Comparative Evaluation of SELECT and Generalized Additive Models for Estimating Gillnet Selectivity of Snow Crab (Chionoecetes opilio) in the East Sea of Korea

This study evaluated snow crab (Chionoecetes opilio) gillnet selectivity using the traditional Share Each Length’s Catch Total (SELECT) framework and a generalized additive model (GAM). Comparative fishing experiments were conducted in the East Sea of Korea using four stretched mesh sizes (210, 240, 255, and 270 mm). Among the candidate SELECT models, the binormal model had the lowest Akaike’s Information Criterion (AIC), although the normal and gamma models also received substantial support (ΔAIC < 2). Under the equal-fishing-power assumption, the ascending L50 increased from 70.47 mm for the 210 mm mesh to 90.60 mm for the 270 mm mesh, indicating a progressive shift in retention toward larger crabs across the tested gear configurations. The final GAM2 and the selected binormal SELECT model produced closely similar fitted selectivity patterns across most of the observed relative-size range and nearly identical in-sample agreement with the observed conditional catch proportions. However, the binormal components should not be interpreted as distinct biological capture mechanisms because capture behavior was not directly observed. Overall, the results support the complementary use of SELECT and GAM for evaluating snow crab gillnet selectivity, while acknowledging uncertainty associated with relative fishing power and the fixed panel arrangement.