 2001. Intraspecific parasitism and nest-site competition in Wood Ducks. Animal Behaviour 61:787-803.
SEMEL, B., P W. SHERMAN, AND S. M. BYERS. 1988. Effects of brood parasitism and nest-box placement on Wood Duck breeding ecology. Condor 90:920-930.
SEMEL, B., P W. SHERMAN, AND S. M. BYERS. 1990. Nest boxes and brood parasitism in Wood Ducks: a management dilemma. Pages 163-170 in Proceedings of the 1988 North American Wood Duck Symposium (L. H. Fredrickson, G. V. Burger, S. P. Havera, D. A. Graber, R. E. Kirby, and T. S. Taylor, Editors). St. Louis, Missouri, USA.
SENECHAL, H. 2003. Etude comparative de l'ecologie de nidification et d'e1evage des couvees de Garrot a oeil d'or (Bucephala clangula) et de Harle couronne (Lophodytes cucullatus) dans un habitat de riviere. Thesis. Universite Laval, Quebec, Canada.
HELENE SENECHAL, (1) GILLES GAUTHIER, (1,3) AND JEAN-PIERRE L. SAVARD (2)
(1) Departement de Biologie and Centre d'etudes Nordiques, Pavilion Vachon, 1045 Avenue de la Medecine, Universite Laval, Quebec, PQ G1V 0A6, Canada.
(2) Science and Technology, Environment Canada, 1141, Route de l'Eglise, C.P. 10100, [9.sup.e] etage, Quebec, PQ, G1V 4H5, Canada.
(3) Corresponding author; e-mail: email@example.com
TABLE 1. Mean [[+ or -] SE] clutch size (sample size) of the host species only and total clutch size (including interspecific parasitic eggs) in incubated nests of Common Goldeneyes and Hooded Mergansers, Ste. Marguerite River, Quebec, Canada, 2000-2002. Host species clutch only Species 2000 2001 2002 Common 9.2 [+ or -] 0.5 9.7 [+ or -] 0.5 10.3 [+ or -] 0.8 Goldeneye (23) (31) (24) Hooded 10.0 [+ or -] 1.2 8.7 [+ or -] 0.5 8.9 [+ or -] 0.7 Merganser (10) (13) (10) Total clutch Species 2000 2001 2002 Common 9.8 [+ or -] 0.6 10.0 [+ or -] 0.6 11.0 [+ or -] 0.9 Goldeneye (23) (31) (24) Hoode 12.7 [+ or -] 0.9 11.5 [+ or -] 1.0 10.4 [+ or -] 1.2 Merganser (10) (13) (10) TABLE 2. Percent nesting success (sample size) of Common Goldeneyes and Hooded Mergansers in nest boxes, Ste. Marguerite River, Quebec, Canada, 2000-2002. Parasitized nests (a) Host species 2000 2001 2002 Common Goldeneye 86 (7) 67 (15) 85 (13) Hooded Merganser 100 (8) 88 (8) 75 (4) Non-parasitized nests Host species 2000 2001 2002 Common Goldeneye 79 (24) 64 (22) 50 (24) Hooded Merganser 100 (2) 67 (6) 67 (6) All nests Host species 2000 2001 2002 Common Goldeneye 81 (31) 65 (37) 62 (37) Hooded Merganser 100 (10) 79 (14) 70 (10) (a) Intraspecific and interspecific nest parasitism. TABLE 3. Mean [[+ or -] SE] percent hatching success (sample size) of Common Goldeneyes and Hooded Mergansers in successful nest boxes, Ste. Marguerite River, Quebec, Canada, 2000-2002. Parasitized nests (a) Host species eggs Species 2000 2001 2002 Common 84 [+ or -] 16 83 [+ or -] 4 79 [+ or -] 8 Goldeneye (5) (2) (7) Hooded 85 [+ or -] 9 85 [+ or -] 9 100 Merganser (6) (7) (2) Parasitized nests (a) Heterospecific parasitic eggs Species 2000 2001 2002 Common 100 76 [+ or -] 10 67 [+ or -] 17 Goldeneye (6) (7) (2) Hooded 93 [+ or -] 7 83 [+ or -] 17 87 [+ or -] 10 Merganser (5) (2) (7) Non-parasitized nests Species 2000 2001 2002 Common 97 [+ or -] 2 91 [+ or -] 2 96 [+ or -] 3 Goldeneye (17) (22) (15) Hooded 90 [+ or -] 5 89 [+ or -] 8 90 [+ or -] 5 Merganser (4) (4) (5) (a) Interspecific nest parasitism. TABLE 4. Characteristics explaining use by Common Goldeneyes and Hooded Mergansers for nest boxes used at least 2 years, Ste. Marguerite River, Quebec, Canada, 1998-2002. '+' indicates a positive effect and '-' indicates a negative effect. Numbers in parentheses are the number of used vs. unused boxes. Species Variables (a) Effect P Common SURFPOND + 0.003 Goldeneye ORIEN S + 0.047 DISWAT - 0.004 DISTREE + 0.032 (37 vs. 27) Hooded CANOP + <0.001 Merganser DISTREE + 0.006 (8 vs. 55) (a) SURFPOND: total surface area of ponds, ORIEN: compass direction faced by the nest box entrance (S = South), DISWAT: distance to the water edge, DISTREE: distance to the nearest tree [greater than or equal to] 10-cm diameter, and CAN OP: canopy height. FIG. 2. Common Goldeneye and Hooded Merganser nests parasitized by a heterospecific, Ste. Marguerite River, Quebec, Canada, 1998-2002. Numbers in parentheses are number of nests. Rate of interspecific parasitism (%) Common Goldeneye Hooded Merganser 1998 (2) (2) 1999 (27) (7) 2000 (30) (10) 2001 (37) (14) 2002 (36) (10) Note: Table made from bar graph. FIG. 3. Nests that contained eggs that were laid parasitically (either intra- or interspecifically) by Common Gondeneyes among all nests initiated by this species (i.e., nests where gondeneyes laid at least 1 egg), and vice-versa for Hoodedn Mergansers, Ste. Marguerite River, Quebec, Canada, 2000-2002. Numbers in parentheses are number of nests. Nests with parasitic eggs (%) Common Goldeneye Hooded Merganser 2000 (39) (18) 2001 (57) (18) 2002 (44) (20) Note: Table made from bar graph.
|Printer friendly Cite/link Email Feedback|
|Author:||Senechal, Helene; Gauthier, Gilles; Savard, Jean-Pierre L.|
|Publication:||The Wilson Journal of Ornithology|
|Article Type:||Brief article|
|Date:||Dec 1, 2008|
|Previous Article:||Lesser snow geese and Ross's Geese form mixed flocks during winter but differ in family maintenance and social status.|
|Next Article:||Movements and habitat use by Red-breasted Merganser broods in eastern New Brunswick.|