Description and analysis of dietary preference of lingcod in the nearshore zone, Oregon

Graduation date: 2010The lingcod (Ophiodon elongatus) is an important groundfish species in both recreational and commercial fisheries of the Northeastern Pacific. It is a large, fast-growing, generalist predator of invertebrates and fishes. In response to concerns that lingcod may limit populations of rockfishes (Sebastes spp.), I compared the diets of 375 adult lingcod from two locations on nearshore reefs along the Oregon Coast with estimates of relative prey availability from dive surveys. Juvenile and adult rockfishes of relatively site-attached species were the most abundant potential prey observed in dive surveys, yet they were the least preferred prey type as determined by Johnson’s (1980) preference analysis. In contrast to the transient, pelagic fishes that comprised 46% of lingcod diet by number, rockfishes comprised only 4.7% of prey items. The implication for marine reserves being considered along the Oregon Coast is that lingcod are unlikely to be a major threat to protected rockfish stocks. However, current preference models that assume unconstrained capacity (the sum of factors that reduce consumption) and equivalence between abundance and availability perform poorly when comparing among many prey types because of scale distortion (where relative consumption is compared with unrealistic levels of prey availability) and large and uneven bias in prey availability measures. By assuming capacity constraints based on maximum observed use prey availability can be more realistically described, and problems of\ud scale distortion and prey abundance/availability mismatch that occur with resource preference models can be solved. The inclusion of resource capacity with a preference index has important advantages when analyzing the resource use of a habitat and prey generalist. Additionally, a standard prey selection model is presented that includes a binary probability term to characterize a fundamental difference between transient and resident prey types. This two-model approach described here is robust to un-aggregated inputs (raw data including zeros) and makes more efficient use of typical field data. For example, this approach provides indirect evidence that lingcod display inverse frequency-dependent prey selection. The two-model approach, when applied to lingcod, provides insight into the likely outcome of changes in prey abundances with respect to predator-prey relationships