Context-dependent seed dispersal by a scatter-hoarding corvid

Authors

Mario B. Pesendorfer, University of Nebraska–LincolnFollow
T. Scott Sillett, Smithsonian Conservation Biology Institute
Scott A. Morrison, The Nature Conservancy, San Francisco, CA
Alan C. Kamil, University of Nebraska-LincolnFollow

Date of this Version

2016

Citation

Published in Journal of Animal Ecology 85 (2016), pp 798–805. doi 10.1111/1365-2656.12501

Comments

Copyright © 2016 Mario B. Pesendorfer, T. Scott Sillett, Scott A. Morrison, and Alan C. Kamil. Published by John Wiley Inc. Used by permission.

Abstract

1. Corvids (crows, jays, magpies and nutcrackers) are important dispersers of large-seeded plants. Studies on captive or supplemented birds suggest that they flexibly adjust their scatter-hoarding behavior to the context of social dynamics and relative seed availability. Because many corvid-dispersed trees show high annual variation in seed production, context-dependent foraging can have strong effects on natural corvid scatter-hoarding behavior.

2. We investigated how seed availability and social dynamics affected scatter-hoarding in the island scrub jays (Aphelocoma insularis). We quantified rates of scatter-hoarding behavior and territorial defense of 26 colormarked birds over a three-year period with variable acorn crops.

3. We tested whether caching parameters were correlated with variation in annual seed production of oaks as predicted by the predator dispersal hypothesis, which states that caching rates and distances should vary with seed abundance in ways that benefit tree fitness. We also tested whether antagonistic interactions with conspecifics would affect scatter-hoarding adversely, as found in experimental studies.

4. Caching behavior varied with acorn availability. Caching distances correlated positively with annual acorn crop size, increasing by as much as 40% between years. Caching rates declined over time in years with small acorn crops, but increased when crops were large. Acorn foraging and caching rates were also negatively correlated with rates of territorial aggression. Overall foraging rates, however, were not associated with aggression, suggesting that reduced dispersal rates were not simply due to time constraints.

5. Our field results support laboratory findings that caching rates and distances by scatter-hoarding corvids are context-dependent. Furthermore, our results are consistent with predictions of the predator dispersal hypothesis and suggest that large seed crops and social interactions among scatter-hoarders affect dispersal benefits for oaks and other masting tree species.