Northern saw-whet owl (Aegolius acadicus brooksi) COSEWIC update and status report: chapter 5

• Previous page
• Table of contents
• Next page

Habitat

Habitat requirements

Breeding habitat

In a systematic survey of habitat use by Haida Gwaii Saw-whet Owls during the breeding season, Gill and Cannings (1997) found that owls were detected more often at points closer to riparian areas and in habitats containing greater amounts of mature and old forest. They found that elevation, distance to saltwater and distance to a change in habitat type did not influence the likelihood of detecting an owl. As well, the condition of male owls in the breeding season appears to be appreciably lower when territories contain less mature and old forest. When the mature and old forest content within 500 m of a territory core falls below 60-70%, significant declines in calling rate, an indirect measure of condition, of the territory occupants have been noted (Holschuh 2004b, Holschuh and Otter submitted). Active owl sites were all found below 300 m elevation despite searches that included higher elevations. Active sites have been mostly in the CWHwh1 (submontane wet hypermaritime coastal western hemlock variant) and CWHvh2 (central very wet hypermaritime coastal western hemlock variant) biogeoclimatic zones. Although some owl sites have been located in the higher elevation CWHwh2 (montane wet hypermaritime coastal western hemlock variant), none have been detected in the mountain hemlock or alpine tundra biogeoclimatic zones on Haida Gwaii. Although survey logistics largely inhibit placing call playback surveys in these high-elevation habitats, calling owls at higher elevations would have been in earshot during numerous surveys conducted at lower elevations.

In habitat surveys of 25 active sites in 2002 and 2003, Holschuh (2004) found most sites to have structurally complex forest, namely mature or old forest types with relatively abundant snags. The mean height of veteran trees was 37.4 m (± 8.74 m S.D.), while the main tree canopy was 28.2 ± 7.24 m and the subcanopy 17.5 ± 5.53 m in height. The main tree canopy generally had the greatest density (25-50% cover), while other tree layers generally had less than 25% cover. The most common tree species of the occupied sites were Western Hemlock (Tsuga heterophylla), Sitka Spruce (Picea sitchensis), Western Redcedar (Thuja plicata), and to a lesser degree Mountain Hemlock (Tsuga mertensiana), Yellow Cedar (Chamaecyparis nootkatensis) and Shore (Lodgepole) Pine (Pinus contorta).

Overall, breeding habitat of the Saw-whet Owls on Haida Gwaii includes mature and old forest; these forests characteristically are structurally complex (Gill and Cannings 1997, Holschuh 2004). The upper canopy is generally composed of older trees that have persisted through disturbances in combination with a younger, regenerating layer of shade-tolerant trees at various levels of establishment (Meidinger 1999). It is these structurally complex old forests that tend to be richest in appropriate nesting cavities, which appear to be a limiting factor across the landscape (Doyle unpubl. data).

Feeding habitat requirements are likely similar to those of the nominate subspecies, where owls forage in openings or along forest edges (Cannings 1993). For instance, an unmated male that was radio-tagged in late April was found along forest edge, >1 km from the territory core, and was presumably foraging along the riparian and road corridors (Holschuh and Otter, unpubl. data). Mature and old forest habitats also provide a more open interior ideal for foraging, while young, dense forests are generally avoided (Cannings 1993).

I calculated available habitat on Haida Gwaii through data from the Haida Gwaii/Queen Charlotte Islands Land Use Plan Background Report (LUP 2003). Assuming few if any owls would be in the Mountain Hemlock or Alpine Tundra zones, areas reported as containing old forests in the Western Hemlock Zone were added together. Using this technique, there were 2927 km² in landscapes with ongoing forest harvest and 2561 km² in landscapes with no forest harvest at present.

Non-breeding habitat

Habitat use by Northern Saw-whet Owls outside the breeding season has not been systematically documented on Haida Gwaii. Nevertheless, data on the diet of birds collected primarily during the fall (road collisions), suggest high levels of marine invertebrate consumption (Hobson and Sealy 1991, Sealy 1999), suggesting that the owls may be shifting to habitats closer to coastal areas during the fall and winter (S. Sealy, pers. comm.).

Habitat trends

The amount of contiguous, suitable old forest habitat is declining on Haida Gwaii largely due to forest harvest. Many of the most productive watersheds on interior Graham Island and northern Moresby Island have been logged extensively, and the regenerating forest currently lacks the structural attributes necessary for high-quality habitat (snags, open flyways for foraging, etc.). Although forestry activities are less focused on eastern Graham Island, much of the area has extensive boggy areas, also unsuitable as habitat during the breeding season (see Figure 4). Currently, approximately 5,500 km² of the total 10,000 km² area of Haida Gwaii is potentially suitable habitat, down from approximately 6,300 km² in 1993, an almost 13% decrease over the last 10 years. This information is derived from a BC Forest Service Forest Inventory (1993) and the Haida Gwaii/Queen Charlotte Islands Land Use Plan Background Report (LUP 2003). Current projections suggest forest harvest rates of about 2,000 ha/year, concentrated mainly in old-growth habitats. Figure 5 shows annual harvest rates over the past century.

Habitat protection/ownership

On Haida Gwaii approximately 3,250 km² are currently managed as Tree Farm Licences (TFL 39, TFL 25, TFL 47) and 4,250 km² are classified as provincial Timber Supply Area. The remaining 2,250 km² of land are protected parks and/or reserves. The largest protected area is Gwaii Haanas National Park Reserve, occupying the southern half of Moresby Island and adjacent islands, with a total area of 1,470 km². The north-eastern corner of Graham Island is protected from resource extraction in Naikoon and Pure Lake Provincial Parks (690 km²). It should be noted that although this area provides areas of good habitat, much of these latter two parks is boggy habitat, which does not provide many of the attributes associated with suitable breeding habitat. There are a further 93 km² of ecological reserves scattered around Haida Gwaii (LUP, 2003).

About 75% of the Haida Gwaii landscape is available for forest harvest and approximately 25% of Haida Gwaii has already been harvested. The areas of highest harvest density are also the most productive watersheds on Haida Gwaii, and may have provided some of the best breeding habitat on the island. The density of owls throughout the patchy harvested landscape appears to be considerably lower (Holschuh 2004a; see Abundance) than in unharvested areas. Although the population in Gwaii Haanas appears to be at higher density than other areas surveyed in the past, the area of Gwaii Haanas may not be sufficient to sustain this endemic subspecies.

Twelve Wildlife Habitat Areas (WHAs) have been proposed for the Northern Saw-whet Owl on the Queen Charlotte Islands (D. Fraser, pers. comm.). These cover 901 ha and correspond to 12 owl breeding territories located through surveys. If approved, these WHAs would exclude forest harvest and road development within their boundaries. Proposed WHAs for Northern Goshawk and Marbled Murrelets might also protect an unknown number of Saw-whet Owl territories if approved.