Bottlenose dolphin abundance in coastal Moreton Bay 2000

Original provider:
University of California, Irvine

Dataset credits:
Vimoksalehi Lukoschek, University of California, Irvine

Abstract:
Marine megafauna populations in coastal waters are increasingly threatened by anthropogenic impacts. Moreton Bay, a large embayment in south-east Queensland, lies adjacent to one of the fastest growing regions in Australia and has a resident population of bottlenose dolphins, <i>Tursiops aduncus</i>. Evaluation of the effectiveness of any proposed management strategy requires robust population abundance estimates.

We estimated abundances of bottlenose dolphins in central eastern Moreton Bay (350 km2) using two commonly used abundance estimation methods for cetaceans: photo-identification mark-recapture and line-transect surveys. Mark-recapture data were analyzed in CAPTURE using a model that allowed capture probabilities to vary between sampling events and between individuals. Based on an estimated 76% of the population identifiable photographically, total abundance estimates were 673 ± 130 s.e. (1997) and 818 ± 152 s.e. (1998). Line-transect data, analyzed using DISTANCE, gave an abundance estimate of 407 ± 113.5 s.e. (2000). These abundance estimates are large compared with many other coastal bottlenose dolphin populations. The line-transect surveys comprised a pilot study, and the lower line-transect abundance estimate is probably best attributable to methodological issues. In particular, smaller mean group size was estimated for the line-transects surveys (2.85 ± 0.29 s.e.) than the mark-recapture surveys (4.87 ± 0.39 s.e., 1997; 5.78 ± 0.73 s.e., 1998), and line-transect group sizes were probably underestimated. In addition, the line-transect detection probability (g(o)) was assumed to be one but was almost certainly less than one. However, the possibility of an actual decline in population size cannot be ruled out. Coefficients of variation (CV) were lower for mark-recapture than for line-transect surveys, however, CVs of line-transect estimates could be lowered through improved survey design. We evaluated the power of these surveys to detect trends in potential population declines for bottlenose dolphins in Moreton Bay and make recommendations for ongoing monitoring strategies.

原始提供方：加利福尼亚大学欧文分校（University of California, Irvine）

数据集贡献者：Vimoksalehi Lukoschek，加利福尼亚大学欧文分校

摘要：近岸海域的海洋巨型动物种群正日益受到人为活动的威胁。摩顿湾（Moreton Bay）是昆士兰州东南部的大型海湾，毗邻澳大利亚发展最快的区域之一，栖息着宽吻海豚（*Tursiops aduncus*）。评估任何拟议管理策略的有效性，均需稳健的种群丰度估算数据。

我们采用鲸类研究中两种常用的丰度估算方法，对摩顿湾中东部（面积350平方千米）的宽吻海豚种群丰度进行了估算：照片识别标记-重捕法与线样带调查法。标记-重捕数据通过CAPTURE软件分析，所用模型允许不同采样事件间以及不同个体间的捕获概率存在差异。基于约76%的种群可通过摄影识别这一前提，总丰度估算值分别为1997年的673 ± 130 标准误（s.e.）和1998年的818 ± 152 标准误。通过DISTANCE软件分析的线样带调查数据，则得到2000年的丰度估算值为407 ± 113.5 标准误。

与其他诸多近岸宽吻海豚种群相比，本次估算的丰度数值偏高。本次线样带调查属于预实验，其较低的丰度估算结果或可主要归因于方法学层面的问题：具体而言，线样带调查估算的平均群规模（2.85 ± 0.29 s.e.）低于标记-重捕调查（1997年为4.87 ± 0.39 s.e.，1998年为5.78 ± 0.73 s.e.），且线样带调查的群规模大概率存在低估。此外，尽管假设线样带的检测概率g(o)为1，但实际几乎肯定小于1。不过，种群规模确实存在实际下降的可能性，这一点无法排除。

标记-重捕法的变异系数（CV）低于线样带调查法，但线样带估算的变异系数可通过优化调查设计得以降低。我们评估了本次调查用于探测摩顿湾宽吻海豚种群潜在下降趋势的效力，并为后续监测策略提出了建议。