Dataset: Disko Bay Environmental and Biological Observations

Sampling Methodology: Between April 20th and May 11th 2011, 27 dilution experiments were conducted at a 250 m deep coastal site, 1 nautical mile south of Disko Island, Western Greenland (N 69 11, W 53 14); (see Figure 1 in Levinsen et al. 2000a). Source water was collected every 2-3 days. Water column profiles of temperature, salinity, in situ PAR and Chl a fluorescence were acquired with a SBE19plus CTD. Water samples were collected with Niskin bottles from the fluorescence maximum at depths ranging between 15 and 40 m, transferred into to 20L polycarbonate carboys using submerged silicone tubing and stored in the dark for transport to the laboratory. 

Experimental design: Phytoplankton growth rates and herbivorous grazer-induced mortality rates were measured using the dilution method (Landry and Hasset, 1982) in a two-point modification using whole seawater (WSW) and a diluted fraction containing 10% WSW.  Triplicate 1.8L bottles of both WSW and 10% WSW were incubated for 24 hours in laboratory vans under cool fluorescence light with a 20:4 light-dark cycle. All bottles were placed inside neutral density mesh screen bags to simulate the light level at sampling depth (10-15 µmol photons m-2 s-1) and were manually rotated every 4 hours.
Data treatment for the dilution experiment based rate estimates follow procedures outlined in Morison & Menden-Deuer (2017). Briefly, phytoplankton growth rates measured in the 10% WSW were considered to be a reasonable estimate for the instantaneous growth rates unaffected by grazing. Thus, phytoplankton growth rate (µ,d-1) was calculated as µ = 1/t * ln(Ct/C0), with Ct and C0 the final and initial Chlorophyll a concentration respectively and t the time elapsed in days. Herbivorous grazing rate (g, d-1) was calculated as the difference between µ measured in the highly diluted (µ10%) and WSW (µWSW) sample: g = µ10% - µWSW. 
Chlorophyll a was extracted from triplicate subsamples collected when bottles were filled initially and in triplicate from each of the triplicate dilution bottles after 24 hours. In addition, the size structure of the initial phytoplankton community was characterized from triplicate size-fractionated Chl a samples (>0.7 GF/F and >20 µm).  The extraction method followed Graff & Rynearson (2011) with the exception of the use of 95% ethanol as a solvent (Jespersen & Christoffersen 1987). The volume filtered ranged from 50 to 200 mL depending on phytoplankton abundance and dilution. 

Temperature treatments: Samples were incubated at 3 temperature treatments: in situ (0ºC), +3C and +6C over ambient. Water temperature for the 3 treatments was maintained as follows: the in situ treatment temperature was maintained through addition of snow to the incubation basin and was 0C (±0.0C). The +3C treatment was left to equilibrate with the ambient walk-in incubator air temperature and was 3.9C on average (±0.2C), and the +6C treatment temperature was maintained by a flow through water bath and was 6.0C on average (±0.2C). Incubation bottles were not acclimated to prior temperature. Thus, the transfer from in situ to incubation temperature could have induced a temperature dependent shock-response in the plankton communities and affected the rates measured.  Although acclimation could have reduced the potential shock induced by the target temperature, delay in commencement of the experiments would have extended the incubation duration and thus could have altered the species composition and nutrient concentrations. The experimental results from the temperature manipulation treatments should be viewed cautiously in light of the impossibility of acclimating whole communities to rapid changes in environmental conditions in a manner that preserves the integrity of the sampled community and minimizes incubation effects, including nutrient limitation and lack of immigration/emigration (see discussion and Grear et al. 2017).