New observations of the distribution, morphology and dissolution dynamics of cryogenic gypsum in the Arctic Ocean

To date, observations on a single location indicate that cryogenic gypsum (Ca[SO4] q2H2O) may constitute an efficient but hitherto overlooked ballasting mineral enhancing the efficiency of the biological carbon pump in the Arctic Ocean. In June–July 2017 we sampled cryogenic gypsum under pack ice in the Nansen Basin north of Svalbard using a plankton net mounted on a remotely operated vehicle (ROVnet). Cryogenic gypsum crystals were present at all sampled stations, which suggested a persisting cryogenic gypsum release from melting sea ice throughout the investigated area. This was supported by a sea ice backtracking model, indicating that gypsum release was not related to a specific region of sea ice formation. The observed cryogenic gypsum crystals exhibited a large variability in morphology and size, with the largest crystals exceeding a length of 1 cm. Preservation, temperature and pressure laboratory studies revealed that gypsum dissolution rates accelerated with increasing temperature and pressure, ranging from 6 % d−1 by mass in polar surface water (−0.5 C) to 81 % d−1 by mass in Atlantic Water (2.5 C at 65 bar). When testing the preservation of gypsum in formaldehyde-fixed samples, we observed immediate dissolution. Dissolution at warmer temperatures and through inappropriate preservation media may thus explain why cryogenic gypsum was not observed in scientific samples previously. Direct measurements of gypsum crystal sinking velocities ranged between 200 and 7000 m d−1, suggesting that gypsum-loaded marine aggregates could rapidly sink from the surface to abyssal depths, supporting the hypothesized potential of gypsum as a ballasting mineral in the Arctic Ocean.


Introduction 36
Climate change in the Arctic Ocean has led to a drastic reduction of summer sea ice extent as 37 well as to a significant thinning of the sea ice (Kwok, 2018;Kwok and Rothrock, 2009). Sea 38 ice strength has reduced, and increased deformation and fractionation result in a progressively 39 increasing sea ice drift speed (Docquier et al., 2017) and sea ice export. Over the past decades 40 the ice export via the Fram Strait alone has increased by 6% and 11% per decade as annual 41 mean, and during the productive spring and summer period, respectively (Smedsrud et al., 42 2017). An increasing amount of sea ice produced in the East Siberian and Laptev Sea melts 43 over the adjacent continental slopes or in the central Arctic Ocean (Krumpen et al., 2019). 44 Overall, the Arctic Ocean sea ice cover has shifted to a predominantly seasonal ice cover. 45 However, although the majority of sea ice diminishes during late summer, the amount of sea 46 ice produced in autumn to winter progressively increases (Kwok, 2018). from melting sea ice (Wollenburg et al., 2018a). This single event was the first and only 59 20°C (Peeken, 2018). One ice-core from station 80 and four bottom slices (10 cm) of ice-161 cores from station 45 were studied to investigate the gypsum crystal morphologies within sea 162 ice. Each section was transferred into a measuring jug with lukewarm tap water for approx. 163 two seconds, and then the jug was emptied over a 32 µm analysis sieve, and repeatedly 164 refilled. This process was continued until all ice was melted. With the aid of a hand shower 165 and a wash bottle the residue on the sieve was rinsed and transferred into a 30 µm mesh-166 covered funnel, dried and transferred into a micropaleontological picking tray for inspection 167 and documentation. For storage, the residue was transferred into pre-weighed labelled 168 micropaleontological slides. 169

Dissolution experiments 170
The aim of our dissolution experiments was to investigate the persistence of gypsum crystals 171 against dissolution in the Arctic water column (water mass trials) and under common 172 biological sample treatment (Formaldehyde trial). 173 Dissolution experiments were carried out on individual gypsum crystals collected from 174 ROVnet samples. Hereby, 5 cryogenic gypsum crystals with different crystal morphologies, 175 and from both size fractions were used in each reaction chamber. Before the start and after the 176 termination of each experiment, pictures of the cryogenic gypsum crystals used were taken 177 with an Axiocam 506 colour camera under a Zeiss Axio Zoom V16 microscope. The weight 178 of the crystals before and after each treatment was determined with a high-precision Sartorius 179 SE2 ultra-microbalance after they had been transferred into a pre-weighted silver boat. The 180 experimental running time of each experiment was 24 hours. 181

Water mass trials 182
The experiments to simulate dissolution within the different water masses and hydrostatic 183 pressure regimes of the Arctic Ocean were carried out with high-pressure chambers installed 184 in a cooling table (Wollenburg et al., 2018b). With a high-pressure pump (ProStar218 Agilent 185 Technologies), peak tubing, and multiple titanium valves a continuous isobaric and isocratic 186 one-way seawater flow of 0.3 ml/min was directed through a set of four serially arranged 187 high-pressure chambers each with an internal volume of 0.258 ml (Wollenburg et al., 2018b). per L and psu-offset. The natural pH of 8.1 after equilibration to the refrigerator's atmosphere 192 (at 2.5 °C and at atmospheric pressure), lowers to pH 8.05 at 2.5 °C at 150 bar (Culberson and 193 Pytkowicx, 1968). Five experiments, with 4 high-pressure chambers were carried out. The 194 Polar Surface (PSW) water corresponding experimental trial was running at -0.5 °C and 3 bar, 195 the experimental Atlantic Water (AW) trial at +2.5 °C and 65 bar, and three experimental 196 Deep Water trials were conducted at -1 °C and 100, 120 and 150 bar, respectively. 197

Formaldehyde trial 198
To study the effect of Formaldehyde treatment on cryogenic gypsum, the crystals were 199 subjected to a Formaldehyde solution of 4% in seawater, which is commonly used to preserve 200 biological samples. The stock solution consisted of 500 ml Formaldehyde concentration of 201 40%, 500 ml aqua dest. and 100 g hexamethylenetetramine, adjusted to a pH of 7.3-7.9. 202 Aliquots of the 20% stock solution were added to the four-fold volume of artificial Arctic 203 Ocean sea water to obtain a final concentration of 4%. 204 The Gypsum crystals were transferred into Falcon Tubes, and the 4% Formaldehyde solution 205 was added. The Falcon tubes were then either stored at 3 °C, or at room temperature. After 206 the experiments, the gypsum crystal-Formaldehyde suspension was washed with deionized 207 water over a 10 µm mesh using a wash bottle, and dried on gauze. As in all formaldehyde 208 trials all gypsum dissolved, no post-experimental weight was determined. The size-specific sinking velocity of cryogenic gypsum was measured in a settling cylinder 213 (Ploug et al., 2008). The cylinder (30 cm high and 5 cm in diameter) was filled with filtered 214 seawater (salinity 32) and surrounded by a water jacket for thermal stabilization at 2 °C. The 215 settling cylinder was closed at both ends, only allowing insertion of a wide-bore pipette at the 216 top. Immediately before measurement, the gypsum was submerged into seawater with a 217 salinity of 32 and a temperature of 2 °C, and then transferred to the settling cylinder with a 218 wide-bore pipette. The gypsum crystals were allowed to sink out of the wide-bore pipette, 219 which was centered in the cylinder. The descent of the pellets was recorded by a Basler 4 220 MPixel Ethernet camera equipped with a 25 mm fixed focal lens (Edmund Optics). The where CD is the dimensionless drag force (equation 2), ρw is the density of seawater (1.0256 g 235 cm -3 , for a salinity of 32 at 2 °C), SV is the measured sinking velocity in cm s -1 , g is the 236 gravitational acceleration of 981 cm s -2 , and ESD is the equivalent spherical diameter in cm. 237 We calculated CD using the drag equation for low Reynolds numbers (White, 1974): where η is the dynamic viscosity (1.7545 × 10 -2 g cm -1 s -1 for a salinity of 32 at 2 °C). Equation 246 2 is valid up to a Reynolds number of 2x10^5 (Vogel and Beety, 1994). The gypsum crystals 247 had Reynolds numbers ranging from 0.77 to 128. The tracking approach works as follows: An ice parcel is traced backward or forward in time 267 on a daily basis. Tracking is stopped if a) ice hits the coastline or fast ice edge, or b) ice 268 concentration at a specific location drops below 50% and we assume the ice to be formed. 269 The applied sea ice concentration product was provided by CERSAT and was based on 270 85 GHz SSM/I brightness temperatures, using the ARTIST Sea Ice (ASI) algorithm. particles, as often found in sea ice (Nürnberg et al., 1994), were essentially absent. for the >63 and >30<63 µm size fraction, respectively. At station 45 the crystal length-width 303 ratio varied between 1.37 and 1.98, measured in the >30<63 µm size fraction of the surface 304 sample, and the >63 µm size fraction of the 10 m sample. The cryogenic gypsum crystals 305 retrieved from the melted ice core drilled at this station were solid and hyaline. In size and 306 shape they resembled the crystals of the 10 m layer at this station, with a mean crystal length 307 of 114.2 µm, mean width of 57.2 µm, and a length-width ratio of 2 (Fig. 4). 308 At station 66, the crystals from 0 m water depth were dominated by large, pencil-like, hyaline 309 and solid crystals with a mean crystal length of 1,355 µm and mean width of 415 µm in the 310 dominating >63 µm fraction (99.25% mass) (Fig. 2B, S3, Tab. 2). These crystals with an 311 average length-width ratio of 3.27 were found as isolated crystals, but very often also as inter-312 grown crystal rosettes with two to more than 10 individual crystals involved ( Fig. S3; Tab. 2). 313 The >30<63 µm size fraction (0.75% mass) was dominated by matte, whitish, rounded 314 gypsum particles and tiny gypsum needles with a mean crystal length of 56.67 µm (Fig. S3,

Experiments to simulate cryogenic gypsum dissolution within Formaldehyde-364 treated biological samples 365
In the Formaldehyde experiments we exposed our set of cryogenic gypsum crystals to a 366 Formaldehyde solution of 4%, which is commonly used to store pelagic samples from the 367 Polar Oceans (Edler, 1979). Irrespective of the temperature at which the sample was stored, 368 all gypsum dissolved within 24 hours. 369

Sinking velocities of gypsum crystals 370 371
The sinking velocity (SV) of the gypsum crystals increased with crystal size (Fig. 7A). Small 372 crystals with an equivalent spherical diameter (ESD) of 200 µm sank with 300 m d -1 while 373 large gypsum crystals with ESDs of 2,000 to 2,500 µm sank with velocities of 5,000 to 7,000 374 m d -1 . The size to settling relationship was best described by a power function (SV = 4239.9 375 ESD 0.839 , R 2 = 0.84). As the power function suggests, the settling velocity levelled off for the 376 largest gypsum crystals (Fig. 7A). The observed excess density of all crystals was smaller 377 than is expected from the density of gypsum (2310 kg/m 3 ). For the visually non porous 378 smaller crystals drag, the deviation of gypsum crystals from round particles, and dissolution 379 may be the main reason for the calculated lower density. 380 However, plotting the excess density as a function of size (Fig. 7B) also showed that the 381 excess density of the gypsum decreased with increasing crystal size. The microscopic images 382 show that large crystals were more porous and had more complex shapes (Fig. S8 A- compared to the small crystals that were more spherical and less porous (Figs. 2, 4-5, S8 D). 384 Hence, the flat settling to size relationship for large gypsum crystals (Fig. 7A), was essentially 385 due to a combination of increased porosity causing decreasing excess density and increased 386 drag due to the complex shapes of the large crystals. However, here we show that gypsum crystals exhibit a strong variability in size and 406 morphology. Particularly large crystals were characterised by more complex shapes (Fig. 2, 5, 407 S3-4) and increased porosity (Figs. S6A-C), compared to the small planar euhedral ( Fig. 2A) 408 and more spherical crystals (Fig. S6D). Euhedral crystal needles larger but otherwise similar to The sea ice microstructure dictating the formation of gypsum crystals in the brine matrix 428 likely varied among ice-floes due to different ages, origins and drift trajectories (Fig. 1B). For 429 example, station 66 was the only station where the sea ice likely formed over the central 430 Nansen Basin only months before our study (Fig. 1B). The surface sample of station 66 had 431 large intergrown hyaline star-shaped gypsum crystals that were observed at no other station. 432 They also showed a considerably higher length-width ratio than crystals from second-year ice 433 of stations 32/80 and 45 ( Fig. 1B; Fig. 2). Accordingly, a close relationship between local sea 434 ice properties and gypsum crystal morphology in the underlying water was evident from the 435 comparison of gypsum crystals collected with the ROVnet with those retrieved from ice cores 436 The small temperature range of the -6.2 to -8.5 °C window, which is also the only gypsum 478 precipitation temperature spectrum applicable in the Arctic Ocean, has been considered one other sea ice precipitates that are quantitatively much more abundant, leading the focus 483 towards other sea ice precipitates (Butler and Kennedy, 2015;Geilfus et al., 2013). Although 484 cryogenic mirabilite and hydrohalite are three and twenty-two times more abundant than 485 gypsum, respectively (Butler and Kennedy, 2015), gypsum is the only sea ice precipitate that 486 survives for one to several days within the Arctic water column. Cryogenic gypsum 487 dissolution increases with increasing hydrostatic pressure and increasing temperatures (Fig.  488   6). However, well preserved cryogenic gypsum crystals were retrieved from algae aggregates 489 collected from 2,146 m water depth, suggesting that either the transport from the surface to 490 this depth was very rapid or that dissolution was decreased and/or prevented once gypsum 491 crystals were included within the matrix of organosulfur compound-rich aggregates 492 (Wollenburg et al., 2018a). Yet, as seawater is usually undersaturated with respect to gypsum 493 (Briskin and Schreiber, 1978a;Briskin and Schreiber, 1978b