Research Paper

INTRODUCTION

The boreal forest of Alberta, Canada is one of the most dissected landscapes in the world (Bayne et al. 2005, Schneider and Dyer 2006, Dabros et al. 2022). Dissection is the first stage in fragmentation processes and occurs as linear features are created on the landscape to access remote areas. Energy sector development in Alberta’s boreal forest has created an extensive network of linear features such as seismic lines, pipelines, transmission lines, and roads (Dyer et al. 2002, Schneider and Dyer 2006, Dickie et al. 2020). Seismic lines are narrow cuts through the forest that are created by machines used to access remote areas for hydrocarbon exploration (Dabros et al. 2018). While the footprint of linear features is extensive, its effects on biodiversity are less clear than the effects of forest fragmentation caused by agricultural conversion. Some species at risk have lower abundance/use along or near linear features, presumably because of edge avoidance (e.g., Canada Warbler [Cardellina canadensis]) (Gregoire et al. 2022), or are affected by increased mortality (e.g., woodland caribou [Rangifer tarandus caribou]) due to higher use and faster movement of predators along and near seismic lines (Latham et al. 2011). However, the assumption from studies on focal species that linear features have negative impacts on all aspects of biodiversity is tenuous. How many species react in a positive versus negative way and the identity of those species is thus important to understand from a land use and reclamation perspective.

People who walk down linear features or look at them from above have a clear perception that they subdivide forest patches, which in theory reduces patch size and increases edge effects (Fischer and Lindenmayer 2007). However, whether narrow linear features split forest patches may depend on whether individual birds include such gaps as part of their territories or if they avoid areas near linear features, thereby creating edge effects. Presumably, the wider a linear feature becomes, the more likely it is to lead to habitat loss and patch isolation for species that rely on mature forests as habitat (Fahrig 2003, Fischer and Lindenmayer 2007, Fahrig et al. 2019). However, all linear features created by the energy sector within forests are vegetated to some degree; they are just in an earlier successional state than in adjacent forests. How vegetated and wide a linear feature has to be to create habitat for early seral species and how wide a linear feature has to be to cause fragmentation for a forest bird species is poorly understood.

In Alberta, the issue of how linear features impact biodiversity is increasingly important as limits on energy development are being implemented until existing linear features are recovered (Government of Alberta 2017). In addition, established policies require that only the area needed for maintenance of pipelines and transmission lines be maintained in a grass-covered state relative to the original construction width (Government of Alberta 1994). The suitability of vegetation on linear features as wildlife habitat likely depends on the function of the feature (Schneider and Dyer 2006); the time since both the feature was constructed and the vegetation was allowed to regenerate (Kemper and Macdonald 2009); the influence of north–south or east–west orientation of the linear feature on irradiance, microclimate, and vegetation growth within the linear feature (Finnegan et al. 2019, Franklin et al. 2021); and the length of the linear feature (Pattison et al. 2016) and its width. Categorical statements about how linear feature width impacts species are coarse because linear feature width varies from ~3 to 14 m for seismic lines, ~15 to 30 m for pipelines, 10 to 30 m for roads, and typically > 30 m for transmission lines (Oberg 2001, Ball et al. 2009, Askins et al. 2012, Dickie et al. 2020, Filicetti and Nielsen 2022). In addition, cumulative effects planning (i.e., shared pipeline and transmission line corridors) can result in shared linear feature rights-of-way that are 80–100 m in width (Kalukapuge, personal observation). Currently, most of the wildlife models used for regulatory decision-making treat linear features as being the same width, or in very broad categories such as soft (vegetated) and hard (non-vegetated) linear features (ABMI 2015, Domahidi et al. 2019), or narrow, intermediate, and wide categories (King et al. 2009, Askins et al. 2012, Mahon et al. 2019, Crosby et al. 2023). Such categories have also been used inconsistently across studies, which makes it difficult to generalize results. As a result, although the impacts of linear features on wildlife have been reported, there is ambiguity about how linear width per se influences wildlife in the boreal forest.

Past research has shown positive, negative, or neutral effects of linear features on forest bird species abundance (Bayne et al. 2016, Leston et al. 2023) and community characteristics, including species diversity, richness, composition, and evenness (Rich et al. 1994, Fleming 2001, Machtans 2006, Askins et al. 2012). In the boreal forest, species associated with early successional habitats were more likely to show increased abundance along seismic lines or pipelines than within forests, while species associated with older forests showed mixed positive and negative responses to linear features (Bayne et al. 2016). Similarly, species associated with early successional habitats or older deciduous forests were more likely to increase in abundance with linear footprint amount and proximity, while species associated with older coniferous upland forests were more likely to decrease in abundance with linear footprint amount and proximity (Leston et al. 2023). As a result, whether or not dissection in the boreal forest has different impacts on birds relative to other ecosystems remains uncertain. For example, Machtans (2006) found that relative to intact deciduous forest in boreal forests of the Northwest Territories, bird communities did not change when an area was experimentally dissected by 8 m wide seismic lines. Similarly, Fleming (2001) found no change in the number of species detected and relative abundance of birds among different width classes of pipelines in the boreal forest of Alberta but did observe differences in alpha diversity. In contrast, in deciduous forests of New Jersey, negative effects for forest birds were observed when linear features were ≥ 16 m wide, whereas linear features ≤ 8 m wide had negligible effects (Rich et al. 1994). In Connecticut, relative abundance of several rare shrubland species increased along narrow transmission lines compared to wider ones, which suggests that in some situations, forest dissection can be positive for species of concern (Askins et al. 2012). Anderson et al. (1977) found that in deciduous forests in Tennessee, sites within a narrow corridor (~12.0 m) were associated with reduced species richness, while sites within a wider corridor (~30.5 m) supported high species diversity and greater total abundance of all bird species combined. Additionally, in the wider corridors, there were both forest and open-country species that were not observed in the adjacent forest. Furthermore, Anderson et al. (1977) indicated that species composition appeared to be similar in wider corridors (61.0 m and 91.5 m) and narrow corridors (12.0 m). In general, the literature suggests that wider linear features are more likely to support the habitat requirements of early successional and shrubland bird species (Confer and Pascoe 2003, King et al. 2009). However, whether the same threshold widths are observed for birds in all ecosystems remains uncertain. In addition, whether such changes are positive or negative depends on conservation priorities and the specific species that use wider linear features as habitats.

We investigated the relationship between linear feature width and both community-level metrics (species richness and composition) and abundance of individual species in songbird communities along forest edges bordering linear features. Our aim was to address the following question: Do all linear features have the same disturbance effect on songbird species richness and diversity, and if not, how much does linear feature width influence the outcome? We hypothesized that songbird communities would change both numerically and compositionally as the width of the linear feature increases. We predicted that as linear features become wider, they will attract songbird species that prefer shrubby vegetation and open habitats. However, if edge effects for forest species are large, then areas with wider linear features might show no change in diversity or richness because no forest birds will be detected at the widest features. However, if edge effects for forest songbird species are small, wider linear features, such as pipelines and transmission lines, may support greater species richness and diversity compared to narrow seismic lines.

METHODS

Study area

Our study sites were located in the boreal forest in areas such as Calling Lake, Cold Lake, Conklin, Fort McMurray, and Lac La Biche in north and northeastern Alberta, Canada. We focused on songbird communities near linear features in mature upland forests characterized by trembling aspen (Populus tremuloides), balsam poplar (Populus balsamifera), and white spruce (Picea glauca). The most common shrubs in these areas included beaked hazelnut (Corylus cornuta), willow (Salix spp.), and alder (Alnus spp.).

Site selection and classification

We used autonomous recording units (ARU) at 156 locations belonging to three general types of soft linear features (Domahidi et al. 2019): seismic lines, pipelines, and transmission line rights-of-way. “Soft” means the linear feature was capable of growing vegetation (e.g., it was not a road). Sites that represented the three linear feature types were systematically selected in a spatially balanced manner based on predefined criteria: (1) at least 55% of the buffer area (150 m) was occupied by upland deciduous or mixedwood forest, and (2) sites were at least 150 m away from other human-related footprints (e.g., roads, well sites, and harvest blocks) to minimize cumulative effects. The sites represented a range of line widths from 4 to ~100 m (Fig. 1). Each site was selected independently of others considering accessibility by vehicles and access permissions, meaning the selection of one site did not influence the spatial selection of another. This allowed for some local areas to contain all three or two linear feature types within the same local area, while others might not. The width of each linear feature was measured as the distance from one edge to another within the area around the ARU. Width measurements were taken in the field. Additionally, the dominant forest type surrounding the linear feature (e.g., mixedwood or deciduous) was ground-truthed on-site.

Bird surveys

We conducted unlimited-distance surveys during late May to early July in 2021 and 2022 via passive acoustic monitoring with ARUs (ARUs: SM2, SM3, SM4; Wildlife Acoustic Inc.) equipped with two omnidirectional microphones. We deployed ARUs at the edge of each linear feature with the aim of detecting birds that were using the linear feature as well as the forest edge. ARUs were attached to a tree at a height of 1.5 m above the ground (Lankau 2014), and the left microphone was consistently oriented toward the footprint to maintain uniformity across all ARU stations. Some of the ARUs were left in the field for multiple days to be used in different analyses that required multiple visits. For each site, we randomly selected a single recording (3 minute) from between 0500 and 0800 hours, which represented the time when birds are most vocally active. This approach is equivalent to single-visit abundance surveys. Detailed ARU protocol for deployments and settings files used are available on WildTrax (WildTrax, Edmonton, AB, Canada). Typically, a single ARU was deployed per linear feature, but when the forest stand surrounding the linear feature transitioned from one type to another (e.g., deciduous to mixedwood), multiple ARUs were deployed on the same linear feature, with a minimum distance of ≥ 300 m maintained between two sampling locations. This avoided the double counting of the same individual by the two ARUs because 300 m is the maximum detection radius for most boreal birds (Bayne et al. 2008, Mahon et al. 2019).

Acoustic data processing

We extracted species and abundance data using WildTrax, which is an acoustic and camera data processing platform and data repository. Identifying species and different individuals within the same species on WildTrax involves several steps. WildTrax generates spectrograms for acoustic recordings, where the X axis represents time and the Y axis represents frequency. The vocalization of a particular species generates a unique spectrogram with a specific length, syllables, and trills, often within a unique frequency range. Therefore, identifying a species requires listening to the recording while simultaneously viewing the spectrogram. Distinguishing different individuals of the same species relies on multiple sound and spectrogram attributes. Key factors include the time and amplitude of individual songs (e.g., if two songs are overlapping, they have to come from more than one individual). The amplitude of the signal provides information on the distance from the bird to the ARU. Multiple amplitudes in a relatively short period are typically from different individuals. We also rely on whether the left or right channel is louder because this helps a transcriber determine direction in some species and differences in song structure between individuals. Numerous studies have suggested that differences in abundance estimates between an observer in the field and a transcriber in WildTrax are no different than two observers estimating these features in the field, and some studies have shown that recordings followed by human listening in the lab provide more reliable abundance and richness data compared to human-conducted point counts (Celis‐Murillo et al. 2009, Sedláček et al. 2015, Van Wilgenburg et al. 2017). Additionally, when recordings are compressed using high-quality techniques, the impact on the ability to accurately identify birds is minimal (Rempel et al. 2005).

We used the “1SPT” (tag per task limit) option in Wildtrax, which allowed us to tag each individual the first time it was either seen or heard in a particular recording, and recorded the number of distinct individuals within each survey in order to determine abundances. We did not account for the detectability of each species. We used data only from passerines (order Passeriformes).

Statistical analyses

Local species richness (α diversity)

We calculated the total species richness for each ARU location using the “vegan” R package (version 2.6.6.1) (Oksanen et al. 2019) in R (R Core Team 2022). Subsequently, we assessed the species richness of different species guilds. The guilds were determined based on species’ associations with different vegetation types for nesting locations, as well as species–vegetation associations for living and foraging, as published by Morrison (1981), Schieck and Song (2006), and the Alberta Biodiversity Monitoring Institute biodiversity browser. To examine the relationship between species richness and linear feature width, we used generalized linear models (GLMs) with Poisson and negative binomial distributions for the dependent variable (species richness) and linear feature width (m) as the predictor. We tested the model diagnostics, including patterns of the model residuals and overdispersion, using the “DHARMa” R package (version 0.4.6) (Hartig 2021).

In the first model, we assessed how total bird species richness detected at each ARU location was influenced by linear feature width. In the second model, we tested the effect of linear feature width on the richness of species that as adults in the breeding season forage or inhabit shrubby vegetation (e.g., Lincoln's Sparrow [Melospiza lincolnii]), prefer shrubs for both nesting and foraging (e.g., Clay-colored Sparrow [Spizella pallida]), or nest on the ground but use shrubby habitats for living and foraging during the breeding season (e.g., LeConte’s Sparrow [Ammodramus leconteii]). The total number of species we labeled shrub-associated in our dataset was 30 (Appendix 1). In our third model, we tested the effect of linear feature width on richness of species that are known to prefer mature forests and forest interiors. The total number of mature forest and forest interior species in our dataset was 15 (Appendix 2).

Changes in species composition (β diversity)

We performed non-metric multidimensional scaling (NMDS) analysis using the “metaMDS” function in the “vegan” package to understand changes in species composition (β diversity) in songbird communities as a function of linear feature width. In the analysis, we used the Bray-Curtis distance matrix, which enabled us to quantify the compositional dissimilarities among different communities based on abundance data (Anderson et al. 2011). We used the “envfit” function in “vegan” to assess the relationship between the environmental variable (linear feature width) and community composition by calculating the squared correlation coefficients as a measure of goodness of fit. Additionally, we used permutational multivariate analysis of variance, with the “adonis” function to evaluate the significance of this relationship. Both analyses were run using 999 permutations to generate empirical p values (Anderson 2001).

We generated the NMDS plot for the bird communities across 156 linear feature sites and then used the “ordisurf” function in the “vegan” package to incorporate contours that represented the different linear feature widths onto the ordination space. “ordisurf” fits a smooth surface for linear feature width (continuous variable) with cross-validation selection of smoothness. A Pearson correlation test between NMDS axis scores and the environmental variable (linear feature width) was performed to evaluate the potential influence of linear feature width on species distribution in the NMDS ordination space.

Species-specific threshold responses

We analyzed 17 species that were detected at ≥ 10% of ARU stations (Table 1). For each species, we ran four GLMs including linear, inverse, square-root, and piecewise regression functions of linear feature width to detect the species-specific threshold responses of birds to linear feature width. The abundance of each species was modeled using a Poisson error distribution with a log link function. We assessed relative model fit using Akaike’s information criterion (AIC) (Burnham and Anderson 2002) and plotted predicted abundance versus width from the model with the lowest AIC for each species. Goodness of fit was assessed using pseudo-R² (% deviance explained by the GLM with the lowest AIC).

RESULTS

Local species richness (α diversity)

Our first model for total bird species showed a statistically significant positive relationship between alpha richness and linear feature width (Z = 3.746, β = 0.005, SE = 0.001, p < 0.001) (Fig. 2). The model exhibited good fit with no outliers in the residuals when generating QQ plot residuals in the package DHARMa (KS test, p = 0.78; dispersion test, p = 0.93; outlier test, p = 1). For birds typically associated with shrubby vegetation, greater alpha richness was associated with wider linear features (Z = 4.681, β = 0.009, SE = 0.001, p < 0.001) (Fig. 2). The model was statistically significant and displayed a good fit with no outliers in the residuals (KS test, p = 0.94; dispersion test, p = 1; outlier test, p = 0.88). The model for alpha richness of mature forest and forest interior birds using a negative binomial GLM was statistically significant (Z = -2.105, β = -0.005, SE = 0.002, p = 0.03), and DHARMa QQ plot residuals (KS test, p = 0.92; dispersion test, p = 0.72, outlier test, p = 0.9) (Fig. 3) showed no outliers or overdispersion. This indicated a negative response of mature forest specialists and forest interior species to increasing linear feature width. Residual plots and model diagnostics for the GLMs are provided in Appendix 3.

Changes in species composition (β diversity)

The NMDS ordination (Fig. 4) (stress = 0.25, width p = 0.001, and r² = 0.30; Pearson correlation estimates, t = 5.279, and p < 0.001) showed clear patterns in the distribution of songbird species as a function of linear feature width. In the ordination plot, there were two distinct but relatively loose clusters for pipelines/transmission lines (orange dots) versus seismic lines (blue dots), indicating that sites within each linear feature type had similar species compositions.

The species that showed a strong association with wider linear features included those that use shrubby or early seral vegetation for nesting and/or living and foraging during the breeding season (e.g., Alder Flycatcher [Empidonax alnorum], Clay-colored Sparrow, Lincoln’s Sparrow, LeConte’s Sparrow, Palm Warbler [Setophaga palmarum], Wilson’s Warbler [Cardellina pusilla], Magnolia Warbler [Setophaga magnolia], and Orange-crowned Warbler [Leiothlypis celata]). Additionally, species known to use shrubby wetlands and moist habitats during the breeding season in Alberta, such as Northern Waterthrush (Parkesia noveboracensis), Swamp Sparrow (Melospiza georgiana), Yellow Warbler (Setophaga petechia), and Yellow-bellied Flycatcher (Empidonax flaviventris), appeared to closely associate with wider linear features.

In contrast, mature forest and forest interior species such as Brown Creeper (Certhia americana), Black-throated Green Warbler (Setophaga virens), Ovenbird (Seiurus aurocapilla), Philadelphia Vireo (Vireo philadelphicus), Bay-breasted Warbler (Setophaga castanea), and Red-breasted Nuthatch (Sitta canadensis) were more likely to be found at points adjacent to narrow linear features (Fig. 4).

Species-specific threshold responses

In general, species that we identified as using shrubby vegetation for nesting and foraging showed an increase in abundance with increasing linear feature width. These species were predicted best by the linear model as the best fit except for the Alder Flycatcher, which was predicted best by the square root model (Fig. 5, Table 1).

Abundance of Common Yellowthroat (Geothlypis trichas), American Robin (Turdus migratorius), and Lincoln’s Sparrow increased along wider linear features according to a piecewise regression model. Abundance of Common Yellowthroat and American Robin did not increase significantly until linear features were at least 50 m wide, while abundance of Lincoln’s Sparrow increased steeply with linear feature width up to a breakpoint of approximately 25 m (Fig. 5). Abundance of White-throated Sparrow (Zonotrichia albicollis) was best predicted by the piecewise regression model; abundance declined with linear feature width up to approximately 65 m, then increased (Table 1). Both the Ovenbird and the Red-eyed Vireo (Vireo olivaceus) showed a sudden drop in abundance when linear feature width exceeded the average seismic line width of 4–8 m (Fig. 5).

Tennessee Warbler (Leiothlypis peregrina) and Swainson’s Thrush (Catharus ustulatus) appeared to have negative responses to linear feature width, and the inverse model best described their trends (Table 1). Abundance of both Dark-eyed Junco (Junco hyemalis) and Ruby-crowned Kinglet (Regulus calendula) increased with linear feature width and reached a threshold at 10–15 m and ~30 m, respectively, beyond which there was a decrease (Table 1). The relative fit of models with different functional forms of linear feature width is presented in Appendix 4. Model coefficients and 95% confidence intervals for the linear feature width effect from the top three models for each species analyzed in this study are listed in Appendix 5.

DISCUSSION

To understand how linear features of varying width influence songbird assemblages, we used ARU surveys to assess the community- and species-level responses to different soft linear features in the boreal forest of Alberta. Our results suggest that linear feature width is a key driver of bird diversity, and species richness and composition. Overall, wider linear features (e.g., pipelines and/or transmission lines) and the forest adjacent to them was associated with greater species richness compared to forest edges that border narrow linear features such as seismic lines (Fig. 2). Additionally, we observed changes in species composition as linear feature width increased. Species associated with shrubby, early seral vegetation and open habitats appeared to prefer wider linear features over narrower ones. Moreover, species that have affinities for shrubby vegetation for nesting and foraging had a close association with wider linear features in the NMDS ordination space (Fig. 4), which suggests that greater species richness in areas with wider linear features is caused, at least partially, by the attraction of early successional and shrub-associated species to such areas.

We suggest that three major factors contribute to the positive relationship between species richness and the width of linear features, as well as the compositional changes observed in surveys done on wider linear features: (1) more heterogeneous and complex vegetation structure, (2) availability of novel habitat conditions required by certain groups, and (3) differential edge effects on selective species.

Structural complexity of vegetation generally correlates positively with bird diversity and species composition (MacArthur and MacArthur 1961), especially for insectivorous bird species (Ferger et al. 2014, Khamcha et al. 2018). Ground-based vegetation surveys and remote sensing analyses have often found that wider linear features such as pipelines and transmission lines have greater variability in vegetation structure than narrower linear features (Clarke et al. 2006, Nuijten et al. 2021). While we did not include vegetation conditions on linear features in our models, our results support several studies on transmission lines that have suggested that shrub-associated birds are attracted to such areas because of changes in vegetation structure on the line per se (Confer and Pascoe 2003, King et al. 2009). Although pipelines have not been surveyed as much as other energy sector linear features, previous bird surveys along pipelines have found that species richness increases with increasing line width (Fleming 2001, Langlois 2017).

We found strong evidence that both the abundance and species richness of birds that rely on shrubby and early seral habitats show a distinct increase as the width of linear features increases. All soft linear features create a vegetation structure that differs from the adjacent forest: linear features tend to have more graminoid and forb cover and often have denser shrubs (Askins et al. 2012, MacDonald et al. 2020). However, narrow linear features such as seismic lines do not seem to attract many early seral bird species, presumably because there is not enough area to support a territory in a long but narrow area. The shrub-associated species we observed to have greater abundance with increasing linear feature width, such as Lincoln’s Sparrow (Sánchez et al. 2022), Common Yellowthroat (Bulluck and Buehler 2006), and Alder Flycatcher (Mahon et al. 2019), are also more likely to be found on or near footprints such as abandoned well sites, which are 1-ha polygonal features that are often characterized by shrubby vegetation and grass. Other species that increased with linear feature width (e.g., LeConte’s Sparrow, Clay-colored Sparrow, and Yellow Warbler) are typically associated with areas that have low canopy cover and vegetation characterized by grass and disturbed vegetation (Westworth and Telfer 1993, Foster et al. 2017). However, species such as the Red-eyed Vireo, which uses shrubs for nesting yet prefers a greater canopy cover and is known for canopy-foraging (Siepielski et al. 2001, Marshall et al. 2002), seem to be abundant and closely associated with narrow linear features.

One novel finding of our study was that species known to nest and forage in moist habitats and shrubby wetlands (e.g., Common Yellowthroat, Northern Waterthrush, Swamp Sparrow, Yellow Warbler, and Yellow-bellied Flycatcher) were more likely to be found in wider linear features, and their abundance increased as the line width increased. These results suggest that a currently unknown mechanism (perhaps associated with linear feature creation or maintenance) is causing an accumulation of water on the linear feature and/or adjacent forest (Raiter et al. 2018). We have observed some areas on wide linear features at the base of hills that resemble conditions found in natural wetlands (Kalukapuge, personal observation).

The declining trend in species richness of mature forest and forest interior species such as the Ovenbird, Canada Warbler, Black-throated Green Warbler, and Brown Creeper (Fig. 3) in response to increasing line width contrasts with the patterns observed for shrubby and early seral species increases. Negative edge effects and/or the lack of suitable habitat conditions in wider linear features presumably causes these species to avoid such areas.

A fundamental challenge in assessing how linear features influence birds is the limitations inherent in using point count/ARU data to measure bird habitat use. Whether the species we observed at greater abundance on wider linear features have territories that are entirely within the linear feature or include the forest edge cannot be established from these types of data. Compositional differences in bird communities associated with wider linear features could simply be due to the combination of birds that use the forest and birds that (1) need shrubs and grass, or (2) are found on the linear feature only. Alternatively, it is possible that wider linear features create larger edge effects caused by spillover of the birds from the linear feature into the forest. Spillover could create cumulative effects whereby birds that live mostly on the linear feature but sometimes use the edge, start to compete for resources with the birds living in the forest. Ultimately, this could give rise to a new bird community through novel species interactions (Anderson et al. 1977). Further work is needed to determine if wider linear features create larger edge effects than narrower linear features (Khamcha et al. 2018, Langlois et al. 2023). If this is the case, then loss of specific species from the forest community may explain the resulting compositional differences we observed as linear features became wider.

Reliance on unlimited-distance, point count-style acoustic surveys captures bird vocalizations at a distance. Therefore, we could not determine which species were singing from the linear feature, were using the edge, or were present only in the forest interior. To address this, we recommend that methods of estimating the distance of birds to the ARU be incorporated into future studies. The intensive sound localization approaches introduced by Gregoire et al. (2022) would be particularly valuable in understanding the use of space by species on linear features and edges. Furthermore, the linear features selected for this study are situated within mature forest stands. Whether linear feature width would have any effect in young, regenerating, and early successional forest stands created by fire or harvesting remains unknown. In this study, we assumed that the detectability of species in wider linear features is the same as in narrower ones. However, in future analyses, it would be insightful to determine if there is any difference in the detectability of different species due to linear feature width.

It is well established that height and density of trees in regrowing forest influences bird species diversity, richness, and abundance (Schieck and Song 2006, Lankau et al. 2013, Zhao et al. 2013, Owen et al. 2020), but the point at which shrubs and trees on a linear feature become dense or tall enough to allow forest birds to begin using those areas as habitat is another key uncertainty. Such information will be essential for regulators charged with setting linear feature recovery targets. Different forest types regenerate at different rates, which may have varying impacts on songbirds (Mahon et al. 2019). Studies have shown mixed results regarding the distinctiveness of bird communities in mixedwood and deciduous forests. Hobson and Bayne (2000) found that bird communities in mixedwood forest stands are distinguishable from those in other forest types. Conversely, some studies have shown that the number of individuals, species richness, and community structure can be similar between mixedwood and deciduous forests (Machtans and Latour 2003, Richmond et al. 2015). While we have assumed minimal or no effects from using two similar upland forest types, given that the mixedwood forest in this study is primarily deciduous dominant, future studies could benefit from incorporating fine-scale forest type covariates such as ecosite classification or tree species composition to provide greater clarity. Additional information on how birds react to linear features that dissect forest types such as spruce or pine is also needed to assess the overall impact of the energy sector on boreal bird communities; this is an area of active data collection.

CONCLUSIONS

We clearly demonstrate that songbirds in mixedwood and deciduous forests respond to the width of linear features in the boreal forest. Therefore, we recommend moving away from the traditional grouping of linear features into vegetated (soft) versus non-vegetated (hard) that are currently used in regulatory decision making and land use planning. Only then will we be able to fully understand how energy sector development is impacting birds and determine what types of restoration and recovery actions are likely to have the greatest benefit. Additionally, if alterations in the species pool in areas of wider linear features prove detrimental to the primary bird community, and those species are not merely moving away from those cleared areas into the forest but are also losing their habitat, resulting in local-level declines, then the restoration and management priorities should focus on wider linear features.

LITERATURE CITED

Alberta Biodiversity Monitoring Institute (ABMI). 2015. Manual for species modeling and intactness (20029). Version 2016-04-14. Alberta Biodiversity Monitoring Institute, Alberta, Canada.

Alberta Biodiversity Monitoring Institute biodiversity browser. https://beta.abmi.ca/biobrowser

Anderson, M. J. 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecology 26(1):32-46. https://doi.org/10.1111/j.1442-9993.2001.01070.pp.x

Anderson, M. J., T. O. Crist, J. M. Chase, M. Vellend, B. D. Inouye, A. L. Freestone, N. J. Sanders, H. V. Cornell, L. S. Comita, K. F. Davies, et al. 2011. Navigating the multiple meanings of β diversity: a roadmap for the practicing ecologist. Ecology Letters 14(1):19-28. https://doi.org/10.1111/j.1461-0248.2010.01552.x

Anderson, S. H., K. Mann, and H. H. Shugart, Jr. 1977. The effect of transmission-line corridors on bird populations. American Midland Naturalist 97(1):216-221. https://doi.org/10.2307/2424698

Askins, R. A., C. M. Folsom-O’Keefe, and M. C. Hardy. 2012. Effects of vegetation, corridor width and regional land use on early successional birds on powerline corridors. PLoS ONE 7(2):e31520. https://doi.org/10.1371/journal.pone.0031520

Ball, J. R., E. M. Bayne, and C. S. Machtans. 2009. Energy sector edge effects on songbird nest fate and nest productivity in the boreal forest of western Canada: a preliminary analysis. Pages 161-170 in Proceedings of the Fourth International Partners in Flight Conference: Tundra to Tropics. McAllen, Texas.

Bayne, E. M., L. Habib, and S. Boutin. 2008. Impacts of chronic anthropogenic noise from energy‐sector activity on abundance of songbirds in the boreal forest. Conservation Biology 22(5):1186-1193. https://doi.org/10.1111/j.1523-1739.2008.00973.x

Bayne, E., L. Leston, C. L. Mahon, P. Sólymos, C. Machtans, H. Lankau, J. R. Ball, S. L. Van Wilgenburg, S. G. Cumming, T. Fontaine, F. K. A. Schmiegelow, and S. J. Song. 2016. Boreal bird abundance estimates within different energy sector disturbances vary with point count radius. Condor: Ornithological Applications 118(2):376-390. https://doi.org/10.1650/CONDOR-15-126.1

Bayne, E. M., S. L. Van Wilgenburg, S. Boutin and K. A. Hobson. 2005. Modeling and field-testing of Ovenbird (Seiurus aurocapillus) responses to boreal forest dissection by energy sector development at multiple spatial scales. Landscape Ecology 20(2):203-216. https://doi.org/10.1007/s10980-004-2265-9

Bulluck, L. P., and D. A. Buehler. 2006. Avian use of early successional habitats: Are regenerating forests, utility right-of-ways and reclaimed surface mines the same? Forest Ecology and Management 236(1):76-84. https://doi.org/10.1016/j.foreco.2006.08.337

Burnham, K. P., and D. R. Anderson. 2002. Model selection and multi-model inference: a practical information-theoretic approach. 2nd edition. Springer, New York.

Celis-Murillo, A., J. L. Deppe, and M. F. Allen. 2009. Using soundscape recordings to estimate bird species abundance, richness, and composition. Journal of Field Ornithology 80(1):64-78. https://doi.org/10.1111/j.1557-9263.2009.00206.x

Clarke, D. J., K. A. Pearce, and J. G. White. 2006. Powerline corridors: degraded ecosystems or wildlife havens?. Wildlife Research 33(8): 615-626. https://doi.org/10.1071/WR05085

Confer, J. L., and S. M. Pascoe. 2003. Avian communities on utility rights-of-ways and other managed shrublands in the northeastern United States. Forest Ecology and Management 185(1-2):193-205. https://doi.org/10.1016/S0378-1127(03)00255-X

Crosby, A. D., L. Leston, E. M. Bayne, P. Sólymos, C. L. Mahon, J. D. Toms, T. D. S. Docherty, and S. J. Song. 2023. Domains of scale in cumulative effects of energy sector development on boreal birds. Landscape Ecology 38:3173-3188. https://doi.org/10.1007/s10980-023-01779-8

Dabros, A., K. L. Higgins, and J. Pinzon. 2022. Seismic line edge effects on plants, lichens and their environmental conditions in boreal peatlands of Northwest Alberta (Canada). Restoration Ecology 30(4):e13468. https://doi.org/10.1111/rec.13468

Dabros, A., M. Pyper, and G. Castilla. 2018. Seismic lines in the boreal and arctic ecosystems of North America: environmental impacts, challenges, and opportunities. Environmental Reviews 26(2):214-229. https://doi.org/10.1139/er-2017-0080

Dickie, M., S. R. McNay, G. D. Sutherland, M. Cody, and T. Avgar. 2020. Corridors or risk? Movement along, and use of, linear features varies predictably among large mammal predator and prey species. Journal of Animal Ecology 89(2):623-634. https://doi.org/10.1111/1365-2656.13130

Domahid, Z., J. Shonfield, S. E. Nielsen, J. R. Spence, and E. M. Bayne. 2019. Spatial distribution of the Boreal Owl and Northern Saw-whet Owl in the boreal region of Alberta, Canada. Avian Conservation and Ecology 14(2):14. https://doi.org/10.5751/ACE-01445-140214

Dyer, S. J., J. P. O’Neill, S. M. Wasel, and S. Boutin. 2002. Quantifying barrier effects of roads and seismic lines on movements of female woodland caribou in northeastern Alberta. Canadian Journal of Zoology 80(5):839-845. https://doi.org/10.1139/z02-060

Fahrig, L. 2003. Effects of habitat fragmentation on biodiversity. Annual Review of Ecology, Evolution, and Systematics 34(1):487-515. https://doi.org/10.1146/annurev.ecolsys.34.011802.132419

Fahrig, L., V. Arroyo-Rodríguez, J. R. Bennett, V. Boucher-Lalonde, E. Cazetta, D. J. Currie, F. Eigenbrod, A. T. Ford, S. P. Harrison, J. A. G. Jaeger, et al. 2019. Is habitat fragmentation bad for biodiversity? Biological Conservation 230:179-186. https://doi.org/10.1016/j.biocon.2018.12.026

Ferger, S. W., M. Schleuning, A. Hemp, K. M. Howell, and K. Böhning‐Gaese. 2014. Food resources and vegetation structure mediate climatic effects on species richness of birds. Global Ecology and Biogeography 23(5):541-549. https://doi.org/10.1111/geb.12151

Filicetti, A. T., and S. E. Nielsen. 2022. Effects of wildfire and soil compaction on recovery of narrow linear disturbances in upland mesic boreal forests. Forest Ecology and Management 510:120073. https://doi.org/10.1016/j.foreco.2022.120073

Finnegan, L., K. E. Pigeon, and D. MacNearney. 2019. Predicting patterns of vegetation recovery on seismic lines: informing restoration based on understory species composition and growth. Forest Ecology and Management 446:175-192. https://doi.org/10.1016/j.foreco.2019.05.026

Fischer, J., and D. B. Lindenmayer. 2007. Landscape modification and habitat fragmentation: a synthesis. Global Ecology and Biogeography 16(3):265-280. https://doi.org/10.1111/j.1466-8238.2007.00287.x

Fleming, W. D. 2001. Effects of pipeline rights-of-way on forest birds in the boreal forest of Alberta. Thesis. University of Alberta, Edmonton, Alberta, Canada.

Foster, K. R., C. M. Godwin, P. Pyle, and J. F. Saracco. 2017. Reclamation and habitat‐disturbance effects on landbird abundance and productivity indices in the oil sands region of northeastern Alberta, Canada. Restoration Ecology 25(4):532-538. https://doi.org/10.1111/rec.12478

Franklin, C. M. A., A. T. Filicetti, and S. E. Nielsen. 2021. Seismic line width and orientation influence microclimatic forest edge gradients and tree regeneration. Forest Ecology and Management 492:119216. https://doi.org/10.1016/j.foreco.2021.119216

Government of Alberta. 1994. Environmental protection guidelines for pipelines. https://open.alberta.ca/dataset/eda09f7a-652a-44ba-970d-617e11e3030b/resource/79a54841-de93-41d5-ab1c-981eec412a39/download/enviroprotectionguidepipelines-il-1994.pdf

Government of Alberta. 2017. Alberta’s approach to achieve caribou recovery. https://open.alberta.ca/dataset/94cbf5b9-03db-411d-b3a4-0bafb058fd29/resource/f6a57700-44df-409b-9e89-d182611e4564/download/altaapproachcaribourecovery-dec2017.pdf

Gregoire, J. M., R. W. Hedley, and E. M. Bayne. 2022. Canada Warbler response to vegetation structure on regenerating seismic lines. Avian Conservation and Ecology 17(2):26. https://doi.org/10.5751/ACE-02262-170226

Hartig, F., and M. F. Hartig. 2021. Package ‘DHARMa’. R package version 0.3.

Hobson, K. A., and E. Bayne. 2000. Breeding bird communities in boreal forest of western Canada: consequences of “unmixing” the mixedwoods. Condor 102(4):759-769. https://doi.org/10.2307/1370303

Kemper, J. T., and S. E. Macdonald. 2009. Directional change in upland tundra plant communities 20‐30 years after seismic exploration in the Canadian low‐arctic. Journal of Vegetation Science 20(3):557-567. https://doi.org/10.1111/j.1654-1103.2009.01069.x

Khamcha, D., R. T. Corlett, L. A. Powell, T. Savini, A. J. Lynam, and G. A. Gale. 2018. Road induced edge effects on a forest bird community in tropical Asia. Avian Research 9:20. https://doi.org/10.1186/s40657-018-0112-y

King, D. I., R. B. Chandler, J. M. Collins, W. R. Petersen, and T. E. Lautzenheiser. 2009. Effects of width, edge and habitat on the abundance and nesting success of scrub-shrub birds in powerline corridors. Biological Conservation 142(11):2672-2680. https://doi.org/10.1016/j.biocon.2009.06.016

Langlois, L. A. 2017. Effects of Marcellus shale gas infrastructure on forest fragmentation and bird communities in northcentral Pennsylvania. Dissertation. Pennsylvania State University, University Park, Pennsylvania, USA.

Langlois, L. A., S. J. Brenner, and M. C. Brittingham. 2023. Collocating pipelines to minimize fragmentation: evaluating ecological costs of a shale gas mitigation practice. Journal of Wildlife Management 87(7):e22468. https://doi.org/10.1002/jwmg.22468

Lankau, H. E. G. 2014. Songbird responses to regenerating seismic lines in the boreal forest. Thesis. University of Alberta, Edmonton, Alberta, Canada.

Lankau, H. E., E. M. Bayne, and C. S. Machtans. 2013. Ovenbird (Seiurus aurocapilla) territory placement near seismic lines is influenced by forest regeneration and conspecific density. Avian Conservation and Ecology 8(1):5. https://doi.org/10.5751/ACE-00596-080105

Latham, A. D. M., M. C. Latham, M. S. Boyce, S. Boutin. 2011. Movement responses by wolves to industrial linear features and their effect on woodland caribou in northeastern Alberta. Ecological Applications 21(8):2854-2865. https://doi.org/10.1890/11-0666.1

Leston, L., E. Bayne, J. D. Toms, C. L. Mahon, A. Crosby, P. Sólymos, J. Ball, S. J. Song, F. K. A. Schmiegelow, D. Stralberg, and T. D. S. Docherty. 2023. Comparing alternative methods of modelling cumulative effects of oil and gas footprint on boreal bird abundance. Landscape Ecology 38(1):147-168. https://doi.org/10.1007/s10980-022-01531-8

MacArthur, R. H., and J. W. MacArthur. 1961. On bird species diversity. Ecology 42(3):594-598. https://doi.org/10.2307/1932254

MacDonald, A., S. F. Bartels, S. E. Macdonald, K. E. Pigeon, D. MacNearney, and L. Finnegan. 2020. Wildlife forage cover and composition on pipeline corridors in Alberta: implications for wildlife conservation. Forest Ecology and Management 468:118189. https://doi.org/10.1016/j.foreco.2020.118189

Machtans, C. S. 2006. Songbird response to seismic lines in the western boreal forest: a manipulative experiment. Canadian Journal of Zoology 84(10):1421-1430. https://doi.org/10.1139/z06-134

Machtans, C. S., and P. B. Latour. 2003. Boreal forest songbird communities of the Liard Valley, Northwest Territories, Canada. Condor 105(1):27-44. https://doi.org/10.1093/condor/105.1.27

Mahon, C. L., G. L. Holloway, E. M. Bayne, and J. D. Toms. 2019. Additive and interactive cumulative effects on boreal landbirds: winners and losers in a multi‐stressor landscape. Ecological Applications 29(5):e01895. https://doi.org/10.1002/eap.1895

Marshall, M. R., R. J. Cooper, J. A. DeCecco, J. Strazanac, and L. Butler. 2002. Effects of experimentally reduced prey abundance on the breeding ecology of the Red‐-eyed Vireo. Ecological Applications 12(1):261-280. https://doi.org/10.1890/1051-0761(2002)012[0261:EOERPA]2.0.CO;2

Morrison, M. L. 1981. The structure of western warbler assemblages: analysis of foraging behavior and habitat selection in Oregon. Auk 98(3):578-588.

Nuijten, R. J., N. C. Coops, C. Watson, and D. Theberge. 2021. Monitoring the structure of regenerating vegetation using drone-based digital aerial photogrammetry. Remote Sensing 13(10):1942. https://doi.org/10.3390/rs13101942

Oberg, P. R. 2001. Responses of mountain caribou to linear features in a west-central Alberta landscape. Thesis. University of Alberta, Edmonton, Alberta, Canada.

Oksanen, J., F. G. Blanchet, M. Friendly, R. Kindt, P. Legendre, D. McGlinn, et al. 2019. Vegan: R Package, version 2.5-6.

Owen, K. C., D. J. Melin, F. A. Campos, L. M. Fedigan, T. W. Gillespie, and D. J. Mennill. 2020. Bioacoustic analyses reveal that bird communities recover with forest succession in tropical dry forests. Avian Conservation and Ecology 15(1):25. https://doi.org/10.5751/ACE-01615-150125

Pattison, C. A., M. S. Quinn, P. Dale, and C. P. Catterall. 2016. The landscape impact of linear seismic clearings for oil and gas development in boreal forest. Northwest Science 90(3):340-354. https://doi.org/10.3955/046.090.0312

Raiter, K. G., S. M. Prober, H. P. Possingham, F. Westcott, and R. J. Hobbs. 2018. Linear infrastructure impacts on landscape hydrology. Journal of Environmental Management 206:446-457. https://doi.org/10.1016/j.jenvman.2017.10.036

R Core Team. 2022. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.

Rempel, R. S., K. A. Hobson, G. Holborn, S. L. Van Wilgenburg, and J. Elliott. 2005. Bioacoustic monitoring of forest songbirds: interpreter variability and effects of configuration and digital processing methods in the laboratory. Journal of Field Ornithology 76(1):1-11. https://doi.org/10.1648/0273-8570-76.1.1

Rich, A. C., D. S. Dobkin, and L. J. Niles. 1994. Defining forest fragmentation by corridor width: the influence of narrow forest‐dividing corridors on forest‐-nesting birds in southern New Jersey. Conservation Biology 8(4):1109-1121. https://doi.org/10.1046/j.1523-1739.1994.08041109.x

Richmond, S., E. Jenkins, A. Couturier, and M. Cadman. 2015. Thresholds in forest bird richness in response to three types of forest cover in Ontario, Canada. Landscape Ecology 30:1273-1290. https://doi.org/10.1007/s10980-015-0183-7

Sánchez, N. V., L. Sandoval, R. W. Hedley, C. C. St. Clair, and E. M. Bayne. 2022. Relative importance for Lincoln’s Sparrow (Melospiza lincolnii) occupancy of vegetation type versus noise caused by industrial development. Frontiers in Ecology and Evolution 10:198. https://doi.org/10.3389/fevo.2022.810087

Schieck, J., and S. J. Song. 2006. Changes in bird communities throughout succession following fire and harvest in boreal forests of western North America: literature review and meta-analyses. Canadian Journal of Forest Research 36(5):1299-1318. https://doi.org/10.1139/x06-017

Schneider, R., and S. Dyer. 2006. Death by a thousand cuts: impacts of in situ oil sands development on Alberta's boreal forest. Pembina Institute, Calgary, Alberta, Canada.

Sedláček, O., J. Vokurková, M. Ferenc, E. N. Djomo, T. Albrecht, and D. Hořák. 2015. A comparison of point counts with a new acoustic sampling method: a case study of a bird community from the montane forests of Mount Cameroon. Ostrich 86(3):213-220. https://doi.org/10.2989/00306525.2015.1049669

Siepielski, A. M., A. D. Rodewald, and R. H. Yahner. 2001. Nest site selection and nesting success of the Red-eyed Vireo in central Pennsylvania. Wilson Bulletin 113(3):302-307. https://doi.org/10.1676/0043-5643(2001)113[0302:NSSANS]2.0.CO;2

Van Wilgenburg, S. L., P. Sólymos, K. J. Kardynal, and M. D. Frey. 2017. Paired sampling standardizes point count data from humans and acoustic recorders. Avian Conservation and Ecology 12(1):13. https://doi.org/10.5751/ACE-00975-120113

Westworth, D. A., and E. S. Telfer. 1993. Summer and winter bird populations associated with five age-classes of aspen forest in Alberta. Canadian Journal of Forest Research 23(9):1830-1836. https://doi.org/10.1139/x93-233

Zhao, Q., E. T. Azeria, M.-L. Le Blanc, J. Lemaître, and D. Fortin. 2013. Landscape-scale disturbances modified bird community dynamics in successional forest environment. PLoS ONE 8(11):e81358. https://doi.org/10.1371/journal.pone.0081358