Bioenergetics individual based model explorations of juvenile coho salmon growth during their first summer on the Oregon shelf

Graduation date: 2010Salmon survival and eventual recruitment success have long been thought to be determined within the first summer following ocean migration. Juvenile growth during this period is largely influenced by ocean conditions such as temperature, prey availability, abundance, and quality. Shifts in these conditions due to climatic perturbations are particularly prevalent in systems controlled by seasonal upwelling events, and consequently can have large influences on summer growth, overwintering survival, and eventual cohort recruitment. Individual Based Simulations (IBM) were performed using a modified Wisconsin bioenergetics model on juvenile coho salmon (Oncorhynchus kisutch) using forcing fields characteristic of the Oregon coast from 41.75°N to 45°N during late spring to summer (June 1st to August 15) of 2000 and 2002. The bioenergetics model used consumption estimated from prey fields developed from June and August U.S. GLOBEC cruises, and a multiple prey foraging model. The consumption rates were converted to realized growth of the juvenile coho using the standard Wisconsin model parameterization for coho salmon, except that respiration components were modified to coho specific rates using a functional relation of Trudel and Welch (2005). Temperatures for the simulations were provided by a Regional Ocean Model System simulation of the Oregon and Washington shelf environment physics, that had bias in temperatures removed using observed SeaSoar field temperatures from June and August cruises. Observed weights of juvenile coho salmon were used to establish the initial weight ranges and to determine the model’s suitability and accuracy to describe juvenile and ‘jack’ growth. Recognizing that several of the dominant prey types of juvenile coho (from stomach content analysis) may be significantly underestimated by the plankton net sample density estimates available from the cruises, we examined through simulation several scenarios that made density corrections to these prey types. The simulations were successful in capturing differences in the juvenile coho growth patterns of 2000 and 2002. In 2000, the spring transition was relatively late, upwelling was delayed, and the mesozooplankton community was comprised of smaller individuals, compared to 2002. The overall lower regional prey biomass and the shift to smaller size plankton in 2000 resulted in delayed coho growth despite a larger mean initial smolt weight. Base case simulations of coho in 2002 (starting with 75g WW juveniles) had 22.9% higher final weights than did fish simulated using 2000 temperatures and prey fields. This indicates substantial potential for interannual variability in growth. However, the base case simulations assume that juvenile coho encounter with prey is random, whereas it is more likely that coho are somewhat able to optimally exploit their prey environment by preferentially locating and feeding in regions have higher prey densities, or better quality prey. We examined several different intensities of “optimal feeding”, where the juvenile coho preferentially fed in regions having higher prey densities. Juveniles that were able (through unspecified behavioral mechanisms) to find regions of high prey density grew much faster during their first summer in the ocean than did fish that randomly encountered prey. Optimal feeding by individual juvenile coho salmon resulted in final sizes of fish in 2000 and 2002 that were similar, as the maximum prey density patches in 2000 were higher than in 2002