Estimating a thermal constant of spawning to explain spawning time of Pacific herring Clupea pallasii across space and time

ABSTRACT: Explaining variation in life history phenology requires us to disentangle environmental-dependent variability from that caused by adaptive change across time and space. Here, we offer thermal time models (models measuring time in temperature units) as tools to understand the spawning dynamics of small pelagic fish, such as Pacific herring Clupea pallasii. We hypothesised that thermal time explains the annual timing of spawning of Pacific herring across space and time. By testing this hypothesis, we identified developmental constants (thermal constants of spawning) that can be used to make spawning time predictions. We examined spatio-temporal changes in Pacific herring spawning time over a 69 yr period (1941-2010) across 6 regions off British Columbia (BC), Canada. We estimated the degree-days (DD, °C-days) from the onset of gonadal maturation to spawning by combining spawning time estimates with distribution-specific temperature estimates. We then fitted models to explore how DD to spawning can be used to explain observed spawning time patterns across space and time and identified temperature-independent sources of variability (e.g. adaptive differences among regions, spawner size). We found that, even though Pacific herring often spawned ∼5 d later with each increasing degree in latitude, the average thermal time in DD to spawning was ∼1700°C-days. We also found that DD to spawning explains linear variation in spawning time across years for some regions of the BC Pacific herring. Thermal time models can aid in predictions of environmental responses and forecasts of life-history phenology in a changing climate.