Research Paper

INTRODUCTION

Knowledge of how a species selects its breeding habitat allows for identification and management of that habitat on local and landscape scales. For forest-dwelling birds, vegetation cues directly signal ultimate factors such as nest site and food availability, and risks of predation and parasitism (Hildén 1965). Information about proximate factors is commonly used in models to predict a species’ habitat or distribution across a landscape, which are then used to inform conservation and management decisions. Whereas most studies modeling forest-dwelling birds use environmental variables representing landcover, and vegetation composition and structure to represent habitat (e.g., Zlonis et al. 2017, Westwood et al. 2019), very few studies also incorporate social cues, such as the influence of intra- and interspecific interactions (Campomizzi et al. 2008). Nevertheless, interactions among organisms are fundamental to determining their realized niche, and thus likely to influence their distribution. Disentangling the relative roles of environmental cues and social information in habitat selection is essential to identify more accurate conservation strategies for habitats and species, particularly with increasing anthropogenic global changes (Thomas et al. 2001, Campomizzi et al. 2008).

A primary mechanism driving clustered distributions in species is conspecific attraction (Campomizzi et al. 2008). Researchers have suggested various reasons that individuals may select habitat based on the presence of conspecifics, including finding more potential mates, benefits of group vigilance, and use of conspecifics as indicators of habitat quality (e.g., resource type and quality, previously successful breeding; summarized by Muller et al. 1997). Social attraction may be occurring at multiple times during habitat selection, during migration (Mukhin et al. 2008), pre-breeding searching and settlement (Ward and Schlossberg 2004), and post-breeding (Nocera et al. 2006, Betts et al. 2008). Although vegetation cues require direct sampling of possible sites by individuals before selecting a breeding location, locating conspecifics may be a more efficient means of site assessment (Hildén 1965, Cornell and Donovan 2010, Valente et al. 2021). Birds, particularly migratory birds, have short breeding seasons, so decisions related to habitat selection must be made quickly, and their breeding communities must reassemble themselves annually (Nocera and Betts 2010). Fletcher and Sieving (2010) reconcile the use of social information (social cues) in landscape ecology studies; landscape features can change the accessibility and value of social information, and this interaction may profoundly affect the efficiency and outcome of habitat selection.

Species density is frequently used to infer habitat quality under the assumption that individuals occur at greater densities in better quality habitats, or in areas where they can achieve higher survival rates and reproduction rates (Fretwell and Lucas 1969, Doligez and Boulinier 2008). However, density alone can be a misleading proxy for habitat quality, and social interactions such as conspecific attraction could lead to an aggregation of individuals from the same species even when the habitat is unsuitable (Van Horne 1983). This behavior has been postulated in the case of Canada Warbler (Cardellina canadensis, CAWA), a species at risk in Canada (Environment Canada 2016), where it occurs in harvested stands in the western boreal forest (Flockhart et al. 2016, Hunt et al. 2017). The Canada Warbler breeding season and its habitats are reasonably well described, but they vary regionally, particularly with different types of land use, forest types, and forest disturbances (i.e., natural: wind, wildfire; human: forest harvest; Haché et al. 2014, Ball et al. 2016). Although there is evidence that habitat use by Canada Warbler in disturbed areas due to forestry activity is influenced by conspecific attraction, it has been found that high conspecific density in these harvested areas leads to low breeding success (Flockhart et al. 2016, Hunt et al. 2017). This evidence suggests that conspecific density might be higher in disturbed (harvested) than undisturbed (unharvested) stands. However, this hypothesis has not been tested either in an undisturbed stand or at the landscape scale.

The use of artificial conspecific cues (song playbacks) has been proposed to test the influence of conspecific attraction in habitat choices (e.g., Betts et al. 2008, Cornell and Donovan 2010, Albrecht-Mallinger and Bulluck 2016), and as a conservation strategy to promote settlement of a species in more suitable areas (Schlossberg and Ward 2004). The objective of our study was to evaluate three proximate cues that may be important for habitat choices made by Canada Warbler: (1) disturbance at the local and landscape level, (2) shrub cover, canopy tree cover, and tree height within each occurrence site, and (3) the presence of vocal conspecific cues. Previous studies used artificial conspecific cues to attract CAWA, but their protocol included calls of shorter duration to increase detectability when the species was already expected to be breeding (Flockhart et al. 2016, Hunt et al. 2017, Westwood et al. 2019). We instead assessed the influence of conspecific cues in habitat choices through the use of conspecific playbacks during the pre-breeding season (at least seven days before breeding). We predicted that (1) where playbacks are not used, individuals will be more attracted to the sites within disturbed landscapes with clusters of other Canada Warbler, and (2) where continuous playback is used in undisturbed habitat, Canada Warbler will preferentially select those habitats.

METHODS

Study area

We conducted the study within Ecoregion 4W (Pigeon River), which is located in the Ontario Shield Ecozone in Northwestern Ontario (Fig. 1). Mixed forest is the most extensive land cover class (33.2%), following by areas of scattered forest (19.3%), water (17.5%), coniferous forest (11.5%), deciduous forest (10.6%), and cutovers (recently harvested forest, 3.6%; Crins et al. 2009). Predominant land uses include timber harvesting, resource-based tourism, mineral exploration, and agriculture; the city of Thunder Bay is the only urbanized community in the ecoregion (Wester et al. 2018).

We selected three Ontario Breeding Bird Atlas (OBBA) plots (10 x 10 km) based on information from the previous OBBA 2001–2005 (Cadman et al. 2007), after pilot visits during May of 2021. We identified areas disturbed by forestry activity and the time since harvest for each plot using the global forest change layer from 2000 to 2020 (Hansen et al. 2013), publicly available through the Global Land Analysis and Discovery webpage from the University of Maryland (https://glad.umd.edu/dataset/gedi/). Based on the percentage of area within the plots with current forestry activity and activity documented in the global forest change layer, i.e., from 2000 to present, we classified the OBBA plots to represent three forested landscapes, one to represent low disturbance (≤ 20% harvested), one to represent medium disturbance (21–50% harvested), and the third to represent high disturbance (51–70% harvested). The majority of disturbed areas dated to current activity or forestry activity not more than 7 years before. We classified stands within the landscapes by time since forest harvest as follows: (1) unharvested, (2) early succession (0–7 years since harvest), (3) mid succession (9–14 years since harvest), and later succession (18–22 years since harvest).

The OBBA plot considered to represent the “low disturbance” landscape was located in part in Quetico (Gwetaming) Provincial Park (Fig. 1). Quetico encompasses 4718 km² with numerous lakes and streams (Ontario Ministry of Natural Resources 2018). Quetico is located in Northwestern Ontario, south of the town of Atikokan, approximately 160 km west of Thunder Bay, and adjacent to the Canada-United States (U.S.) boundary. The park occupies a zone of transition between boreal forests to the north, mixed forests to the south, and Great Plains forests to the west and southwest. The other two plots were located entirely in the Dog River Matawin Forest Management Unit (Fig. 1), which is managed by Resolute Forest Products, Inc. in a long-term (2020–2030) lease. The “medium disturbance” landscape was located near Kabaigon, where forest management was done during a previous planning period (before 2020), and the “high disturbance” landscape was near Shebandowan, where there are recent post-harvested areas (≤ 20 years) and current timber extraction.

Bird surveys

Observations of occurrence and social aggregation

All survey methods were approved by the Lakehead University Animal Care Committee (Animal Utilization Protocol #1468679). We conducted bird point count surveys in each 10 x 10 km landscape during the breeding season from late May to early July of 2021 and 2022. We located the point counts randomly within each year since harvest category: unharvest = 83, early = 37, mid = 11, late = 14 points, a total of 145 points, 48 in the Quetico, 48 in the Kabaigon, and 49 in the Shebandowan area plots. Point counts were done at sites approximately 250–300 m apart from each other. The observations were conducted in favorable weather conditions, never during precipitation events or with wind > 20 km/h, both of which reduce the detectability of singing birds (Cadman et al. 2007). After a five-minute point count, the observers (commonly two) used speakers or mobile devices to apply a short playback protocol using Canada Warbler songs to increase the detectability of the species and to register the number of individual conspecific responses; the observers registered the total number of territorial males detected by sight or sound at each site.

The short playback protocol was modified from Flockhart et al. (2016) and Hunt et al. (2017) as follow: (1) 30 s of conspecific playback, (2) 1 min of silence, (3) 30 s of playback, and (4) 1 min of silence. The repetition of the playback and the silent periods helped to reduce the bias of artificial calls and the possible effect of individuals approaching just out of curiosity. The observers counted only the males that responded during or after the second playback period of conspecific vocalizations. We considered a site to have social aggregation when we found more than one male within a 60-m radius.

Canada Warbler pre-breeding settlement experiment

To test the influence of conspecific cues on Canada Warbler settlement, we used artificial conspecific playback during the pre-breeding season to implement an experimental protocol adapted from Betts et al. (2008) and Cornell and Donovan (2010) and followed the recommendations from Ahlering et al. (2010). We ensured that the study sites were vacant by conducting 5 min passive surveys (no conspecific songs) and 1 min conspecific playback during 2021 and 2022 breeding seasons. We considered a site acceptable for inclusion if Canada Warbler were not detected within 60 m of the playback. We selected 12 pairs of vacant sites to establish the experimental protocol, 12 for treatment (7-day conspecific playbacks), and other 12 for control (no playbacks); we located the sites at least 250 m from each other, in each landscape (10 x 10 km area) during the pre-breeding season (late May to mid-June of 2022 and 2023). We selected the sites using a gradient of shrub cover from high to low to assess the influence and importance of shrub cover in Canada Warbler pre-breeding settlement.

Our treatments consisted of Canada Warbler male vocalizations played for 8 h daily (from 05:00 to 13:00) using an automated speaker system (Fig. 2; Cupiche-Herrera et al. 2023) over a 7-d treatment period. To reduce habituation to playbacks, the soundtracks contained 3 min of vocalizations (calls or songs) followed by 30-s gaps between vocalizations of 1 min; we also programmed 10 min of continuous silence between soundtracks, and the songs and calls were alternated every 20 min; we used the Canada Warbler vocalizations publicly available in www.xeno-canto.org (see list of authors and recordings codes in Appendix 1, Table A1). During the visits to the sites (before and after the 7-d treatment period), we used the short playback protocol, as used for occurrence detection, to assess whether Canada Warbler was present at the treatment sites. We did not install the automated speaker system in control sites, which we only visited after the 7-d treatment period. At these sites, we used a point count with an initial 5-min silent observation followed by the short playback protocol (Fig. 2) to increase the Canada Warbler detectability; we used the short protocol once per control site to avoid the attraction of males from surrounding sites.

Vegetation survey

To assess potential vegetation cues for breeding site selection, the following vegetation data were collected at all sites where point counts were conducted: forest type (conifer, mixed, deciduous), tree canopy cover, shrub cover, and tree canopy height. We also noted the main tree and shrub species for the sites where we found Canada Warbler (Appendix 1, Table A2). To identify the forest type in each site we used observations in the field, as well as the Ontario Land Cover Compilation (15 m resolution layer) produced by the Ontario Ministry of Natural Resources (2023), publicly available through Ontario GeoHub (https://geohub.lio.gov.on.ca/); we recorded coniferous forest when conifer trees were ≥ 75% of the canopy cover; deciduous forest when deciduous trees were ≥ 75% of the canopy cover; and mixed forest when conifer trees were > 25% and deciduous trees > 25% of the canopy cover (Ontario Ministry of Natural Resources 2023). Canopy tree cover (trees > 5 m) and shrub cover (1–5 m) were measured with a spherical densiometer and the application “Canopy Capture” (2018) for radii of 50 m. We extracted tree canopy height average from each site through the Canopy Height 30 m resolution layer (Potapov et al. 2021), publicly available through the Global Land Analysis and Discovery webpage from the University of Maryland. We collected the data at the end of June and early July of each year of study (2021, 2022, 2023).

Data analysis

We modeled three types of Canada Warbler responses: occurrence, social aggregation (sites where we detected more than one individual), and settlement (Canada Warbler response after artificial conspecific cue during the pre-breeding season). These responses were analyzed against potential predictors: year, disturbance at the site level (where point counts were done), disturbance at the landscape level, forest type, tree canopy cover, shrub cover, tree canopy height, and treatment (the latter only used as a predictor for Canada Warbler settlement, see Table 1). We used a GLM (generalized linear model) with a logistic function structure using the package lme4 in the program R, version 3.6.3 (R Core Team 2021) to do the analysis. For numerical variables, we did a correlation analysis to reduce collinearity among them (Appendix 1, Tables A3, A4). Then we compared the best model sets for the three responses to test which habitat cues are more influential for Canada Warbler occurrence and social aggregation, and whether Canada Warbler settlement depends on habitat cues and/or playback cue (treatment); we used an information-theoretic approach to select the most parsimonious model for each response by contrasting values of Akaike’s Information Criterion corrected for small sample sizes (AICc; Burnham and Anderson 2002).

RESULTS

During 145 five-min point counts conducted in 2021 and 2022, we assessed 54 sites where Canada Warbler males were present and counted a total of 84 male individuals. In 2021, Canada Warbler was found at 33 sites and we counted 53 individuals, while in 2022, we detected Canada Warbler at 21 sites and counted 31 individuals. During the playback experiment, we detected a total of 13 individuals at treatment sites where the conspecific playback cues were implemented: we detected seven individuals in 2022 and six in 2023. Conversely, only three individuals were detected at control sites, all of them during the 2023 experiment period.

Occurrences were higher in areas where shrub cover was > 55% (Table 2, Fig. 3A). Shrub cover was the main predictor for all types of Canada Warbler responses and appeared in all top ranked models (Table 3). The dominant shrub species were Mountain Maple and Beaked Hazel (Acer spicatum and Corylus cornuta, Appendix 1, Table A2). Canopy cover and height were also important predictors of occurrence and social aggregation (Table 3, Fig. 3B, C). The dominant tree species included Paper Birch and White Spruce (Betula papyrifera and Abies balsamea Appendix 1, Table A2). Local-scale disturbance conditions (site) influenced Canada Warbler occurrence and settlement, but no such influence occurred at the landscape scale (Appendix 1, Table A5). Conspecific vocalizations (i.e., the treatment effect or effect of playbacks of calls or songs) appeared in the top ranked model for Canada Warbler settlement during the pre-breeding season (Table 3). Time since harvest also emerged as a key variable in the top models of social aggregation and Canada Warbler settlement. Early harvest has a positively influenced social aggregation (Table 3, Fig. 4A), while unharvested conditions favored settlement of Canada Warbler through the playback experiment (Fig. 4A).

Canada Warbler occurrences and social aggregation were more frequently encountered in mixed forest, followed by deciduous forest (Fig. 4B). Although we observed more occurrences in undisturbed sites (n = 32), social aggregation was higher in disturbed sites at both local and landscape scales (Fig. 4C, D); social aggregation was especially prevalent in recently harvested areas (< 9 years; Fig. 4A). Of the 22 sites across the three landscapes where we found ≥ 2 Canada Warbler individuals, 16 sites (73%) were in harvested areas. Conversely, of 32 sites where we recorded just one individual, 22 (69%) were located in undisturbed sites.

DISCUSSION

Our results suggest that both social and vegetation cues influence Canada Warbler habitat choices in the ecoregion 4W of Northwestern Ontario. Shrub cover emerged as one of the main predictors, with Canada Warbler detections more frequent in areas where shrub coverage exceeded 55%. Canopy cover and height also were variables predicting Canada Warbler occurrences and social aggregation, which indicate the preference of Canada Warbler for structured vertical stratification suggesting that even when Canada Warbler may occur in harvested areas, it still relied on old-growth forest features. In addition, Canada Warbler detections were higher in areas with mixed and deciduous forest; these results are consistent with what Haché et al. (2014) reported, where Canada Warbler was more common in areas with tall trees, mixed and dense forest.

Our study found that recently harvested sites supported higher levels of social aggregation, potentially because of the transient increase in undergrowth following timber extraction, which temporarily boosts resource availability. Previous studies in the western boreal forest have reported higher density of Canada Warbler in post-harvest cuts (regenerated areas > 5 years since disturbance; Ball et al. 2016, Hunt et al. 2017) and areas of light partial harvest (Becker et al. 2012). Another study in the Canadian Maritimes found no difference in predicted density of CAWA between areas managed for forest harvesting and protected areas (Westwood et al. 2019). Abundance or density is often used to infer habitat quality, because individuals occur at greater densities in favorable areas where they achieve higher survival and reproduction (Fretwell and Lucas 1969, Harrison et al. 2005, Doligez and Boulinier 2008). However, density alone can be misleading, because of conspecific attraction that can lead to the aggregation of individuals in unsuitable habitats.

The higher social aggregation of Canada Warbler in harvested sites could be due to the combination of high shrub cover availability (Reitsma et al. 2009) and the attraction to conspecifics used as a shortcut to the best ultimate cue, given the limited time individuals have to assess habitat and search for mates (Hunt et al. 2017). It is also possible that areas disturbed by forestry activity allow for easier detectability of conspecifics because of their calls carrying over longer distances. Even though landscape-scale disturbance (at the scale of the Ontario Breeding Bird Survey plots) did not appear as an important predictor, we observed Canada Warbler social aggregation was higher in the two OBBS plots with medium and high levels of disturbance. The use of conspecific cues to locate habitat may be especially valuable to reduce the costs of searching and choosing settlement, especially when the time spent to search is reduced (Greene and Stamps 2001, Fletcher 2006, Alhering et al. 2010). First-time breeders (often arriving later) are more likely to respond to social cues; experiments with conspecific attraction have shown this to be the case (Ward and Schlossberg 2004, Nocera et al. 2006, Betts et al. 2008).

We found conspecific vocalizations strongly influence the Canada Warbler settlement during the pre-breeding season. Canada Warbler individuals are more prone to settle in unoccupied, less disturbed sites with the stimulus of artificial conspecific song and call cues. This effect underlines the importance of acoustic communication in settlement decisions of Canada Warbler, suggesting that vocal cues might play a critical role in intra-species interactions and territory establishment. Therefore, artificial conspecific cues could be a useful tool in conservation and management of Canada Warbler, promoting their settlement and increasing their population density in undisturbed or low disturbance areas (Schlossberg and Ward 2004, Schepers and Proppe 2016), as well as potentially expanding its geographical range or “reintroduction” in areas where extirpation is suspected (Anich and Ward 2017). Because knowledge gaps about the long-term efficacy of conspecific broadcasts as an attraction strategy remain, caution should be used when introducing conspecific broadcasts into unoccupied habitat.

Our study highlights the need for targeted conservation strategies that consider both the structural complexity of forest habitats and the dynamics introduced by disturbance such as logging. Management practices that promote a mosaic of harvested and unharvested might benefit Canada Warbler populations by providing a variety of ecological niches. Furthermore, the role of conspecific acoustic cues should not be overlooked in conservation planning. Acknowledging the sociality of many bird species, conspecific attraction is likely to play an important role in habitat choices for many migratory birds, and should be considered and incorporated into modeling, management, and conservation frameworks.

Further research is needed to determine the impact of forest harvest and conspecific attraction on Canada Warbler populations, including comparing breeding success in harvested areas to unharvested habitats, primarily to determine whether areas with high social aggregation are productive for Canada Warbler or pose a risk to its survival rates. It is required to understand the long-term impacts of different logging practices on Canada Warbler and to explore the potential benefits of varying the age and spatial distribution of harvest patches. Additionally, experimental studies focusing on the role of specific vocalizations and their effectiveness in different habitat types could refine our understanding of Canada Warbler habitat use and social structure.

LITERATURE CITED

Ahlering, M. A., D. Arlt, M. G. Betts, R. J. Fletcher, J. J. Nocera, and M. P. Ward. 2010. Research needs and recommendations for the use of conspecific-attraction methods in the conservation of migratory songbirds. Condor 112:252-264. https://doi.org/10.1525/cond.2010.090239

Albrecht-Mallinger, D., and L. Bulluck. 2016. Limited evidence for conspecific attraction in a low-density population of a declining songbird, the Golden-winged Warbler (Vermivora chrysoptera). Condor 118:451-462. https://doi.org/10.1650/CONDOR-15-217.1

Anich, N. M., and M. P. Ward. 2017. Using audio playback to expand the geographic breeding range of an endangered species. Diversity and Distributions 23:1499-1508. https://doi.org/10.1111/ddi.12635

Ball, J., P. Sólymos, F. K. A. Schmiegelow, S. Haché, J. Schieck, and E. M. Bayne. 2016. Regional habitat needs of a nationally listed species, Canada Warbler (Cardellina canadensis), in Alberta, Canada. Avian Conservation and Ecology 11(2):10. https://doi.org/10.5751/ACE-00916-110210

Becker, D. A., P. B. Wood, and P. D. Keyser. 2012. Canada Warbler use of harvested stands following timber management in the southern portion of their range. Forest Ecology and Management 276:1-9. https://doi.org/10.1016/j.foreco.2012.03.018

Betts, M. G., A. Hadley, N. Rodenhouse, and J. J. Nocera. 2008. Social information trumps vegetation structure in breeding-site selection by a migrant songbird. Proceedings of the Royal Society B 275:2257-2263. https://doi.org/10.1098/rspb.2008.0217

Burnham, K. P., and D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. Second edition. Springer-Verlag, New York, New York, USA.

Cadman, M. D., D. A. Sutherland, G. G. Beck, D. Lepage, and A. R. Couturier. 2007. Atlas of the breeding birds of Ontario, 2001–2005. Bird Studies Canada, Environment Canada, Ontario Field Ornithologists, Ontario Ministry of Natural Resources, and Ontario Nature, Toronto, Ontario, Canada. http://www.birdsontario.org/atlas-2/book/

Campomizzi, A., J. Butcher, S. Farrell, A. Snelgrove, B. Collier, K. Gutzwiller, M. Morrison, and N. Wilkins. 2008. Conspecific attraction is a missing component in wildlife habitat modelling. Journal of Wildlife Management 72:331-336. https://doi.org/10.2193/2007-204

Cornell, K., and T. Donovan. 2010. Scale-dependent mechanism of habitat selection for migratory passerine: an experimental approach. Auk 127(4):899-908. https://doi.org/10.1525/auk.2010.09171

Crins, W. J., A. P. Gray, P. W. C. Uhlig, and M. C. Wester. 2009. The ecosystems of Ontario, Part I: ecozones and ecoregions. Technical Report- 01. Science and Information Branch, Inventory, Monitoring and Assessment Section. Ontario Ministry of Natural Resources, Peterborough Ontario, Canada. https://files.ontario.ca/mnrf-ecosystemspart1-accessible-july2018-en-2020-01-16.pdf

Cupiche-Herrera, V. J., G. I. Balan-Medina, J. D. Cú-Vizcarra, A. A. Mora-Roche, J. Rodríguez-Valenzuela, and B. McLaren. 2023. Low-cost automated call box system to conduct playback experiments for wildlife research and management. HardwareX 14:e00418. https://doi.org/10.1016/j.ohx.2023.e00418

Doligez, B., and T. Boulinier. 2008. Habitat selection and habitat suitability preferences. Pages 1810-1830 in S. E. Jørgensen and B. D. Fath, editors. Behavioral ecology. Vol. 3. Encyclopedia of Ecology. Elsevier, Oxford, UK. https://doi.org/10.1016/B978-008045405-4.00015-X

Environment Canada. 2016. Recovery strategy for Canada Warbler (Cardellina canadensis) in Canada (Proposed). Species at Risk Act Recovery Strategy Series. Environment Canada, Ottawa, Ontario, Canada.

Fletcher, R. J. 2006. Emergent properties of conspecific attraction in fragmented landscapes. American Naturalist 168(2):207-219. https://doi.org/10.1086/505764

Fletcher, R. J., and K. E. Sieving. 2010. Social information use in heterogeneous landscapes: a prospectus. Condor 112:225-234. https://doi.org/10.1525/cond.2010.090236

Flockhart, D., G. W. Mitchell, R. G. Krikun, and E. M. Bayne. 2016. Factors driving territory size and breeding success in a threatened migratory songbird, the Canada Warbler. Avian Conservation and Ecology 11(2):4. https://doi.org/10.5751/ACE-00876-110204

Fretwell, S. D., and H. Lucas, Jr. 1969. On territorial behavior and other factors influencing habitat distribution in birds. I. Theoretical development. Acta Biotheoretica 19:16-36. https://doi.org/10.1007/BF01601953

Greene, C. M., and J. A. Stamps. 2001. Habitat selection at low population densities. Ecology 82(8):2091-2100. https://doi.org/10.1890/0012-9658(2001)082[2091:HSALPD]2.0.CO;2

Haché, S., P. Solymos, T. Fontaine, E. M. Bayne, S. Cumming, F. K. A. Schmiegelow, and D. Stralberg. 2014. Analyses to support critical habitat identification for Canada Warbler, Olive-sided Flycatcher, and Common Nighthawk: Final report 1 and 2. Boreal Avian Modelling Project, Edmonton, Alberta, Canada.

Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyukavina, D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy, A. Egorov, L. Chini, C. O. Justice, and J. R. G. Townshend. 2013. High-resolution global maps of 21st-Century forest cover change. Science 342:850-853. https://doi.org/10.1126/science.1244693

Harrison, R. B., F. K. A. Schmiegelow, and R. Naidoo. 2005. Stand-level response of breeding forest songbirds to multiple levels of partial-cut harvest in four boreal forest types. Canadian Journal of Forest Research 35:1553-1567. https://doi.org/10.1139/x05-076

Hildén, O.1965. Habitat selection in birds: a review. Annales Zoologici Fennici 2:53-75.

Hunt, A., E. M. Bayne, and S. Haché. 2017. Forestry and conspecifics influence Canada Warbler (Cardellina canadensis) habitat use and reproductive activity in boreal Alberta, Canada. Condor 119:832-847. https://doi.org/10.1650/CONDOR-17-35.1

Mukhin, A., N. Chernetsov, and R. Kishkinev 2008. Acoustic information as a distant cue for habitat recognition by nocturnally migrating passerines during landfall. Behavioral Ecology 19:716-723. https://doi.org/10.1093/beheco/arn025

Muller, K. L., J. A. Stamps, V. V. Krishnan, and N. H. Willits. 1997. The effects of conspecific attraction and habitat quality on habitat selection in territorial birds (Troglodytes aedon). American Naturalist 150:650-661. https://doi.org/10.1086/286087

Nocera, J. J., and M. G. Betts. 2010. Special section: social information and avian habitat selection. Condor 112(2):222-224. https://doi.org/10.1525/cond.2010.100005

Nocera, J. J., G. J. Forbes, and L. A. Giraldeau. 2006. Inadvertent social information in breeding site selection of natal dispersing birds. Proceedings of the Royal Society B 273(1584):349-355. https://doi.org/10.1098/rspb.2005.3318

Ontario Ministry of Natural Resources. 2018. Quetico Gwetaming park management plan. Ontario Parks, Thunder Bay, Ontario, Canada. https://www.ontario.ca/page/quetico-provincial-park-management-plan-published-2018

Ontario Ministry of Natural Resources. 2023. Ontario land cover compilation v.2.0. Ontario Forest Research Institute, Ontario Ministry of Natural Resources, Sault Ste. Marie, Ontario, Canada. https://geohub.lio.gov.on.ca/documents/7aa998fdf100434da27a41f1c637382c/about

Potapov, P., X. Li, A. Hernández-Serna, A. Tyukavina, M. C. Hansen, A. Kommareddy, A. Pickens, S. Turubanova, H. Tang, C. E. Silva, J. Amstron, R. Dubayah, J. B. Blair, and M. Hofton. 2021. Mapping global forest canopy height through integration of GEDI and Landsat Data. Remote Sensing of Environment 253:112165. https://doi.org/10.1016/j.rse.2020.112165

Pulliam, H. R. 1988. Sources, sinks, and population regulation. American Naturalist 132:652-661. http://www.jstor.org/stable/2461927 https://doi.org/10.1086/284880

R Core Team. 2021. R: A language and environment for statistical computing. Version 3.6.3. R Foundation for Statistical Computing, Vienna, Austria.

Reitsma, L., M. Goodnow, M. T. Hallworth, and C. J. Conway. 2009. Canada Warbler (Cardellina canadensis), version 2.0. In A. Poole, editor. The birds of North America. Cornell Lab of Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bna.421

Schepers, M. J., and D. S. Proppe. 2016. Song playback increases songbird density near low to moderate use roads. Behavioral Ecology 28:123-130. https://doi.org/10.1093/beheco/arw139

Schlossberg, S. R., and M. P. Ward. 2004. Using conspecific attraction to conserve endangered birds. Endangered Species Update 21:132-138. https://public.websites.umich.edu/~esupdate/wholeissueoctdec2004/schlossberg.pdf

Thomas, C. D., E. Bodworth, R. Wilson, A. Simmons, Z. Davies, M. Musche, and L. Conradt. 2001. Ecological and evolutionary processes at expanding range margins. Nature 411:577-581. https://doi.org/10.1038/35079066

Valente, J., G. LaGrande-Rolls, J. Rivers, A. Tucker, R. Fisher, and M. G. Betts. 2021. Conspecific attraction for conservation and management of terrestrial breeding birds: current knowledge and future research directions. Ornithological Applications 123:duab007. https://doi.org/10.1093/ornithapp/duab007

Van Horne, B. 1983. Density as a misleading indicator of habitat quality. Journal of Wildlife Management 47:893-901. https://doi.org/10.2307/3808148

Ward, M. P., and S. Schlossberg. 2004. Conspecific attraction and the conservation of territorial songbirds. Conservation Biology 18:519-525. https://doi.org/10.1111/j.1523-1739.2004.00494.x

Wester, M. C., B. L. Henson, W. J. Crins, P. W. C. Uhlig, and P. A. Gray. 2018. The ecosystems of Ontario, Part 2: ecodistricts. Science and Research Technical Report TR-26. Ontario Ministry of Natural Resources and Forestry, Science and Research Branch, Peterborough, Ontario, Canada.

Westwood, A. R., C. Staicer, P. Sólymos, S. Haché, T. Fontaine, E. M. Bayne, and D. Mazerolle. 2019. Estimating the conservation value of protected areas in Maritime Canada for two species at risk: the Olive-sided Flycatcher (Contopus cooperi) and Canada Warbler (Cardellina canadensis). Avian Conservation and Ecology 14(1):16. https://doi.org/10.5751/ACE-01359-140116

Zlonis, E. J., H. Panci, J. D. Bednar, M. Hamady, and G. J. Niemi. 2017. Habitats and landscapes associated with bird species in a lowland conifer- dominated ecosystem. Avian Conservation and Ecology 12(1):7. https://doi.org/10.5751/ACE-00954-120107