Page 17 - Demo

New Directions in Marine Mammal Behavioral Noise Exposure Criteria5Marine Research of the National Academy of Marine ResearchDOI: 10.29677/MR.202406_4(1).0001 In a recent re-evaluation of the original Southall et al. (2007) approaches, Southall et al. (2021) adapted earlier response severity approaches and developed new analytical methods aimed to improve assessments of acute exposures and potential behavioral responses, with a more explicit distinction of field and captive scenarios. A brief synthesis of these developments and associated conclusions is given here. For behavioral criteria, Southall et al. (2021) diverged from hearing-based distinction of impulsive and non-impulsive noise categories and proposed practical operational categories of sources, applying the revised severity assessments. Southall et al. (2021) also derived discrete severity scales for captive and field exposure contexts and segregated responses along categorical \ As discussed above, repeated mortal strandings of marine mammals in association with military sonarsaccelerated the pace of science, including with the use of experimental methods with both simulations of and full-scale operational industrial sources, to try and understand the nature, likelihood, and mitigation of these effects. Many subsequent studies exposed tagged, free-ranging individuals to military sonar and other noise sources in controlled exposure experiments (Southall et al., 2016; Booth et al., 2022). In evaluating the results, the Southall et al. (2007) response severity scale has been applied and revised for practical applications using expert elicitation methods (Miller et al., 2012; Southall et al., 2019b, 2023a). These and other studies have presented exposure response results in probabilistic functions that clearly demonstrate the probabilistic nature of response in those conditions and have been integrated directly into regulatory approaches (e.g., the NMFS functions applied to U.S. Navy MFAS sources). It should be noted that these initial approaches, which were both entirely focused on acute events and severity assessment outcomes to identify exposures associated with events that could be evaluated relative to response significance, were (as noted by the authors) quite U.S. centric.While they have served to inform approaches in other nations, in addition to their data limitations in many areas, there are fundamental differences in regulatory approaches in other countries, many of which are more focused on population-level than the individual consequences evaluated as \arine Mammal Protection Act. Nevertheless, the increasingly quantitative and more comprehensive and appropriate acoustic and contextual characterization of studies following the Southall et al. (2007) criteria that applied aspects of it, provided considerable steps forward, such that some broader re-assessment was warranted.behavioral response exposure criteria was the evaluation of discrete responses of specific individuals or in some cases groups with an estimated number of individuals and the relation of responses of specified severity to measured or estimated RL. This approach presumes that an individual or individuals are exposed to noise at a specific level and that the response of assessed severity is the result of that exposure. Results across individuals within specified categories and species groups are then evaluated as a function of exposure RL to evaluate patterns and probabilistic relationships of exposure and response. At the time of the Southall et al. (2007) criteria, many of the available data used were observational data from responses measured or inferred during ongoing, uncontrolled industrial activities (much of which was related to seismic survey assessments discussed above), although a small number of experimental studies with simulations of noise in field or laboratory conditions were evaluated. At the time, these were aggregated by mammal groups according to their functional hearing (not behavioral sensitivity) and by their impulsive or non-impulsive characteristics (at the point of the sound source). As such and as noted extensively by Southall et al. (2007), the extensive species, contextual, and source specific differences resulted in large variation in response probability within these broad categories. However, certain patterns, particularly sensitive species groups, and relationships began to be evident. The derivation of the behavioral severity scale as applied to exposure-response conditions at least established a structured approach with which exposure events that could be identified and described in certain parameters (including but as clearly recommended for future studies not limited to RL) could be assigned. Finally, Southall et al. (2007) made many specific recommendations to improve and advance the basis for developing probabilistic response functions for different animal groups and noise types, notably increasing the rigor andcomplexity of noise exposure metrics and the use of experimentally controlled measurements.