Beyond the mean: Variation in incubation temperature influences timing of hatch for lake trout (Salvelinus namaycush)

Abstract

Lake trout (Salvelinus namaycush) is an ecologically and economically important fall-spawning species that inhabits northern lakes at a wide range of latitudes and with widely varying temperature dynamics. However, fall spawning also means that adult behavior is largely divorced from the conditions at hatching. Embryonic development must therefore be responsive to external stimuli in a way that consistently results in hatch occurring during a viable period despite considerable variability in phenology. Despite extensive work on lake trout development and culture, hatch timing remains difficult to predict precisely by either days-to-hatch or thermal units (TU)-to-hatch without experience with a particular system and strain. We hypothesized that this problem may be due in part to a non-linear response to temperature variability during development, coupled with the potential for poorly controlled variability in past studies due to differences in hysteresis and thermal inertia among experimental setups. To test this hypothesis, we reared individually identified embryos from controlled parental crosses in temperature-controlled incubators set to maintain a range of constant or variable temperatures and measured both the timing and size of all individuals at hatch. Our results show that temperature variation has a pronounced effect on the timing of hatching, with embryos reared in a system with a 3 °C daily swing hatching an average of six days earlier than those raised in a constant environment with the same mean temperature. These findings raise questions about the utility of TU as a predictor of hatch timing, especially in natural systems that are often inherently variable.