MEPS 678:197-209 (2021)  -  DOI: https://doi.org/10.3354/meps13898

ABSTRACT: Understanding the relative performance of data collection methods is critical for the production of robust results in any biological field study. This is particularly relevant for monitoring studies of rare and inconspicuous species, such as some cetaceans. Here, we compared how passive acoustic and visual survey methods performed when surveying a multispecies cetacean community in a large-scale survey area and whether the simultaneous application of these methods leads to complementary results. For both methods we compared (1) frequency of unscheduled survey stoppages, (2) detection ranges, (3) success in identifying species, (4) precedence of either method in synchronous detections, (5) detection rates, and (6) influence of sea state on detection rates. Data were collected during 4 surveys conducted between 2015 and 2017, using simultaneous visual and acoustic methods in the Santos Basin, Brazil. A total of 1492 h of simultaneous sampling yielded 617 detections, of which 46.5% (n = 287) were exclusively acoustic, 18.5% (n = 114) were exclusively visual, and 35% (n = 216) were corresponding detections of the same group made by each method, resulting in 108 paired detections. The simultaneous application of visual and acoustic methods was complementary. Acoustics were more efficient in detecting cetaceans—detecting first and further from the vessel and with a greater detection rate—whereas visual observation was more accurate for species identification. When used together, detection rates per species increased, reducing potential biases. By improving acoustic classification through visually confirmed sightings, this integrated approach has the potential to provide a reliable sound library for cetaceans in the Southwest Atlantic Ocean.

KEY WORDS: Comparison · Bioacoustics · Detection rate · Line transect · Marine mammal · Passive acoustic monitoring · South Atlantic Ocean