2.3. Sample analysis
To measure ice salinity, water δ18O composition, and nutrient and chl.a concentrations, the second half of the core was trimmed into a rectangle with a cross-section of 4.5 cm × 2.5 cm and then sliced into 12–18-cm-thick sections. To avoid contamination during sampling and handling, the outer 0.3 cm of each ice section was removed with a stainless steel plane. The ice sample was then put in a polyethylene bag (GL Science Inc., Japan) and melted in a cold room at +4℃. Snow samples were melted in the same bags and in the same cold room. Snow and sea ice melt samples were analyzed with the same protocol as for under-ice water samples, except for nutrient analyses; in snow and sea ice melt samples, nutrient samples were filtered with a 0.22-μm Durapore PVDF Membrane filter (MILLEX GV Filter unit, Merck Millipore Ltd., Germany). For ice algae community assemblage, sea ice melt water for the bottom ice section for JARE60 KU1 and 2 was subsampled into 100-mL polyethylene screw-cap vial (AS ONE Corporation, Tokyo, Japan), and an aliquot of 5-mL of Lugol’s solution was added to preserve the sample.
The salinities of under-ice water and melted snow and sea-ice samples were measured with a conductivity sensor (Cond 315i, WTW GmbH, Germany). The oxygen isotopic ratio of under-ice water and melted snow and sea-ice samples were determined by mass spectrometry (DELTA plus; Finnigan MAT, San Jose, CA) via the equilibration method. The δ18O value (‰) of a sample was calculated against the18O/16O ratio of standard mean ocean water. The standard deviation (SD) on δ18O values was 0.026‰ based on analyses of 10 subsamples of a reference water with δ18O = 0.241‰. Nutrient concentrations (NO3 + NO2, NH4+, PO43−, and Si(OH)4) were measured with an auto-analyzer system (Quattro; Bran+Luebbe, Norderstedt, Germany) according to the spectrophotometric method reported by the Joint Global Ocean Flux Study (JGOFS, 1994). SDs on nutrient concentrations were 0.53, 0.18, 0.07, and 1.1 μmol L−1 for NO3 + NO2, NH4+, PO43−, and Si(OH)4, respectively, calculated from 10 subsamples of a reference water sample with respective known concentrations of 52.53, 9.99, 5.00, and 117.5 μmol L−1. Chl.a concentrations were determined by a fluorometer (Model 10AU, Turner Designs, Inc., Sunnyvale, CA) using the method of Parsons et al. (1984). Standards (0.28–282.30 μg L−1 chl.a) prepared from a liquid chl.a standard (Wako Pure Chemical Industries Ltd., Osaka, Japan) by stepwise dilution withN ,N -dimethylformamide were used to calibrate the fluorometer before analyses. The cell counting for ice algae community assemblage was examined with a microscope (Olympus, BH-T, Tokyo, Japan) 10× oculars and 40× objective.