40-20 Effects of Temporal and Environmental Factors on Seasonal Spiny Dogfish Abundance in the Northwest Atlantic

Skyler Sagarese , School of Marine and Atmospheric Sciences, Stony Brook University, Stony Brook, NY
Michael G. Frisk , School of Marine and Atmospheric Sciences, Stony Brook University, Stony Brook, NY
Katherine Sosebee , Population Dynamics branch, NOAA, NEFSC, Woods Hole, MA
Paul Rago , Population Dynamics branch, NOAA, NEFSC, Woods Hole, MA
Regional fluctuations in spiny dogfish, Squalus acanthias, abundance in the Northwest Atlantic are thought to result from variations in the migratory patterns of juvenile and mature dogfish rather than locally driven population dynamics. However, the complex spatial behavior of spiny dogfish complicates deciphering the driver(s) of population fluxes. To elucidate the underlying mechanisms of spiny dogfish migratory behavior we modeled the occurrence and presence of neonate, juvenile, and adult spiny dogfish using hurdle models fitted with Generalized Additive Models (GAMs). Models were built using data from the Northeast Fisheries Science Center (NEFSC) Bottom Trawl Survey conducted in spring (1968-2009) and fall (1963-2009). GAMs were built separately for juveniles and adults of each sex using only data after 1980 due to the lack of sexing specimens prior to this period. The variations in dogfish abundance with time (year and Julian day) and environment (depth, bottom temperature and zenith) were investigated by modeling: (1) the probability of capturing dogfish based on presence/absence data and (2) dogfish presence based on positive abundance. Seven variables (temporal, spatial and environmental) were excluded from analyses due to high correlations with other covariates. Optimal GAM models were chosen based on the lowest AIC. Preliminary results indicated inclusion of temporal and environmental factors improved our explanation of spiny dogfish abundance and distribution. GAMs revealed that the combined effects of all variables influenced neonate occurrence in both seasons and their presence in spring while depth and bottom temperature were significant predictors in the fall. For juvenile and adult dogfish (males and females), the combined effects of all variables were important predictors of occurrence and presence in both seasons. All stages were tolerant of a wide range of environments but showed preferences for certain environmental conditions. Preliminary GAM results indicated neonates captured during fall predominantly occurred in shallower, warmer waters later in the season compared to spring captured neonates. Conversely, juveniles inhabited warmer, deeper waters than adults in both seasons. Our results will help decipher trends in abundance from changes in decadal and seasonal distributions of spiny dogfish.