MEPS 642:179-190 (2020)  -  DOI: https://doi.org/10.3354/meps13337

ABSTRACT: It is commonly assumed in surveys that the likelihood of capturing or observing individuals of a given species is constant. Yet evidence is building that catchability, or the likelihood of catching an individual present at a site, can vary. We used 5465 paired trap-video samples collected along the southeast US Atlantic coast in 2015-2018 to estimate trap catchabilities of 6 reef fish species (gray triggerfish Balistes capriscus, red porgy Pagrus pagrus, vermilion snapper Rhomboplites aurorubens, black sea bass Centropristis striata, red snapper Lutjanus campechanus, white grunt Haemulon plumierii) as the ratio of trap catch to standardized site abundance from corresponding videos. Species-specific trap catchabilities were then related to 2 primary predictor variables: water temperature and percent of the visible bottom consisting of rocky substrate. Water temperature strongly influenced trap catchabilities for all species after standardizing for all other variables. The 4 warm-water species displayed strong positive relationships between catchability and temperature; of these species, the smallest absolute increase in catchability occurred for vermilion snapper (0.0 at ~14°C to 0.05 at ~28°C) and the largest occurred for white grunt (0.0 at ~14°C to 0.49 at ~28°C). The 2 cooler-water species displayed either a dome-shaped (red porgy) or negative relationship (black sea bass) between catchability and temperature. Furthermore, trap catchabilities for all species declined substantially (42-80%) as the percent hard bottom of the site increased. Only when catchability is properly accounted for can results be considered unbiased and subsequent management advice be considered accurate.

KEY WORDS: Detection · Detectability · Capture probability · Detection probability · Trap · Survey · GAM