Acorn (Melanerpes formicivorus) behavior varies in response to intra- and interspecific calls

Josephine Collier1, Skye Hoolihan2, Ethan Jakob3, Hannah St. John4

1University of California, San Diego; 2University of California, Santa Cruz; 3 University of California, Santa Barbara; 4University of California, Davis

Acorn are a community-oriented species that, like other , uses and responds to vocalizations as a form of communication. The calls of certain competitors, predators, and intraspecific individuals change the behavior of a listener in different ways. The behaviors of three colonies were observed in response to four recordings: Steller’s jay, sharp-shinned hawk, an acorn woodpecker distress call, and a control. An ethogram was recorded for pre-call and post-call behavior of each colony. Behavioral response varied between colonies, possibly due to environmental factors and developed behaviors, leading to inconclusive results.

Keywords: acorn woodpecker (Melanerpes formicivorus), focal sampling, playback, auditory signaling, interspecies relationships

INTRODUCTION Cheney 2003). Birds are heavily reliant on auditory signals due to the vast number of Organisms that are at risk for predation or species-specific songs and calls that provide resource theft must be aware of the them with information about other birds potential interspecific threats to their around them. Birds will use auditory stimuli survival. Staying keyed into the warnings of to identify other birds and mark their predatory or competitive behavior gives an territories, both of which are crucial in advantage to the listener and allows them to discerning friend from foe (Sibley et al. 2001). take preventative measures. One species that heavily relies on auditory communicate using primarily auditory, signals is the acorn woodpecker (Melanerpes olfactory, or visual signals that inform their formicivorus); their cooperative tendencies behavior (Lakin & Hendrie 2020). Auditory and strict territoriality result in frequent and signals function well for terrestrial necessary communication. Acorn woodpeckers organisms that may live in habitats with live in social groups averaging five or six limited sightlines, as visual signals do not individuals and each colony defends a travel as effectively through these areas. territory averaging six hectares (MacRoberts Auditory signals allow an individual to & MacRoberts 1976). These territories spread its message effectively, but an contain all necessary resources and do not unintended listener, such as a predator, may overlap with other colonies (MacRoberts catch on and harm the signaler (Seyfarth & 1970). To survive with decreased food

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 1/8

sources during the winter, acorn intruder species. We hypothesized that a woodpeckers have evolved an extremely competitor’s call, when compared to a unique behavior: the creation and use of designated control call, would cause acorn granaries, trees with holes drilled into them woodpeckers to group up for territory for the purpose of nut storage for the winter, defense, therefore increasing their activity most notably acorns (Sibley et al. 2001). and time spent visible. Alternatively, we These granaries are fiercely defended and theorized that a call indicating a predator in are important markers of the acorn the area (a predator or distress call) would woodpecker’s territory (Drake et al 2020). To cause the woodpeckers to hide, thus defend these territories, acorn woodpeckers lowering their activity and time in sight more will “supplant” intruders, meaning that they than a control call. will make aerial passes at the intruder to persuade them to leave or face pursuit METHODS (MacRoberts 1970). There are many species of intruders, both We conducted our research at the competitors and predators, that are known University of California’s James San Jacinto to invade granaries and are often challenged Mountains Reserve (33.8081° N, 116.7769° by defensive acorn woodpeckers. Acorn W) in Riverside County from May 5 to 9, woodpeckers will alert each other of 2021. At 1,623 to 1,692 m in elevation, the intruder activity when territories are reserve experiences hot and dry summers, invaded (Doherty et al. 1996). This list of and cool and snowy winters, which allows intruders includes the Steller’s jay for the success of the mixed coniferous (Cyanocitta stelleri), a generalist competitor, forest that the reserve is situated in (James who has been known to steal acorns from San Jacinto Mountains Reserve 2021). This granaries (Mumme 1985). The sharp- mixed coniferous forest includes pines shinned hawk (Accipiter striatus) is a (Pinus) and oaks (Quercus), both highly important significant predator of acorn woodpeckers, genera of trees for acorn woodpeckers. and when they are detected in woodpecker We observed acorn woodpeckers at three territory, woodpeckers will hide in their granaries (Fig. 1). Site A (33.81032, roost until it is safe to carry on with daily -116.77486) had an elevation of 1,657 m; activities (MacRoberts 1976). In cases where The primary granary was a dead pine tree, a acorn woodpeckers are caught by predators, snag, with two adjacent snags at the bottom they will send out a distress call. There are of a slope. At the slope’s base was a campsite two main hypotheses on the purpose of this and a trail with minimal foot traffic. There call: that the scream’s purpose is to convene were few oak trees in the area compared to other woodpeckers to protect the attacked the abundant pines. Site B (33.80878, individual, and that the call attracts secondary -116.77730) was at an elevation of 1,655 m predators to cause the original predator to and had the most open space. Site C abandon the prey (Koenig et al. 1991). (33.80632, -116.77463) was at a higher Given the need to defend their territory, elevation than the other sites at 1,707 m and we wanted to know how acorn woodpeckers was the most heavily forested with an even would respond to the calls of various mix of oak and pine.

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 2/8

the call, we observed our individuals for 15 minutes to analyze their post-call behavior. We rotated and observed each site in the morning and the evening for four days, for a total of 12 different observation periods (four at each site). We created an ethogram to study the distribution of acorn woodpecker activity before and after each different type of call, over our observed period. The ethogram shows the behaviors organized into three categories: active, stagnant, and out of view.

Table 1. Description of the different Acorn Woodpecker behaviors recorded during observation Figure 1. Satellite map of UC James San Jacinto periods. Mountains Reserve, Lake Fulmor, Idyllwild, California. There are three experimental sites within the territory of three different Acorn Woodpecker colonies on the reserve. Sites are labelled A (red), B (green), C (blue). The elevation of each site is 1657 m (A), 1655 m (B), and 1707 m (C). Site A is near a campsite and a footpath, site B is next to a dirt road, and site C is adjacent to a trail heading upslope. The points were marked and the map was made using Google Earth and Gaia GPS.

Each morning, from about 9:00 am to 10:30 am, and evening, from about 4:00 pm

to 5:30 pm, we would go to one site and give the woodpeckers a 10-minute acclimation We specifically looked at stagnant period to allow them to get used to our behavior, which includes perching, pecking, presence. We then selected two individuals and preening, and compared it against active from each group and recorded their actions behavior, which includes all other visible for 30 minutes (Table 1), from about 10 to 30 actions: interactions, calling, drumming, m away. If an acorn woodpecker flew out of moving, climbing, and flying. We also sight, we would choose a new individual. compared proportion of time woodpeckers After those 30 minutes, we would play a were not visible and time they were visible randomly selected 45-second call of either a before and after each call. We took written Steller’s jay (Keller 1989), a sharp-shinned notes during each observation to characterize hawk (Budney 1993), an acorn woodpecker acorn woodpecker activity as the call was distress call, or a chorus of Pacific tree frogs played. For another measure for activity (Pseudacris regilla). The call was played level, we also calculated the rate of change using a phone connected to a JBL Clip 3 of activity before and after each call was Bluetooth speaker at full volume. Following played. The rate of change of activity was

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 3/8

calculated in actions per minute and represents these rates before and after each call was played. This was conducted to see if there was a significant difference in changes in activity rates based on the different calls we played. We used JMP statistical software v15 (SAS Institute 2019) for all statistical tests. We ran an ANOVA to compare the rate of change of activity before and after each call at each site. To compare certain behaviors before and after each call at each site to see if there Figure 2. Ethogram of acorn woodpecker behavior. was a significant relationship between the Acorn woodpecker pre-call behaviors were recorded behavior before and after calls were played, for 30 minutes and compared to the 15-minute post- we ran 24 different Pairwise Fisher’s call observations for four different recorded calls: Exact Tests. Steller’s jay, sharp-shinned hawk, acorn woodpecker distress call, and Pacific tree frog control call. The recorded behavior categories were active, stagnant, RESULTS and when the woodpeckers were out of view. Ethogram shows that primarily more time is spent out 3.1 Statistical of view, except for the distress call, where the majority of their behavior was stagnant (N = 12). Acorn woodpeckers spent the most amount of time out of view, followed by Table 2. Acorn woodpecker stagnant vs active (α = 0.004; N = 12); asterisks indicate values of perching in all cases (Fig. 2). There was no significance. Yellow highlight indicates an increase. difference in the rate at which the Blue highlight indicates a decrease. These were 12 woodpeckers performed activities between different Pairwise Fisher’s Exact Tests that the pre-call period and the post-call period demonstrated how proportion of stagnant behavior across all call types (N=12, F=2.8, p=0.13). (behavior of woodpeckers not in movement) to active behavior (behavior of woodpeckers in movement) For the twelve Pairwise Fisher’s Exact Tests, changes from before to after each call at each location. looking at the proportion of time woodpeckers spent being “stagnant,” we found that nine showed a difference between the pre- and post-call periods (Table 2). When it came to the proportion of time in view versus out of view, eight out of the twelve tests displayed a relationship between woodpecker absence and a call being played (Table 3). Although our results from most of our Pairwise Fisher’s Exact Test were significant, the amount by which they differed and the direction in which the results shifted varied greatly.

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 4/8

Table 3. Acorn woodpecker presence vs absence (α = site C, and the woodpecker returned a flight 0.004; N = 12); asterisks indicate values of display, supplanting the jay from the area. significance. Yellow highlight indicates an increase. Blue highlight indicates a decrease. These were 12 When the sharp-shinned hawk call was different Pairwise Fisher’s Exact Tests that played, we observed avoidance behavior by demonstrated how time out of sight (time we were the woodpecker colonies at all three sites. unable to see the woodpecker) to proportion of time They would fly to further trees, especially in eyesight (time we were actually able to see the where they were less visible, such as in woodpecker) changes from before to after each call at each location. densely foliated pine trees. We also noted that they called less frequently in the post- call periods. When we played the recorded distress call, we noticed that multiple individuals would fly to a tree closest to the viewing area and appear to look for the source of the noise. In two of the three events, these individuals would give off their own call during and directly after the playback. We also observed the woodpeckers flying away after their lookout 3.2 Observational period, eventually returning later during the We observed varying responses to the four post-call period. Finally, for the Pacific tree different calls that were not consistent frog call we observed no shared reactions across observation events. The three sites between playback events. had very similar colony sizes: site A had six to DISCUSSION eight woodpeckers, site B had approximately four woodpeckers, and site C had eight While we expected to see no reaction to woodpeckers. During the Steller’s jay call we the control call, the mixed response we observed two sites where multiple observed may have been because it was a woodpeckers came together on a central loud and unexpected noise. The inconsistent tree, close to the viewing area, and looked results may also point towards natural and for the source of the call while also unrelated shifts in acorn woodpecker vocalizing. We did not observe this grouping behavior during our selected time scales. behavior at site B and the woodpeckers were In response to the Steller’s jay call, we much more silent in the post-call period. expected acorn woodpeckers to increase During one observation period we observed their activity and the rate at which they up to seven Steller’s jays came to the nearby would change behaviors, as well as spend trees, even landing on the central tree. more time in view. Contrary to what we Rarely would other birds land on the trees thought, the woodpeckers seemed to stay in the woodpeckers primarily perched on. view roughly the same amount of time During a separate observation period, a before and after the call, seeming to not Steller’s jay was observed making an mind the potential presence of a Steller's jay aggressive flight display at a woodpecker at nearby. Jays are known to steal from acorn

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 5/8

storages, and woodpeckers will supplant the likely a preventative measure, the response jays (MacRoberts 1970). We had personally might be stronger with a visual cue. observed a Steller’s jay take an acorn from Predators such as hawks have been known an active granary prior to our research, and to elicit prey birds' alarm responses in we had expected more instances of this studies where visual cues were used, such as while we recorded data. Jays were present models or video representation (Butler at our three sites, but rarely interacted with et al. 2017). the woodpeckers or the granaries. One Acorn woodpeckers have been known to reason that could explain the lack of jay support the “predator attraction interactions is that the granaries have fewer hypothesis,” where a predator catches a acorn stores at this time of the year. This woodpecker who then screams, often seasonal decrease could also explain the resulting in the arrival of a secondary woodpeckers’ neutral response to the jay predator, distracting the primary threat call. While acorn woodpeckers still defend (Koenig et al. 1991). Based on this, we did their preferred trees during the spring, not expect woodpeckers to approach the defense events are much less likely as there played distress call. However, we observed is little motive for outsiders to come into the the woodpeckers flying into nearby trees territory. Insects are much more available at and appearing to search for the source of the the end of winter, and the woodpeckers distress call. We also had one instance where themselves spend more time flycatching in the grouped woodpeckers responded with the spring and summer (MacRoberts & their own distress call. This contradicts the MacRoberts 1976). Acorns contain less distress call study because they did not nutrients than insects and therefore provide observe a conspecific response to the a lower reward in the warmer months, distress call. We propose that the habitat perhaps lowering the interspecies differences affect the behavioral response. competition for acorns (Koenig et al. 2008). Acorn woodpeckers are disproportionately Hawks, as raptors, do not perform calls for studied in oak woodland ecosystems and the same reasons that a songbird might; these studies are not representative of hawks mainly call for breeding or territorial populations in different ecosystems. Acorn purposes (Sibley et al. 2001). Despite this, woodpeckers are heavily dependent on we expected the acorn woodpeckers to acorns but do not require extensive amounts perform avoidance behaviors, such as hiding of oak stands to subsist themselves, so they in dense trees and remaining silent for a are able to thrive in mixed woodland time, to keep their location hidden. The environments (Koenig et al. 2008; Koenig & woodpeckers did become less active at two Benedict 2002). Mixed coniferous forests are of the three sites we studied, however, there more dense than oak woodland or savanna, were no consistent results regarding the and this can affect an ’s expected amount of time the woodpeckers were behavior due to changes in cover. In exposed present. We observed that after the hawk areas, like those that have been burned, call was played, the woodpeckers were less animals tend to be more vigilant because active and did not call as frequently as they are more visible to potential predators before. While this reduced activity is most (Drake et al. 2020). Additionally, hawks call

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 6/8

less frequently in open areas, meaning that ACKNOWLEDGMENTS prey species will have to be more reliant on visual predator cues. Dense foliage impacts This work was performed at the University how far visual signals can be conveyed, of California’s James San Jacinto Mountains making audio signals even more crucial for Reserve, doi:10.21973/N3KQ0T and animals to be aware of. This difference in composed at the University of California coverage may provide some explanation to Steele/Burnand Anza Borrego Desert the observed differences in behavior Research Center, doi:10.21973/N3Q94F. between these ecosystems. Special thanks to Dr. Tim Miller and Dr. Sarah Overall, we found that reactions to Kingston for your guidance in completion of intruder calls differed between colonies this research project. more than expected. This may be due to differences in the woodpecker colonies REFERENCES themselves. Hierarchical relationships, Budney, G. 1993. Sharp-shinned hawk calls colony size, territory size, or granary status (northern). Merlin Bird ID by Cornell Lab Version could all impact the manner in which the 1.8.2. woodpeckers respond to an intruder’s call. The colonies also had individual differences Butler, N., R. Magrath, and R. Peters. 2017. Lack of alarm calls in a gregarious bird: models and videos in environmental factors such as nearby of predators prompt alarm responses but no alarm roads and foot traffic (Fig. 1). Furthermore, calls by zebra finches. Behavioral Ecology and each site had unique compositions of Sociobiology 71:1–13. vegetation; for example, while site B had more open grassland, sites A and C were Drake, E., A. Farid, Y. Hanna, and J. Ku. 2020. Acorn woodpeckers (Melanerpes formicivorus) exhibit much denser with a mixture of both oak and more predator avoidance behavior post-fire. pine trees. California Ecology and Conservation Research 4(1). Along with taking visual behavior data, examining the change in call rates by the Griesser, M. 2013. Do warning calls boost survival of woodpeckers themselves after an intruder signal recipients? Evidence from a field experiment in a group-living bird species. Frontiers in Zoology call could yield a better picture of the group 10(1):1–5. response. Intraspecific warning calls are a common signaling strategy used by birds to James San Jacinto Mountains Reserve. 2021. Natural allow a colony to identify and address a History of James Reserve, accessed on 13 May 2021 threat. Analyzing this behavior in acorn from https://james.ucnrs.org/natural-history/ woodpeckers could reveal the auditory JMP®, Version 15. SAS Institute Inc., Cary, NC, 1989- responses to an intruder in addition to the 2019. visual data we gathered (Griesser 2013). Auditory signaling should be further studied Keller, G. A. 1989. Steller’s jay calls (coastal). Merlin to better understand the complexities that Bird ID by Cornell Lab version 1.8.2. have been presented through conspecific Koenig, W. D., and L. S. Benedict. 2002. Size, insect and interspecific behavioral responses of parasitism, and energetic value of acorns stored by acorn woodpeckers. acorn woodpeckers. The Condor 104:539–547.

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 7/8

Koenig, W. D., M. T. Stanback, and P. N. Hooge. 1991. MacRoberts, M. H., and B. R. MacRoberts. 1976. Distress calls in the acorn woodpecker. The Condor Social organization and behavior of the acorn 93:637–643. woodpecker in central coastal California. Ornithological Monographs 21:1–115. Koenig, W. D., J. P. McEntee, and E. L. Walters. 2008. Acorn harvesting by acorn woodpeckers: annual Mumme, R. L., and A. de Queiroz. 1985. Individual variation and comparison with genetic estimates. contributions to cooperative behavior in the acorn Evolutionary Ecology Research 10:811–822. woodpecker: effects of reproductive status, sex, and group size. Behaviour 95:290–313. Lakin, R., and C. Hendrie. 2021. Effects of alarm call playback on the behaviour of wild European Seyfarth, R. M., and D. L. Cheney. 2003. Signalers and herring gulls. Learning and Motivation 73:101691. receivers in animal communication. Annual Review of Psychology 54:145–173. MacRoberts, M. H. 1970. Notes on the food habits and food defense of the acorn woodpecker. The Sibley, D. A., Elphick, C., & J. B. Dunning, Jr. 2001. The Condor 72:196–204. Sibley Guide to Bird Life & Behavior. Alfred A. Knopf, Inc., New York, New York, U.S.A.

CEC Research | https://doi.org/10.21973/N3C963 Spring 2021 Vol. 5 Issue 4 8/8