Use of Song Types by Male Carolina Chickadees (Poecile carolinensis)

Date of Award

January 2017

Degree Type

Closed Access Thesis

Degree Name

Master of Science (MS)


Biological Sciences

First Advisor

Gary Ritchison

Second Advisor

Luke E. Dodd

Third Advisor

Charles L. Elliott


North American chickadees (Poecile spp.) are a closely related group in which vocalizations are very similar. The combinatorial 'chickadee' call, given by all members of the genus, has been well studied, especially in Black-capped chickadees. Chickadee song has received much less study, particularly in Carolina chickadees. Carolina chickadee songs vary widely with geography and few detailed accounts their song repertoires are available. Also, the use of song types among breeding stages has not been studied in Carolina chickadees. Therefore, my objectives were to determine the size of the song type repertoires of Carolina Chickadees in east-central Kentucky by examining variation in the characteristics of male songs and to examine singing rates during different breeding stages to gain insight into the possible function(s) of the songs of male Carolina Chickadees. I studied Carolina chickadees at the Blue Grass Army Depot in Madison County, Kentucky, where nest boxes were placed on forest edges and were occupied by eight focal Carolina chickadees.

A parabolic microphone was used to record male Carolina chickadee song bouts between 30 March 2016 and 24 May 2016. Spectrographic analysis of song syllables was done to examine variation in fundamental frequency and to categorize syllables. The order (Method 1) and number of syllables (Method 2) was used to assess song type using two methods.

Analysis revealed nine syllable types and three song types using Method 1 (order of syllable types) or 20 song types using Method 2 (order and number of syllable types). Mean repertoire size for Method 1 and Method two were 1.625 ± 0.324 songs (range: 1-3) and 6.125 ± 1.46 songs (range: 3-13), respectively. Individual repertoire size and the number of songs recorded from individuals were not correlated for Method 1 (p = 0.27, r = 0.45), but were correlated for Method 2 (p = 0.022, r = 0.781), indicating that the repertoire, as defined by Method 2, was inadequately sampled. The use of faint songs, songs up to 12 syllables, and songs with inverted frequency pattern have been mentioned briefly in Carolina Chickadee literature and were used by males in my study. If the number of syllables in a song is not important for song categorization (Method 1), repertoire sizes for males in my study were small (<6) and the large number of syllable types I identified may facilitate communication that would typically be accomplished by having a large number of song types. Alternatively, if the number of syllable types is important to Carolina Chickadees (Method 2), song types may be the primary unit of communication. The importance of song length in Carolina Chickadees is in need of further study.

The singing rates of male Carolina Chickadee (as defined by Method 1) among breeding stages were examined. Singing rates (songs/hour) were highest in the nest building/egg-laying stage (123.3 ± 30.0 songs/hour) for all song types; the start of incubation coincided with a decrease in singing rates. Song 1, defined by Method 1, had the highest use overall and among breeding stages and was used by all individuals. Analyses of Song 2 and Song 3 were not possible as only two bouts of Song 2 and three bouts of Song 3 were observed. Although my sample size was small and conclusive analysis of song function was not possible, continued use of song after pairing (i.e. throughout all Carolina chickadee breeding stages) could mean that song is used to stimulate the female while she is fertile or to defend the territory, rather than to attract a mate.

This document is currently not available here.