“…While attempting to parameterize (e.g., lethal temperature) these simulations, we discovered that the required behavioural, ecological and physiological data – as well as field data required to test their predictions – often do not exist or are of dubious relevance. These issues likely reflect the difficulty of determining depth, temporal T b profiles and behaviour of overwintering ectotherms (Sinclair, 2001a), the logistical challenge of quantifying time series of physiological responses to chronic rather than acute temperature exposures (Huang and Tu, 2008; Sinclair et al ., 2015), and the incomplete information on cues (e.g., environmental, innate clocks) used for mid‐winter emergence (Heath, 1962; Bishop and Echternacht, 2004; Lutterschmidt et al ., 2006; Nordberg and Cobb, 2016). Our simulations thus rely on unvalidated parameter values, but nonetheless provide “approximate answers” (Tukey, 1962) to ecologically relevant questions.…”