Fine-scale Temporal Overlap Between Avian Scavengers

K. Brogdon, C. Marneweck, D. Jachowski, S. Harris


Raptor species have similar active periods, and often similar ecological niches when they coexist. However, due to factors such as competition and risk from confrontation, some avoidance behavior would be expected, which would seem to contradict activity period data. We used camera traps to capture pictures of golden eagles (Aquila chrysaetos), bald eagles (Haliaeetus leucocephalus), and red-tailed hawks (Buteo jamaicensis), in order to observe levels of co-appearance and obtain more detailed temporal activity data. We paired bald eagles with golden eagles and with red-tailed hawks to compare activity periods. We found that both pairs showed high activity overlap (0.90 for golden eagles and 0.75 for red-tailed hawks). However, the pairs rarely co-occurred in the same photograph, and the median time interval between a bald eagle leaving and another arriving at the carcass was 68 minutes and 90 minutes for golden eagles and red-tailed hawks respectively.. This suggests a finer-scale temporal partitioning that is not observed when only investigating activity overlap and  demonstrates the limitations of activity periods for analyzing interspecific interactions, as actual behavior may be significantly more complex.


For scavengers, this lowers the energy cost of scavenging for all guilds involved not only by reducing direct predation, but also decreasing competition and lowering the rate of other violent encounters (Killengreen et al. 2008).  Although such partitioning is most typical between predator-prey or mesopredator-apex predator species, it also appears among species with very few directly antagonistic behaviors towards each other, such as among various species in the genus Corvus (Baglione & Canestrari, 2009). This demonstrates its potential role in creating and solidifying ecological niches.

For instance, a study on temporal partitioning between Eurasian Lynxes (Lynx lynx), which are apex predators, and wolverines (Gulo gulo), a scavenger that often uses lynx kills as a food source, found no evidence of temporal partitioning between them in terms of time of day. However, it was also observed that these two species were never observed together, despite being equally present at the scavenging sites (Mattison et al. 2011). A study observing temporal and spatial partitioning between red foxes (Vulpes Vulpes) and wolves (Canis lupus) showed little temporal partitioning or spatial partitioning in terms of daily activity patterns but also observed the two species together very rarely. (Torretta et al. 2016).

In this study we examined the relationship between fine-scale temporal overlap and activity periods of avian scavengers in the eastern United States, using two commonly co-occurring pairs as examples. We expected to find high activity overlap, but low co-occurrence, as well as the presence of fine-scale partitioning.


Of 91,754 images, bald eagles occurred in 6,264, golden eagles in 10,839, and red-tailed hawks in 757. There was very high overlap between bald and golden eagle activity periods (overlap coefficient = 0.90, 95% CI = .87- .89). However, bald eagles co-occurred in the same image as golden eagles only 166 times (2.65% of bald eagle images, 1.53% of golden eagle images). The median time interval between a golden eagle arriving at the carcasses after a bald eagle was 68 minutes (range = 0 – 309 minutes).

There was also high overlap between the bald eagle and red-tailed hawk activity periods (overlap coefficient = 0.75, 95% CI = .72 – .80). Yet, bald eagles co-occurred in the same image as red-tailed hawks only twice (0.03% of bald eagle images, 0.26% of red-tailed hawk images). The median time interval between a red-tailed hawk arriving at the carcasses after a bald eagle was 90 minutes (range = 40 – 355 minutes).

Fig. 1. Activity overlap of bald eagles (Haliaeetus leucocephalus) and golden eagles (Aquila chrysaetos).

Fig. 2. Activity overlap between bald eagles (Haliaeetus leucocephalus) and red-tailed hawks (Buteo jamaicensis)

Materials and Methods

Photographs were taken using motion-activated wildlife cameras baited with carcasses. Carcasses were replaced once every few weeks, and were usually white-tailed deer carcasses from hunters. Photos were tagged using the program Digikam to add metadata to the images. The relevant data recorded were species present, quantity of each species, and time of day.

The bald eagle is the apex avian predator in this system, and the bald-golden eagle pair represents two similarly-sized birds, with average weights of 4-7kg and 3-6kg respectively (O’Gara 1994). The red-tailed hawk, with an average weight of 1kg, represents (Preston 2001) a pair with a weight discrepancy. We extracted the overlap coefficient for each pair, and also looked at the time intervals between the last photograph of a bald eagle visit and a subsequent first photograph of its pair (either golden eagle or red-railed hawk) on the same day.

We performed analysis using R version 4.0.1 (R Core Team 2020) using the packages camtrapR (Jürgen et al. 2016), lubridate (Grolemund & Wickham 2011), and overlap (Linkie and Ridout 2009).


Our study was based off observations of some species being present with other species more frequently than others, and wanting to dive deeper into this information. In doing so, we discovered that there was more depth to this question than we had initially anticipated. Not only was there a large discrepancy between activity periods and the expected co-appearance, but both bird species altered their behavior to minimize interactions. These results could merit further investigation into the specific behaviors of birds and the specific advantages gained from temporal partitioning, and highlight the intricacies of avian scavenger interspecific interactions.


Baglione, V., & Canestrari, D. (2009). Kleptoparasitism and Temporal Partitioning of Sympatric Corvids Foraging in a Refuse Dump. The Auk, 126(3), 566-578. doi:10.1525/auk.2009.08146

Garrett Grolemund, Hadley Wickham (2011). Dates and Times Made Easy with lubridate. Journal of Statistical Software, 40(3), 1-25. URL

James L. Hayward, Joseph G. Galusha, Shandelle M. Henson “Foraging-Related Activity of Bald Eagles at a Washington Seabird Colony and Seal Rookery,” Journal of Raptor Research, 44(1), 19-29, (1 March 2010)

Jürgen Niedballa, Rahel Sollmann, Alexandre Courtiol, Andreas Wilting (2016). camtrapR: an R package for efficient camera trap data management. Methods in Ecology and Evolution 7(12), 1457-1462.

Killengreen, S. T., Strømseng, E., Yoccoz, N. G., & Ims, R. A. (2011). How ecological neighbourhoods influence the structure of the scavenger guild in low arctic tundra. Diversity and Distributions, 18(6), 563-574. doi:10.1111/j.1472-4642.2011.00861.x

Martin Ridout, Matthew Linkie (2009). Estimating overlap of daily activity patterns from camera trap data. Journal of Agricultural, Biological, and Environmental Statistics 14(3), 322-337.

Mattisson, J., Persson, J., Andrén, H., & Segerström, P. (2011). Temporal and spatial interactions between an obligate predator, the eurasian lynx (lynx lynx), and a facultative scavenger, the wolverine (gulo gulo). Canadian Journal of Zoology, 89(2), 79.

O’Gara, Bart W., “EAGLES” (1994). The Handbook: Prevention and Control of Wildlife Damage. 61.

Preston, C. R. (2000). Red-tailed hawk (1st ed., p. 8). Mechanicsburg, PA: Stackpole Books.

R Core Team (2020). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL

Torretta, E., Serafini, M., Puopolo, F., & Schenone, L. (2016). Spatial and temporal adjustments allowing the coexistence among carnivores in liguria (N-W italy). Acta Ethologica, 19(2), 123-132. doi: