Rapid changes in thermal environments are threatening many species worldwide. Thermal acclimatisation processes may partially buffer species from the impacts of these changes, but currently the knowledge about the temporal dynamics of acclimatisation remains limited. Acclimatisation phenotypes are typically determined in laboratory conditions that lack the variability and stochasticity that characterize the natural environment. Through a Distributed Lag Non-linear Model (DLNM), we use field data to assess how the timing and magnitude of past thermal exposures influence thermal tolerance. We apply the model to two Scottish freshwater Ephemeroptera species living in natural thermal conditions. Model results provide evidence that rapid heat hardening effects are dramatic and reflect high rates of change in temperatures experienced over recent hours to days. In contrast, temperature change magnitude impacted acclimatisation over the course of weeks but had no impact on short term responses. Our results also indicate that individuals may de-acclimatise their heat tolerance in response to cooler environments. Based on the novel insights provided by this powerful modelling approach, we recommend its wider uptake among thermal physiologists to facilitate more nuanced insights in natural contexts, with the additional benefit of providing evidence needed to improve the design of laboratory experiments. 

The data are a collection of Serratella ignita and Baetis rhodani Critical Thermal maxima (CTmax) values. 

S. ignita (Poda 1761) and B. rhodani (Pictet 1843) are common and ubiquitous freshwater invertebrates across Europe. Both species usually produce a single generation per year, with eggs overwintering (hatching: ~March/April) and adults emerging from April to September, depending on local thermal conditions (see, López-Rodríguez et al., 2009; Raddum & Fjellheim, 1993) with non-overlapping generations in Scotland (Maitland, 1965; Morgan & Egglishaw, 1965).  Larvae of both species were collected from four sites in Scotland, in rivers upstream and downstream from two natural lakes (Rescobie Loch and Loch Morie) and two artificial reservoirs created by damming (Loch of Lintrathen and Glenquey Reservoir), giving a total of 8 sampling locations . Larval sampling took place within a circle of 5 m radius from the loggers, and within the same riffle, using kick sampling. Subsequent local-scale spatial mapping of temperatures suggests that within-riffle temperatures are generally homogenous. Once collected, larvae were placed in containers filled with river water and immediately transported to the laboratory in cooler bags (average trip 2h by car). Throughout the transport, we assumed minor variation in the water temperature since the last temperature experienced in the field. CT_max values were established in experiments using high performance Grant Optima TX150 circulating water baths with programmable heating settings. Larvae were always handled using plastic spoons, to prevent damage. They were placed in 8 plastic cups (7x15 cm), one for each location. Additional plastic cups were used if the number of individuals exceeded 30. Cups were filled with respective site water in order to preserve thermal conditions and minimise physiological stress. The cups were then placed in the water bath. Prior to experiments, all individuals were equilibrated at 10°C for 10 minutes. During each thermal trial, samples were subjected to a constant water temperature increase of 0.1°C/minute, which was monitored and controlled by Grant TX150 programmable circulating water bath with C2G cooling attachment (Grant Instruments, Shepreth, UK). CT_max values for each individual were recorded when no physical response occurred after three consecutive prods (Becker & Genoway, 1979); these individuals were then removed from cups while the remaining individuals continued to be monitored. All the CT_max experiments were performed on the day of sample collection. No behavioural interactions between individuals were noted during trials. 

Water temperature was continuously recorded at each sampling location with HOBO Pendant UA-002-64 data loggers (Onset Corp. Bourne, MA, USA, accuracy +/-0.2°C) on a time step of 15 min over the 28 days prior to each larval sample collection. Loggers were placed carefully in locations with running water (riffles) to have good mixing through the water column, and in areas with representative and homogenous morphology of each stream far from sources of potential thermal alteration such as confluences.