Research Paper

INTRODUCTION

Habitat can be defined as the resources and conditions present in an area that produce occupancy — including survival and reproduction — by a given organism (Hall et al. 1997). Habitat use represents the choices an animal makes about what habitat to occupy and use and a variety of variables such as microclimate, vegetation, food abundance, and predation risk, have been shown to be important in this choice (Hutto 1985, Block and Brennan 1993). The choices happen at multiple temporal and spatial scales, such as between seasons or between days, or the used land cover type versus a specific foraging site (Johnson 1980, Hutto 1985). Understanding the habitats an organism uses and whether there is choice is critical to understanding their life history.

Caprimulgids and Common Nighthawks in particular (Chordeiles minor: hereafter nighthawks) are aerial insectivores that are experiencing some of the steepest declines among Canada’s bird guilds (Nebel et al. 2020). Nighthawks are putatively declining at a rate as high as 6.6% each year (COSEWIC 2018, Sauer et al. 2017) and were listed as Threatened in Canada (Species at Risk Act 2002). However, northern populations of nighthawks within the boreal ecoregion, a vast conifer-dominated forest, have been found to be more abundant than southern populations, particularly in recently burned areas (Knight et al. 2021, Kolbe et al. 2024) which in large part led to a revision in their status to Special Concern (COSEWIC 2018). Our understanding of the biology of nighthawks in the boreal forest is limited relative to more southerly parts of the species range, however, given the substantial population in the boreal region, this knowledge may be crucial for the management of the species.

Many Caprimulgids are known to sit on roads at night (Quesnel 1986, Poulin et al. 1998, Jackson 2003, Camacho 2013). Within the Caprimulgidae, nighthawks belong to the subfamily Chordeilinae, while the subfamily Caprimulginae is composed of true nightjars (hereafter, nightjars). Nightjars are known to use roads for thermoregulation (Camacho 2013, De Felipe et al. 2019), foraging (Jackson 2003) and reducing predation risk (Camacho et al. 2017), but the reasons for why nighthawks frequent roads at night are unclear. It has been assumed that nighthawks use roads to roost, where roosting is defined as settling in a location to sleep, rest or digest. Poulin et al. (1998) speculated that nighthawks used roads as bachelor roosts, but this was based only on an absence of females among other behavioral observations. This hypothesis, even if true, provides little insight into why nighthawks choose roads as night roost sites.

Thermoregulation is a plausible explanation for nighthawks’ road use given that Camacho (2013) found evidence that nightjars use roads for this purpose. During cold weather, Red-necked nightjars (Caprimulgus ruficollis) used asphalt roads which were warmer than other available substrates (Camacho 2013). If road use for thermoregulation is important, roads should be warmer during nights than non-road sites so that birds derive a benefit.

Foraging has been ignored as an explanation because nighthawks do not typically forage during periods of true night when solar influence is minimal (Mills 2008). Among Caprimulgids, the nighthawks forage by “hawking”: capturing flying insects during non-stop flight, typically crepuscularly, at dawn and dusk (Brigham and Barclay 1992). The caprimulgids known as nightjars, e.g., Common Poorwills (Phalaenoptilus nuttallii), forage by “sallying”: waiting for flying prey to come close and then flying a short distance from a tree or ground perch to catch it before returning to their perch. They do this at dusk, dawn, as well as during the night. Nighthawk wings are relatively long and narrow, while the wings of nightjars tend to be shorter and broader. Long, narrow wings are better for sustaining relatively fast flight, while short, broad wings are better for taking off from the ground and maneuvering through clutter. These anatomical differences may reflect the preferred feeding method of each subfamily, but there is no reason to think that either subfamily is entirely restricted to that method. Nighthawks are known to extend their foraging period using artificial lights (Shields and Bildstein 1979, Foley and Wszola 2017), and this behavioral flexibility may extend to their choice of foraging method. Flexibility in foraging behavior by aerial insectivores was reported by Ratcliffe and Dawson (2003) who found that little brown bats (Myotis lucifugus) and northern long-eared bats (Myotis septentrionalis), traditionally thought to be strict aerial foragers and surface gleaners, respectively, engage in the opposite style of foraging given the opportunity.

The use of roads by both nighthawks and nightjars poses a potential mortality risk caused by passing vehicles (Jackson 2003, Camacho 2013). Determining the factors behind nighthawks’ use of sites on roads, potential reasons for their use of roads, and quantifying mortality risk due to vehicles will collectively inform decisions about conservation and management.

We predicted that nighthawks would roost on roads in proportion to their general abundance near a road location at both coarse and fine scales (first and second order site use, sensu Johnson 1980) and when there were fewer potential roosts available nearby. We predicted that nighthawks would use sites for roosting with higher and denser verge vegetation because this should provide “backing” (an object that acts as a visual screen and physical obstacle behind an animal) and be a potential aid for avoiding detection and capture by predators (Wang and Brigham 1997, Camacho 2014). We predicted that on-road temperature would affect the use of roads and their choice of specific sites along a road. We also predicted that the frequency of nighthawks using the road would increase with lunar illumination because nighthawks may use roads as foraging sites, like nightjars (De Felipe et al. 2019). The uncluttered space of the road provides a clear view of the sky against which backlit insects could be more easily spotted, as well as room for birds to sally and capture prey. Finally, we expected that nighthawk mortality would be negligible because the number of vehicles using the road we sampled was negligible. To test this we surveyed for dead nighthawks and deployed quail carcasses to assess scavenging rates and correct our traffic mortality estimates.

METHODS

Study sites

Our study occurred near the Goldcorp Musselwhite Mine (52.60 N; 90.44 W) in northwestern Ontario, Canada which lies in the boreal shield ecozone, where mature forest is composed of coniferous trees such as white spruce (Picea glauca), black spruce (Picea mariana), jack pine (Pinus banksiana), and deciduous trees such as trembling aspen (Populus tremuloides), white birch (Betula papyrifera), and willow (Salix sp.) colonize forest gaps. The boreal shield covers almost 200 million hectares in Canada, ranging from Saskatchewan to Newfoundland. In 2011, a 141,000 ha fire burned south of the mine for ~34 km on either side of a gravel road (henceforth “road”) that leads to it. The fire was mostly stand-replacing. The private road to the mine traverses the burned area and is connected to a public gravel road that continues southeast for approximately 200 km to Pickle Lake, Ontario. The vegetation in the unburned forest we sampled was representative of mature boreal shield. Woody vegetation within the rocky landscape of the burned area was composed primarily of early successional species like jack pine, willow, birch, and alder (Alnus sp.); young, equal-age jack pine were particularly abundant. Although many of the charred, dead trunks of the previous forest stand remained upright, many had also toppled over and canopy cover was low relative to the unburned forest. The vegetation along the roadside tended to be willow and alder interspersed with herbaceous plants such as asters (Aster sp.), fireweed (Chamaenerion angustifolium), and yarrow (Achillea millefolium).

Surveys

We conducted abundance surveys (point counts) from June to August in 2015 and 2016 at points separated by at least 500 m along the road. In 2015, our point counts (n = 673) were each 5 min long and conducted most nights, provided weather conditions allowed (i.e., low wind, low precipitation). In 2016, we increased point count (n = 96) duration to 6 min to match the Canadian Nightjar Survey Protocol (Knight et al. 2019) and conducted the surveys weekly. The point count transects usually consisted of a dozen point counts. The transects began at a point randomly selected using the sample function in program R (R Core Team 2016), moved in a random direction chosen using a coin toss, and began approximately 30 min before sunset. We stopped surveying approximately 90 min after sunset, which was when we perceived that nighthawks ceased vocalizing. We undertook each transect along the road in either burned or unburned forest habitat. We alternated the burn habitat type surveyed nightly. We defined burned forest as areas with charred trees and new woody plant growth within the limits of the 2011 fire. We counted all nighthawks detected, whether seen or heard.

In addition to point counts, we conducted road surveys from June to August in 2015–16. In 2015, we performed these each night (n = 51 surveys) following the completion of the point count survey. In 2016, we reduced the frequency of road surveys (n = 17) to every third night because we only conducted point counts weekly and because we wanted to avoid habituating nighthawks and potentially influencing mortality rates due to nighthawk–vehicle collisions. Also in 2016, we sampled at random times of the night, not just at the time immediately after when point counts ended. Each road roost survey consisted of driving 2-3 circuits of the burned forest (34 km) road, at ~30 km/hr while scanning with a Brinkmann Q-beam 800-2500-0 spotlight. We did not survey the road in unburned forest because we never found nighthawks on the road during the 30 nights that we returned from abundance surveys during the time nighthawks were most often found on the burned portion of the road (23:00–1:00). On our surveys, we recorded the location of each nighthawk found sitting on the road, its position relative to the road edge (perpendicular, parallel, or angled), the cardinal direction of the closest road edge, the direction the bird was facing, and the distance from the road edge. Sex was determined when possible, but the diagnostic male white throat patch and tail band was not always apparent. We cannot rule out the possibility of double-counting some individuals but we did not count individuals in the same place on the road on a subsequent trip on a given night and think it highly unlikely to have been common. We began each road survey either from the mine entrance or from the location where the last point count was completed. Whenever we drove along the road (day or night), we recorded any dead nighthawks discovered.

We used lunar phase data obtained for the Opapimiskan Lake Airport at Musselwhite Mine. Lunar phase is defined as the percent of the moon disk illuminated. This oscillates between 0% and 100% approximately biweekly, which means it takes about four weeks (29.5 days) to complete a full cycle. At 0% and 100% illumination, the lunar phases are described as “new” and “full” moons, respectively. As illumination grows from new to full moon, 50% illumination is referred to as “first quarter.” Conversely, as the full moon’s illumination shrinks back to a new moon, 50% illumination is referred to as “third quarter.” All protocols involving the birds were approved (Protocol #15-02) by the University of Regina’s President’s Committee on Animal Care.

We sampled vegetation along the burned portion of the road in August 2016 near to each set of roost locations (a group of roosts (1–14) within a 200 m stretch of road that did not overlap). Given that roost locations were usually clumped, we randomly chose one site from within each roosting clump to serve as the vegetation survey location (n = 84). Each vegetation survey at a roost clump was paired with a vegetation survey at a random site along the road, chosen using a random distance and bearing from the paired site. We generated the random numbers using the sample function in R. The roadside vegetation was never cut or trimmed. At each vegetation survey point, we measured the height of the tallest plant (±1 cm) within 1 m of the road. We also measured density using a Robel (Robel et al. 1970) pole with nine markings spaced 10 cm apart, placed in the vegetation 1 m from the road edge. We viewed the pole from the center of the road. We selected the side of the road for vegetation surveys by assigning to each survey point randomly generated numbers representing the east or west side of the road. We assessed potential nighthawk roost availability by walking into the forest from the edge of the road perpendicularly for 100 equidistant steps along a compass bearing. We classified each stepping place as a potential roost site or not. Potential roost sites were defined on the basis of day roost sites we discovered at the study site: clear of branches for at least 60 cm and located at a spot with canopy cover <50%. The criteria were estimated visually and all assessments were made by GJF.

Surface temperature

We used iButtons (Maxim model no. DS1921G) to measure surface temperature both on and off the road every 20 minutes from sunset to sunrise. We placed iButtons level with the soil surface and left them from June to August 2016. We placed two on-road and two off-road. We placed the off-road iButtons within 10 m of the on-road sites, but several meters from the edge of the road. The two pairs of on-/off-road iButtons were approximately 20 km apart. We selected the sites based on nighthawk roost locations in 2015. We placed one pair of iButtons in a location commonly used by nighthawks, and the second pair in a location with no known use. We chose the off-road sites using the same potential roost location criteria; the one associated with where birds roosted was in a potential roost site, and the other one was not.

Scavenging experiment

We deployed Common Quail (Coturnix coturnix) carcasses to assess how quickly road-killed nighthawks might be scavenged given that they are similar in size. We purchased dead adult males from Bry-Conn Dev. (Androssan, AB, Canada) and froze them. We performed trials in June and July 2016. The first in the morning (~10:00 h), and the second in the evening (~19:30 h). We placed one quail every km (Trial 1, n = 15 quail; Trial 2, n = 21 quail) along the road from a random starting point and alternated one on the eastern versus western edge of the road. We returned every 12 hours for 60 hours to determine if carcasses had been removed.

Data analysis

We tested whether the location of nighthawks on the burned portion of the road was related to their abundance by measuring the distance from the mine entrance to each road detection and each point count detection. We compared the distances of the road locations and point count locations using a t-test. We used a generalized linear model (GLM) with a binomial family and a logistic link to test whether any factors measured were associated with road use (i.e., vegetation height, density, and roost availability) by nighthawks. We used a binomial family because nighthawks were present or not (i.e., 0 or 1) at a given site (i.e., roost vs. random). We used a Chi-squared test to assess whether birds roosted close (<1 m) or far (>1 m) from the edge of the road and if position (perpendicular vs parallel) relative to the closest road edge were consistent. We used a GLM with a Poisson distribution and logarithmic link to assess whether lunar phase or cloud cover were related to the abundance of nighthawks on roads. Given the nature of the cycle, lunar phase had to be fit as a quadratic term. We used a Poisson family because the response variable were abundance data with no overdispersion. We used a t-test to examine the difference in temperature recorded from 22:00–05:00 h of on- and off-road surfaces. We used a Kaplan-Meier survival analysis (q = d / d + l; where q = survival rate (i.e., carcass permanence) at time i, d = number of deaths (i.e., carcass disappearance) at time i, l = number survived) to calculate the removal rate of quail carcasses from the road. We used base packages in R version 3.3.0 (R Core Team 2016) and employed an alpha = 0.05 for all statistical tests.

RESULTS

There was a statistically significant difference between road use and point count locations where we detected birds (t = 2.37; df = 540; p = 0.02; Fig. 1). None of the site use factors were significant, including vegetation height (z = 1.76; p = 0.08), density (z = 0.96; p = 0.34), or the availability of off-road roosts (z = 0.03; p = 0.97). Although not significant, as vegetation height increased, the probability that a nighthawk used the site also increased. We found nighthawks less than one meter from the road edge significantly more often than other distances (Χ-squared = 8.21; df = 1; p = 0.004). Sixty-three percent of 117 nighthawks on the road were one meter or less from the edge of the road. Nighthawks were also significantly more likely to sit perpendicular to the direction of the road and face toward the center (Χ-squared = 113.7; df = 2; p = <0.001). Eighty-one percent (n = 108) of nighthawks were oriented toward the middle of the road and only 15% faced toward the vegetation. Just over half of those 15% (n = 9) were 1 meter or less from the edge. The number of nighthawks found roosting on any given night was significantly positively correlated with lunar phase (z = -2.30; p = 0.02; Fig. 2), but not cloud cover (z = -0.001; p = 1.00) or year (z = -0.10; p = 0.32). The more moon face illuminated, the more likely we found nighthawks on the road. There was a significant difference between the mean nighttime (22:00-05:00 h) on- and off-road surface temperature (t = 20.5; df = 4594; p = <0.001; Fig. 3). On-road surfaces were always warmer, but both surface temperatures declined at a similar rate overnight. There was no difference in surface temperatures between roost and non-roost areas.

Of the birds on the road that we could sex, more were males than females (9 females, 25 males). Seven of the females were observed on the road in June during the egg-laying period with the other two seen in the first two weeks of July. We twice observed a nighthawk sitting on the road sally up to capture a moth. On five occasions from 10-23 June, we observed females appearing to ingest grit from the road. Females comprised 58% of the sexed individuals seen on the road in June, but only 9% thereafter.

We used the 2015 vehicle entry log kept by the mine to determine traffic rates on the road. We did not include our use of it because both our frequency and speed (~30 km/hr) were atypical of normal road traffic. Between May 31 and August 10 (72 days), 398 vehicles traveled the road, or 5.5 per day. However, only 36 of those, or 0.5 per day, traveled the road between 22:00 and 05:00 (the time when nighthawks used the road). In 2015 there was a mean of 3.2 nighthawks using the road per day and in 2016 there was a mean of 3.7 nighthawks per day (cumulative mean of 3.5 nighthawks/day). In each year, we recorded only one vehicle-caused nighthawk road mortality, and we assume road traffic in 2016 was not significantly different from 2015, which we are confident in due to our extensive time spent on the road. Based on our trials with quail, there was a 44% carcass removal rate within 12 hours of placement which increased to 69.4% by 24 hours and 91.7% by 60 hours. We drove the road at least once per 24 hours on 91% of the days on-site (n = 110 days over both years). Correcting for undetected road-killed birds due to scavengers removing the carcasses, we estimate that 1.7 birds were killed by passing vehicles each year (detected deaths + (miss probability * detected deaths)). This results in a mortality risk of approximately 1% per vehicle passing during their road use period ((nighthawk mortality per night / mean nighthawks on road per night) / passing vehicles during activity period).

To estimate nighthawk mortality, we used the formula (N = O / 1 - p) where O was the number of detected carcasses and p was the probability a carcass had persisted (assumed to be 24 hours here). This provided 2 / (1 - 0.694) = 6.5 nighthawks fatalities. To estimate the mortality risk for each individual, we used the formula (P = N / M * S) where M was the mean number of road-roosting birds each night and S was the number of nights on-site. This gave 6.5 / (3.5 * 110) = 0.016, meaning 1.6% of roosting events resulted in a fatality. Finally, to estimate the risk of a vehicle killing a roosting nighthawk, we used the formula (K = N / V * S) where V is the average number of vehicles per night. This gave us 6.5 / (0.5 * 110) = 0.118, or a 12% risk of a vehicle killing a roosting nighthawk.

DISCUSSION

We found no evidence that any of the vegetation parameters we measured predicted nighthawk occurrence. The fine scale roost sites (third order use, sensu Johnson 1980) used by nighthawks were not predicted by the local abundance of birds based on the point count data. Thus, areas that had more nighthawks did not mean nearby road sections were used more. However, this does not apply to coarse scale site use (first order use, sensu Johnson 1980). We found no nighthawks roosting on the road in areas of unburned forest where nighthawk abundance was also significantly lower (G. J. Foley, unpublished data). It is unknown how far nighthawks traveled to roost sites. However, data for three Caprimulgus spp. illustrate that distances traveled from roost to foraging sites differ with species and habitat type but are less than 5 km; Egyptian Nightjars (C. aegyptiacus 1000–4000 m; Wasserlauf et al. 2023), European nightjars (C. europaeus; 600–3400 m; Evens et al. 2018) and Red-necked Nightjar (C. ruficollis; 500–1400 m; Camacho et al. 2014).

Nighthawks may receive thermoregulatory benefits from roosting on the road if surface temperatures on-road are warmer than off-road (Camacho 2013). Temperatures both on- and off-road declined steadily overnight, but on-road temperatures were always warmer. There was no difference in surface temperature between sites that were used and those that were not. We do not think thermoregulation can be the sole explanation for road use. Anecdotally, nighthawks are seldom, if ever, recorded on paved roads, which would provide greater thermoregulatory benefits than gravel roads (Camacho 2013). Additionally, of the nighthawks detected on randomized transects, 68% were detected roosting on the road between 23:00 and 01:00. During the coldest and thus most thermally challenging period - the pre-dawn hours - nighthawks tended to leave the road and roost elsewhere. Once roads get cold in the middle of the night they are not a good thermoregulatory option because the heat transfer rate of road substrates is higher than that of other materials/sites (e.g., a branch), so the road surface removes more heat from birds than other surfaces (Bakken and Gates 1975).

Other potential uses of roads that have been proposed include ingesting grit (small stones used in digestion) or dust bathing (Poulin et al. 1998). We observed no dust bathing, but did record five instances of apparent grit ingestion. This has been observed regularly in nightjars (Jenkinson and Mengel 1970), despite grit being used less frequently by insectivores than granivores (Gionfriddo and Best 1996). Females may also use grit to supplement calcium reserves needed for egg development (Turner 1982, Barclay 1994). Incubation is carried out primarily by females (Tomkins 1942, Rust 1947, Brigham 1989), and is the likely explanation for the male skewed sex ratio observed later in the season. The timing of our observations of grit-ingestion coincides with the egg-laying period of local nighthawks (Brigham et al. 2020), which is consistent with females using grit from gravel roads to supplement calcium reserves. However the low level of grit-ingestion observations and the number of males roosting on roads suggests that this cannot be the principal function. Male nighthawks used the road all summer and were detected almost three times as often as females, and yet we had no observations of them ingesting grit (Poulin et al. 1998, Jackson 2003).

Poulin et al. (1998) speculated bachelor male nighthawks used roads as roost sites, but this also seems an unlikely explanation. First, nighthawks raise a single clutch each season, therefore, the potential fitness reward of staying at their reproductive site diminishes quickly as the season progresses, yet males continued to use the road. Second, nighthawks used the road more in the early parts of the night (68% between 23:00–01:00; note that sunset ranged from 20:00 to 21:33 and sunrise from 05:00 to 06:27 during the study period). If they were just roosting (i.e., resting or sleeping), there should be no temporal clumping; they should be present all night long unless they are disturbed by passing vehicles and leave to go elsewhere or the road surface becomes too cold. Third, the locations of nighthawks on the road were not correlated with nearby roost availability. If nighthawks were using the road to roost on, fewer might be expected on roads in areas with more potential off-road roosts nearby. On the other hand, if they are using the road for a different reason, such as grit-ingestion or foraging, the number of nearby roosts should make little difference. Fourth, the locations of nighthawks on the road were not correlated with point count abundance. If nighthawks were using road locations solely based on temperature and temperature was consistent across the surface, abundance would be expected to predict location. Finally, using a road as a roost site should be thermally beneficial during the day as well as the night. Even if traffic caused nighthawks to leave a roost on the road for an off-road site, similar detection rates of on-road birds early in the day and early in the night would be expected. This is not the case; we never found any nighthawks on the road during the day and, while the phenomenon of nighthawks using roads at night is well-known to those familiar with the birds, nighthawks on roads during the day has, to our knowledge, never been reported.

Aldridge and Brigham (1991) calculated that nighthawks’ short, crepuscular foraging period may not allow them to meet their daily energetic requirements, unless this requirement is overestimated or their capture success rate is greater than 100%. The use of torpor would reduce energy requirements but, while capable of heterothermy, nighthawks do not appear to use it commonly (Firman et al. 1993, Fletcher et al. 2004). Daily energetic requirements could be reduced if an energy-saving strategy distinct from torpor is used, e.g., taking advantage of the thermoregulatory benefits of roads.

This led Aldridge and Brigham (1991) to speculate that nighthawks may be capturing multiple prey items per attack. A third possibility, however, is that nighthawks are obtaining additional food outside of the typical foraging period. Traditionally, the foraging period has been assumed to be limited to crepuscular periods, the time immediately before and after dusk and dawn, unless extended by their ability to forage under artificial lights (Shields and Bildstein 1979, Foley and Wszola 2017). Nighthawks’ standard foraging strategy of hawking prey relies on sufficient light levels whereas nightjar sallying reduces their need for high ambient light levels and moonlight appears to be sufficient (Mills 1986, Jetz et al. 2003, Jackson 2007, Smit et al. 2011).

Nighthawks and nightjars are morphologically similar, the most obvious difference being wing shape. Despite this, nighthawks may not be limited to a hawking foraging technique. After their crepuscular foraging, nighthawks may extend foraging into the night by sallying. While this has not been reported, most observations of foraging by nightjars occur because nightjars call and forage simultaneously, alerting a potential observer to their whereabouts. Nighthawks also call and forage concurrently, but apparently only while hawking insects. After darkness grounds them, calling ceases but if they continue foraging by sallying, their silence would make them less detectable than nightjars.

Nightjar activity is known to be correlated with the lunar cycle; nightjars are more active the closer the moon is to complete illumination (Mills 1986, Brigham 1992, Brigham et al. 1999, Jetz et al. 2003, Ashdown and McKechnie 2008, Woods and Brigham 2008, Jackson 2003, Smit et al. 2011). Nightjars are also known to use roads as sallying locations because the wide, open space of the road provides a clear sky backdrop against which to backlight insects (Jackson 2003). We found that the number of nighthawks using the road was positively correlated with increasing lunar illumination mirroring C. ruficolis (De Felipe et al. 2019). Shortly after their crepuscular foraging bout was over, nighthawks were also found on the road more often (23:00–01:00). This mirrors the peak foraging period of nightjars (Jackson 2003, Smit et al. 2011) and when insects are typically most abundant at night (Racey and Swift 1985). Most nighthawks were less than 1 m from the road edge and even more were both perpendicular to the road and facing away from the roadside vegetation, another behavior similar to that reported for nightjars who forage from roads (Quesnel 1986, Camacho 2014). Finally, we directly witnessed two nighthawks sallying from the road to capture moths - items normally considered to be consumed more by nightjars (Brigham and Barclay 1992). Despite previously published suggestions that nighthawks’ foraging period is strictly crepuscular regardless of the lunar cycle (Aldridge and Brigham 1991, Brigham and Fenton 1991, Brigham and Barclay 1992, Brigham et al. 1999), we argue that nighthawks likely extend their foraging period by sallying and that roads provide a good site to sally from. Additionally, nighthawks roosting on roads would receive a thermoregulatory benefit because on-road temperatures are consistently warmer than off-road. However, this leaves the question of why these birds do not use asphalt roads unanswered? Asphalt surfaces, especially in the boreal region, should provide the same or greater benefits yet they do not appear to be used. This assumes asphalt roads differ from gravel roads only in substrate type? Conditions adjacent to roads (e.g., availability of nesting habitat and protective cover) might also play a role.

We did not find any significant correlations between nighthawk road site use and vegetation, but nighthawks may be using road sites based on prey availability or the amount of visible sky. We did not directly assess either of these aspects, so any further discussion is speculative. Sites on roads that nighthawks potentially use as sallying sites should have a clear view of the sky to better observe backlit prey items. Jackson (2003) and Camacho (2014) found that nightjars were most abundant on-road sections where vegetation was present on only one side. This has been proposed to be because no vegetation on one side maximizes the amount of visible sky while the vegetation side provides protection from predators (Camacho et al. 2017). The open spaces created by the recent burn has allowed pioneering and fire-reliant plants to experience intense but low regrowth, potentially making for greater sky visibility than in unburnt areas.

Nighthawks foraging on roads would be exposed to mortality risk by vehicles. Due to low traffic rates on the road, mortality risk was low, although per vehicle risk was moderately high. Further, vehicles typically traveled down the center of the road while nighthawks almost always sat near the edge. On busier roads, the chance of being struck likely increase, but mortality rates from collisions probably also increase because drivers are less likely to drive down the center of the road. Keeping in mind that our study was localized to a small portion of the boreal forest, our results may be cautiously extrapolated to other boreal locations with suitable nighthawk habitat. We acknowledge that we did not assess searcher efficiency which can be important for mortality estimates. As a consequence, our mortality estimate could be higher. Many roads in the boreal likely have similar rates of night time traffic use. More highly traveled roads are often paved which nighthawks do not appear to use. Therefore, we think it unlikely that vehicles are a significant source of nighthawk mortality in the boreal forest.

CONCLUSION

At fine spatial scales, nighthawks did not roost on roads near to where they were most abundant, but at coarse scales site use was related to abundance. Nighthawks did not use road roosts based on vegetation density or potential roost availability nearby. Road surface temperatures were higher than off-road, but the timing of road use by nighthawks suggests thermoregulation may not be the primary reason for road use. Our estimate of mortality due to vehicle strikes in the boreal forest suggests that this is not a substantial conservation concern. Roads do provide a substrate for nighthawks to ingest grit at, particularly by females during the egg-laying period, and roosting on roads at night potentially allows for sallying to extend the foraging period, a behavior not previously reported for nighthawks.

LITERATURE CITED

Aldridge, H. D. J. N., and R. M. Brigham. 1991. Factors influencing foraging time in two aerial insectivores: the bird Chordeiles minor and the bat Eptesicus fuscus. Canadian Journal of Zoology 69(1):62-69. https://doi.org/10.1139/z91-010

Ashdown, R. A. M., and A. E. McKechnie. 2008. Environmental correlates of Freckled Nightjar (Caprimulgus tristigma) activity in a seasonal, subtropical habitat. Journal of Ornithology 149:615-619. https://doi.org/10.1007/s10336-008-0309-7

Bakken, G. S., and D. M. Gates. 1975. Heat-transfer analysis of animals: some implications for field ecology, physiology, and evolution. Pages 255-290 in D. M. Gates and R. B. Schmerl, editors. Perspectives of biophysical ecology. Ecological Studies, vol 12. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-87810-7_16

Barclay, R. M. R. 1994. Constraints on reproduction by flying vertebrates: energy and calcium. American Naturalist 144(6): 1021-1031. https://doi.org/10.1086/285723

Block, W. M., and L. A. Brennan. 1993. The habitat concept in ornithology. Theory and applications. Pages 35-91 in D. M. Power, editor. Current ornithology vol. 11. Plenum Press, New York, USA. https://doi.org/10.1007/978-1-4757-9912-5_2

Brigham, R. M. 1989. Roost and nest sites of Common Nighthawks: are gravel roofs important? Condor 91(3):722-724. https://doi.org/10.2307/1368127

Brigham, R. M. 1992. Daily torpor in a free-ranging goatsucker, the common poorwill (Phalaenoptilus nuttallii). Physiological Zoology 65(2):457-472. https://doi.org/10.1086/physzool.65.2.30158263

Brigham, R. M., and R. M. R. Barclay. 1992. Lunar influence on foraging and nesting activity of common poorwils (Phalaenoptilus nuttallii). Auk 109(2):315-320. https://doi.org/10.2307/4088200

Brigham, R. M., and M. B. Fenton. 1991. Convergence in foraging strategies by two morphologically and phylogenetically distinct nocturnal aerial insectivores. Journal of Zoology 223(3):475-489. https://doi.org/10.1111/j.1469-7998.1991.tb04778.x

Brigham, R. M., R. C. A. Gutsell, R.S. Wiacek, and F. Geiser. 1999. Foraging behaviour in relation to the lunar cycle by Australian Owlet-nightjars Aegotheles cristatus. Emu 99(4):253-261. https://doi.org/10.1071/MU99031

Brigham, R. M., J. Ng, R. G. Poulin, and S. D. Grindal. 2020. Common Nighthawk (Chordeiles minor), version 1.0. Pages in A. F. Poole, editor. Birds of the World. Cornell Lab of Ornithology, Ithaca, New York, USA. https://doi.org/10.2173/bow.comnig.01

Camacho, C. 2013. Behavioural thermoregulation in man-made habitats: surface choice and mortality risk in Red-necked Nightjars. Bird Study 60(1):124-130. https://doi.org/10.1080/00063657.2012.753400

Camacho, C. 2014. ‘Bodyguard’ plants: predator−escape performance influences microhabitat choice by nightjars. Behavioural Processes 103:145-149. https://doi.org/10.1016/j.beproc.2013.11.007

Camacho C, S. Palacios, P. Sáez, S. Sánchez, and J. Potti. 2014. Human-induced changes in landscape configuration influence individual movement routines: lessons from a versatile, highly mobile species. PLoS ONE 9(8):e104974. https://doi.org/10.1371/journal.pone.0104974

Camacho, C., P. Sáez‐Gómez, J. Potti, and J. M. Fedriani. 2017. Nightjars, rabbits, and foxes interact on unpaved roads: spatial use of a secondary prey in a shared‐predator system. Ecosphere 8(1):e01611 https://doi.org/10.1002/ecs2.1611

COSEWIC. 2018. COSEWIC assessment and status report on the Common Nighthawk (Chordeiles minor) in Canada 2018. Ottawa. https://www.canada.ca/en/environment-climate-change/services/species-risk-public-registry/cosewic-assessments-status-reports/common-nighthawk-2018.html

De Felipe, M., P. Sáez-Gómez, and C. Camacho. 2019. Environmental factors influencing road use in a nocturnal insectivorous bird. European Journal of Wildlife Research 65(1):1-7. https://doi.org/10.1007/s10344-018-1237-3

Evens, R., N. Beenaerts, T. Neyens, N. Witters, K. Smeets, and T. Artois. 2018. Proximity of breeding and foraging areas affects foraging effort of a crepuscular, insectivorous bird. Scientific Reports 8:3008 https://doi.org/10.1038/s41598-018-21321-0

Firman, M. C., R. M. Brigham, and R. M. R. Barclay. 1993. Do free-ranging Common Nighthawks enter torpor? Condor 95:157-162. https://doi.org/10.2307/1369397

Fletcher, Q. E., R. J. Fisher, C. K. R. Willis, and R. M. Brigham. 2004. Free-ranging common nighthawks use torpor. Journal of Thermal Biology 29(1):9-14. https://doi.org/10.1016/j.jtherbio.2003.11.004

Foley, G. J., and L. S. Wszola. 2017. Observation of Common Nighthawks (Chordeiles minor) and bats (Chiroptera) feeding concurrently. Northeastern Naturalist 24(2):26-28. https://doi.org/10.1656/045.024.0208

Gionfriddo, J. P., and L. B. Best. 1996. Grit-use patterns in North American birds: the influence of diet, body size, and gender. Wilson Bulletin 108:685-696.

Hall, L. S., P. R. Krausman, and M. L. Morrison. 1997. The habitat concept and a plea for standard terminology. Wildlife Society Bulletin 25:173-182.

Hutto, R. L. 1985. Habitat selection by nonbreeding, migratory land birds. Pages 455-476 in M. L. Cody, editor. Habitat selection in birds. Academic Press, Orlando, Florida, USA.

Jackson, H. D. 2003. A field survey to investigate why nightjars frequent roads at night. Ostrich 74(1-2):97-101. https://doi.org/10.2989/00306520309485374

Jackson, H. D. 2007. A review of the evidence for the translocation of eggs and young by nightjars (Caprimulgidae). Ostrich 78(3):561-572. https://doi.org/10.2989/OSTRICH.2007.78.3.2.313

Jenkinson, M. A., and R. M. Mengel. 1970. Ingestion of stones by Goatsuckers (Caprimulgidae). Condor 72(2):236-237. https://doi.org/10.2307/1366641

Jetz, W., J. Steffen, and K. E. Linsenmair. 2003. Effects of light and prey availability on nocturnal, lunar and seasonal activity of tropical nightjars. Oikos 103(3):627-639. https://doi.org/10.1034/j.1600-0706.2003.12856.x

Johnson, D. H. 1980. The comparison of usage and availability measurements for evaluating resource preference. Ecology 61(1):65-71. https://doi.org/10.2307/1937156

Knight, E. C., K. C. Hannah, R. M. Brigham, J. McCracken, G. Falardeau, M. F. Julien, J. S. Guenette, A. Manthorne, A. Sidler, G. J. Foley, et al. 2019. Canadian Nightjar Survey Protocol. http://wildresearch.ca/programs/nightjar-survey/

Knight, E. C., A. C. Smith, R. M. Brigham, and E. M. Bayne. 2021. Combination of targeted monitoring and Breeding Bird Survey data improves population trend estimation and species distribution modeling for the Common Nighthawk. Ornithological Applications 123(2). https://doi.org/10.1093/ornithapp/duab005

Kolbe, S. R., G. J. Niemi, A. M. Bracey, M. A. Etterson, and A. R. Grinde. 2024. Incorporating weather in counts and trends of migrating Common Nighthawks. Avian Conservation and Ecology 19(1):9. https://doi.org/10.5751/ACE-02621-190109

Mills, A. M. 1986. The influence of moonlight on the behavior of goatsuckers (Caprimulgidae). Auk 103(2):370-378. https://doi.org/10.1093/auk/103.2.370

Mills, A. M. 2008. Latitudinal gradients of biologically useful semi-darkness. Ecography 31:578-582. https://doi.org/10.1111/j.0906-7590.2008.05346.x

Nebel, S., J. Casey, M. A Cyr, K. Kardynal, E. Krebs, E. Purves, M. Bélisle, R. M. Brigham, E. Knight, C. Morrissey, and R. Clark. 2020. Falling through the cracks: a renewed call for a roadmap to conserve aerial insectivores in North America. Avian Conservation and Ecology 15(1):23. https://doi.org/10.5751/ACE-01618-150123

Poulin, R. G., L. D. Todd, and R. M. Brigham, R.M. 1998. Male Common Nighthawk use of gravel roads at night. Prairie Naturalist 30:85-90.

Quesnel, V. C. 1986. Why do nightjars sit on the roads at night? Living World, Journal of the Trinidad and Tobago Field Naturalists 19-23.

R Core Team. 2016. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.r-project.org/

Racey, P. A., and S. M. Swift. 1985. Feeding ecology of Pipistrellus pipistrellus (Chiroptera: Vespertilionidae) during pregnancy and lactation I. Foraging behaviour. Journal of Animal Ecology 54(1):205-215. https://doi.org/10.2307/4631

Ratcliffe, J. M., and J. W. Dawson. 2003. Behavioural flexibility: the little brown bat, Myotis lucifugus, and the northern long-eared bat, M. septentrionalis, both glean and hawk prey. Animal Behaviour 66(5):847-856. https://doi.org/10.1006/anbe.2003.2297

Robel, R. J., J. N. Briggs, A. D. Dayton, and L. C. Hulbert. 1970. Relationships between visual obstruction measurements and weight of grassland vegetation. Journal of Range Management 23(4):295-297. https://doi.org/10.2307/3896225

Rust, H. J. 1947. Migration and nesting of nighthawks in northern Idaho. Condor 49(5):177-188. https://doi.org/10.2307/1364338

Sauer, J. R., D. K. Niven, J. E. Hines, D. J. Ziolkowski, K. L. Pardieck, J. E. Fallon, and W. A. Wink. 2017. The North American breeding bird survey, results and analysis 1966 - 2015. Version 2.07.2017. Laurel, Maryland, USA.

Shields, W. M., and K. L. Bildstein. 1979. Bird versus bats: behavioral interactions at a localized food source. Ecology 60(3):468-474. https://doi.org/10.2307/1936065

Smit, B., J. G. Boyles, R. M. Brigham, and A. E. McKechnie. 2011. Torpor in dark times: patterns of heterothermy are associated with the lunar cycle in a nocturnal bird. Journal of Biological Rhythms 26(3):241-248. https://doi.org/10.1177/0748730411402632

Species at Risk Act. 2002. http://laws-lois.justice.gc.ca/eng/acts/s-15.3/page-1.html

Tomkins, I. R. 1942. The “injury-feigning” behavior of the Florida nighthawk. Wilson Bulletin 54:43-49.

Turner, A. K. 1982. Timing of laying by swallows (Hirundo rustica) and Sand Martins (Riparia riparia). Journal of Animal Ecology 51(1):29-46. https://doi.org/10.2307/4308

Wang, K., and R. M. Brigham. 1997. Roost-site characteristics of Common Poorwills, Phalaenoptilus nuttallii, in Saskatchewan. Canadian Field Naturalist 111(4):543-547. https://doi.org/10.5962/p.358255

Wasserlauf, Y., A. Gancz, A. Ben Dov, R. Efrat, N. Sapir, R. Dor, and O. Spiegel. 2023. A telemetry study shows that an endangered nocturnal avian species roosts in extremely dry habitats to avoid predation. Scientific Reports 13:11888 https://doi.org/10.1038/s41598-023-38981-2

Woods, C. P., and R. M. Brigham. 2008. Common Poorwill activity and calling behavior in relation to moonlight and predation. Wilson Journal of Ornithology 120(3):505-512. https://doi.org/10.1676/06-067.1