Incorporating temperature-driven seasonal variation in survival, growth, and reproduction into population models for small fish

Abstract

Seasonal variation in survival and reproduction can be a large source of prediction uncertainty in models used for conservation and management. A seasonally varying population model is developed that incorporates temperature-driven differences in mortality and reproduction for the small fish species the sheepshead minnow Cyprinodon variegatus. A temperaturedependent growth rate function is developed for the von Bertalanffy constant, K, measured at various temperatures and fit to a logistic curve. The value of K at each temperature estimates the duration of mobile life stages. Stage-specific mortality is modeled as a power function of size that includes a temperature-dependent modification for extraneous ecological factors such as predation. Seasonal reproduction is described through temperature-dependent functions of embryo survival and stage duration. Model sensitivity analysis indicates that the largest influence on the population growth rate is K. Cessation of reproduction at lower temperatures is also a critical driver of population growth rate. Population projections used either seasonal or constant parameters for 16 hypothetical populations and demonstrated the potential for long-term error propagation in population projection using constant rates. Models used for conservation and management should employ realistic temporal resolution that will allow for seasonal effects of temperature on growth, reproduction, and survival to be incorporated into population projections.