This report summarizes the results of satellite-tagging efforts conducted prior to Submarine Command Course (SCC) training events from August 2021 and 2022 on the Pacific Missile Range Facility (PMRF) off the island of Kaua’i. Twenty-one delphinids were satellite tagged with the intent of observing their behavior before, during, and after the SCCs and estimating received levels of mid-frequency active sonar (MFAS). Of these, 16 individuals remained in the area or with direct paths between their location and sources of surface ship hull-mounted MFAS such that received levels could be estimated using propagation modeling. These included six short-finned pilot whales (Globicephala macrorhynchus), one false killer whale (Pseudorca crassidens), five melon-headed whales (Peponocephala electra), one rough-toothed dolphin (Steno bredanensis), and two common bottlenose dolphins (Tursiops truncatus). The analytical methods developed and described in Henderson et al. (2021) were applied to the resulting data; these included smoothing the animal tracks in 5-min steps using the R-package crawl, calculating the 95% confidence error ellipse around each step location, and then modeling transmission loss along multiple radials through the error ellipse from the sea surface to the seafloor. Dives were also modeled using behavior logs from the tags so that the depth of the animal at the time of each transmission per 5-min bin could also be considered when estimating the received levels (RLs). The median RL for each 5-min bin was calculated based on the aggregation of all transmissions and radials per bin, this plus two times the standard deviation (± 2*SD) are reported as the range of possible exposure values. Dive behavior variables were also analyzed within diel period (dawn, day, dusk, night) and across SCC phase (before, Phase A, Interphase, Phase B, after) to determine if dive behavior changed relative to the presence of training activity or MFAS.
For all animals, the maximum median received sound pressure levels ranged from 72.6 to 147.7 dB re 1 μPa. In fact, as many of the animals had moved out of the area prior to the onset of MFAS, maximum median RLs were at or below 100 dB re 1 μPa for eight of the animals. The other eight were on or near the range during periods of MFAS and therefore had relatively higher estimated RLs, although no estimated RL exceeded 156.2 dB re 1 μPa (including +2*SD). Some differences in dive behavior were noted as well, including changes in median dive depth and median dive duration between SCC phases, as well as changes in dive rates or the amount of time spent at the surface. However, there were no consistent changes in behavior either within or across species; these variables differed between the SCC phases in a variety of ways across individuals. In addition, no changes in horizontal movement were observed relative to periods of MFAS. All tagged animals were also compared to Cascadia Research Collective’s photo-identification catalog; most were matched to existing animals in the catalog, with one new bottlenose dolphin and two new rough-toothed dolphins added. Of the pre-existing identifications, the false killer whale, three of the pilot whales, two of the bottlenose dolphins, and one of the rough-toothed dolphins all belong to known Hawaiian island populations, and the latter two belong to Kaua’i island-specific populations. These data, aggregated with previously tagged or photo-identified animals from Kaua’i and the other Hawaiian Islands, build the basis for understanding long-term consequences of exposure at the population level as well as the potential to identify and understand the impacts of repeated exposures. As most of the tagged animals in this study are island (or Islands) residents, it is possible they have been exposed to MFAS previously. Since none of the animals demonstrated consistent or overt responses, it is also possible they have habituated to (or at least tolerate) MFAS and other training-associated sounds. However, with additional future data and continued improvements in analytical methods for satellite tag data, these aggregated data may be able to be used in finer-scale behavioral response analyses, and some of these uncertainties may be addressed.
Citation:
Henderson, E.E., C. Martin, A.E. Harnish, M.A. Kratofil, R.W. Baird, and S.W. Martin. 2024. Received Level Estimation, Behavioral Response, and Diel Behavior Analyses for Delphinids Tagged at the Pacific Missile Range Facility in 2021-2022. Naval Information Warfare Center Pacific.
Download PDF