Michael A. Carson

Victoria Natural History Society

Victoria, British Columbia

October 1994

Summary

This report documents the occurrence and distribution of Black- headed Grosbeaks (BHGR) in the Greater Victoria area during the week of June 1-8, 1994 in the early breeding season.

The survey was undertaken by 18 volunteers covering 29 areas of likely habitat (low-lying open deciduous woodland or forest edge). It was supplemented by additional coverage of unknown extent by birders who utilize the Victoria Natural History Society's Rare Bird Alert, though, in fact, the only reports provided from the alert system coincided with observations made in the regular survey.

Only 8 of the 29 areas reported BHGR occurrence, the species being an uncommon breeder in southeastern Vancouver Island. Female BHGR observations totalled only 3, this small number reflecting the weaker, less frequent song of the female and its less spectacular plumage. A total of 19 is the estimate for the number of male BHGR observed in the survey.

Three areas accounted for 80% of the observations: the lower, southeast slopes of Mount Newton, the woodland edges of Elk and Beaver Lake Park, and the Blenkinsop Lake area. The latter area accounted for at least 7 male BHGR, more than one third of the 19 records of male BHGR in the survey.

This clustering of observations is attributed to the relative scarcity in the Victoria region of large areas of suitable habitat, together with the tendency of the species to return each year for breeding to the general area of their own birth.

The survey confirms the major importance of the linear willow- alder habitat along Lochside Trail in the vicinity of Blenkinsop Lake for BHGR breeding, a finding long known in a general way among local birders.

Any conversion of the Blenkinsop-Lochside Trail from its current status as a quiet dead-end country walk into a major recreational through corridor, with trail widening (and vegetation removal) to accomodate multi-purpose trail use is likely to have a negative impact on this breeding habitat, and may drastically reduce the already low numbers of this breeding species in the Greater Victoria area.

Background

Black-headed Grosbeak (BHGR) is indicated as an uncommon summer breeding species in the Southeastern Vancouver Island checklist (Gates and Taylor, 1989). It overwinters in Mexico. The species breeds in open deciduous vegetation, near surface water, and, as such, is found in the checklist area primarily in the more extensive lowland parts around Cowichan Bay and Duncan rather than in the Victoria area.

Within Greater Victoria, regular reports of BHGR exist for the Blenkinsop Lake area, especially on Lochside Trail to the north of the lake, but elsewhere observations have been spotty. Up to six singing male BHGR were frequently observed on (or from) this part of Lochside Trail in the summer of 1993, and over the years the Blenkinsop Lake area has provided an easily accessible venue for birders wishing to enjoy the colourful plumage and even more colourful song of this species.

The mixed hedgerow-tree habitat presently enjoyed by the grosbeaks and other species along this part of Lochside Trail is threatened, however, by plans for a combined equestrian, pedestrian and cycle artery for through traffic along the trail using a proposed bridge over Blenkinsop Lake. Ultimately, as this artery grows in popularity, widening of the trail will become essential in order to allow safe coexistence of all the different types of through traffic. Such widening will thin the hedgerow vegetation to a point where it may no longer provide sufficient cover for nesting species such as the grosbeaks.

With this perspective, it seemed appropriate to the Parks and Conservation Committee of the Society to undertake a survey of BHGR during the breeding season throughout all likely habitats in the Greater Victoria area. The purpose of the survey was to determine the number of BHGR present during the breeding period and their distribution. This would permit more accurate assessment of the importance of the Blenkinsop-Lochside Trail population.

The survey

The survey was undertaken by 18 volunteers during the period June 1 to June 8. This period was chosen for two reasons.

The first is that BHGR is a relatively late spring migrant, and, prior to this time, it seems likely that some of the birds observed are simply passing through the area. This suspicion is confirmed by the absence of observations in early June in areas where BHGR were seen during May. There is some possibility that migrants are still moving through even in the first week of June: in 1993 a BHGR was seen by Jeff Gaskin on the top of Mount Tolmie on June 7 and on top of Mount Douglas on June 14, neither of which can be considered likely breeding habitats. It was decided not to delay the survey further into June, however, because of the second factor.

The second point is that experience with previous surveys indicated that the number of BHGR heard during the spring and summer decreased steadily over time, presumably because of the decline in courtship songs during the summer as both males and females become more involved with incubation, brooding and feeding of young. Indeed, breeding among some birds may start quite early: David Fraser reported seeing a female grosbeak on Lochside Trail south of Martindale Road carrying a caterpillar in her bill as early as May 13 this year! It is hard to believe that she was actually feeding young, however, given the usual time lag of at least two weeks between start of nest construction and hatching of eggs. Perhaps this was simply part of courtship feeding. A female carrying flies in her bill was seen at Rithet's Bog on June 26, 1993; this is more consistent with the view that breeding took place in early-mid June.

The various observations above indicate that it is possible that both breeding birds and migrants are present during the first week of June, but this week seemed to be the best compromise.

The survey was not restricted to a specific day. Observers were free to choose whichever day was convenient. The only disadvantage of this multi-day approach is if there is inter-area movement of grosbeaks during the week allowing the same bird to be counted more than once. It was assumed that the risk of this was minimal, given that the birds should be establishing and using territories by this time.

In addition to this survey of specific areas, notice of the survey was given on the Rare Bird Alert so that any person observing BHGR during that week, in areas not being covered by the survey, could phone in their observations.

Observations were split by sex. This is easy when the bird is seen, but the chances of a bird being detected by sight in the thick deciduous vegetation of June are low. Most birds are heard first and seen later, if at all. It is potentially more difficult to distinguish sex when the bird is simply heard because both sexes sing. However, the normal songs of the two sexes are quite different, though the song of the male is quite variable and it does sing while incubating eggs on the nest. According to Weston (1947) males alternate the incubation with females, sitting for about 20 minutes in an average incubation period, so that without such vocalization it would be possible to miss a male BHGR at such times.

The results: inventory accuracy

Sampled areas

In areas where only one or no grosbeaks were observed, there is clearly some chance that birds were simply missed. This could arise for several reasons. One is that birds may have been temporarily in other parts of their territories away from the route being taken by the observer. A second is that the bird may not have been singing at the time of the survey.

As an example, Barbara Begg could find no BHGR along Creswell Road during the first week of June, but did observe one in the cottonwoods there between June 17 and 23, and heard one on July 4. Whether the bird was actually there during the survey or simply wandered in later is not known. As a second example, repeated surveys along the full length of Lochside Trail in the Blenkinsop Valley north of the lake in the first two weeks of June indicated numbers of males to range from as high as 5 to as low as two. These two extremes were recorded under virtually identical survey conditions: between 9.00 and 10.30 on overcast, windless mornings.

It is difficult to see how this statistical uncertainty could be removed from any survey without repeated coverage through each area. This is beyond the resources of any amateur naturalist group. However, almost all areas surveyed were in the early-mid morning and some areas were, in fact, repeated once or twice.

In cases where more than one male was recorded, there is the same possibility of underestimation. However, in this case, there is the danger of overestimation because of the same bird being counted twice. This arises because of the large territory that the birds apparently use.

In the Blenkinsop Valley, for example, one male was seen to fly right across the valley for a distance of about 500 metres. Thus, when two birds are observed at a spacing along the trail of only 150-200 m, the question arises as to whether the second observation is simply the result of movement of the first bird. Weston (1947) reports BHGR nests in California being spaced as close as 100 m, so such close spacing of two males is certainly possible.

The problem of double-counting is to some extent minimized by the powerful song of the male BHGR enabling it to be heard up to 150 metres away in many cases: thus a second bird that is 300 m away from the first bird can frequently be heard singing at the same time as the first bird. At spacings of more than 300 metres, however, this assurance no longer exists: in this case verification of two birds requires back-tracking to ensure that the first bird is still there. Though back-tracking along a trail is a simple enough precaution to take, back-tracking through head- high swampy grasses is less attractive. Moreover the problem is complicated by the fact that the volume of a male BHGR song is quite variable: in some cases it may not be heard until less than 50 m away.

The distribution of male BHGR observations in the Blenkinsop Valley is shown in the attached map. A to E were observed on the morning of June 4 in a traverse of Lochside Trail from McKenzie to Lohbrunner; and F, G, H and I were observed the next morning in traversing from the "Trail Head" to Royal Oak Drive. G and J were observed the next evening in traversing the same stretch in reverse. There is a chance that one or more of the birds seen in the area north of Lohbrunner was the same as a bird seen south of the Trail Head on the previous day, but it is thought that this is unlikely. In two years of birding along Blenkinsop Trail no BHGR has been seen (by me) in or flying through the open area between the Trail Head and Lohbrunner. Some of the observations within these two areas were, however, thought to be the result of double- counting. It was felt that J and I were probably the same bird (though the distance between them was 350 m) and that G and H were also the same bird. Observations E and F were also regarded as the same bird. The estimate of 7 males for the Blenkinsop Lake area is therefore possibly a conservative one. A separate estimate for just Lochside Trail between the lake and the Trail Head made by Ron Satterfield during the early part of June put the figure at 5 males, two more than the highest figure found in the survey and used here.

In the Newton Heights area (on the southeast lower slopes of Mt Newton) it is also possible that the total of 3 males recorded for that area is too low. Three were noted by Sheila Mosher on a single walk, and a fourth in a different location on another day. A fifth was heard from Haldon Road by a separate observer but probably corresponds to one already recorded. Three of the birds seen (two males and one female) were found on a 10-acre property north of Haldon Road, the two males apparently feeding in the trees together. Three birds (two males and a female) were similarly reported from a sunflower seed feeder in Dean Park Estates on the Miyasaki residence, about 1000 metres north of the Haldon Road property, the two males again feeding together. It has been assumed here that these two groups of birds are actually the same. The Dean Park birds were seen fairly regularly from May 12 on, but were not seen after June 2, perhaps because nesting had already begun in the lower area.

The problem of counting is thus complicated considerably by the apparently variable (and large) territories established by the birds. Reference to Weston (1947) and Bent (1968) provided no information on the extent of BHGR territories. The problem is compounded by the fact that not all males appear to have distinct territories. This is indicated by the pair of males at Newton Heights and at Dean Park Estates. Similarly two males and one female were seen by Colleen O'Brien, moving around together on the woodland edge of Beaver Lake near Jennings Lane. The sociological significance of these male pairs is unclear. From an inventory standpoint, it casts doubt on the assumption that one song means one territory and only one male.

Unsampled areas

The present survey clearly did not cover all likely areas of BHGR breeding occurrence in Greater Victoria. Many small pockets of open deciduous woodland in moist low-lying areas could not be sampled because of insufficient observers. Among these might be noted the shores of Tod Inlet, stands of alder and cottonwood on the east side of Martindale Valley (where one male BHGR was singing in May), low areas around Prospect Lake and Trevlac Pond, and the margins of Florence Lake and Langford Lake.

Suitable habitat in most of these areas is, however, generally quite small in extent. The survey did include many such small habitats and in general most of these small pockets of deciduous woodland did not reveal grosbeaks. As indicated in the next section, most of the BHGR observations were clustered in a few large areas of suitable habitat. Thus, though the total number of male BHGR obtained for the Greater Victoria area is obviously a minimum value, the true number may not be significantly greater than this.

The results: distribution pattern

The main survey produced observations of 19 (or perhaps 20) males and 3 females, recognizing that this may be a slight underestimate for reasons given in the first section dealing with the accuracy of the inventory. Females are, of course, much more difficult to observe because of their weaker, shorter and more infrequent song, so that no special significance should be attached to the female count. The only BHGR observations reported to the Rare Bird Alert during this week were all in areas covered by the main survey.

The survey data show three distinct clusters of BHGR: the Newton Heights area of Central Saanich, the woodland edges of Beaver-Elk lakes and the Blenkinsop Lake area. These three areas provided 80% of the male BHGR heard or seen in the regular survey. It is probably not coincidence that the appearance of a Rose-breasted Grosbeak (the eastern equivalent of the Black-headed) during the week of the survey (and only the fourth record for Vancouver Island) coincided with one of these three clusters, on the east side of Beaver-Elk Lake park.

All three areas correspond to classic BHGR habitat as described by Bent (1968) and others: low-lying terrain, near surface water, covered with open woodland (or woodland edge) of mostly deciduous trees. Willows, alder and maples were the source of most bird observations.

The reason for the preference for this habitat does not appear to have been established. Bent (1968) notes that, despite the thick bill, seeds account for less than 15% of food intake. Almost 60% of food intake is animal matter (53% being insects, especially beetles), while vegetable matter is dominated by elderberry and orchard fruit where available.

Whatever the underlying reason for this habitat preference, such habitat is increasingly scarce on the Saanich Peninsula and this must be regarded as a major constraint on BHGR numbers here. Yet from the findings of this survey, habitat does not appear to be the major constraint because several large areas of suitable habitat showed no grosbeaks. According to Jeff Gaskin, the grounds of the University of Victoria and Queen Alexandra Hospital, as well as Swan Lake, all provided one or two observations last summer during the same period, but none was found there in this survey. Other sightings in previous years from late June on may have simply been post-breeding dispersal. Bent (1968) indicates that incubation lasts only 12 days and the nestling period a similar time, so that by late June breeding may be essentially over, at least for some birds.

The distinct clustering of BHGR observations is also presumably related to bird biology. According to Bent (1968), BHGR seem to repeat their nesting at the same sites each year. In addition, there is a tendency, in some species, for offspring to return to the general area of their birth the following year. This has certainly been documented for Northern Orioles (Sealy, 1980) and may well apply to grosbeaks too. Such behaviour would seem to offer many advantages. In particular, assuming that the driving force in the breeding period is the urge to find a mate, the best chances of success are likely to be in areas where breeding has occurred in past years. For this and other reasons, new breeding adults would be expected to return to the place of their own birth to breed. In this way, once a cluster becomes established it would tend to perpetuate itself. With this kind of argument it is difficult, in fact, to understand why birds would try to disperse far away from their own general breeding area until such time as numbers become too large to accommodate adequately-sized territories.

Isolated observations at places such as Albert Head Lagoon may be the result of accidental wandering. In other cases, such as Rithet's Bog, this year's sighting in June (1 male), repeats sightings in the previous year (about 6 males throughout June in 1993). The much smaller number this year may be related to over- wintering mortality or to lack of breeding success last year. The large number of BHGR present at Rithet's Bog in June 1993 was somewhat surprising in view of the relatively small amount of deciduous tree growth there.

Conclusions

The 1993 June 1-8 survey produced about 20 male Black-headed Grosbeak observations in the Greater Victoria area. Not all of these were necessarily breeding birds. Additional males may have been missed because of the limited amount of observer time possible, but judging by the large number of negative observations in likely habitats it is thought that undetected birds were relatively few.

Most of the birds were concentrated in three distinct clusters: one on the south-east lower slopes of Mount Newton; another in the woodland edges around Beaver and Elk Lakes; and the last around Blenkinsop Lake and Lochside Trail immediately north of the lake.

It is difficult to overemphasize the significance of the Blenkinsop-Lochside population. The area accounted for seven males (more than a third of all male BHGR observations in the survey), five of the seven males occupying sites along Lochside Trail itself.

It can only be hoped that the trail-side vegetation in the Blenkinsop area is left intact in the future. Present plans to widen the trail to make it a multipurpose recreational corridor can certainly be expected to jeopardize the Black-headed Grosbeak population. Indeed, rather than hedgerow thinning, it would be advantageous to have additional tree plantings along parts of the trail, especially where gaps have already appeared.

Acknowledgements

This survey was made possible through the participation of the following volunteer observers: David Allinson, Jerry and Gladys Anderson, Gerry Ansell, Barbara Begg, Darren Copley, Aziza Cooper, David Fraser, Jeff Gaskin, Tom Gillespie, Gordon Hart, Anne Knowles, Ev Miyasaki, Sheila Mosher, Colleen O'Brien, Jim Phillips and Ron Youngash. Thanks are also due to Bryan Gates for providing coverage on the Rare Bird Alert.

References

Bent, A.C., 1968. Life Histories of North American Birds. U.S. National Museum Bulletin No. 237: Cardinals, Grosbeaks, Buntings, Towhees, Finches, Sparrows and Their Allies, Part 1.

Gates, B.R. and K. Taylor, 1989. Checklist of birds: Victoria and Southeastern Vancouver Island. Victoria Natural History Society.

Sealy, S.G., 1980. Reproductive responses of Northern Orioles to a changing food supply. Canadian Journal of Zoology, v. 58, p. 221-7.

Weston, H.G. Jr., 1947. Breeding behaviour of the Black-headed Grosbeak. Condor, v. 49, p. 54-73.