Flight Burst Frequency of Foraging African Grey Parrots
Written for a class at The University of Arizona. Based on less official, side observations while studying Grey parrots foraging in Cameroon, Africa, summer 1997.
Introduction
Foraging animals experience a trade-off between food acquisition and predator avoidance behaviors. Flight from a predator is energetically costly and becomes more costly the greater the distance to a refuge. Therefore, the decision to flee must be efficiently weighed. Ydenberg and Dill (1986) developed a qualitative flight initiation distance model which predicts that the greater the distance to the refuge the sooner flight would be initiated. This suggests that an upper limit may be set on the distance from a refuge that an animal can forage. At this point, it is predicted that the cost of predator avoidance is greater than any benefit gained by foraging at the site.
This study aimed to test Ydenberg and Dill ‘s (1986) prediction by measuring the frequency of flight ‘bursts’ in Grey parrots (Psittacus erithacus). The Grey parrots observed in this study forage extensively on soil, grasses, and other ground plants at exposed soil patches in Bolou savanna, a rainforest clearing. Are flight bursts influenced by foraging and predation risk in the Grey parrot?
Can differences in flight burst frequencies of Grey parrots be correlated with characteristics of the foraging patches? I observed Grey parrot foraging patterns in Lobéké Reserve, southeastern Cameroon, and found that Greys concentrated their foraging activity in selected bare soil patches of a large swamp clearing. Although greatly exposed to predators, the parrots forage extensively on soil, grasses, and other ground plants. Grey parrots frequently fly off in energetically costly flight1bursts’ while foraging in large groups (200–800 parrots) on the ground. These flight bursts are rather loud, attention-grabbing initiations of flight done by the majority of the birds in the area.
Studies of how parrots compensate for these trade-offs (i.e., flight bursts, habitat selection, vigilance) may provide insight to the social structure of these birds and their ecological communities.
Because flight bursts are energetically costly, I assumed that the bursts serve an adaptive function.
Two hypotheses were proposed and tested; that flight burst frequency~ influenced by (1) distance to refuge and (2) presence of trapping activities. Correlations between flight burst frequency and site characteristics may support one of the hypotheses mentioned above. They may help explain why Grey parrots engage in energetically costly flight bursts while foraging.
Study Area
Lobéké Reserve occupies an ~2000 sq. km. area of semi-deciduous evergreen and swamp forest that lies within the northwestern comer of the Congo basin in southeastern Cameroon (Fig. 1).
Two rainy seasons exist with peaks in April and October; annual precipitation is 1600–1700 mm per year (WCS 1996). The study area was Bolou savanna (~ 1 sq. km.), one of several swamp clearings in Lobéké (Fig. 2). The area is composed of a variety of tall grasses, various water-plants, bare soil patches, and both tlov.’ing and stagnant water areas. Along the edges of, and interspersed within, the savanna are groups of large shrubs, palms, and deciduous trees. Although the area was recently designated as a reserve, pigeon trappers frequented the savanna.
Study Subject
Grey parrots, widely distributed across equatorial Africa (Forshaw 1989) are common in the international pet trade (Dändliker 1992), however, little is known about their ethology and ecology. The ground foraging behavior and geophagy observed in this study has not yet been described in the literature. Forshaw (1989) and Fry et al. (1988) both report that Grey parrots primarily feed on fruits and seeds in the trees. Socially, Greys are well known for their vocal mimicry, but laboratory studies demonstrate cognitive and communicative abilities that extend beyond mere vocal imitation. Todt (1975) and Pepperberg (1994) suggest that these skills develop best in a social context. It is not understood how Greys use such abilities in nature.
Methods
All observations occurred from late June to mid-August during the long rainy season. I concentrated sampling efforts at the study area between 0630 and 1100. I observed foraging directly from a blind and noted locations of vocalizations and flight directions of nearby groups. I recorded the frequency distribution of foraging Greys at these selected sites within Bolou savanna.
By noting the time of day for each occurrence, I recorded (1) the amount of time that at least one parrot remained on the ground and (2) the flight burst frequency of Grey parrots foraging in my area of direct visual observation.
Assessment of predation risk
I assessed the potential predation risk to predators at each site where flight burst activity was recorded. (1) I determined distance to refuge (i.e., surrounding deciduous trees and palms) in 4 compass directions (see Cowlishaw 1997). For any distance greater than 100 feet, I designated it as >100 feet (Cowlishaw 1997). I plotted the flight burst frequency against the mean distance to the refuge. (20 I noted presence or absence of trapping activity conducted by my own research group at the immediate location. (Diana May and I, along with three local hires, trapped Grey parrots on two occasions for measuring and banding purposes. We used U-shaped net tied to a string. The string was pulled by someone in a nearby blind. In both instances, we tied 1–3 captive parrots to buried sticks at the site to attract wild parrots to the site.) Using only data from Site VII, I created a bar graph to illustrate the effects of trapping activities on flight burst frequency.
Predicted results
In large areas (i.e., greater distance to refuge), there is usually less cover from predators but visibility increases. I predicted that with increased distance to refuge, the frequency of flight bursts would increase. I predicted that the frequency of flight bursts would increase in the presence of our trapping activities, but the amount of time spent foraging would decrease. Because the number of flight bursts recorded at each site was inconsistent and the number of bouts recorded is low, any correlations found may not be significant.
Results
The results show types of foraging sites, environmental conditions that affect site use, and frequency of flight bursts at the sites. I witnessed avian predation on two separate occasions. I was not able to identify the species. Otherwise, my view overhead was obstructed, therefore, I was not able to see any potential aerial predators. Also, according to our field guides, snake predation is also possible.
Patterns of activity
Ground foraging events occurred between 0730 and 1100, typically around 0900 and averaging 40 minutes in length. Parrots leave their roosts (see Dändliker 1992) before dawn to gather in perimeter trees of Bolou savanna. Single parrots, pairs, and small groups cross the clearing (or savanna) periodically. Groups steadily increase in size; vocal activity also increases. The parrots congregate in “assembly points” surrounding the patch in which the entire flock will eventually forage (Ward and Zahavi 1973).
When one parrot lands on the ground to feed, others land within seconds. Next, groups from other areas of the clearing join the large flock. The group on the ground rapidly increases in size as the Greys swoop in from trees in the immediate area, landing within approximately one meter of the parrots already feeding. Although parrots on the ground rarely vocalize, parrots in the trees vocalize extensively, especially when in large groups (200–800). Small groups (<50) engaged in foraging activity, including both the individuals on the ground and in the surrounding trees, are often almost silent.
Patterns of habitual use
Parrots foraged in sites with similar characteristics. They landed in saturated to supersaturated soil bordered by grasses. Approximately 40–60% of the perimeter of this area is bordered by palms, deciduous trees, and shrubs located within 5–20 meters of the bare soil patch perimeter.
The Site Frequency graph (Fig. 3) illustrates the weekly distribution of birds at each foraging site (Fig. 2). In the first four weeks, foraging activity was dispersed over Sites I-V. Their pattern of behavior showed that they alternated between these sites, especially Sites I, II, and III, daily. This is in contrast to their behavior in the last four weeks in which they concentrated their feeding at a selected site for multiple days, especially at Site VII, and including Sites VIII and IX. After a persistent rain during Week Four, the Greys concentrated all foraging at Site VII. When the moisture had visibly dried, the parrots foraged exclusively at Sites VIII and IX.
The parrots never foraged at Site Z. Site Z was the only other bare soil patch that I observed in Bolou savanna. It resembles Sites VIII and IX in vegetative qualities, but the distance to refuge was greater than 100 feet in all directions. Interestingly, Site Z was frequented by hundreds of green-fruited pigeons almost daily. The pigeons also foraged at sites frequented by the parrots.
Patterns of foraging behavior
Parrots engaged in geophagy at all identified group foraging sites in Bolou savanna. The parrots ate flowers, leaves, stems, roots, and possibly seeds of ground plants; they also stripped tree branches and masticated its bark. They ate plants along the edges of bare soil areas, which are typically occupied by shorter, possibly younger, plants. The ground plants consumed by the Grey include: Cyperus spp., Rynchospore corymbosa, Eleocharis acutangula, Oldenlandia alnafolia, and Echinochloa cruspavonis. The Greys also consumed the seeds of the following tree species: Celtis tessmannii, Myrianthus arboreus, and Pterocarpus soyauxii (Table 1).
Predation Risk Variables
I measured the distance to the nearest refuge (i.e., deciduous trees and palms) in 4 compass directions to assess predation risk at each site. The averaged distances and ranges, respectively, were measured as follows: Site II, 72 ft, 30->100 ft; Site VII, 62 ft, 15->100 ft; Site VIII, 81 ft, 25->100 ft, Site IX, 35 ft, 23->100 ft (Table 2).
Our research group trapped parrots for banding and measuring twice at Site VII. In the absence of trapping the foraging patch was not altered, and my field guide and I remained hidden in a blind. When we were trapping parrots, the patch was altered with the addition of (1) a net, (2) a “preferred” species of ground plant (according to our local trappers), (3) 1–2 teaser parrots (described in methods), and (4) 4–5 additional people hiding in the surrounding bushes, or blinds.
Patterns in flight burst frequency
When we engaged in trapping activities at Site VII, the 0.465 min-1; Site VII, 0.248 min-1, 0.392 min-1, 0.355 min-1; Site VIII, 0.183 min-1; and Site IX, 0.811 min-1. The total foraging times were also recorded for each foraging bout. (Table 2).
I plotted the flight burst frequencies of foraging Grey parrots against the foraging patches’ mean distance to the refuge (Fig. 4). The result was a linear correlation described by y = -0.0135x + 1.2266 (R2 = 0.8793). I also created a bar graph to illustrate the effect of trapping activities on flight burst frequencies at Site VII (Fig. 5).
Discussion
Flight bursts may be initiated for several reasons. Because flight is energetically costly, and loud, large flight bursts can draw the attention of predators, I postulate that the behavior has an adaptive function (i.e., predator avoidance). One possible reason is to signal to the potential predators that they are alert to possible pursuit and therefore would be more difficult to capture than those remaining on the ground. This reason is analogous to the slotting behavior of Thomson’s gazelles ( see Caro 1996). Another possibility is that parrots in less advantageous positions (i.e., one of low patch quality, greater distance to refuge, less visibility for a spotting predator, flock’s perimeter) make false alarm calls to gain a more advantageous position by rearranging the individuals; that is, they engage in selfish herd behavior (Alcock 1998). Flight bursts might help increase their range of vigilance.
In addition, although not recorded quantitatively, it appeared that birds on the ground were more likely to engage in flight bursts than birds in the surrounding trees. This observation provides further support that the studied behavior does indeed serve as a predator avoidance mechanism.
Once the parrot has been detected by a predator, it must use avoidance tactics to stay alive. African Grey parrots possess a cryptic grey body with red tail feathers that are absolutely brilliant when exposed and spread in flight. By bursting into group flight, the tails become the predominant feature. Not only are the tail feathers an expendable part of the bird. but they also are easily removed (pers. obs.) and do not cover any major organs. Therefore, the parrot can engage in these flight bursts to avoid capture and lose its tail with relatively low cost. Because flight to cover is not a necessary component of flight bursts, the parrots can return to their food more quickly as part of the group or choose to return to the nearby trees. The dilution effect (Alcock 1998) is an integral part of these flight bursts. Potential topics of research on this behavior include: why the bursts are initiated, who can and does initiate, when the bursts are initiated. Such studies may help to reveal social structure in the Grey parrots.
The results obtained did not support my hypothesis that the flight burst frequency of Grey parrots foraging increases as the mean distance to refuge increases. In fact, the frequency decreased in a beautifully linear fashion in relation to an increase in distance (y_ = -0.0135x + 1.2266; R2 = 0.8793; see (Fig. 4). However, as predicted, the frequency of flight bursts did increase when trapping activities were present (Fig. 5). The frequencies recorded in the presence of trapping activities at Site Vll were 0.448 min-1 and 0.465 min-1; the frequencies in its absence at the same site were 0.248min-1, 0.392 min-1, and 0.355 min-1.
Many possible reasons exist for the results obtained. Only one habitat variable was measured in this study. It is possible that a variable exists that is a stronger influence on the parrots’ flight behavior than the distance to the refuge (e.g., height of surrounding grasses, area of foraging patch, actual presence of a predator(!). These variables were not kept constant. An interesting possibility may be that Greys in the trees (i.e., the refuge) are initiating the flight bursts with alarm calls. Perhaps, the increased distance negatively affects alarm call frequency. Also, if the ability of foraging parrots to spot potential predators increases with distance from refuge cover, then perhaps the parrots are better able to judge their optimal cost/benefit ratio for initiation of flight (see Cowlishaw 1998).
That the flight burst frequency did increase in the presence of trapping suggest that the parrots were aware of a potential threat; thus, the birds were ‘flighty’. This may not lead directly to the conclusion that the flight bursts was low, the discussed results may not even be statistically significant.
Given that ground foraging behavior puts the parrots at risk for predation, we propose that such foraging behavior has an important adaptive function. In another study, we suggest the function is related to nutritional needs (Bentley et al. 1997). We witnessed avian predation on two separate occasions. According to our field guides, snake predation is also possible. In this study, I propose that, given the energetic costs of frequent initiations of flight, flight bursts serve as a predator avoidance tactic. Predation might also be countered possible sentinel behavior, large group size, selection of foraging site, as well as other adaptive behaviors, such as alarm calling.
Bibliography
Alcock. J. 1998. Animal Behavior: An Evolutionary Approach, Sixth Edition Sinauer Associates, Sunderland, MA.
Bentley, C. S , D. L. May, I. M. Pepperberg. 1997. Foraging Behavior of Grey parrots in Cameroon. Poster presentation.
Caro, T. M. 1996. The function of slotting in Thomson’s gazelles: Some tests of the predictions. Animal Behavior 34: 663–684.
Cowlishaw, G. 1997. Trade-offs between foraging and predation risk determine habitat use in a desert baboon population. Animal Behavior 53: 667–686.
Dändliker, G. 1992. The Grey Parrot in Ghana: A population survey, a contribution to the biology of the species. a study of its commercial exploitation and management recommendations. A Report on CITES project S-30, 132 pp. ClTES Secretariat, Lausanne, Switzerland.
Forshaw, J.M. 1989. Parrots of the World. Blandford, London.
Fry, C. H., S. Keith, and E. K. Urban. 1988. The birds of Africa. Vol. 3. Academic Press, London.
Janson. C. H. 1998. Testing the predation hypothesis for vertebrate sociality: prospects and pitfalls. Behaviour 135: 389–410.
Kacelnik, A. and M. Bateson 1996 . Risky theories-the effects of variance on foraging decisions. American Zoologist 36: 402–434.
Lima, S. L. and L. M. Dill. 1990. Behavioral decisions made under the risk of predation: A review and prospectus. Can. J. Zool. 68:619–640.
May, D. L. 1996. The behaviour of African Grey Parrots in the rainforest of the Central African Republic. Psittascene 8:8–9.
Mysterud, A. and R. A. Ims. 1998. Functional responses in habitat use: availability influences relative use in trade-off situations. Ecology 79( 4 ): 1435–1441.
Pepperberg, I. M. l 994. Vocal learning in Grey parrots (Psittacus erithacus): effects of social interaction, reference, and context. Auk 111:300–313.
Todt, D. 1975. Social learning of vocal patterns and modes of their applications in Grey parrots. Z. Tierpsychol. 39:178–188.
Ward, P., and A. Zahavi. 1973. The importance of certain assemblages of birds as “informationcentres” for food-finding. Ibis 115:517–534.
WCS. 1996. The Lobéké Forest. Summary of Activities 1988–1995. Wildlife Conservation Society, Bronx, New York. 217 pp.
