Biotic Parameters

Chlorophyll a was extracted and measured following the Jespersen and Christoffersen (1987) protocol. Briefly, 500 ml seawater was filtered on (A/E) glass fiber filters (∼1 μm pore size) under low vacuum and extracted in ethanol (96%) in darkness. These filters catch single-celled cyanobacteria down to the size of Synechococcus that are abundant in the Baltic Sea; Prochlorococcus are not found in the Baltic Sea and Cyanobium only at times account for minor portions of the Chl a (Komárek et al., 1999; Haverkamp et al., 2009; Bertos-Fortis et al., 2016; Celepli et al., 2017). Chl a concentrations were measured using a Turner fluorometer, average values of technical duplicates are presented. Samples for phytoplankton community composition were collected and counted microscopically after preservation with acid Lugol’s solution following (Utermöhl, 1931, 1958; Legrand et al., 2015). Briefly, 500 ml sample for phytoplankton abundance were fixed using 2% Lugol’s solution and kept in darkness until processing. Two to ten milliliters of sample were subsequently transferred into sedimentation chambers and counted using an Olympus CK X41 microscope. For each sample, a minimum of 300 cells were counted and identified to genus or species level when possible. Phytoplankton biomass was calculated based on biovolume (Olenina et al., 2006) and carbon content (Edler, 1979).

Bacterial abundance was preserved in duplicates with formaldehyde as described in Lindh et al. (2015) and enumerated in a flow cytometer (Facs Calibur Becton Dickinson in 2011–2012 and Partec Cube8 in 2013–2014) as described in Giorgio et al. (1996). Bulk bacterial abundance samples were stained with SYTO13 (Life Technologies) during 2011–2012 and SYBR Green during 2013–2014 (Life Technologies). Briefly, cells were thawed in darkness at room temperature, stained using SYTO13 or SYBR Green (5 μM final concentration) and incubated in darkness for 15 min, before enumeration. Averages of technical duplicates are presented.

For bacterial heterotrophic production estimates, we diluted commercial ³H-leucine (Perkin Elmer; 1 mCi ml^-1) to 1 μM using cold leucine (Gasol et al., 1998). Then the samples were incubated with lukewarm ³H-leucine (40 nM final concentration) for 2 h at approximate in situ temperatures in at least triplicates with one killed control [trichloric acid (TCA); 5% final concentration; Sigma-Aldrich] following the leucine incorporation and centrifugation protocol described in Smith and Azam (1992). We assumed a conversion factor of 0.86 for the transformation from cellular carbon to protein, a factor of 0.073% leucine in total proteins and an isotope dilution of 2 according to Simon and Azam (1989) as presented in previous LMO reports [see for example Lindh et al. (2015) and Baltar et al. (2016)]. Average values of technical replicates are presented.

Extracellular enzymatic activity and substrate uptake rate constants (K) were determined biweekly from March 2012 to December 2013, and monthly during 2014. Extracellular enzymatic activities; β-glucosidase, leucine aminopeptidase (LAPase), and alkaline phosphatase (APase) were determined in technical quadruplicates according to the fluorometric enzyme assays and conditions described in Baltar et al. (2016). Substrate uptake rate constants (K) were determined in technical triplicate 10 or 30 ml samples with one formaldehyde killed control (Sigma-Aldrich; 1.4% final concentration) using ³H labeled L-leucine and glucose (final concentrations 0.5 and 1.0 nM, respectively) and ¹⁴C labeled L-amino acid mix (dissolved free amino acids, DFAA), acetate, and pyruvate (final concentrations 1.0, 10, and 10 nM, respectively, PerkinElmer). Samples were incubated in the dark at in situ temperature for 1 h (a linear relationship between K and incubation time ranging between 30 min and 3 h was determined) and killed by adding formaldehyde (Sigma-Aldrich; 1.4% final concentration) 10 min prior to filtration through 0.2 μm polycarbonate filters (25 mm diameter, GVS Life Sciences). Filters were rinsed with 2 ml 0.9 M NaCl three times and placed in scintillation vials. Three milliliters of Ultima Gold^TM XR liquid scintillation cocktail (Sigma) was added, samples were incubated in the dark for 18 h and counted in a liquid scintillation counter (Wallac WinSpectral 1414). Disintegrations per minute (DPM) of negative controls were subtracted from sample mean DPM and K was calculated for mean DPM using Equation (1) according to a first-order reaction type, where A = total radioactivity added (DPM ml^-1), a = incorporated radioactivity (DPM ml^-1), and t = time (h). Hence, K illustrates the substrate uptake rate constant of a specific substrate over time.

To investigate the potential of limiting nutrients (C glucose, N ammonium, P phosphate) for bacterial growth, we performed nutrient limitation assays. Bacterial nutrient limitation was determined by aliquoting 250 ml of seawater to acid washed and Milli-Q rinsed polycarbonate bottles and adding 24 μM carbon (C) as glucose, 4.2 μM nitrogen (N) as ammonium (NH₄Cl), and/or 0.6 μM phosphorus (P) as phosphate (NaH₂PO₄). As in previous experiments (Pinhassi et al., 2006), these additions were chosen since heterotrophic bacteria typically have lower C:N:P ratios than phytoplankton, e.g., 45:10:1 for bacteria in the Bothnian Sea (Zweifel et al., 1993) or see ratios reported by Fagerbakke et al. (1996) from a broad variety of environmental and laboratory settings (Fagerbakke et al., 1996). Nutrients were added in duplicate (C, N, P, and CP) or single (CN, NP, and CNP) treatments compared to control treatments (K) and incubated in the dark for 24 h at 16°C 2012 until summer 2013, after summer 2013 bottles were incubated at approximate in situ temperatures. After 24 h, differential responses to nutrient addition were determined by measuring bacterial heterotrophic production as described above. Optimally, these nutrient enrichment bioassays were done in the dark as to measure the short-term response of the heterotrophic bacteria to the specific experimental changes in nutrient availability. However, in reality, multiple biological activities that are associated with diurnal variations in light are likely to influence the heterotrophic bacterial production also in such dark incubations, like changes in production/consumption of DOC by phytoplankton or mixotrophic flagellates. The precise influence of such factors on measures of short-term bacterial activities remains unknown, but the minor changes in the controls compared to the cases where enrichments have an effect indicates that, overall, the bioassay approach provides interpretable results. We consider a response in bacterial heterotrophic production to nutrient additions, in comparison to control incubations, as a proxy for growth limitation and the results are presented as bacterial heterotrophic production subtracted by the bacterial heterotrophic production of the control treatment.