Philipp Assmy

Paul A. Dodd

Pedro Duarte

Jens Dujardin

Ashley Elliott

Mar Fernández-Méndez

Agneta Fransson

Mats A. Granskog

Kate Hendry

Robert Hodgson

Hanna Kauko

Svein Kristiansen

Melanie J. Leng

Amelie Meyer

CJ Mundy

Lasse Mork Olsen

Ilka Peeken

Marthe Sandbu

Jago Wallenschus

Anette Wold

The refrozen lead was sampled on five locations along a transect from the edge towards the centre of the lead. The transect was repeated 13 times between 7 May and 3 June 2015.

Data include:

- ice and snow thickness

- bulk salinity

- chlorophyll a concentration and standing stock

- MAA (mycosporine-like amino acids) concentrations

- HPLC algal pigment concentrations (added in version 2 - see column header explanations in methods below)

In version 3 added:

- particulate organic carbon (POC) and nitrogen (PON)

- d18O and barium data (mainly one day available)

In addition, sample ID numbers for taxonomy, particulate and dissolved matter (CDOM) absorption (to facilitate linking to other data sets).

See [Kauko et al. 2017]( http://doi.org/10.1002/2016JG003626) and [Kauko et al. 2018]( https://doi.org/10.3389/fmars.2018.00199) for more sampling details.

The thick ice dataset contains results from second-year and first-year sea ice samples collected during the N-ICE2015 cruise with ice corers with 9 or 14 cm diameter (Mark II coring system, KOVACS enterprise, Roseburg, USA).

Data include:

- ice and snow thickness

- bulk salinity

- inorganic nutrients (nitrate plus nitrite, phosphate and silicate)

- chlorophyll a concentrations

In version 5 added:

- HPLC algal pigment concentrations

- MAA (mycosporine-like amino acids) concentrations

- particulate organic carbon (POC) and nitrogen (PON)

- biogenic silica (BSi)

- dissolved organic carbon (DOC) and total dissolved nitrogen (TDN)

- salinity values measured with a salinometer

- d18O

- barium

In addition, sample ID numbers for taxonomy, particulate and dissolved matter (CDOM) absorption, gases, and dissolved inorganic carbon (DIC) (to facilitate linking to other data sets).

IMPORTANT NOTE: The data that has been added in version 5 for thick ice or version 3 for the refrozen lead have not been extensively worked on (except tracer data) and therefore the user is advised to carefully control these data before using them.

Large part of the bulk nutrient concentrations were very low and within measurement uncertainty (see methods below) and are marked with 'wmu' in the dataset. Brine nutrient concentrations were not measured.

Samples were collected from January until June 2015. The data file contains information on the date, hour, latitude, longitude and ice depth of each sample.

Two types of ice cores were collected for sampling: biological ice cores and chemical/tracer ice cores. In this dataset (version 5), the chemistry/tracer ice cores are included that contained data on nutrients on the days/coring sites where biological ice cores were taken (the dataset includes data on all variables from those ice cores). In other words, more chemistry/tracer ice core data exist (especially later in the campaign, two-three chemistry/tracer cores were taken per sampling event, where e.g. nutrients and some other chemistry/tracer samples were taken from different ice cores). These data will be added to this dataset in a later version. In the meanwhile, contact the authors Paul Dodd (lab salinity and d18O), Kate Hendry (barium), Agneta Fransson (gases and DIC) and Mats Granskog (d18O) for these data. For most sampling days and coring events, all ice core types exist and it is possible to gather information for most variables.

See [Olsen et al. (2017)]( https://doi.org/10.1002/2016JG003668) for more sampling details. See also [Granskog et al. 2017]( https://doi.org/10.1002/2016JC012398) and [Hendry et al. 2018]( https://doi.org/10.1029/2017JC013668).

#### Methods

Bulk salinity:

Salinity in the melted ice core samples was measured with WTW Cond 3110 probe (WTW Wissenschaftlich-Technische Werkstätten GmbH, Weilheim, Germany)

Chlorophyll a:

Chl a samples were collected on 25 mm GF/F filters (Whatman), extracted in 100% methanol for 12 h at 5°C on board the ship and measured fluorometrically with an AU10 Turner Fluorometer (Turner Design, Inc.). Phaeopigments were measured by fluorescence after acidification with 5% HCl. Calibration of the Turner fluorometer was carried out following the JGOFS protocol35 (Knap et al., 1996). Unit mg/m3 (for the refrozen lead in addition mg/m2).

MAAs:

The MAA method is described in [Elliott et al. (2015)]( https://doi.org/10.3354/meps11540) and based on the method by [Carreto et al. (2005)]( https://doi.org/10.1007/s00227-004-1447-y). In the thick ice file the unit is mg/L and in the refrozen lead file the unit is mg/m3.

HPLC:

The HPLC method is described in [Tran et al. (2013)]( https://www.biogeosciences.net/10/1909/2013/bg-10-1909-2013.pdf). In the thick ice file the unit is ng/L and in the refrozen lead file the unit is mg/m3. Header definitions: Chla - chlorophyll a; Chl b - chlorophyll b; Chlc1_Chlc2 - chlorophyll c1 + chlorophyll c2; Chlc3 - chlorophyll c3; MgDVP - magnesium 2,4-divinylpheoporphyrin a5 monomethyl ester; Chla_ide - chlorophyllide a; Phide_a - pheophorbide a; Phytin_a - pheophytin a; Phytin_b - pheophytin b; Fucox - fucoxanthin; 19But - 19’butanoyloxyfucoxanthin; 19Hex - 19’hexanoyloxyfucoxanthin; Perid - peridinin; Dinox - dinoxanthin; Gyroxdiester - gyroxanthin diester; Neox - neoxanthin; Lutein - lutein; Prasino - prasinoxanthin; Antherax - antheraxanthin; Allox - alloxanthin; Violax - violaxanthin; Zeax - zeaxanthin; DD - diadinoxanthin; DT - diatoxanthin; Astax - astaxanthin; Echineone - echineone; Alpha_car - alpha carotene; Beta_car - beta carotene

Inorganic nutrients:

Sections of the ice cores were melted in cleaned plastic containers with lids. Samples for the inorganic nutrients nitrate and nitrite (NO3 + NO2), phosphate and silicate were were collected in 20 mL scintillation vials,fixed with 0.2 ml chloroform and stored refrigerated until sample analysis approximately 6 months later. Concentrations of these nutrients were measured on a modified Scalar auto-analyzer following Bendschneider & Robinson (1952) and RFA Methodology, for NO3+NO2, and Grasshoff (1965), for PO4 and SiOH4. Unit µmol/L.

Measurement of nitrite in seawater The method is based on that nitrite reacts colorimetrically with aromatic amin and forms a diazonium ion in acidic medium. The diazonium ion connects to a new aromatic amine and forms an azo. The absorption of the color is measured spectrophotometrically at 540 nm on a modified Alpkem Flow Solution IV autoanalyzer or a modified scalar autoanalyzer (Bendschneider & Robinson 1952 RFA Methodology). The detection limit for nitrites is 0.06 µmol/L (Scalar) and is significant to the second decimal place.

Measurement of nitrate in seawater The method is based on the chemical reduction of nitrate to nitrite by means of cadmium in the presence of copper ions. Then nitrite forms a diazonium ion with an aromatic amine in an acidic environment, which is diverted to a new aromatic amine and an azo dye is formed. Absorption of the color is measured spectrophotometrically at 540 nm on a modified Alpkem Flow Solution IV autoanalyser or a modified scalar autoanalyser (Bendschneider & Robinson 1952 RFA Methodology). The amount of nitrate is calculated as the difference between total reduced nitrite (this method) and NO2 measured without cadmium reduction (see Method U3_3 above). The detection limit for nitrate is 0.4 µmol/L (Scalar) and is significant to the first decimal place.

Measurement of phosphate in seawater The method is based on the phosphate reacts with molybdate to form phosphomolybdate in acid medium (pH < 1) providing a yellow dye. This dye is reduced by ascorbic acid to a blue dye, and absorbance is measured spectrophotometrically at 810 nm on a modified Alpkem Flow Solution IV autoanalyzer or a modified Scalar autoanalyzer (Grashoff 1965). The detection limit for phosphate is 0.06 µmol/L (Scalar) and is significant to the second decimal place.

Measurements of silicate in seawater The method is based on the silicate reacts with molybdate to form a silikomolybdat (yellow dye) in an acidic environment (pH = 1.5-2). Oxalic acid is added to the silikomolybdat which then is reduced with ascorbic acid. The result is a blue compound which is measured spectrophotometrically at 810 nm on a modified Alpkem Flow Solution IV autoanalyser or a modified Scalar autoanalyzer (Grashoff 1965). The detection limit for silicate is 0.7 µmol/L (Scalar) and is significant to the first decimal place.

Detection limits for nutrients:

* Nitrate detection limit 0.4 µmol/L

* Nitrite detection limit 0.06 µmol/L

* Phosphate detection limit 0.06 µmol/L

* Silicate detection limit 0.7 µmol/L

For POC (µg/L), PON (µg/L), DOC (µM), TDN (µM), and BSi (µmol/L) methods see the corresponding [water column dataset]( https://doi.org/10.21334/npolar.2016.3ebb7f64).

For lab salinity, d18O (‰) and barium (nM) methods, see the files [in this dataset]( https://doi.org/10.21334/npolar.2016.516bc529) and articles [Granskog et al. 2017]( https://doi.org/10.1002/2016JC012398) and [Hendry et al. 2018]( https://doi.org/10.1029/2017JC013668).

#### Related datasets

Physical data from the same coring cites [can be found here]( https://doi.org/10.21334/npolar.2017.c3db82e3).

Water column biogeochemistry is [here]( https://doi.org/10.21334/npolar.2016.3ebb7f64).

Phytoplankton and ice algal [abundance here]( https://doi.org/10.21334/npolar.2017.dc61cb24).

Particle absorption coefficients [here]( https://doi.org/10.21334/npolar.2017.35978199).

CDOM absorption coefficients [here]( https://doi.org/10.21334/npolar.2017.f46970ba).

CEOS IDN DIF

9.8.4