| Summary: | Effective conservation is often an inter-disciplinary endeavor, even drawing from seemingly disparate disciplines within the field of biology. As a case in point, physiological indicators, such as mass, fat mass, and cortisol levels, are rapidly becoming important links between disturbance and population-level effects of disturbance. However, to complete the link from behavioral and physiological responses seen in individuals to population-level effects also requires integration of data from short-term behavioral and physiological studies as well as long term population monitoring. Historically, physiological studies have mainly focused on finding universal equations that describe physiological processes, so results from these studies tend to emphasize strong relationships and understate uncertainty. In contrast, ecologists and population modelers often must acknowledge uncertainty, not only because the complex study systems are fraught with unexplained variability, but also because the level of resulting uncertainty has important implications for species management, decision making, and conservation. To provide the best possible information, researchers need to also quantify and incorporate uncertainty in physiological measurements and models when conservation decisions are at stake. The elephant seal model is the first case study and most complete marine mammal model linking disturbance of individuals to population-level effects, but the models do not incorporate uncertainty in important physiological parameters (fat mass and wean mass). However, 95% posterior intervals for wean mass range from ± 10 to 25% of mean values for at least 20% of the measurements in northern elephant seals (Mirounga angustirostris). In addition, preliminary results from validation studies show fat mass 95% posterior intervals are at least ± 8% of mean estimates. We will demonstrate how the uncertainty in physiological metrics is carried forward in northern elephant seal conservation models, potentially leading to different ...
|
|---|