Cameras Show Sciurids Visiting White-Headed Woodpecker Nests Without Depredating Contents

Nest predation has long been recognized as an influential process in the reproductive biology of birds (Lack 1954; Martin 1993). Before the use of video surveillance in the 1990s, ecologists' understanding of nest predator identity was hampered by logistical difficulties of viewing nest predation. Since the 1990s, studies with video surveillance have improved our understanding of nest predation and informed management (e.g., Conner et al. 2010). Yet most video surveillance studies have focused on open-cup nesting birds. Although nest predation in cavity nesters is lower, nest predation is nevertheless considered the leading cause of nest failure in this guild (Mahon and Martin 2006; Paclik et al. 2009). Information on nest predators for primary cavity excavators (PCEs) is important because PCEs are keystone species whose cavities are used by many secondary cavity users. Thus, PCE nest success can have ramifications for overall ecosystem health (Lindenmayer et al. 2000; Drever et al. 2008).

Most past studies of nest survival in PCEs have inferred nest predator identity on the basis of signs left at nests, occupancy of abandoned cavities, or casual observations on the abundance of putative predators. This bias exists despite past research with open-cup nesters that demonstrates that such inferences are unreliable (Staller et al. 2005; Ball and Bayne 2012). To address some of these information gaps, we undertook an exploratory study of nest predation for an at-risk PCE in North America, the white-headed woodpecker Picoides albolarvatus. The white-headed woodpecker occurs in dry pine forests in western North America and is thought to face nest predation mostly from diurnal rodents in the family Sciuridae (hereafter, sciurids), including golden mantled ground squirrel Callospermophilus spp., pine squirrel Tamiasciurus spp., and chipmunks Tamias spp. (Wightman et al. 2010; Hollenbeck et al. 2011; Kozma and Kroll 2012). Past studies have reported nest success as low as 39% (Hollenbeck et al. 2011), and nest predation is considered one of the primary factors limiting populations (Mellen-McLean et al. 2013). Concerns over nest depredation by sciurids have prompted managers to consider reducing shrub cover in dry forests of the northwestern United States because shrubs provide habitat for sciurids. It is hoped that by reducing sciurid habitat, their populations will be reduced and woodpecker nest survival will increase. In an effort to document proportions of nests depredated by different sciurid species, we set remote cameras at white-headed woodpecker nests and obtained surprising footage of sciurids visiting nests without depredating eggs or nestlings. These observations contrast hypotheses in the literature. Here we report sciurid visits to active nests, rates of parental nest defense, and sciurid visits to recently fledged and failed nests.

Our observations took place in central Washington (approximately 46°48′N, 121°03′W). Forest cover varied on the basis of aspect, elevation, and longitudinal distance from the Cascade Crest. Common tree species included grand fir Abies grandis, ponderosa pine Pinus ponderosa, and Douglas-fir Pseudotsuga menziesii. Our study area had been harvested for timber at least once in the last 50 y and was open to livestock grazing (sheep and cattle). Most areas had also been burned with prescribed fire in the last 10 y. Additional details on study sites and their management history are available in Lorenz et al. (2015).

We searched for white-headed woodpecker nests in May and June 2014 using systematic searches and behavioral cues. We set a passive infrared trail camera within ∼10 m of cavities to capture still images and video clips of animals visiting nest sites. Forests in this region were open with little shrub cover and it was possible to see for many hundreds of meters through the forest. Therefore, cameras had an unobstructed view of nest cavities and the snag above and below the cavity, as well as the area behind and on either side of the nest without us having to disturb the surrounding vegetation. We used passive infrared cameras manufactured by RECONYX, Inc. (Holmen, WI; n = 2), Bushnell Outdoor Products (Overland Park, KS; n = 4), Wild Game Innovations (Grand Prairie, TX; n = 3), and Moultrie Inc. (Alabaster, AL; n = 1). Although passive infrared cameras are subject to bias from false negatives (camera does not trigger with motion), we used them in our study as an affordable (< $500), off-the-shelf option compared to continuous video recording systems, which can be expensive (> $1,000) and require considerable set-up time. We minimized false negatives by setting the trigger sensitivity to the highest possible setting. To avoid leading predators to nests we followed standard bird-nest monitoring protocols when checking cameras (Ralph et al. 1993). When there were fewer active nests than cameras we set cameras at all nests. When there were more active nests than cameras we prioritized nests at which high-quality images could be obtained with minimal disturbance to nests. Whenever possible, we left cameras at nests for 5 d after fledging or failure.

We first set cameras to take still photos because we assumed predator visits would always lead to nest failure. Within the first few days of monitoring we found that putative predators sometimes visited nests without causing predation and afterward we set cameras to record video for the longest duration possible for each unit (30–60 s) to document predator behavior. We monitored nest stage and contents with a digital inspection camera (Cen-tech Inc., Camarillo, CA) or by observing the behavior of the nestlings and parents (Jackson 1976). For nine nests we used a hole saw (Ibarzabal and Tremblay 2006) to open nest cavities within 5 d of fledging. We counted and color-banded all nestlings and placed very-high-frequency radio transmitters (Advanced Telemetry Systems, Istanti, MN) on one or two nestlings from each brood (n = 8 broods). We then used radiotelemetry within 1 wk of nest fledging to locate family groups and verify that all nestlings fledged. Our bird-handling protocols were approved by the University of Idaho Animal Care and Use Committee (protocol number 2011-30) and complied with the Ornithological Council Guidelines for the Use of Wild Birds in Research (Fair et al. 2010).

After the nest season we reviewed all footage for images, video, or sounds of visitors and parent woodpeckers (also referred to as “captures”). Visitors often caused the camera to trigger multiple times on the same visit. For these cases, we considered all captures by the same taxa within the same 1-h block of time as a single visit for the taxa other than parent white-headed woodpeckers. For parent white-headed woodpeckers, we considered all captures by the same sex within the same minute as a single visit. Given the potential subjectivity with these definitions, we present both the number of independent visits and the total number of photos/videos by different taxa.

We set cameras at 17 active white-headed woodpecker nests for 179 observer days (4,296 observer hours) and obtained reliable footage from 16 nests. Four nests failed and 12 succeeded with no partial depredations. Twelve percent of captures occurred during the incubation period and 88% during the nestling period. Among 153,986 photos and 16,335 videos (3,880 min of video), 86% had no animal in view (false positives). We grouped the remaining 21,075 photos and videos into 5,201 visits. White-headed woodpeckers were the most common visitor, accounting for 94% of visits. The most common nonparental visitors were chipmunks (31%), Douglas squirrels (22%), and western bluebirds Sialia mexicana (15%) (Table 1, Table S1). Footage showed sciurids either climbing nest snags and approaching within ∼10 cm of the nest entrance (in two cases, Douglas squirrels ran over/across the cavity entrance; Video S1), or pausing at or sniffing and investigating the cavity entrance (Video S2). In these cases, sciurids did not enter the nest cavity; they climbed near or to the cavity entrance, and then climbed back down the nest snag. Most (73%) captures of sciurids occurred in the morning before 1100 hours (Figure 1). We did not observe mantled ground squirrels climbing nest snags (Table 1). Most (71%) nest snags climbed by chipmunks and Douglas squirrels were successful and no depredations or partial predations occurred.

Two visits by chipmunks elicited reactions by parents that were captured on camera (Table 1). In both cases, chipmunks were in the background on the ground or on a log behind the nest. We did not observe audible or visible reaction by parents to 33 chipmunks or Douglas squirrels climbing on the nest snag (Figure 2; Videos S1, S2). One Douglas squirrel that was climbing down a white-headed woodpecker nest snag was attacked by a nuthatch within ∼5 cm of the woodpecker's nest cavity entrance.

In contrast, both visits by diurnal raptors elicited responses by parents. In the case of the Cooper's hawk Accipiter cooperii, parents called overhead for the 8 min that the hawk attempted to scale the nest snag. For the visit by a male American kestrel, the male parent woodpecker flew at and attacked the kestrel while the kestrel was attempting to reach into the cavity with a foot. Elk Cervus elaphus, western bluebirds, and Williamson's sapsuckers Sphyrapicus thyroideus were also sometimes met with aggression by adults. In all but two cases, elk were ignored by parents. In one case, the female parent began calling and then entered the cavity when elk approached, perching with head in the entrance. In the second case, both adults called overhead while elk fed near their nest cavity, similar to their reaction to the Cooper's hawk (Figure 2). For bluebirds and the sapsucker, adults engaged in chases through the air or struck out from inside the cavity when approached. We obtained 1,190 photos and videos at night but did not document nocturnal predator activity. Captures at night were of grazing mule deer Odocoileus hemionus (n = 12), elk (n = 1142), or cattle (n = 36) in the background behind or below the nest.

Four nests failed during our monitoring. At one failed nest we obtained three photos of a weasel at the cavity on the day the nest failed (481 nest; Table S3). We suspect that the weasel depredated the nest but we had set the camera to take still photos and did not obtain photos of the weasel with nestlings in its mouth. At the second failed nest (Nile 3.5 nest; Table S3), we suspect that parents abandoned their nest with nestlings. Footage indicated that parents stopped attending their nest with three nestlings for a 61.4-h period. Sixty-three hours after the last parental visit with food, a Douglas squirrel entered the cavity and removed a dead nestling. We do not know the fate of the second nestling. Later that afternoon (70.4 h after last parental visit with food), we visited the cavity and extracted the third dead nestling, which did not have outward indications of injury.

At the third failed nest, a chipmunk entered the cavity the day before it failed (East Devil nest; Table S3). The nestlings were estimated to be 20–23 d old and the chipmunk remained inside the nest cavity for 1–5 min (the exact duration is unknown because of a malfunctioning date and time stamp on the camera). The chipmunk exited the cavity with nothing in its mouth and audible nestlings were fed by adults for the next 7 h and until evening of the same day. In the next footage the following morning at approximately 0730 hours, the nest was quiet and western bluebirds were visiting and entering the cavity. It is unlikely, though possible, that the nest fledged because nestlings were 2–4 d from typical fledging age. The nest may have failed because of injury to nestlings when the chipmunk had entered the previous day, although we could not be certain. We obtained no footage from the fourth failed nest (Umtanum nest; Table S3) on the day of nest failure.

We monitored 11 nests after fledging or failure for a total of 624 observer hours. For successful nests, time to first visit on average was 8.5 h (± 9.6 h). For failed nests, time to first visit was 27.2 h (± 19.8 h). Western bluebirds and chipmunks were the most common nonparental visitors to recently vacated nests (Table 2). Five of seven successful nests were used by chipmunks or squirrels on average 16.9 h after fledging. In one extreme case, a chipmunk entered the Oak Lowest nest within 20 min of fledging (we had radiotagged all nestlings and confirmed successful fledging by radiotelemetry). Our footage showed that the chipmunk used the cavity for denning for the next two nights, after which point we removed the camera. This nest cavity entrance had been investigated several times by a chipmunk during the nestling stage.

Ground squirrels, chipmunks, and pine squirrels are documented nest predators for many small birds (Renfrew and Ribic 2003; Kirkpatrick and Conway 2010; Cox et al. 2012). Anecdotal evidence suggests that they are nest predators for white-headed woodpeckers as well (Wightman et al. 2010). We were therefore surprised to observe 23 visits by sciurids to successful white-headed woodpecker nests from the incubation to late nestling stages in which predation did not occur. Some successful nests were used within hours of nest fledging by denning sciurids. Our observations show that sciurids may be knowledgeable of the locations of woodpecker nests, visit them on occasion, and use them soon after fledging while not acting as nest predators. We call for additional studies that identify woodpecker nest predators using video surveillance and determine proportions of nests depredated by different species.

We also encourage research to test prevailing hypotheses about how woodpecker nests are protected. For white-headed woodpecker, researchers have hypothesized that nests may be protected from sciurid depredation by 1) being placed in areas of low sciurid abundance (which leads to low encounter rates by sciurids; Wightman et al. 2010; Hollenbeck et al. 2011; see also review by Paclik et al. 2009), 2) being placed in areas where sciurids have a difficult time locating nest sites (such as areas of high tree density, which also leads to low encounter rates by sciurids; Hollenbeck et al. 2011), and 3) parental attentiveness/defense (Hollenbeck et al. 2011; Kozma and Kroll 2012). Nests did not seem protected by low encounter rates in our study. Chipmunks and Douglas squirrels investigated or ran past the entrance to 10 nest cavities that had no losses of eggs or nestlings. It is likely that many more sciurids could have encountered nests if they had chosen to do so. On surveys for sciurids after the nesting season we counted a mean of 7.9 sciurids within 100 m of nests during a 10-min survey period (Table S2). Most nest cavities in our study were low to the ground (mean cavity height was 2.2 m) and occurred in areas with woody debris cover (Table S2). This would have placed nests in a level of the forest and habitat type associated with frequent use by chipmunks and grounds squirrels (Converse et al. 2006; Sullivan et al. 2012).

This is not the first study to observe nonaggressive interactions between woodpeckers and squirrels at nest sites. At least three studies have described squirrels (Tamiascuirus hudsonicus or Glaucomys volans) using cavities within the same tree as woodpeckers during the breeding season, without nest depredation (Kilham 1968; Conner et al. 1996; Walters and Miller 2001). Reasons that sciurids chose not to depredate such nests require more study. It is possible that sciurids are interested in the use of cavities and are willing to wait until cavities are vacated by woodpeckers before occupying them. Nonetheless, undefended eggs and nestlings contain high amounts of calories (Carey et al. 1980). In past research it was assumed that sciurids would not bypass such an easy meal. This assumption is used in part to explain the presumed need for white-headed woodpeckers to nest in areas with low sciurid abundance. One explanation is that sciurids are interested in consuming woodpecker eggs and nestlings but are intimidated by the possibility of parental nest defense. During the incubation stage, parents may “block” the nest entrance (Kilham 1968) and meet visitors to the cavity entrance with a blow to the head. For the nestling phase, aggressive parental nest defense may have occurred early in the nesting cycle and was recalled by rodents that visited later in the nesting season. Yet this poorly explains many sciurid visits in our study (Videos S1, S2). Rodents that encountered effective nest defense would not have returned later in the nesting cycle and risked getting attacked—if the defense had been effective, rodents would have avoided nest cavities altogether.

The nest-defense hypothesis would also be supported if nestlings defend themselves, but our anecdotal observations indicate that nestling defense is ineffective in white-headed woodpecker. While banding ∼200 woodpecker nestlings we observed that they do not defend themselves from human handlers, unlike adult woodpeckers. Also, during capture, nestling woodpeckers press into the bottom of the cavity and cling with their claws to the cavity (Jackson 1976). This makes nestlings easier to handle than adults, but would also make them easier to subdue and kill. Thus, the nestling defense hypothesis is not well supported. Nestling body size is another possible deterrent to nest predation by sciurids. However, laboratory studies by Bradley and Marzluff (2003) demonstrate that rodents as small as mice are capable of subduing birds larger than a nestling white-headed woodpecker (55–70 g).

Assuming that rodents are indeed common nest predators for this species—which must be verified in future studies—we hypothesize that nests in our study may have been protected from rodent predation by 1) hesitation on the part of sciurids to enter an occupied cavity when they were unsure of the occupant, 2) disinterestedness to consume eggs or nestlings (because other food was available and more easily obtainable), or 3) hesitation to linger on snags because of the risk of predation on themselves. We encourage future studies that test these hypotheses. We feel more confident in concluding that the encounter rate hypotheses did not protect nests in our study. Sciurids were clearly aware of several nests in this study that succeeded in fledging young. This does not mean that sciurids never depredate white-headed woodpecker nests, but it does indicate that more research is warranted before management actions are taken to reduce sciurid habitat aimed to increase woodpecker nest success.

Please note: The Journal of Fish and Wildlife Management is not responsible for the content or functionality of any supplemental materal. Queries should be directed to the corresponding author for the article.

Table S1. Number of visits by vertebrate animals to white-headed woodpecker Picoides albolarvatus nests (n =16) captured on passive infrared cameras and parental reactions to these visits, obtained from nests monitored in central Washington in 2014.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S1 (35 KB DOCX).

Table S2. Methods and summary of results for vegetation and point count surveys conducted at 16 white-headed woodpecker Picoides albolarvatus nests after the breeding season in central Washington in 2014.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S2 (14 KB DOCX).

Table S3. Raw data on all visits by vertebrates captured on passive infrared cameras to 16 white-headed woodpecker Picoides albolarvatus nests monitored in central Washington in 2014. Note that the date and time stamp on some cameras malfunctioned, resulting in an incorrect date in the spreadsheet. We obtained all footage between May 20 and July 7, 2014.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S3 (1451 KB XLSX).

Video S1. Video clip showing Douglas squirrel Tamiasciurus douglasii climbing up (squirrel's tail appears in video at 00:00 s) and then down (00:04–00:05 s) an active white-headed woodpecker Picoides albolarvatus nest snag (nest contains three nestlings) on June 25, 2014 in central Washington. Nest cavity entrance appears as a small black hole (approximately the size of the squirrel's head) in the upper quarter of the snag and faces the camera. The squirrel ran over the nest cavity entrance between 00:04 and 00:05 s in this video clip. No depredations were observed at this nest over the course of our monitoring and all thee hatched nestlings survived to fledging.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S4 (968 KB AVI).

Video S2. Video clip of chipmunk Tamias spp. examining the cavity entrance of a white-headed woodpecker Picoides albolarvatus nest containing three nestlings on July 6, 2014 in central Washington. No depredations were observed at this nest over the course of our monitoring and all thee hatched nestlings survived to fledging.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S5 (1004 KB AVI).

Reference S1. Mellen-McLean K, Wales B, Bresson B. 2013. A conservation assessment for the white-headed woodpecker (Picoides albolarvatus). U.S. Department of Agriculture, Forest Service, and U.S. Department of Interior, Bureau of Land Management. 41 p. Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S6; also available at https://digital.osl.state.or.us/islandora/object/osl:80689 (1216 KB PDF).

Reference S2. Ralph CJ, Geupel GR, Pyle ., Martin TE, DeSante DF. 1993. Handbook of field methods for monitoring landbirds. Gen. Tech. Rep. PSW-GTR-144. Albany, California: Pacific Southwest Research Station, Forest Service, U.S. Department of Agriculture; 41 p.

Found at DOI: http://dx.doi.org/10.3996/052017-JFWM-039.S7; also available at https://www.fs.fed.us/psw/publications/documents/psw_gtr144/psw_gtr144.pdf (1450 KB PDF).

We thank A. J. Woodrow for loaning cameras and C. Clark for assisting with the painstaking process of reviewing footage. We also thank T. Kogut for photo editing and R. Huffman and J. St. Hilaire for logistical support. We appreciate thoughtful comments by D. Leonard, A. White, an anonymous reviewer, and the journal editors who improved the manuscript.

Any use of trade, product, website, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government.