High calf mortality in bottlenose dolphins in the Bay of Islands, New Zealand–a local unit in decline

Abstract

Bottlenose dolphins (Tursiops truncatus) in the Bay of Islands, New Zealand, have been studied for almost two decades. Since 2003, fewer than 150 dolphins visited the bay during each season and the local unit has declined 7.5% annually from 1997 to 2006. The causes of decline are unclear but probably include mortality and emigration. Here, we used a long-term database to estimate reproductive parameters of female bottlenose dolphins including recruitment rates. A total of 704 surveys were conducted in which 5,577 sightings of 408 individually identified dolphins were collected; of these 53 individuals were identified as reproductive females. The calving rate increased between periods (1997-1999 = 0.13, CL = 0.07-0.21; 2003-2005 = 0.25, CL = 0.16-0.35 calves/reproductive female/year). A 0.25 calving rate suggests that on average, a female gives birth only once every four years, which is consistent with the estimated calving interval (4.3 yr, SD = 1.45) but still is lower than values reported for other populations. Conversely, apparent mortality rates to age 1+ (range: 0.34-0.52) and 2+ (range: 0.15-0.59) were higher than values reported elsewhere. The high apparent calf mortality in conjunction with a decline in local abundance, highlight the vulnerability of bottlenose dolphins in the Bay of Islands. Long-term studies are required to understand the causes of high calf mortality and the decline in local abundance. Meanwhile, management should focus on minimizing sources of anthropogenic disturbance and enforcing compliance with current legislation.

Joint estimation of survival and breeding probability in female dolphins and calves with uncertainty in state assignment

“…Like Tezanos‐Pinto et al (), we were unable to discriminate the age of the young more precisely than YOY or calf (age 1–3). For this reason, the dataset was filtered so it only retained the females sighted with young individuals since the YOY state, in order to determine the calf's age in subsequent resightings.…”

“… Tezanos‐Pinto et al () supposed that all calves never resighted were alive (first line) and then supposed that all calves never resighted were dead (second line). …”

“…We estimated some reproductive parameters (e.g., young survival probability) with descriptive methods in order to compare them to the estimations provided by the CR multievent methods. The approach used in this study was derived from those described in Tezanos‐Pinto et al (). First‐year calf survival probability was calculated as the number of observed YOY that survived their first year divided by the total number of YOY observed over the full study period.…”

“…For this reason, the dataset was filtered so it only retained the females sighted with young individuals since the YOY state, in order to determine the calf's age in subsequent resightings. When a female previously sighted with a calf was resighted alone, Tezanos‐Pinto et al () assumed that the calf was dead as it did not reach weaning age. When a female was not resighted, the fate of the calf was unknown, so we did not consider this calf in the estimation of young survival probability using the descriptive approach.…”

“…While much information is available concerning its distribution and abundance (Currey, Rowe, Dawson, & Slooten, 2008;Laporta, Fruet, & Secchi, 2016), reproductive parameters are more difficult to study, as it needs more field efforts and more detailed information on the reproductive state of the individuals. The vast majority of methods used to estimate reproductive parameters in this species did not account for detection probability (Baker et al, 2018;Brough et al, 2016;Haase & Schneider, 2001;Tezanos-Pinto, Constantine, Mourão, Berghan, & Scott Baker, 2014) and thus are subject to the detection issues we detailed above. To address this, we used multievent CR models (Pradel, 2005) to estimate the probability of adult female survival and calf survival (age-dependent), breeding probability (dependent on calf survival), female detection probability, and calf observation probability (given the state of the female, to account for uncertainty in breeding state assignment).…”