• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Habitat Use By Riparian and Upland Birds in Old-Growth Coastal British Columbia Rainforest

Posted on: Friday, 30 September 2005, 06:00 CDT

By Shirley, Susan M

ABSTRACT.-The value of riparian habitats to birds differs among ecosystems. I tested whether riparian habitat near large streams and rivers in the Pacific Northwest supported a higher abundance and diversity of birds than adjacent upland forest. From 1996 to 1998, I surveyed breeding birds at four 9-ha sites in coastal western hemlock forest on western Vancouver Island, British Columbia. Five species of forest generalists dominated both riparian and upland bird communities: Winter Wren (Troglodytes troglodytes), Chestnut- backed Chickadee (Poecile rufescens), American Robin (Turdus migratorius), Swainson's Thrush (Catharus ustulatus), and Pacific- slope Flycatcher (Empidonax difficilis). Species richness and total abundance were similar over the riparian-to-upland gradient, whereas abundances of riparian specialists and aerial foragers declined with distance from the river. To explore whether vegetation composition and structure explained bird distribution patterns, I sampled three locations along both riparian and upland transects at each site. Riparian areas had higher densities of deciduous trees; conifer and snag densities were higher in upland areas. Salmonberry (Rubus spectabilis) cover was marginally higher in riparian areas and blueberry (Vaccinium spp.) cover was higher in upland areas. There was little effect of distance from the river on most bird species, but there were stronger associations of birds with specific vegetation attributes. Tree and snag densities explained the most variation in abundance of aerial foragers, and eight of nine individual species, whereas distance from the river and shrub cover were important predictors of Hammond's Flycatcher (Empidonax hammondii) abundance. Apart from riparian specialists and a few species with strong vegetation associations, bird assemblages in riparian and upland habitats of this moist forest type were dominated by similar sets of generalist species. Received 1 December 2003, accepted 26 April 2005.

Riparian habitats are influenced by both stream channel processes and the adjacent upland vegetation (Brinson et al. 1981, Naiman et al. 1993). Topography, plant communities, hydrologie regimes, and soil type typically distinguish riparian areas from upland areas. Riparian habitats are heavily influenced by seasonal changes in water flow, and alluvial soils in riparian habitats tend to be wetter than soils in uplands. Riparian plant communities have diverse vegetation structures, high edge: area ratios, and are dominated by woody vegetation. These features are common to all riparian habitats, but vary greatly depending on geographical location. Riparian ecosystems often support high bird diversity and abundance (Thomas et al. 1979, Knopf et al. 1988, Anthony et al. 1996) because of their complex vegetation structure (LaRue et al. 1995, Wiebe and Martin 1998), high plant diversity (Bull 1978, Raedeke 1988), and proximity to water.

There is a strong bird diversity gradient from riparian to upland habitats in southwestern and agricultural regions of the U.S. (e.g., Stauffer and Best 1980, Szaro 1980, Knopf 1985, reviewed in Knopf and Samson 1994), where bird diversity is higher in riparian and lower in upland areas. In contrast, studies of mature, undisturbed stands in forests with greater annual precipitation (McGarigal and McComb 1992, Murray and Stauffer 1995, Wiebe and Martin 1998) have shown equal or lower diversity in riparian habitats compared with upslope habitats; these studies (McGarigal and McComb 1992, Murray and Stauffer 1995, Wiebe and Martin 1998), however, focused on riparian areas associated with small (<5 m wide) mountain streams. Some riparian areas show greater diversity near larger streams and rivers (Knopf and Samson 1994, Lock and Naiman 1998), and the avian use of riparian habitat relative to upland habitat along larger streams and rivers in the Pacific Northwest has not been well examined. I studied avian habitat use along larger streams and rivers within continuous undisturbed forest of the Pacific Northwest.

My first objective was to test the hypothesis that bird species diversity and abundance is higher in old-growth riparian habitat associated with large streams and rivers than in adjacent old- growth upland habitat. In the Pacific Northwest, riparian zones tend to be dominated by deciduous trees, whereas uplands are dominated by conifers (McGarigal and McComb 1992, Pearson and Manuwal 2001). Disproportionate use of riparian habitats by birds should be reflected in a decline in species richness and abundance with increasing distance from the river, but association with riparian habitat may vary among species and guilds. Riparian specialists that rely on the stream or river as a food source and nest near streams should decline in abundance with increasing distance from the river. Riparian forest edges along streams support a higher invertebrate biomass (Murakami and Nakano 2002), due to higher densities of aquatic insects (Murakami and Nakano 2002), and possibly greater primary productivity (Ranney et al. 1981). Aerial foragers such as flycatchers may respond to emergent aquatic insects near streams and rivers (Gray 1993) and should occur at highest densities near the water's edge. Conversely, conifer specialists may increase with distance from the river due to increasing conifer densities (McGarigal and McComb 1992, Pearson and Manuwal 2001).

Second, I explored how variation in vegetation composition and structure from riparian to upland habitat explains distribution patterns of several species. Studies conducted in other temperate coniferous forests have revealed differences in vegetation structure and composition between riparian and upland (McGarigal and McComb 1992, Pearson and Manuwal 2001). If the structural and species attributes of riparian vegetation communities are the primary predictors of bird diversity and abundance, then use of riparian habitats should be related to the prevalence of these structures relative to upland areas. Alternatively, bird diversity and abundance should not differ between riparian and upland habitats that are similar in vegetation structure and composition.

METHODS

Study area.-The study was conducted in three valleys on the west coast of Vancouver Island, British Columbia, between Ucluelet in the north and Bamfield in the south (48 5' N, 125 5' W). Four sites of continuous old-growth forest were selected along the Nahmint (n = 2; 2 km apart), Taylor (n = 1), and Klanawa rivers (n = 1). The sites were embedded within a mosaic of forest patches of different ages across a landscape in which the amount of primary forest varied from 50 to 70%. I classified rivers to stream order at a 1: 50,000-map scale based on branching following Kuehne (1962). The Nahmint and Taylor rivers (fourth order) and the Klanawa River (fifth order) ranged in width from 15 to 57 m. Study sites were located in the Western Vancouver Island ecoregion, within the moist-to-very-wet maritime biogeoclimatic subzones of coastal western hemlock (Klinka et al. 1991, Nuszdorfer and Boettger 1994). The forest is dominated by amabilis fir (Abies amabilis), western hemlock (Tsuga heterophylla), and western red cedar (Thuja plicata). Red alder (Alnus rubra) and bigleaf maple (Acer macrophyllum) occur at their greatest densities adjacent to the rivers, but were also scattered throughout the forest in moister areas. The understory is dense, highly stratified, and contains shrubs such as salmonberry (Rubus spectabilis), red huckleberry (Vaccinium parvifolium), salal (Gaultheria shallon), and devil's club (Oplopanax horridus), with Alaskan (Vaccinium alaskense) and oval-leaf blueberry (V. ovalifolium) predominating in the upland areas. The climate is cool and wet in winter and warm and dry in late summer (July-September). Annual precipitation in the area averages 3,100 mm, and daily temperatures average 3.2C in January and 15.6C in July (Environment Canada Climate Data Services).

Vegetation sampling.-I sampled vegetation during 1995 within 20- m-radius circular plots (0.13 ha) using a procedure modified from James and Shugart (1970). I accounted for the large size of trees locally by increasing the plot radius (Mueller-Dombois and Ellenberg 1974, Bryant et al. 1993). To improve the accuracy of visual estimation over a large area, each plot was divided into four quadrats. I sampled vegetation in each of the four quadrats and then calculated means of the four quadrats for each variable. Plots were placed at three stations 150 m apart along two 500-m transects: one in riparian and one in upland habitat. Transects were oriented parallel to the river, 20 m (riparian) and 160 m (upland) from the river, for a total of six plots per site. Because previous studies indicate that both floristics and structural attributes play roles in avian habitat selection (MacArthur and MacArthur 1961, Holmes and Robinson 1981, Robinson and Holmes 1984), I focused on 10 variables representing broad measures of vegetation characteristics: density of deciduous and coniferous trees, snag density, volume of coarse woody debris (CWD), species richness, total percent cover of all shrubs and forbs, and percent cover of the twodominant shrubs (salmonberry and blueberry). Density (number/ha) was recorded for coniferous trees, deciduous trees, and snags within the entire 20-m- radius plot. Trees <3 m in height and ferns were treated as shrubs. Richness and percent cover of shrubs was sampled in a 10-m-radius subplot nested within 20-m circular plots. Richness and percent cover of forbs was sampled using four 1-m^sup 2^ quadrats placed at the center of shrub plots. CWD, defined as fallen logs >10 cm in diameter, was sampled at the point of intersection along the circumference of 20-m plots; I recorded diameter and length to calculate volume (m^sup 3^/ha) of CWD (Van Wagner 1968, Thomas et al. 1979).

Bird sampling.-Details of bird sampling can be found in Shirley (2002). Briefly, birds were censused using a full-plot, area-search method (Slater 1994). A 9-ha grid was established at each site by running a 450-m line parallel and adjacent to the river's edge and nine perpendicular lines extending 200 m from the river. Grid lines were set 50 m apart and flagged at 25-m intervals. Censusing was conducted at each site by at least two observers who walked the grid lines from 05:00 to 10:00 (PST) on days without rain or high winds. We censused birds at each site four times each breeding season from 1 May to 15 July so that each site was censused once approximately every 2 weeks. Birds of prey and flyovers were not included in the censuses. I varied the order in which sites were sampled, and three to four observers rotated among sites and grid lines. To avoid double-counting, vocal and visual observations were recorded on site maps that were later evaluated to calculate number and relative abundance of bird species with respect to distance from the river. The numbers of observations over the four censuses in each year were averaged to provide a mean number of species and individuals per species for each site. I categorized observations into four distance categories from the river's edge (0-50, 51-100, 101-150, and 151- 200 m) and calculated relative abundance for each distance category as the abundance averaged over 3 years and four sites.

I analyzed bird abundances by selected guilds and by individual species. I focused the guild analysis on two guilds that I predicted would show a gradient in abundance from riparian to upland: riparian specialists and aerial foragers. I assigned species to guilds after Hatler et al. (1978), Ehrlich et al. (1988), and Campbell et al. (1990, 1997) (Appendix). For individual species, I restricted my analysis to those with >5 observations in each year (nine species). When estimating species richness by site, I minimized the impacts of transient species by excluding species that were likely migrants and species observed in only one census session during each year (Willson and Comet 1996).

Habitat associations.-I used Akaike's Information Criterion corrected for small samples (AICC) to select suitable models of association between vegetation variables and avian abundance (Burnham and Anderson 2002). I used multiple regression and estimated the residuals to model species richness and abundance of avian guilds and individual species as a function of vegetation variables. Models were based on a priori hypotheses of those vegetation variables that may be associated with a guild's or species' abundance. For each model, I computed AIC^sub c^ and ΔAIC^sub c^. Model likelihoods were standardized to sum to 1 and expressed as Akaike weights (ω). The Akaike weight can be considered as the weight of evidence supporting a given model as the best model; the higher the Akaike weight, the stronger the model. To identify plausible models for each species or guild, I ranked the Akaike weights of models in a given set to produce evidence ratios (i.e., the weight of the best model divided by that of a given model; Burnham and Anderson 2002). Evidence ratios express the likelihood of the selected model relative to other models.

Data analysis.-Vegetation attribute and avian abundance data were tested for normality using the Shapiro-Wilks statistic (Shapiro and Wilk 1965) before conducting paired t-tests and ANOVAs. Homogeneity of variances for one-way ANOVA and repeated-measures ANOVA were tested using the Levene and Bartlett-Box F tests, respectively (Norusis 1994). Species abundance data that violated these assumptions were either log (y + 1) or rank transformed (Conover and Iman 1982). The α level of significance was set at 0.10 to minimize the high biodiversity cost of making a type II error in resource management decisions (Toft and Shea 1983, Dayton 1998). I also define a level of "marginal significance" as 0.15 > P > 0.10. All data were analyzed using SPSS for Windows 6.1.4 (Norusis 1994).

To compare riparian and upland means for each of the 10 vegetation variables, I used paired t-tests because riparian and upland habitats were paired by site for each of the four sites. Rather than correcting for multiple tests using the standard Bonferroni method, which has several disadvantages when gauging the effects of variables in ecological research (Nakagawa 2004), I present effect sizes as recommended by Hurlbert (1994). I evaluated the biological significance of the results using established criteria (Cohen 1988) where a small effect size = 0.2, medium = 0.5, and large = 0.8.

To compare abundances (all species combined, two guilds, and nine individual species) by distance category from the river and among years, I used a two-way, repeated-measures ANOVA, with year and distance from river's edge as fixed effects. Because the same sites were censused over 3 years, I treated year as a repeated variable in a model that specified polynomial contrasts to detect linear or quadratic trends over time (Gurevitch and Chester 1986, von Ende 1993). Because there were no significant year effects, data were pooled across years for all comparisons except for American Robin (Turdus migratorius), which showed a significant distance-by-year interaction. I then tested for differences in pooled abundance (one- way ANOVA; all species combined, two guilds, and eight species) across four 50-m intervals from the river's edge-with distance from edge as a fixed factor (n = 48). For American Robin, I performed the above analysis for each year separately; however, results are presented for all years together (Fig 1). Effect sizes for the oneway ANOVAs were calculated using the Eta squared method (Levine and Hullett 2002).

Because there was a significant year effect for species richness, I did not pool data across years. I compared species richness across the four 50-m intervals from the river's edge and over time using two-way, repeated-measures ANOVA with year and distance from edge as fixed factors (n = 16).

RESULTS

Vegetation.-Four of 10 measures of vegetation structure and composition-density of coniferous and deciduous trees, shrub- species richness, and blueberry cover-differed between riparian and upland habitats (Table 1). Effect sizes were medium for coniferous tree density and large for deciduous tree density, reflecting substantial biological differences. Riparian habitats had nearly five times the density of deciduous trees compared with upland areas, while upland areas had greater conifer density and percent blueberry cover. Snag density and salmonberry cover were greater in upland and riparian areas, respectively; effect sizes were large, but these differences were only marginally significant due to a small sample size. CWD and forb cover were not statistically different between habitats, but effect sizes were medium and could indicate biological significance.

Avian abundance and diversity.-During 1996-1998,1 recorded 645 observations of 36 species. For all sites combined, there were >20 observations for 9 species, accounting for 80% of all observations. The five most abundant species were Winter Wren (Troglodytes troglodytes), Chestnut-backed Chickadee (Poecile rufescens), American Robin, Swainson's Thrush (Catharus ustulatus), and Pacific- slope Flycatcher (Empidonax difficilis). For all years and in each distance interval, assemblages were dominated by these five species, which composed 53-58% of total observations-with only minor variations in their abundance rankings. Except for one forest interior species (Pacific-slope Flycatcher), these species are forest generalists. Winter Wren and Chestnut-backed Chickadee were the dominant species in upland sections and were replaced, in part, by American Robin and Swainson's Thrush near the river. Riparian specialists generally occurred close to the river and at low abundances, with the exception of Hammond's Flycatcher (Empidonax hammondii).

Because the dominant species were forest generalists, total abundance did not differ with distance from the river (F^sub 3,44^ = 1.26, P = 0.30; Fig. 1A). As expected, abundances of riparian specialists and aerial foragers declined with distance from the river (Fig. 1B, C; riparian specialists: F^sub 3,44^ = 7.98, P < 0.001; aerial foragers: F^sub 3,44^ = 5.23, P = 0.027). Of the nine species for which I had sufficient data for analysis, only two varied significantly in abundance across the distance intervals: Swainson's Thrush (F^sub 3,44^ = 2.85, P = 0.10; Fig. IG) and Hammond's Flycatcher (F^sub 3,44^ = 11.74, P = 0.001; Fig. 1K) were more common near rivers (all other species: P ≥ 0.31, effect sizes < 0.07).

FIG. 1. Relative abundance (mean number of observations/site/ year) and standard deviations (error bars) at 50-m intervals from the river's edge for (A) all species combined, (B) riparian specialists, (C) aerial foragers, (D) Winter Wren, (E) Chestnut- backed Chickadee, (F) American Robin, (G) Swainson's Thrush, (H) Pacific-slope Flycatcher, (I) Golden-crowned Kinglet, (J) Varied Thrush, (K) Hammond's Flycatcher, and (L) Hairy Woodpecker, 1996- 1998, western Vancouver Island, British Columbia, Canada (n= 48).

TABLE 1. Vegetation characteristics of riparian and upland habitats, western Vancouver Island, British Columbia, 1995. Of 10 variables, four (deciduous and coniferous tree densities, shrub species richness, and percent cover of blueberry species) differed (P < 0.10) between riparian and upland habitats (n = 12 for all tests).

FIG. 2. Means and standard deviations (error bars) of avian species richness at 50-m intervals from the river's edge, 1996- 1998, western Vancouver Island, British Columbia, Canada.

Species richness did not differ with distance from the river (F^sub 3,12^ = 0.15, P = 0.93) and there was no significant interaction between distance and year (F^sub 6,24^ = 0.61, P = 0.62). Species richness, however, differed among years (F^sub 2,24^ = 6.28, P = 0.028); 14% more species were detected in 1997 than in 1996 and 1998 (Fig. 2).

Habitat associations.-For most bird species, associations with specific vegetation attributes were stronger than with distance from the river. The best model for aerial forager abundance was one showing a negative relationship with conifer density (Table 2). This model was strongly supported (ω^sub i^ = 0.850), being 14 times more likely than the second-best model in the set. Abundance of riparian specialists was predicted equally by four single-variable models showing positive relationships with salmonberry, percent shrub cover, and shrub species richness, as well as a negative relationship with distance from the river. Support for all four models, however, was weak (all four ω^sub i^ < 0.250) and the best model was only 2.5 times more likely than the next-best model in the set.

TABLE 2. Habitat model selection using Akaike's Information Criterion for abundance of avian guilds and species in undisturbed forest sites, western Vancouver Island, British Columbia, 1996- 1998. Selected models with the highest likelihood are shown. Secondary models are included if they were nearly equal to the first model.

The abundances of four species (American Robin, Hairy Woodpecker [Picoides villosus], Pacific-slope Flycatcher, and Winter Wren) were best predicted by single-variable models showing negative relationships with conifer density (Table 2). Models for Hairy Woodpecker and Pacific-slope Flycatcher had moderate support (ω^sub i^ = 0.533 and ω^sub i^ = 0.545), being 3-4 times more likely than the second-best models in the sets. The models for American Robin and Winter Wren had weaker support (ω^sub i^ = 0.405 and ω^sub i^ = 0.311) and were only twice as likely as the second model in the set. Chestnut-backed Chickadee abundance was best predicted by two single-variable models showing negative relationships with conifer (ω^sub i^ = 0.387) and snag densities (ω^sub i^ = 0.376); both models had almost equal support and were 3 times more likely than the third model in the set. Golden-crowned Kinglet (Regulus satrapa) abundance was best predicted by a model showing a positive relationship with conifer density; however, this model was relatively weak (ω^sub i^ = 0.334) and only 1.3 times more likely than the second model in the set. Hammond's Flycatcher abundance was best predicted by two singlevariable models representing a negative relationship with distance from the river (ω^sub i^ = 0.396) and a positive relationship with percent shrub cover (ω^sub i^ = 0.385). Both models had almost equal support and were 5 times as likely as the third model in the set. Swainson's Thrush abundance was best predicted by three single-variable models showing negative relationships with percent blueberry cover (ω^sub i^ = 0.266) and density of coniferous trees (ω^sub i^ = 0.244) and a positive relationship with density of deciduous trees (ω^sub i^ = 0.229). The models had almost equal support, although support for any one was quite weak. The best model for Varied Thrush (Ixoreus naevius) abundance showed a positive relationship with snag density (ω^sub i^ = 0.457). This model had moderate support, being 4 times as likely as the second model in the set.

DISCUSSION

Species diversity and abundance along the riparian gradient.- Contrary to my original predictions, species abundance and diversity of birds were similar along a distance gradient away from the river. Although species richness varied among years, total abundance remained similar during the study. Other studies in coniferous forests of the Pacific Northwest have also found that riparian areas do not support higher numbers of bird species or individuals (McGarigal and McComb 1992, Murray and Stauffer 1995, Pearson and Manuwal 2001). In contrast, studies in more arid or agricultural environments (Carothers et al. 1974, Stevens et al. 1977, Wauer 1977, Stauffer and Best 1980) found large differences in diversity and abundance between riparian and upland habitats. McGarigal and McComb (1992) proposed three hypotheses to account for the regional difference in these patterns: (1) high stream density and availability of water in upland areas in the Pacific Northwest, (2) a less pronounced microclimatic gradient (moderated by maritime influences) in northwestern forests, and (3) a more subtle transriparian gradient in vegetation structure. Higher rainfall and less variation in annual temperatures on western Vancouver Island compared with Washington and Oregon may produce an even less pronounced transriparian gradient.

In this study, riparian habitats had greater densities of deciduous trees, and the understory was dominated by salmonberry. In contrast, upland areas had higher densities of coniferous trees, and blueberry species dominated the shrub understory; uplands also tended to have greater snag densities. While low statistical power may have limited my ability to detect statistically significant differences in some attributes, my results are consistent with those of other studies that evaluated vegetation structure across the transriparian gradient of forests in the Pacific Northwest (McGarigal and McComb 1992, Pearson and Manuwal 2001). McGarigal and McComb (1992) attributed lower bird species richness in riparian as opposed to upland areas to the scarcity of conifers found along streams; however, in my study, conifers were not as scarce along riparian areas, perhaps accounting for the similarity in species richness between the two habitats. The lack of a strong gradient in vegetation structure from riparian to upland is also reflected in the distribution of the most common bird species. Abundances of eight of the most common bird species, as well as abundance of the aerial foraging guild, were associated most closely with densities of certain canopy and understory species rather than distance from the river. Complex topography, combined with consistently moist conditions, provides suitable habitat for most species across the riparian-upland gradient that I studied, and it probably accounts for the lack of strong riparian effects at the community level.

The large fourth- or fifth-order streams and rivers in my study area contrast with the smaller, second-order streams that were the focus of some previous studies in northern forests (McGarigal and McComb 1992, Wiebe and Martin 1998). In those studies, riparian forests supported equal or fewer species and individuals compared with surrounding uplands (McGarigal and McComb 1992, Wiebe and Martin 1998). Studies of larger-order streams, however, have indicated that they support denser, more complex riparian vegetation communities and greater avian density, species richness, and abundance (Knopf 1985, Lock and Naiman 1998). Lock and Naiman (1998) found greater species richness and abundance along larger rivers where the riparian habitat contained a higher ratio of deciduous to coniferous vegetation; in my study, however, avian species richness and abundance were similar across the riparian to upland gradient, even along larger streams. Most species used both riparian and upland habitats, whereas only a few species specialized in either habitat. In northwestern forests, these specialists represented a small fraction of the overall community.

Habitat selection.-Of the 36 species recorded, five occurred only near the river. Four of these riparian specialists-Common Merganser (Mergus merganser), American Dipper (Cinclus mexicanus), Belted Kingfisher (Ceryle alcyon), and Spotted Sandpiper (Actitis macularius)-depend on stream invertebrates and/or fish as food resources and they nest in adjacent riparian vegetation or riverbanks (Enns et al. 1993, Campbell et al. 1997). The remaining species, Willow Flycatcher (Empidonax traillii), rarely occurs in mature forest except in riparian areas. In the coastal western hemlock zone, Willow Flycatchers more commonly occur in marshes and early successional clearcuts (5-10 years of age) associated with young red alder and willow trees (Enns et al. 1993, Campbell et al. 1997).

Five species occurred only at single sites in upland sections of forest: Fox Sparrow (Passerella iliaca), Mutton's Vireo (Vireo huttoni), Olive-sided Flycatcher (Contopus cooperi), Spotted Towhee (Pipilo maculatus), and Yellow Warbler (Dendroica petechia). All of these species are rare in mature forests and may select large patches of open, deciduous vegetation in forest interiors.

As predicted, I found that aerial foragers declined in abundance with increasing distance from the river. Aerial foragers include Hammond's Flycatcher, a species that occupies a wide variety of habitats (Willson and Comet 1996). Throughout much of the Pacific Northwest, the species is an upslope specialist that is associated strongly with conifers at sites characterized by relatively open canopies (Sakai and Noon 1991, McGarigal and McComb 1992); however, farther north in Alaska it favors deciduous stands (Willson and Comet 1996), and in the forests of Vancouver Island (Waterhouse and Harestad 1999, Shirl\ey 2002) and southeastern British Columbia (Kinley and Newhouse 1997), this species is largely restricted to mixed riparian forests that include large deciduous trees and conifers. Whereas Hammond's Flycatcher may use riparian habitat, it is sympatric with the Pacific-slope Flycatcher in old-growth forest (Campbell et al. 1997) and its distribution may reflect some habitat partitioning between the two species. In southern Colorado, where Hammond's and Cordilleran (Empidonax occidentalis) flycatchers co- occur, Hammond's Flycatcher densities were approximately onehalf those of the Cordilleran Flycatcher (Beaver and Baldwin 1975). In these areas of overlap, Hammond's Flycatcher inhabited aspen habitat, while the Cordilleran Flycatcher used aspen-conifer habitat. Behavioral observations by Sakai and Noon (1991) suggested that some competition likely occurs between the Hammond's and Pacific-slope flycatchers, but it does not result in competitive exclusion of one species by the other. Pacific-slope Flycatcher abundance differed little along the gradient of distance away from the river, a reflection of its association with large conifer trees in both riparian and upland habitats. Interestingly, studies farther south in the Pacific Northwest (McGarigal and McComb 1992, Pearson and Manuwal 2001) reported that Pacific-slope Flycatchers are associated with riparian habitats, whereas Hammond Flycatchers are associated with upland habitats. The reason for this difference is unclear, but may relate to differences in species composition and size distribution of trees in the two forest habitats. For example, in the previous studies, large-diameter conifers required by Hammond's Flycatchers (Sakai and Noon 1991), such as Douglas-fir (Pseudotsuga menziesii), occur in greater numbers in upland forests.

Several of the dominant species in my study, including Chestnut- backed Chickadee and Golden-crowned Kinglet, showed associations with conifer or snag density. An exception was the Swainson's Thrush, which, although a forest generalist, was more abundant in riparian habitat. This species is widespread on the west coast (Campbell et al. 1997) and often forages in salmonberry and devil's club in riparian habitats. The positive association with deciduous trees suggests that these structures or some other closely associated vegetation may be an important influence on habitat selection for this species. Furthermore, virtually all Swainson's Thrush nests encountered incidentally during surveys (n = 8) were found in salmonberry (SMS pers. obs.), suggesting a strong association with this shrub for nesting habitat and/or food.

Management implications.-Patterns of avian diversity and abundance in riparian communities often have been explained by dramatic gradients in microclimate and vegetation structure or composition (Carothers et al. 1974, Stevens et al. 1977, Dickson 1978, Szaro 1980). Where these gradients are subtle, as in forests of the Pacific Northwest, the patterns disappear (Wiebe and Martin 1998, Pearson and Manuwal 2001) or they may reverse-upland areas supporting greater diversity and abundance than riparian areas (McGarigal and McComb 1992). Northwest forests generally lack a strong microclimatic gradient from riparian to upland (Brosofske et al. 1997), and ephemeral streams or ponds occur in virtually every upland area. These factors create a fine-scale habitat mosaic in which patches dominated by conifers are interspersed with deciduous trees and shrubs that provide habitat for species more typical of deciduous-tree-dominated riparian areas.

Recent discussions of land management practices to preserve native biodiversity of forest species include using a landscape- level approach that protects both riparian and upland habitats to ensure connectivity across the landscape (McGarigal and McComb 1992, Wiebe and Martin 1998). Maintaining connectivity may prevent isolation of remnant forest patches (Fahrig and Merriam 1985, Saunders and de Rebeira 1991, Gonzalez 2000); however, the lack of upland specialists in my study argues against placing too much emphasis on upland areas per se. Unmanaged riparian areas not only provide habitat for those few species that associate with specific features at the river's edge, they also contain habitat elements such as large conifers and snags important to many common forest species.

My study was limited in several ways that should be considered in the development of land use plans. First, although the contrast between riparian and upland habitats is known to be subtle in the moist forests of the Pacific Northwest (McGarigal and McComb 1992, Pearson and Manuwal 2001), differences in structure and composition of vegetation in my study may have been obscured by the small sample size and resulting lack of statistical power. Statistically, several of the vegetation characteristics that I measured were marginally or non-significant, but their medium to large effect sizes indicate possible biological significance. second, comparisons in this study were limited to measures of avian abundance. Future work should focus on measures of relative fitness or productivity in the two habitats. Third, my study was conducted during the breeding season; work is also needed to assess avian distributions during other seasons, as the relative value of riparian and upland habitats may differ between periods of migration and other seasons (Harris 1984, Wiebe and Martin 1998). For example, riparian habitats may provide critical habitat for Neotropical migrants as they travel between their wintering and breeding grounds (Stevens et al. 1977, Finch 1991), and riparian habitat may be important for the survival and population stability of migratory species during the breeding season. The diversity and density of some migrants may be greater in riparian corridors because they are easy to follow and/or provide diverse foraging habitats (Wiens 1989, Wiebe and Martin 1998).

ACKNOWLEDGMENTS

I thank K. G. Beal, D. J. Huggard, L. G. Barrett-Lennard, and two anonymous reviewers for comments on earlier drafts of this manuscript, and K. T. Fort, S. Frioud, S. L. Hicks, D. Lewis, R. Maraj, G. Matscha, F Pouw, and S. Weber for their hard work in the field. I also thank G. Matscha and C. M. Ferguson for their help in processing the vegetation data. I am grateful to W. French and the engineering group at Weyerhaeuser Canada for their cooperation in locating field sites. Financial support was provided by the Habitat Conservation Trust Fund, Forest Renewal British Columbia, a University of British Columbia graduate fellowship to SMS, and a Natural Sciences and Engineering Research Council operating grant to J. N. M. Smith.

LITERATURE CITED

ANTHONY, R. G., G. A. GREEN, E. D. FORSMAN, AND S. K. NELSON. 1996. Avian abundance in riparian zones of three forest types in the Cascade Mountains, Oregon. Wilson Bulletin 108:280-291.

BEAVER, D. L. AND P. H. BALDWIN. 1975. Ecological overlap and the problem of competition and sympatry in the Western and Hammond's flycatchers. Condor 77:1-13.

BRINSON, M. M., B. L. SWIPT, R. C. PLANTICO, AND J. S. BARCLAY. 1981. Riparian ecosystems: their ecology and status. U.S. Department of the Interior, Fish and Wildlife Service, Kearneysville, West Virginia.

BROSOFSKE, K. D., J. CHEN, R. J. NAIMAN, AND J. F. FRANKLIN. 1997. Harvesting effects on microclimatic gradients from small streams to uplands in western Washington. Ecological Applications 7:1188-1200.

BRYANT, A. A., J. P. SAVARD, AND R. T. MCLAUGHLIN. 1993. Avian communities in old-growth and managed forests of western Vancouver Island, British Columbia. Technical Report Series, no. 167. Canadian Wildlife Service, Nanaimo, British Columbia, Canada.

BULL, E. V. 1978. Specialized habitat requirements of birds: snag management, old growth, and riparian habitat. Pages 74-82 in Proceedings of the workshop on nongame bird habitat management in the coniferous forests of the western United States (R. M. DeGraaf, Tech. Coord.). General Technical Report PNW-64, USDA Forest Service, Pacific Northwest Forest and Range Experiment Station, Portland, Oregon.

BURNHAM, K. P. AND D. R. ANDERSON. 2002. Model selection and multimodel inference, 2nd ed. Springer-Verlag, New York.

CAMPBELL, R. W. 1990. The birds of British Columbia. Royal British Columbia Museum, Victoria, and Canadian Wildlife Service, Ottawa, Ontario, Canada.

CAMPBELL, R. W., N. K. DAWE, I. MCTAGGART-COWAN, J. M. COOPER, G. W. KAISER, M. C. E. McNALL, AND G. E. J. SMITH. 1997. The birds of British Columbia, vol. 3: flycatchers through vireos. University of British Columbia Press, Vancouver, Canada.

CAROTHERS, S. W., R. R. JOHNSON, AND S. W. AITCHISON. 1974. Population structure and social organization of southwestern riparian birds. American Zoologist 14:97-108.

COHEN, J. 1988. Statistical power analysis for the behavioral sciences, 2nd ed. Erlbaum, Hillsdale, New Jersey.

CONOVER, W. J. AND R. L. IMAN. 1982. Analysis of covariance using the rank transformation. Biometrics 38:715-724.

DAYTON, P. K. 1998. Reversal of the burden of proof in fisheries management. Science 279:821-822.

DICKSON, J. G. 1978. Forest bird communities of the bottomland hardwoods. Pages 66-73 in Proceedings of the workshop: management of southern forests for nongame birds (R. M. DeGraaf, Ed.). General Technical Report SE-14, USDA Forest Service, Southeastern Forest Experiment Station, Atlanta, Georgia.

EHRLICH, P. R., D. S. DOBKIN, AND D. WHEYE. 1988. The birder's handbook: a field guide to the natural history of North American birds. Simon and Schuster, New York.

ENNS, K. A., E. B. PETERSON, AND D. S. MCLENNAN. 1993. Impacts of hardwood management on British Columbia wildlife: problem analysis. Forestry Canada Pacific Forestry Centre, Victoria, British Columbia, Canada.

FAHRIG, L. AND G. MERRIAM. 1985. Habitat patch connectivity and population survival. Ecology 66:1762-1768.

FINCH, D. M. 1991. Popul\ation ecology, habitat requirements, and conservation of Neotropical migratory birds. General Technical Report RM-205, USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

GONZALEZ, A. 2000. Community relaxation in fragmented landscapes: the relation between species richness, area and age. Ecology Letters 3:441-448.

GRAY, L. J. 1993. Response of insectivorous birds to emerging aquatic insects in riparian habitats of a tallgrass prairie stream. American Midland Naturalist 129:288-300.

GUREVITCH, J. AND S. T. CHESTER. 1986. Analysis of repeated measures experiments. Ecology 67:251-255.

HARRIS, L. D. 1984. The fragmented forest. University of Chicago Press, Chicago, Illinois.

HATLER, D. E, R. W. CAMPBELL, AND A. DORST. 1978. Birds of Pacific Rim National Park. British Columbia Provincial Museum, Victoria, Canada.

HOLMES, R. T. AND S. K. ROBINSON. 1981. Tree species preferences of foraging insectivorous birds in a northern hardwoods forest. Oecologia 48:31-35.

HURLBERT, S. H. 1994. Old shibboleths and new syntheses. Trends in Ecology and Evolution 9:495-496.

JAMES, F. C. AND H. H. SHUGART. 1970. A quantitative method of habitat description. Audubon Field Notes 24:727-736.

KINLEY, T. AND N. J. NEWHOUSE. 1997. Relationship of riparian reserve zone width to bird density and diversity in southeastern British Columbia. Northwest Science 71:75-86.

KLINKA, K., J. POJAR, AND D. V. MEIDINGER. 1991. Revision of biogeoclimatic units of coastal British Columbia. Northwest Science 65:32-47.

KNOPF, F. L. 1985. Significance of riparian vegetation to breeding birds across an altitudinal cline. Pages 105-111 in Riparian ecosystems and their management: reconciling conflicting uses (R. R. Johnson, C. D. Ziebell, D. R. Patton, P. F. Folliott, and R. H. Hamre, Eds.). General Technical Report RM-120, USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

KNOPF, F. L., R. R. JOHNSON, T. RICH, F. B. SAMSON, AND R. C. SZARO. 1988. Conservation of riparian ecosystems in the United States. Wilson Bulletin 100:272-284.

KNOPF, F. L. AND F. B. SAMSON. 1994. Scale perspectives on avian diversity in western riparian ecosystems. Conservation Biology 8:669- 676.

KUEHNE, R. A. 1962. A classification of streams, illustrated by fish distribution in an eastern Kentucky creek. Ecology 43:608-614.

LARUE, P., L. BELANGER, AND J. HUOT. 1995. Riparian edge effects on boreal balsam fir bird communities. Canadian Journal of Forest Research 25:555-566.

LEVINE, T. R. AND C. R. HULLETT. 2002. Eta squared, partial eta squared, and misreporting of effect size in communication research. Human Communication Research 28:612-625.

LOCK, P. A. AND R. J. NAIMAN. 1998. Effects of stream size on bird community structure in coastal temperate forests of the Pacific Northwest, USA. Journal of Biogeography 25:773-782.

MACARTHUR, R. H. AND J. W. MACARTHUR. 1961. On bird species diversity. Ecology 42:594-598.

MCGARIGAL, K. AND W. C. MCCOMB. 1992. Streamside versus upslope breeding bird communities in the central Oregon Coast Range. Journal of Wildlife Management 56:10-23.

MUELLER-DOMBOIS, D. AND H. ELLENBERG. 1974. Aims and methods of vegetation ecology. Wiley, New York.

MURAKAMI, M. AND S. NAKANO. 2002. Indirect effect of aquatic insect emergence on a terrestrial insect population through predation by birds. Ecology Letters 5:333-337.

MURRAY, N. L. AND D. F. STAUFFER. 1995. Nongame bird use of habitat in central Appalachian riparian forests. Journal of Wildlife Management 59:78-88

NAIMAN, R. J., H. DECAMPS, AND M. POLLOCK. 1993. The role of riparian corridors in maintaining regional biodiversity. Ecological Applications 3:209-212.

NAKAGAWA, S. 2004. A farewell to Bonferroni: the problems of low statistical power and publication bias. Behavioral Ecology 15:1044- 1045.

NORUSIS, M. J. 1994. SPSS advanced statistics 6.1. SPSS, Inc., Chicago, Illinois.

NUSZDORFER, F. AND R. BOETTGER. 1994. Biogeoclimatic units of the Vancouver Forest Region: southern Vancouver Island and Sunshine Coast. Province of British Columbia, Ministry of Forests, Vancouver, Canada.

PEARSON, S. AND D. A. MANUWAL. 2001. Breeding bird response to riparian buffer width in managed Pacific Northwest Douglas-fir forests. Ecological Applications 11:840-853.

RAEDEKE, K. J. 1988. Streamside management: riparian wildlife and forestry interactions. Institute of Forest Resources, University of Washington, Seattle.

RANNEY, J. W., M. C. BRUNER, AND J. B. LEVENSON. 1981. The importance of edge in the structure and dynamics of forest islands. Pages 67-96 in Forest island dynamics in man-dominated landscapes (R. L. Burgess and D. M. Sharpe, Eds.). Springer-Verlag, New York.

ROBINSON, S. K. AND R. T. HOLMES. 1984. Effects of plant species and foliage structure on the foraging behavior of forest birds. Auk 101:672-684.

SAKAI, H. F. AND B. R. NOON. 1991. Nest-site characteristics of Hammond's and Pacific-slope flycatchers in northwestern California. Condor 93:563-574.

SAUNDERS, D. A. AND C. P. DE REBEIRA. 1991. Values of corridors to avian populations in a fragmented landscape. Pages 221-240 in The role of corridors (D. A. Saunders and R. J. Hobbs, Eds.). Surrey Beatty & Sons, Chipping Norton, New South Wales, Australia.

SHAPIRO, S. S. AND M. B. WILK. 1965. An analysis of variance test for normality (complete samples). Biometrika 52:591-611.

SHIRLEY, S. 2002. Forest fragmentation and regrowth: use of riparian and upland forest by birds in managed and unmanaged mature coastal British Columbia rainforest. Ph.D. dissertation, University of British Columbia, Vancouver, Canada.

SLATER, P. J. 1994. Factors affecting the efficiency of the area search method of censusing birds in open forests and woodlands. Emu 94:9-16.

STAUFFER, D. F. AND L. B. BEST. 1980. Habitat selection by birds of riparian communities: evaluating effects of habitat alterations. Journal of Wildlife Management 44:1-15.

STEVENS, L. E., B. T. BROWN, J. M. SIMPSON, AND R. R. JOHNSON. 1977. The importance of riparian habitat to migrating birds. Pages 156-164 in Importance, preservation and management of riparian habitat: a symposium (R. R. Johnson and D. A. Jones, Eds.). General Technical Report RM43, USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

SZARO, R. C. 1980. Factors influencing bird populations in southwestern riparian forests. Pages 403-418 in Workshop proceedings: management of western forests and grasslands for nongame birds (R. M. DeGraff and N. G. Tilghman, Eds.). General Technical Report INT-86, USDA Forest Service, Intermountain Forest and Range Experiment Station, Ogden, Utah.

THOMAS, J. W., C. MASER, AND J. E. RODIEK. 1979. Riparian zones. Pages 40-47 in Wildlife habitats in managed forests: the Blue Mountains of Oregon and Washington (J. W. Thomas, Ed.). USDA Agriculture Handbook, no. 553. U.S. Government Printing Office, Washington, D.C.

TOFT, C. A. AND P. J. SHEA. 1983. Detecting community-wide patterns: estimating power strengthens statistical inference. American Naturalist 122:618-625.

VAN WAGNER, C. E. 1968. The line intersect method in forest fuel sampling. Forest Science 14:20-26.

VON ENDE, C. N. 1993. Repeated-measures analysis: growth and other time-dependent measures. Pages 113-137 in Design and analysis of ecological experiments (S. M. Scheiner and J. Gurevitch, Eds.). Chapman and Hall, New York.

WATERHOUSE, F. L. AND A. S. HARESTAD. 1999. The value of riparian forest buffers as habitat for birds in the coastal western hemlock zone, British Columbia: a pilot study. Research Section, Vancouver Forest Region, British Columbia Ministry of Forests, Nanaimo, Canada.

WAUER, R. R. 1977. Significance of Rio Grande riparian systems upon the avifauna. Pages 165-174 in Importance, preservation and management of riparian habitat: a symposium (R. R. Johnson and D. A. Jones, Eds.). General Technical Report RM-43, USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado.

WIEBE, K. AND K. MARTIN. 1998. Seasonal use by birds of stream- side riparian habitat in coniferous forest of northcentral British Columbia. Ecography 21:124-134.

WIENS, J. A. 1989. The ecology of bird communities. Cambridge University Press, Cambridge, United Kingdom.

WILLSON, M. F. AND T. A. COMET. 1996. Bird communities of northern forests: patterns of diversity and abundance. Condor 98:337- 349.

SUSAN M. SHIRLEY1

1 Dept. of Zoology, Univ. of British Columbia, 6270 University Blvd., Vancouver, BC V6L 1V8, Canada; e-mail: Shirley@zoology.ubc.ca

APPENDIX. Guild classification of avian species (after Hatler et al. 1978; Ehrlich et al. 1988; Campbell et al. 1990, 1997).

Guild/Species

Aerial foragers

Hammond's Flycatcher (Empidonax hammondii)

Olive-sided Flycatcher (Contopus cooperi)

Pacific-slope Flycatcher (Empidonax difficilis)

Willow Flycatcher (Empidonax traillii)

Riparian specialists

American Dipper (Cinclus mexicanus)

Belted Kingfisher (Ceryle alcyon)

Common Loon (Gavia immer)

Common Merganser (Mergus merganser)

Hammond's Flycatcher (Empidonax hammondii)

Spotted Sandpiper (Actitis macularius)

Warbling Vireo (Vireo gilvus)

Yellow Warbler (Dendroica petechia)

Copyright Wilson Ornithological Society Sep 2005

Source: Wilson Bulletin, The

More News in this Category