GrowChinook: an optimized multimodel and graphic user interface for predicting juvenile Chinook salmon growth in lentic ecosystems

Linked foraging and bioenergetics models allow for increased understanding of fish growth potential and behavior by incorporating prey availability coupled to environmental conditions including temperature and prey visibility. To inform our understanding of growth and vertical migration patterns of Chinook salmon (Oncorhynchus tshawytscha) inhabiting lentic ecosystems, we linked foraging and bioenergetics models to create GrowChinook ( http://growchinook.fw.oregonstate.edu/ ). This multimodel design and optimization routine has broad applications in examining growth potential and predicting habitat use in stratified environments. We demonstrate the use of GrowChinook for the spring–summer rearing period in three Willamette River basin reservoirs, Oregon, USA. These reservoirs support juvenile spring Chinook salmon that exhibit a novel reservoir-reared life history that includes larger juvenile fish compared with nearby stream-reared subyearlings. Model outputs of predicted growth and depth use patterns based on observed prey distributions and environmental conditions were corroborated by observed empirical size and growth data from other years. Our simulations support diel vertical migration as a tactic that increases growth potential and contribute to understanding juvenile Chinook salmon growth in stratified systems.

Bioenergetic and limnological foundations for using degree-days derived from air temperatures to describe fish growth

Degree-days (DD) are an effective metric for quantifying the thermal opportunity for ectotherm growth. There is strong empirical evidence to suggest that DD are useful for describing fish growth and that immature growth increases linearly with DD. However, fish ecology lags behind other disciplines in the widespread adoption of DD. We provide (1) a foundation for the observed linear relationship between immature fish growth and DD and (2) justification for using DD derived from air temperatures as a proxy for DD derived from water temperatures in fish science. We use bioenergetics models and both simulated and empirical water temperatures to show that immature annual and interannual fish growth are approximately linear with water DD. We then use simulated and empirical data to show that air and surface water temperatures are often highly correlated and that immature fish growth is also approximately linear with air DD. By connecting the dots among air temperature, water temperature, and fish growth, we lay the foundation for wider adoption of DD in fish science.

Larval Feeding: Mechanisms, Rates, and Performance in Nature

Larvae of many marine invertebrates must capture and ingest particulate food in order to develop to metamorphosis. These larvae use only a few physical processes to capture particles, but implement these processes using diverse morphologies and behaviors. Detailed understanding of larval feeding mechanism permits investigators to make predictions about feeding performance, including the size spectrum of particles larvae can capture and the rates at which they can capture them. In nature, larvae are immersed in complex mixtures of edible particles of varying size, density, flavor, and nutritional quality, as well as many particles that are too large to ingest. Concentrations of all of these components vary on fine temporal and spatial scales. Mechanistic models linking larval feeding mechanism to performance can be combined with data on food availability in nature and integrated into broader bioenergetics models to yield increased understanding of the biology of larvae in complex natural habitats.

Dependence of feeding rates on body mass when food density is limiting to growth

Bioenergetics models are commonly used to predict effects of changes in metabolic rates and food availability on growth. However, food intake rate generally is assumed to vary as Wd, where d = 2/3, an assumption based on observations from feeding trials in laboratory studies. Further, the von Bertalanffy growth function (VBGF) is specifically integrated using this assumption. We argue that when considered from an ecological perspective, d is highly uncertain, dependent on how swimming speed, reactive distance, and prey biomass varies ontogenetically with the growth of a predator. Incorrectly specifying d leads to incorrect predictions of consumption and metabolism, especially at younger ages that are typically under-sampled. Three alternate means of detecting departures from d = 2/3 are provided, the most promising of which involves fixing initial length of the generalized VBGF to the length at endogenous feeding and directly calculating von Bertalanffy parameters (L∞, K, t0). Using this approach, it may be possible to more accurately estimate consumption and metabolism and to characterize lifetime growth.