Abstract
The Population Consequences of Disturbance (PCoD) framework provides a conceptual framework which can be used to forecast a plausible range of population-level outcomes given a specific set of input data. To implement such frameworks for a species of interest requires significant baseline knowledge of foraging patterns, lifehistory, and demographic parameters. However, for many marine mammal populations, current knowledge is lacking and such ‘data poor’ situations mean that any such forecasts have significant uncertainty associated with them. Given these uncertainties there is merit in identifying the data gaps that need to be filled in order to better parameterise the models. However it may take decades to fill these gaps and, in the meantime, undetected population declines may occur. In this report we focus on identifying methods for monitoring populations that are subject to disturbance that may also provide insights into the processes through which disturbance may affect these populations. In addition, we aim to identify priorities for monitoring to inform future PCoD analysis of the potential effects of Navy activities on marine mammal populations. Therefore, our ultimate objective is to identify a suite of variables that can provide information on changes in demography or health, together with the methodologies that can be used to measure these variables.
To identify and address the knowledge gaps highlighted above, we conducted a comprehensive survey of the literature to identify suitable response variables which could be monitored using established survey techniques or techniques that are currently in development. Following the initial literature review, we held a workshop with a small number of experts on monitoring approaches to develop this list further and to identify the current state of utility and feasibility of the different approaches for Navy relevant marine mammal species groups. This report summarises the results of the literature review, and the outputs from the workshop. In addition, we explore the methods and/or techniques required to collect appropriate datasets and the feasibility of using them to monitor different species and populations (section 2) with sufficient precision to avoid false positive results (i.e. results that suggest a population is in decline when it is not). Therefore, using existing PCoD benchmark models, we explore the potential for different demographic parameters to provide early warning indicators of population decline and explore the potential to detect change and limit the proportion of false positive results (section 3). In order to realistically assess what methods might be feasible to conduct, it was considered crucial to assess the monitoring infrastructure that currently exists. As such, we also summarize US Navy marine mammal monitoring in terms of the approaches already in use, the platforms for research available (e.g. vessels, aerial, fixed sensors etc.) and species/populations which are currently (or have recently been) monitored (section 4). Building from this foundation, we assess the potential for current monitoring practice to inform a PCoD analysis (section 5.3) using the lessons learned from the literature review and sensitivity analysis phases.
Using existing PCoD models, we determined that changes in certain demographic variables are strongly correlated with changes in abundance or population status, and can therefore provide some early warning of future changes in abundance. In particular, the proportion of immature animals in a population might provide a reasonable early indicator of population decline. We also explored the ratio of mothers to calves/pups but determined that there was a high risk of false positives (i.e. predicting a decline when there is none). We observed that demographic parameters tend to be most commonly estimated from monitoring using established approaches such as visual surveys and capture-recapture. In addition, both vertical and lateral photogrammetry appear to be viable methods to determine important demographic parameters. Monitoring body condition might be a suitable approach to identify ‘unhealthy’ animals (though determining causation may be difficult) and is a particularly attractive route for monitoring PCoD and a range of methods are in development to explore this topic area. In general, monitoring individual health and physiological variables was determined to be important in informing elements of the PCoD framework, primarily via photogrammetry, remote tissue sampling, direct handling and individual tracking approaches. The continued development of remote tissue sample libraries and analytical approaches to improve our understanding of stress response, physiology and –omics fields is critical. Additionally, it is important to continue the use (and development) of PAM techniques to monitor cetacean populations to better understand the relative and absolute indices that can be derived to inform PCoD.
We recommend that, where possible, monitoring programs are developed to specifically inform future PCoD analyses, which requires a clear set of objectives regarding the purpose of the monitoring. Monitoring programs should be identified which can provide reference or control populations against which observed patterns can be compared.
We have highlighted that, where possible, the integration of new technologies into existing Fleet/SYSCOM monitoring efforts might provide significant added value. The inclusion of novel approaches into monitoring programs where infrastructure exists means a cost-effective increase in what can be achieved by a given program.