Effect of Artificial Concentrated Feeding Area Resource Depression on the Territoriality of the Anna Hummingbird (Calypte anna)

Linda Mahoney and Kathleen Kuechler

Department of Biological Sciences

Saddleback College, Mission Viejo, Ca 92692

Abstract:

Territorial behaviors such as chases and gorget displays are often used by Anna’s hummingbirds to defend their feeding territory. The intensity of such a display is determined by the quality of food resources in said territory, which in turn dictates the amount of energy that can be expended to defend a territory from intruders (Carpenter et al. 1989). This study compared the frequency of high-intensity territorial displays when resource availability was either “superabundant” (this isn’t mentioned until discussion) (Carpenter 1987) or depressed in a resident population of Anna’s hummingbirds whose main food supply was a spatially concentrated locale of artificial feeders. It was hypothesized that the hummingbirds would exhibit a greater frequency of high-intensity territorial displays when they received high-quality resources (high resource availability?) due to increased energy uptake, allowing them to exert more energy defending their feeding territory. It was found that Anna’s hummingbirds exhibited high-intensity territoriality at a frequency of 0.1327±0.0129(±s.e) when receiving high-resource food and 0.1142±0.02893(±s.e) when receiving low-resource food. There was no statistically significant difference in the frequency of high-intensity territoriality under the two conditions (p=0.2883,one-tailed t test). These results were most likely attributed to elements of Carpenter and MacMillan’s 1976 Threshold model as well as Myers et al argument of competition density.

Introduction:

One consequence of bird evolution is the development of a social organization structure that uses territoriality as one of the primary mechanisms for interspecies and intraspecies interactions (Brown 1969). These interactive dynamics determine individual fitness, with the crux of an individual’s fitness resting on the regulation of their (its, his/her; pronoun needs to match) energy budget (Carpenter et al. 1989). In other words (change transition “In order to achieve maximum fitness level” or something similar), individuals must balance their expenditure of energy with their ability to acquire energy. Territorializing areas with sufficient amounts of nourishment ensures that individuals obtain the energy they need to maximize fitness. However, these behaviors most often occur when the fitness benefits outweigh the energy costs (Brown 1964).

Male hummingbirds vigorously territorialize feeding areas, particularly with regards to other males, encompassing easy access to abundant high-quality food sources with the expectation that a female will enter their territory seeking a stable nesting site (Sibley 2001) (this sentence seems disjointed). Female hummingbirds additionally exhibit feeding territorial behaviors, but primarily in the resource obtainment for and defense of their nesting site (Sibley 2001) (awkward at the end of sentence; maybe switch around “resource obtainment” and “defense”). Territoriality displays can either be an energetically low-cost or high-cost expenditure, whereby the hummingbird exerts either a minimal or maximal amount of energy to perform their (match pronoun to “the hummingbird”) intended behavior. According to Brown’s 1969 and Ewald and Carpenter’s 1978 studies, territorial exhibits such as attacking or long chases are considered (a) high energy-cost expenditures as the defending bird literally chases an invading bird away from their (match pronoun) territory. In short chases, or those in which the defender needs not (or “does not need to)exit its territory before successfully driving an intruder away, threats or gorget displays and vocalizations are all considered low-cost expenditures.

In order to ensure that enough energy resources are available to meet their energy needs, individuals defend high-quality food sources, or those with increased sucrose concentrations, with a greater frequency of high-cost displays than areas with food sources containing lower sucrose concentrations (This sentence seems wordy. Try breaking it into two sentences. Also, what concentrations are considered “high-quality” or “low-quality?” It’s very ambiguous)(Brown 1964; Powers 1987; Ewald and Carpenter 1978). As the quality of food sources decreases, hummingbirds invest less energy into their territoriality displays, thus exhibiting a greater low-cost to high-cost behavioral frequency (Ewald and Orians 1983; Ewald and Carpenter 1978). According to previous studies, whether a hummingbird will use high-cost or low-cost territoriality behaviors is predicated upon the availability of high-quality food sources (Ewald and Carpenter 1978; Powers 1987; Ewald and Orians 1983).

The studies previously discussed focus on hummingbirds obtaining nourishment from both natural and artificial sources, such as bird feeders, in which resources are ostensibly scattered in a random pattern, as no mentions are made pertaining to the spatial distribution of resources. In an effort to differentiate those feeding area distributions from the arrangement which existed in this study, previous researchers have termed areas with mixed source, widely spaced feeding areas “natural/artificial decentralized feeding areas,” or NADFA, and areas with a centrally located and artificial source as “artificial concentrated feeding areas”, or ACFA. In this study, researchers are interested in studying whether the frequency of high-cost territorial displays and resource availability (This is the first time resource availability is specifically mentioned and it is the independent variable in your experiment. In the previous paragraph, you expand a lot on resource quality rather than resource availability. Based on your definition of resource quality and how you actually controlled for resource availability in the methods, it appears that they are two separate factors that can affect territorial behavior. I think it might be beneficial to provide greater background on resource availability (specifically resource depression). You allude to other previous studies. How do they define resource availability? Do they define what is high/low?) so often studied in NADFA also exists in ACFA with a dense population of feeding hummingbirds. Though the density of hummingbirds and space limitations differ from previous studies, the type of behaviors displayed by individuals can only exist as far as their energy budget, thus it is hypothesized that the same relationship between high-resources and a greater frequency of high-intensity behaviors will be exhibited.

Method and Materials:

This study was conducted at a residential three-and-a-half acre avocado grove located in Bonsall, CA, for a duration of four days during mid-March 2010. The daytime temperatures averaged between 22.8-23.9oC and wind speeds averaged 3.5mph. The hummingbird feeders were located on the residences’ back porch, facing the lower half of the avocado grove. Based on years of observations and the abundance of occupied and abandoned nests discovered on the premises by grove owners, the home range (Brown and Orians 1970) of the population of Anna’s hummingbirds studied remains within the grove’s perimeter and possibly within adjacent lots. These “resident” hummingbirds have been provided with nine large commercial feeders containing a consistent supply of an approximately 1.16M sucrose solution for the past six years. Each feeder has seven feeding stations resembling a red and yellow flower and has the capacity to hold 0.960L of solution. While these resident hummingbirds are not tagged and no official counts have been made, it is estimated by grove owners that the Anna’s population numbers between 100-200 individuals, depending on the season. Other species of hummingbirds, such as Rufous, Black-Chinned, and Calliope, have additionally been observed residing on the grove but only in seasonal durations. Researchers of this study took great care in assuring that videotaping was completed prior to the migrational introduction of non-Anna’s species.

The residence’s partially enclosed back porch uses five large evenly spaced pillars for support; two outside pillars and three middle pillars. Attached to the trim between each middle pillar are three hooks for the suspension hummingbird feeders. Nine total feeders are supported by this arrangement. As the resident hummingbirds are accustomed to nine feeders (or 63 feeding stations) at any given time, researchers considered this arrangement “high-resource” availability. When only three feeders were provided, a 66% reduction in resources, it was considered “low-resource” availability.

Maintaining a consistent sucrose content for each day of the study, the researchers designated the first and third day of the study as high-resource and the second and fourth day as low-resource. The three? spaces immediately between the middle pillars were videotaped for 40 minutes each day (simultaneously? At different times?) resulting in a total of eight hours of footage: four hours of high-resource footage and four hours of low-resource footage. Afterwards the footage was analyzed and each incident of territoriality exhibited next to a feeder quantified and categorized as either a high-intensity or low-intensity display based on the behavioral descriptions of previous studies (Ewald and Orians 1982; Ewald and Carpenter 1978; Brown 1969). According to Brown and Orians 1970’s study, a territory is defined as “…a fixed area, which may change slightly over a period of time, [in which] acts of territorial defense by the possessors, which evoke escape and avoidance in rivals so that…the area becomes an exclusive area with respect to rivals.” The intent of this study was to focus on the territorial displays exhibited strictly at the ACFA, therefore researchers did not examine the territorial spatial distributions of areas outside the ACFA’s perimeter. Thus, researchers defined the area surrounding one feeder as a territory (how large was this area?), in accordance with the above mentioned description, based on observations that the hummingbirds will claim and defend one feeder at a time.

For purposes of quantifying low-and-high-intensity chase occurrences, researchers designed the following: (1) low-intensity chases were those which remained within the scope of the camera lens, since a defending hummingbird needs only chase an intruding hummingbird a relatively short distance in order to ensure the invader leaves the territory; and (2) high-intensity chases were considered those which continued beyond the scope of the camera lens, as more energy was needed by defenders to chase intruders the longer distance. Gorget displays, typically the first territorial behavior exhibited by hummingbirds before increasing the severity of their warnings (Sibley 2001), and consequently the associated energy allocation, are distinctive enough low-cost behaviors that researchers needed only to use conventional descriptions in order to recognize and quantify occurrences. According to Ewald and Orians 1982 study, gorget displays are “an energetically inexpensive method of defense in which the owner moves its head from side to side while facing the intruder.” Males additionally flash the fuchsia colored iridescent feathers on their crowns to signal a warning to intruders (Sibley 2001). While vocalizations, or announcement, are another significant and frequently used (this part seemed to be a little repetitive so I crossed out first part) low-cost territorial behavior, due to the density of hummingbirds studied it was improbable to accurately determine which hummingbird vocalized a specific chirp recorded on video. For that matter, it would have been improbable to accurately assign chirps if the data had been collected in situ, because of the nearly non-stop activity occurring around each of the nine feeders. Thus, researchers eliminated the quantification of vocalizations based on the fact that most often vocalizations accompany gorget displays and/or chases (Sibley 2001).

The data collected quantified the number of high-intensity chases and attacks, and the number of low-intensity chases and gorget displays exhibited per day. These numbers were then compared as a frequency of high-intensity displays exhibited per low-intensity display exhibited. The resulting frequencies for high-resource and low-resource allocation were then accessed using a one-tailed t-test to determine if any statistical significance difference resulted.

Results:

A total of eight hours of video was taken over the course of four days. Each forty-minute recording segment was defined as one section with three sections recorded per day. As the video was analyzed, the number of high-intensity and low-intensity territorial displays were recorded for each section, as seen below in (Table 1). Then the frequencies of high-intensity displays were calculated for each section by dividing the number of high-intensity displays by the total number of territorial displays. (To be honest, I don’t think you need Table 1 at all. Figure 1 and Figure 2 sum up this information nicely and more directly. And you would include how exactly the frequencies were calculated in the methods section)

Table 1. Raw data of high-and-low-intensity displays on either high-resource days or low-resource days and calculated frequencies of high-intensity displays.

High-Resource
Section / High-Intensity Display / Low-Intensity Display / Frequency of High-Intensity Display
1 / 14 / 113 / 0.1102
2 / 3 / 31 / 0.0882
3 / 23 / 119 / 0.1620
4 / 7 / 34 / 0.1707
5 / 6 / 6 / 0.1224
6 / 11 / 11 / 0.1429

Low-Resource

Section / High-Intensity Display / Low-Intensity Display / Frequency of High-Intensity Display
1 / 3 / 58 / 0.0492
2 / 2 / 31 / 0.0606
3 / 1 / 23 / 0.0417
4 / 11 / 47 / 0.1897
5 / 18 / 79 / 0.1856
6 / 13 / 69 / 0.1585

From the calculated frequencies, a one-tailed t test was performed to see if the high-resource days exhibited a higher frequency of high-intensity territorial displays than the low- resource days. After the first two days of video were analyzed, it appeared that the data would be statistically significant and support the hypothesis; however, after looking at the data as a whole over the four days, the results were not as clear-cut. As seen in Figure 1, Anna hummingbirds exhibited high-intensity territoriality at a frequency of 0.1327±0.0129(±s.e) when receiving high-resource food and 0.1142±0.02893(±s.e) when receiving low-resource food. There was no statistically significant difference in the frequency of high-intensity territoriality under the two conditions (p=0.2883, one-tailed t test).

Figure 1. Average frequency of high-intensity territorial displays for Anna hummingbirds receiving either a high-resource or a low-resource food supply. Hummingbirds exhibited high-intensity territoriality at a frequency of 0.1327±0.0129(±s.e) when receiving high-resource food and 0.1142±0.02893(±s.e) when receiving low-resource food. There was no statistically significant difference in the frequency of high-intensity territoriality under the two conditions (p=0.2883, one-tailed t test).

Researchers then compared the number of territorial displays on each territory to see if there would be any statistical difference in the total number of displays. The video of the high-resource days was reanalyzed, recording only territorial displays that occurred around a one feeder territory. The resulting data is displayed in Figure 2. When receiving high-resource food, hummingbirds displayed a total of 18 high-intensity and 121 low-intensity territorial behaviors. When receiving low-resource food, they displayed a total of 43 high-intensity and 307 low-intensity territorial behaviors (Figure 2). There was no statistically significant difference in the number of territorial displays between high-resource and low-resource food supplies (p=0.4980, chi-square test).