The widespread use of vocal communication in the animal kingdom has led me to investigate both the application of communication to wildlife conservation and the impact of human technology on communication. Acoustic signals can contain a wealth of information about the signaler, and while individually distinctive vocalizations, or “acoustic fingerprints”, have been used as a tool for the conservation of bird populations, few studies have investigated the potential of a bioacoustic tool for use with mammals. I explored such an application by testing for the presence of acoustic fingerprints and age-related information in the alarm calls of two ground squirrel species with varying levels of sociality: the highly social Belding’s ground squirrel (Spermophilus beldingi) and the relatively asocial golden-mantled ground squirrel (Spermophilus lateralis). I recorded alarm calls from both species, measured acoustic variables using Praat software, and subjected these variables to neural network and discriminant function analysis. Both methods demonstrated the presence of acoustic fingerprints and age-related information in the alarm calls of each species. The temporal stability of acoustic fingerprints was tested in S. lateralis, and this information does appear to be stable over time. In addition, both analysis methods revealed sex-related information in S. beldingi alarm calls. These results together indicate that a bioacoustic tool utilizing the multiple layers of information contained in animal signals is feasible for use with mammals, and may be a valuable addition to the conservationist’s tool kit.

In addition to containing information, the acoustic structure of animal signals is adapted to their function and for effective transmission through the environment. Anthropogenic noise can cause signals to be masked, rendering them undetectable to a receiver. By disrupting communication, masking can directly influence population dynamics. Masking also reduces the effective range, or the distance a signal can travel and still be detectable. I investigated the effects of road noise on a mammalian communication system, again employing S. beldingi as a model. I recorded mean and peak noise amplitudes from locations alongside Highway 120 in Yosemite National Park with different traffic speeds, and at the center of nearby ground squirrel colonies. From these measurements, I calculated the signal to noise ratio (SNR) at each location for a receiver at multiple distances from the sender to determine the effective range. The results indicated that road noise does have the potential to mask the alarm calls of this species, but mainly at peak amplitude levels and only for roadside locations. The effective range of alarm calls produced alongside the road was reduced significantly for all traffic levels. Such signal range reductions can have serious fitness repercussions in the form of increased predation risk if this species is unable to respond through genetic or behavioral mechanisms to maintain signal efficacy. The negative fitness consequences of masking in this species may also trigger cascading effects in populations of its predators. In this way, noise from roads placed in close proximity to protected areas may decrease the effectiveness of the reserve.