Bald Eagle Peck Order on Winter Range, Squamish BC.
By Jorma Jyrkkanen
Old Baldy’s Winter Work
He launches forth in a slow arching dive,
With talons tensing and steeling determination,
Preparing to strike the approaching foe.
Faster now, it will be a hit or miss
For he must have food.
A miss it is, the foe took the bluff, and
It is fish for lunch once again. JJ
Introduction
This scenario is typical of events in the daily life of eagles feeding on salmon in late winter when food is short. The sequence can be summed up as follows:
1. An eagle feeds while several stand nearby waiting for opportunity.
2. A small eagle lands nearby and joins the party of hopeful optimists.
3. A larger eagle launches from a nearby cottonwood and dives toward the feeding eagle.
4. The feeding eagle looks up and sizes up the approaching bird and its own appetite and confidence.
5. The diver may strike the feeding or defending eagle or may displace it or decide to pass it by and perhaps have another go at it when the situation is more opportune such as when the feeding eagle has had enough that it will vacate the fish more readily.
6. If a struggle takes place, often a small thief will make off with the remains especially if a small portion is all that remains. However, a serious impact can occur and I have seen eagles knocked over several rolls as far as two meters from the point of impact.
7. This behavior works for even small attacking eagles because of the physical impossibility of feeding and defending at the same time. They may however fail to displace the defending bird if it decides to make a stand.
I was so intrigued with this feeding behavior that I decided to try to find a way to study it and to see if there was a way to determine the dynamics of who was attacking who. I developed a scoring and statistical technique for analysis which is I think unique since I could find nothing at the time of my work that was similar being applied to raptors.
Here I thought was one of Darwin’s (1859) postulates on natural selection being acted out very aggressively. Was I witnessing a struggle or competition for resources, and was survival of the fittest the outcome? Here was a chance to test Darwin’s Theory of Natural Selection at least in part. I had already seen enough research to fully convince me of its global nature, but here was another chance to quiet the skeptics. To establish that indeed a struggle amongst surplus variants does take place and that there is sequential reduction in the numbers of progeny as they get older, supports Darwin’s theory. I had shown that there was reduction in numbers with age of juveniles in another paper.
Every addition to the list of species that follows Darwin’s postulates is of course substantiation of the universality of the Theory. This in turn profoundly influences how we think about such esoteric topics as the future of humanity and the meaning of god.
Selection of the fittest in Squamish is carried out by an environment that is unstable in the short term. It may be very harsh as it was when food was scarce in the winter of 1979/80 or it may be gentle as it was in the winter of 1978/79 when fish were available throughout most of the winter. Thus it may be said that fluctuating environments are at work to exert selective pressures on the eagle populations and this would work if my research hypothesis is correct, by limiting food supply and forcing a struggle for remaining fish that subsequently affects eagle survival.
To tackle this question, I sought to determine if the following research hypothesis (Hi): u1-u2<0 or (Hi): u1-u2>0 had any validity.
Is there a change in frequency of displacement attacks between eagles of different age classes beyond what would be expected purely on the proportions those age classes formed of the sample population. Is aggressiveness a function of age?
The null hypothesis (Ho): u1-u2=0 is that there was no difference.
My sample time was short unfortunately so while I include a field test of the method, replicates are needed to affirm the true population direction and dynamics of the peck-order. Perhaps I will have time in the future to make these observations.
Methods
I used the aging technique of Gerrard et. al. (1978) discussed in the other articles on this web site. I used a modified set of criteria which I found was consistently observable and separable into four distinct plumage categories.
Up to first year juveniles (J1) had brown bodies, eyes, and beak. Second year juveniles (J2) had brown beak but lightening, much white on the belly and often around the under-wing coverts and tail coverts. Third year juveniles (J3) had an osprey-head look to them and had yellowing beaks and eyes. Fourth year birds, (J4) the survivors moving to recruitment age, or as I call them Grads, had yellow beaks, yellow eyes and almost white heads and tails and already had brown bodies. They had varying degrees of brown remaining on their tails and heads. Adults uniformly had white heads and tails and brown bodies and yellow beaks and eyes.
It is possible that this plumage classification scheme is not absolutely faithful to age and that it takes some birds more or less time to move through them than a single year for each category, but for mechanistic purposes it works to use these identifiable differences to log the chances to survival.
In a January 1979 sample of all of the plumage classes taken prior to the analysis of attack behavior, the various age classes formed a certain proportion of the sample population. By multiplying the proportion by the number of eagles in a feeding group, one could then determine the expected number of attacks by or upon a particular age group or plumage group as would be predicted purely by chance.
If something other than chance were operative, then one would expect that there would be a statistically significant deviation from this expected. I used Chi-squared to determine agreement with the null hypothesis or not. An assumption is that by chance alone, random skewing of age-class representation could occur and interpretation would therefore be in error. Replicates would overcome this difficulty of course.
I then observed a feeding group and tallied the number of attacks by eagles of each plumage class on a polygon where the apexes successively represented each age class (figure 1). From this observed number of attacks, a statistical comparison between expected and observed could be made.
Attack Polygon Used for Field Behavior Recording
Two were in use and the action was fast. One for attacks by and the other for attacks on.
The analysis was done by comparing the observed number of attacks to the expected number of attacks by determination of the amount of chi-squared contribution by each age class to total chi-squared. If the chi-squared contribution varied between age or analogous plumage classes, then there was assumed to be some sort of change in the attack sequence with age thus supporting the research hypothesis and thereby supporting Darwin’s Theory that competition was taking place.
An age group may be aggressive and consequently their feeding interactions may be looked at from the point of view of which age class is attacking which age class more than expected. Selection may be trickier than that however, and it may operate by timid individuals withdrawing or resorting to other strategies such as thievery. Thus the data were analyzed from two points of view. These were ‘Attacks by’ and ‘Attacks on’.
This two way approach helps to understand the dynamics of the interactions more clearly.
Thus one can answer the three sub-questions; ‘were some age groups attacked more, the same or less than expected purely by proportional representation or chance alone’?
Results
Because this research was preliminary, I selected chi-squared at the alpha = .05 or .01 risk level if the data rose to that level of confidence, with 4 degrees of freedom (5 Age-classes-1) and did a two tailed test (more or less options covered).
Data from a feeding observation on the 23rd of January, 1979, taken from the new Highway Bridge across the Mamquam River are presented in the polygon format of Figure 1. Note that this polygon method also permits analysis of attacks within an age group.
The most significant contribution was by the J1 and also by J4. Both groups showed large deviations from expected. The brown up to yearling juveniles only attacked other eagle age classes 3X when the expectation was for 11X. Grads attacked 6X when the expectation was for 2 to 3. Thus, the youngest age classes were the least aggressive towards others while the oldest juveniles were the most aggressive.
Interestingly, in this sample when food was just becoming difficult to find, there were no attacks by eagles on their own age classes amongst the juveniles. This immediately suggests that competition was between age classes and not within when one looks at the picture from the point of view of attacks by an age class.
Adults and the two intermediate aged juvenile categories were about as aggressive as was expected by proportional representation alone. Total chi-squared was not significantly different than expected however. This finding isn’t essential to the research hypothesis because it is only a difference between age groups which needs to be demonstrated, not an overall significant finding. More samples would shed light on this question however.
Chi-squared Table II shows the results of the analysis for the ‘Attacks on’ viewpoint. Who got hit the most or ‘who were the victims’.
These very preliminary results indicate that indeed there is a peck order in wintering eagles and the bullies are the older Juveniles. There is some degree of protection provided for the brown juveniles in their first year but not when push coles to shove when salmon are in short supply.
An interesting surprise observation was made once when adults and older juveniles vacated the winter range leaving behind only brownies to struggle over very scarce fish. The brownies exhibited surprisingly violent aggression within their own age group over these scarce resources.
This observation shows that while peck order between age groups may occur when more than one age group is present, that the aggression can turn inwards within an age group when competition is more severe. Its almost as if there are within-age class alliances when the aggression is external but those break down when than source of competition is removed.
Conclusion
Most age classes of eagles were about as aggressive as would be expected by the proportion they formed of the sample population based on a single observation, which of course needs replication numerous times to be valid.
The exceptions were the youngest and the oldest juveniles. The youngest were least aggressive while the grads where the most aggressive. In human terms, babies are born nice while teens are mean.
Because the youngest group were docile or perhaps because they were granted immunity by virtue of their adolescent plumage eliciting parental or maternal reactions, this group was conferred immunity to some extent from attacks by older eagles of all age classes thereby giving it some advantage during these feeding struggles.
One would expect that this parental type response by older eagles to brownies would result in higher survival of brown juveniles on winter range than otherwise. I found that this group had quite high mortality in other studies and so they did need a hand in this regard. The apparently high mortality of this age class is probably due more to inexperience in all aspects of being an eagle. This social elicitation adaptation therefore gives them some advantage in an otherwise harsh time in their life histories.
Adults were non-discriminating in both aggression and targets for that aggression with future observations needed to test the preliminary hypothesis that they possibly pick on other adults more than juveniles.
The grads are attacked the least and also attack the most. Thus they are both feared by all age classes and reinforce their position by being the most aggressive. The fear is based on real potential for injury for I have seen individuals dealing severe blows in these feeding displacement attacks on at least three occasions over the course of the study.
While replication was desperately needed, this preliminary test of a new technique gave support for the research hypothesis that there was a peck order amongst eagles between age classes. Competition was demonstrated in this survey but it was complex.
Eagles apparently have a complex social order that provides protection for the youngest age classes and severe testing for older juveniles and adults also. The dominant eagles seem to be those juveniles approaching recruitment age or adulthood. As with humans, young adults are the most fit and in tests of fitness, they do rank highly. An interesting hypothesis to test is whether or not thievery confers increased fitness on smaller eagles. It is so common that I am sure it plays a role in natural selection.
Adults are tested by all age groups including their own and any weak or timid individual would be quick to be excluded from a feeding group. This is also true to some extent for other age groups. The ramifications of being excluded are serious.
It could lead to starvation or other losses related to weakness or loss of condition. While at the BC Fish and Wildlife Office in terrace, we had several reports of juvenile eagles starving to death in winter. Within any feeding group, there will be individuals that are lucky Fishers or better Fishers or are by size adapted to fishing a particular size class of prey fish better than those that are not. Feeding in a group means that the luck of those specialist fishers will be shared by the group if the social skills are up to the task.
Threatening take-overs of fish are part of the social behaviors that an eagle must master to survive to become healthy adults. Thus playing in the game of attack and bluff is essential to benefit from the group’s success.
It is my belief that side-line players in the eagle feeding game perish more often than active participants.
By showing that there was competition between eagles and especially between age-classes and by demonstrating shrinkage between age classes through my life history studies, I have found support for Darwin’s postulates of natural selection.
In particular, through all of my Bald eagle studies, I have repeatedly found support for the postulates that there is surplus production, variants exist, there is a struggle for survival through competition, natural selection and survival of the fittest varieties. I have also found that there is a struggle to survive accidental mortality and that some death is due to chance alone, independent of fitness. Being hurled into a tree or power line by a severe downburst for example might fall into this category as would eating a goose that had died of lead poisoning or an animal that had been poisoned for Coyotes. More replicates will determine greater confidence in the specifics of the attack sequence dynamics.
Humor
Do eagles have a sense of humor? You bet they do. I once observed an eagle land on a branch next to another. It then slowly sidled up to the eagle next to it and gently gave a shove with its shoulder. When this didn’t work, it shoved harder, until a pushing match ensued and there was great shoving back and forth and finally the original bird took off and yielded the roost to the interloper.
References
Darwin, Charles. 1859. The Origin of Species. (Penguin Classics Edition, 1968. Editor J. W. Burrow.)
Gerrard, Jonathan, Douglas W. A. Whitfield, Peter Gerrard, P. Naomi Gerrard and William Maher. 1978. Migratory movements and plumage of subadult Saskatchewan Bald eagles. Canadian field Naturalist 92(4):375-382.
Modeling Bald Eagle Populations Using Leslie Matrix
http://jorma-jyrkkanen.livejournal.com/155630.html
More from me. Bald Eagle Habitat Use.
http://jorma-jyrkkanen.livejournal.com/153205.html
http://jorma-jyrkkanen.livejournal.com/223726.html
http://jorma-jyrkkanen.livejournal.com/153742.html
http://jorma-jyrkkanen.livejournal.com/153205.html
http://jorma-jyrkkanen.livejournal.com/196043.html
http://jorma-jyrkkanen.livejournal.com/221327.html
http://jorma-jyrkkanen.livejournal.com/199710.html
http://jorma-jyrkkanen.livejournal.com/157115.html
http://jorma-jyrkkanen.livejournal.com/223295.html
http://jorma-jyrkkanen.livejournal.com/222115.html
See also:
Bald eagle blood chemistry jyrkkanenconservationbiology Wordpress.com
http://jyrkkanenconservationbiology.wordpress.com/2009/04/19/bald-eagle-haliaeetus-leucocephalus-blood-chemistry/
Bald Eagle Subadult Demography Jorma Jyrkkanen wordpress.com
http://jyrkkanenconservationbiology.wordpress.com/2009/04/19/bald-eagle-haliaeetus-leucocephalus-subadult-demography/
© 1982 and 2012. Jorma Jyrkkanen. All rights reserved.
tags: Bald eagles, ethology, zoology, avian behavior, research, winter range, Jorma Jyrkkanen
No comments:
Post a Comment