Experimental setup

H. akashiwo (CCMP 3374) was isolated from Narragansett Bay June 10, 2010 when in situ water temperature was 21.2°C. After isolation the culture was maintained at 15°C. All experiments were conducted with cells that were grown in autoclaved, 0.2 µm sterile-filtered seawater (30-31 ppt) amended with F/2 media without silica (Guillard 1975). All cultures were maintained under a light intensity of 150 µmol photons m-2 s-1 and a 12:12 h light: dark cycle. Preliminary experiments to determine the characteristics of this strains’ growth patterns were used to maintain cultures in exponential phase by transferring cultures as needed every 4 to 10 days (depending on growth temperature), resulting in cell densities of 500-24,000 cell mL-1.  To avoid convolution of thermal response and effect of unconditioned media just after transfer culture transfers were restricted from 1 day prior and 1 day post change in temperature (Grabski and Tukaj 2008 ).

Temperature treatments

Temperature was manipulated in the experiment through a series of sequential temperature shifts. Beginning with a culture which had remained at 15 °C since collection, every four days a triplicate set of the most extreme, current temperature treatments were split with one fraction retained at its current treatment and one fraction shifted a temperature step outward (i.e. further towards the temperature extremes).  Only the cultures growing at the highest and lowest temperatures were split and transferred to new temperatures, but all other cultures were continually maintained in exponential growth phase. Cultures maintained at 15 °C throughout the duration of the experiment served as an acclimated control, and as a reference for the temporal consistency in acclimated rates. Including all temperature steps and extrema, the nine treatments included: 6, 8, 10, 12, 18, 22, 25, 28, 31, and the control at 15°C. It took 20 days to complete the individual temperature shifts at 4-day intervals to reach the most extreme temperature treatments.

Dedicated incubators were used for control (15°C, Model 2015 Low Temperature Incubator, VWR Scientific); 4, 6 (I-41LLVL, Percival Scientific); 8, 18, 22 (I-36LLVL, Percival Scientific); and 10 °C (Environ Air, Holman Engineering). Twenty-five, 28, and 31 °C treatments were accomplished with clear 10 L baths controlled by a coupled aquarium heater and thermostat (Fluval, Tru Temp), housed in an illuminated incubator. Light intensity was consistently controlled across temperature treatments.

Population and growth rate measurements

To quantify changes in cell size, population growth rate, and volumetric growth rate, the abundance and size (Equivalent Spherical Diameter, ESD) distribution of each culture were measured with a 100 µm aperture on a Beckman Coulter Multisizer 3 (Beckman Coulter, Brea California; Kim and Menden-Deuer 2013). The default bin size was the instrument standard of 0.2 µm. Each treatment set was measured daily for 15 days following the initial transfer to the target temperature. Experiments were terminated after 15 days because the objective of these experiments was to establish the response to relatively short-term temperature fluctuations.

Statistical analysis

Mean and standard deviation of size distribution and cell counts as measured by the Coulter Counter were estimated from the raw data by a gaussian distribution, fit with maximum likelihood estimation to the frequency of particles greater than 8 µm. Measurements that immediately followed culture inoculation and those from the growth were omitted. Cell volumes were estimated with equivalent spherical diameter (ESD) and spherical shape approximation. Cell biomass was calculated using ESD measured with the coulter counter and calculated cell volume with the relationship of pg C cell^-1=0.288x volume^0.811 (Menden-Deuer and Lessard 2000). Total biomass was approximate using a culture and time specific mean ESD and cell count.

Specific growth rate was calculated by linear regression of natural log transformed abundance. Production rates were calculated by linear regression of natural log transformed total estimated biomass for each culture. Specific growth rates were used to fit a thermal reaction norm modified from Norberg (2004), where a, b, and z are shape parameters, w the thermal niche width, T  a given temperature, and k(T) the specific growth rate at that temperature.

k(T) = a * e^^bT [1-(T-z)/w/2)^²]

To quantify the effect of time (acclimation) on growth rate the initial growth rates, measured in the first three days (µ₀; Δ time<3 days), as well as final rates (µ_f, 7< Δ time<15 days) were calculated for each temperature treatment.

These experiments were conducted at the University of Rhode Island, Graduate School of Oceanography in the laboratory of Dr. Susanne Menden-Deuer