Non-linear phenomena (NLP), including biphonation, deterministic chaos, sidebands, and subharmonics, have been identified in the spectral structures of both terrestrial and marine mammal calls. Monitoring health using NLP in speech aids human diagnostics and has been explored in primate, canine, and feline calls. Signature whistles from 16 bottlenose dolphins at the U.S. Navy’s Marine Mammal Program were recorded during times of known health status. NLP occurred in 56% of whistles recorded during normal health, and 55% during abnormal health. While NLP were common additions to whistles, health status did not have an effect on the likelihood of a whistle to contain an NLP. No consistent trend of NLP use was found amongst individuals; however, significant differences in the persistence and persistence adjusted per health status per individual indicated that individuals had different patterns of NLP use. Although unrelated to health, this study further emphasises the prevalence of NLP in dolphin signature whistles and highlights the need for further investigation into communicative value and underlying mechanisms of NLP production in dolphin whistles.
Dolphin, vocal biomarker, whistle, non-linear phenomena, health