ACOUSTIC TEMPORAL VARIATION, CALL ORDER, AND TRAVEL IN RESPONSE TO ANTHROPOGENIC NOISE IN WILD SIAMANG (SYMPHALANGUS SYNDACTYLUS)

Author

Justin D'Agostino, Southern Illinois University CarbondaleFollow

Date of Award

6-1-2021

Degree Name

Doctor of Philosophy

Department

Anthropology

First Advisor

Sutton, David

Abstract

Communication is a behavioral strategy used by species in the competition for survival and reproduction. Animals produce signals that are sent to receivers and occasionally transmit information that can be beneficial in a variety of contexts including avoiding predators, locating food resources, and maintaining the pair bond between partners. Researchers are interested in if auditory signals are variable and if the variability is biologically meaningful to receivers. Some species, such as siamang, sing duets as their form of acoustic communication and these duets have been studied for a variety of purposes such as investigating their structure, variability, and functionality. Siamang are a species of gibbon that combine a finite number of call units into structurally complex sequences in rule-governed ways, hereby conveying different contextual situations. Some wild gibbon species have demonstrated variation in the temporal structure of their duets and the distinctiveness in acoustic timing is unique enough for researchers to distinguish among individuals and groups based on their duet contributions. The process of coordinating and producing a uniquely timed duet could function in relation to the formation and maintenance of the pair bond. This dissertation analyzed the temporal variation and call order of wild siamang (Symphalangus syndactylus) duets that are comprised of four distinct sequences comprised of several note types. Previous research has provided detailed descriptions of captive siamang duets that include category descriptions of four call types: grunting, introduction, interlude, and great call. While these descriptions provide a general understanding of the structure of siamang duets, they can be confusing and difficult to study empirically due to their arbitrary nature. For instance, the siamang great call has been described as occurring in over 30 variants and in stable, or typical, and unstable, or atypical forms. The stable or typical form of the great call has been described to occur only after the male’s first bitonal scream. This previous research did not measure the timing of sequences through entire duets or conduct individual note analyses. Also, the most detailed investigations of siamang duets were conducted with captive animals and therefore wild siamang duets have not been quantitatively analyzed. This dissertation research measures the temporal variation of wild siamang duet call types, and the call order of individual notes in the US-II phrase. The findings will provide a more objective description of the siamang duet as well as determine if it has a more stereotyped or flexible nature. Also, in the wake of globalization, duets and other behaviors (such as travel) could be sensitive to anthropogenic disturbances. So, I am interested in how wild siamang might be impacted by human presence. The three main objectives of this dissertation are: 1) document the temporal variation in the call types of wild siamang; 2) document variation in the call order of wild siamang US-II phrases; and 3) document how wild siamang travel is impacted by anthropogenic noise. The study was conducted at 4 remote field sites in northern Sumatra, Indonesia from August 2017 through May 2018. With the assistance of a research team, I recorded acoustic and behavioral data from 10 siamang groups and two individual females. I collected acoustic data using a combination of autonomous recording units and handheld recording devices while following the study subjects on foot. I marked GPS points during behavioral scans and conducted experimentally induced noise playback experiments. I analyzed the audio data with the software program Raven Pro 1.6 and spectrograms were produced with a 512-point (11.6 ms) Hann window (3 dB bandwidth = 135 Hz), with 50% overlap and a 512-point DFT, yielding time and frequency measurement precision of 2.9 ms and 43.1 Hz. I conducted statistical testing in R Studio Version 1.3.959 in conjunction with the statistics lab at Southern Illinois University. In Chapter 3, I performed multilevel regression analysis in the form of a Generalized Linear Mixed Model (GLMM) in order to investigate temporal variation the time spent singing call types (grunting, introduction, interlude, and great call) accounting for differences between songs, groups, and field sites. I used a Bayesian hierarchical model to avoid calculating estimates using data with repeated and imbalanced sampling, and explicitly model variation within individuals, between songs, between groups, and field sites. In Chapter 4, I used the Levenshtein Distance (LD) method for investigating structural variation in the call order of US-II phrases in the siamang duet. The LD method is a quantitative technique for measuring the similarity of sequences and to quantify the differences between strings of data. In Chapter 5, I used the Friedman test to compare the means of distances traveled in 5-minute intervals in the following conditions: pre-playback, cicada (control), traffic, jackhammer, and music. I used post hoc tests with Bonferroni correction to analyze the pairwise comparisons of the five experimental conditions. Overall, the results suggest support of previous studies on the variability of the gibbon duet and animal travel in response to noise. In Chapter 3, I found some temporal variation in call types to exist at all levels of the modeling but only statistically significant when comparing the field sites of Ketambe and Kutapanjang. Relating to the functionality of total call duration, I did not find significant differences in the behavioral categories before or after any of the calls regardless of their total duration. In Chapter 4, I found US-II phrases are not as stereotyped as previous descriptions have suggested. The mean LD distances were higher than those compared by other duetting primates. Also, the results from one-way ANOVA and Games-Howell Post Hoc Testing showed significant mean differences between groups. In Chapter 5, I found siamang are impacted by noise as they traveled further after the anthropogenic noise playback experiments compared to pre-playback travel. However, the pairwise comparisons were not statistically significant and therefore I could not determine which noise condition caused the change in travel distance. Collectively, the results of this dissertation suggest that there is considerable variability in both wild siamang call types and call order in a previously described stereotyped phrase, and that wild siamang travel away is impacted by introduced anthropogenic noise. What this contributes to the field of vocalization studies is further support that precise measurement of song parameters gives a better understanding of song structure than somewhat arbitrary written descriptions. It is important to have descriptions of notes produced in songs, but to understand how those notes relate to each other, an individual note analysis and rigorous statistical methodology is necessary. The variation found in this dissertation could be due to the unique siamang response to their partner’s timing in the process of forming the pair bond. But as the siamang duet is thought to have multiple functions, the temporal variation could also be transmitting other information such as individual and group identity while functioning in territorial defense. In addition, anthropogenic noise caused the animals to change their travel behavior and this suggests it could be harmful if faced on a regular basis. Anthropogenic noise can be detrimental if it occurs in home ranges of species that are territorial and facing shrinking habitats due to continued human encroachment. In order to understand if these findings are biologically meaningful and to have firmer conclusions, future long-term studies will need a larger data set over longer periods of time. It is also important to have a larger data set and compare more siamang groups that live in various noise levels and measure variables such as caloric intake, daily travel, copulations, and number of offspring.