Comment on bg-2021-149 Anonymous Referee # 3 Referee comment on " Biogeochemical controls on wintertime ammonium accumulation in the surface layer of the Southern Ocean

Ammonium (NH4) is an important macronutrient in marine ecosystems and the dynamics of its production, utilisation, and regeneration are reasonably well studied within the marine microbial food web. However, how these dynamics play out in the Southern Ocean is not well understood and this is especially so during the winter months when conditions in this region are challenging due to large storms, low temperatures, limited light availability, and the presence of sea ice. In their paper, Smith et. al provide a detailed snapshot of NH4 concentration and dynamics (uptake and oxidation rates) in the surface water and winter mixed layer during a winter voyage in the Atlantic sector, bordering the Indian sector, of the Southern Ocean. To better understand these dynamics, the authors investigate links between macronutrient concentrations, microbial community composition and biomass, net primary production, particulate organic matter, and nitrogen isotopic fractionation. This is a substantial data set to both analyse and interpret and I commend the authors for their very thorough analysis of the data and its links to the available literature on this topic.


Abstract 17
The production and consumption of ammonium (NH₄⁺) are essential upper-ocean nitrogen 18 cycle pathways, yet in the Southern Ocean where NH₄⁺ has been observed to accumulate in 19 surface waters, its mixed-layer cycling remains poorly understood. For surface samples 20 collected between Cape Town and the marginal ice zone (MIZ) in winter 2017, we found that 21 NH₄⁺ concentrations were five-fold higher than is typical for summer, and lower north than 22 south of the Subantarctic Front (SAF; 0.01-0.26 µM versus 0.19-0.70 µM). Our observations 23 confirm that NH₄⁺ accumulates in the Southern Ocean's winter mixed layer, particularly in 24 polar waters. NH₄⁺ uptake rates were highest near the Polar Front (PF; 12.9 ± 0.4 nM day -1 ) and 25 in the Subantarctic Zone (10.0 ± 1.5 nM day -1 ), decreasing towards the MIZ (3.0 ± 0.8 nM day -26 1 ) despite high ambient NH₄⁺ concentrations, likely due to low sea surface temperatures and 27 light availability. By contrast, rates of NH₄⁺ oxidation were higher south than north of the PF 28 (16.0 ± 0.8 versus 11.1 ± 0.5 nM day -1 ), perhaps due to the lower light and higher iron 29 conditions characteristic of polar waters. Augmenting our dataset with NH₄⁺ concentration 30 measurements spanning the 2018/2019 annual cycle reveals that mixed-layer NH₄⁺ 31 accumulation south of the SAF likely derives from sustained heterotrophic NH₄⁺ production in 32 late summer through winter that outpaces NH₄⁺ consumption by temperature-, light, and iron-33 limited microorganisms. Our observations thus imply that the Southern Ocean becomes a 34 biological source of CO 2 to the atmosphere for half the year not only because nitrate drawdown 35 is weak, but also because the ambient conditions favour net heterotrophy and NH₄⁺ 36 accumulation. 37

Introduction 38
The Southern Ocean impacts the Earth system through its role in global thermohaline 39 circulation, which drives the exchange of heat and nutrients among ocean basins (Frolicher et  40  solubility pumps (Sarmiento & Orr, 1991;Volk & Hoffert, 1985) and through the release of 43 deep-ocean CO 2 to the atmosphere during deep-water ventilation (i.e., CO 2 leak; Broecker & 44 Peng, 1992; Lauderdale et al., 2013;Sarmiento & Toggweiler, 1984). Upper Southern Ocean 45 circulation is dominated by the eastward-flowing Antarctic Circumpolar Current (ACC) that 46 consists of a series of broad circumpolar bands ("zones") separated by oceanic fronts. Southern 47 Ocean fronts can drive water mass formation (Ito et al., 2010) and nutrient upwelling that 48 supports elevated biological activity (Longhurst, 1998;Sokolov & Rintoul, 2007). 49 Concentrations of the essential macronutrients, nitrate (NO 3 -) and phosphate (PO 4 3-), are 50 perennially high in Southern Ocean surface waters, in contrast to most of the global ocean. 51 Consumption of these nutrients, and thus primary productivity in the Southern Ocean, is limited 52 by numerous (often overlapping) factors, including temperature, light, micronutrient 53 concentrations, and grazing pressure (e.g., Boyd  replenishes the nutrients required for phytoplankton growth but the low temperatures and light 66 levels impede biological activity (Rintoul & Trull, 2001). Once the mixed layer shoals in spring 67 and summer, phytoplankton begin to consume the available nutrients until some form of 68 limitation (usually iron; Mtshali et al., 2019;Nelson et al., 2001) sets in. This balance between 69 wintertime nutrient recharge and summertime nutrient drawdown is central to the role of the 70 Southern Ocean in setting atmospheric CO 2 (Sarmiento & Toggweiler, 1984). 71 Iron limitation, which sets in following the spring/early summer bloom, causes phytoplankton 72 to increase their dependence on recycled ammonium (NH₄⁺; Timmermans et al., 1998), which 73 has a far lower iron requirement than NO 3 assimilation (Price et al., 1994). The extent to which 74 phytoplankton rely on NO 3 versus NH₄⁺ as their primary N source has implications for 75 Southern Ocean CO 2 removal since phytoplankton growth fuelled by upwelled NO 3 -( "new 76 production") must be balanced on an annual basis by the export of sinking organic matter 77 ("export production"; Dugdale & Goering, 1967), which drives CO 2 sequestration (i.e., the 78 biological pump; Volk & Hoffert, 1985). By contrast, phytoplankton growth on NH₄⁺ or other 79 recycled N forms ("regenerated production") yields no net removal of CO 2 to the deep ocean 80 (Dugdale & Goering, 1967;Eppley & Peterson, 1979). To-date, considerable research has 81 focused on NO 3 cycling in the Southern Ocean  of regenerated N within the seasonally-varying mixed layerincluding the production of NH₄⁺ 86 and its consumption by phytoplankton uptake and nitrification (the microbial oxidation of NH₄⁺ 87 to nitrite (NO 2 -) and then NO 3 -)remains poorly understood. 88 NH₄⁺ is produced in the euphotic zone as a by-product of heterotrophic metabolism (i.e., 89 ammonification; Herbert, 1999) and as a consequence of grazing by zooplankton (Lehette et al.,90 2012; Steinberg & Saba, 2008), and is removed by phytoplankton uptake (in euphotic waters) 91 and nitrification (mainly in aphotic waters). Heterotrophic bacteria can also directly consume 92 NH₄⁺ (Kirchman, 1994) and have been hypothesized to do so at significant rates in the Southern 93 Ocean mixed layer in winter (Cochlan, 2008;Mdutyana et al., 2020). NH₄⁺ assimilation by 94 phytoplankton, in contrast to NO 3 consumption, requires relatively little energy (Dortch, 1990) 95 such that NH₄⁺ is usually consumed in the surface ocean as rapidly as it is produced (Glibert,96 1982; La Roche, 1983) precision was ± 0.04 μM and the detection limit was 0.04 μM. 252

Chlorophyll-a concentrations 253
Chlorophyll-a concentrations ([chl-a]) were determined shipboard using the nonacidified 254 fluorometric method (Welschmeyer, 1994 The NPP and N uptake filters were fumed with hydrochloric acid in a desiccator for 24 hours to 263 remove inorganic C, then dried for 24 hours at 40°C and packaged in tin cups. Filters to be 264 measured for 15 N were dried in the same way as the NPP/N uptake filters, but not acidified. 265 Samples were analysed using a Delta V Plus isotope ratio mass spectrometer (IRMS) coupled 266 to a Flash 260 elemental analyser, with a detection limit of 0.17 μmol C and 0.07 μmol N and 267 precision of ±0.
Here, M is the species of interest (C, NH₄⁺, NO 3 -, or urea); ρM is the uptake rate of that species The specific carbon fixation rate (V C ) was calculated as ρC/POC and the specific uptake rate of 286 total N (V Ntot ) was calculated as ρN tot /PON (where ρN tot = ρNH 4 + + ρNO 3 -+ ρUrea). The f-ratio 287 (Eppley & Peterson, 1979), used to estimate the fraction of NPP potentially available for 288 export, was then calculated as: 289 No urea uptake experiments were conducted at the underway stations at 50.7ºS and 55.5ºS 291 (both AZ); here, the f-ratio was calculated omitting V urea . For the other AZ stations at which 292 urea uptake was measured, including V urea decreased the fraction of new-to-total production by 293 only 4-8% compared to f-ratio calculations based on V NO3 and V NH4 . 294

Ammonia oxidation rates 295
The azide method of McIlvin and Altabet (2005)  Microphytoplankton and microzooplankton groups (>5-10 μm) were identified and counted in 310 a subsample (20 mL) from each 250 mL amber bottle using the Utermöhl technique (Utermöhl,311 1958) and following the recommendations of Hasle (1978). Plankton groups and individual 312 species were counted and identified using an inverted light microscope (Olympus CKX41) at 313 200x magnification. 314 Cells were also enumerated using an LSR II flow cytometer (BD Biosciences) equipped with 315 blue, red, violet, and green lasers. Here, our focus was on enumerating pico-and nanoplankton. 316 Prior to flow cytometric analysis, 1 mL of each sample was incubated with 10 µL of 1% (v/v) 317 SYBR Green-I, which stains DNA, at room temperature in the dark for 10 minutes (Marie et  318 al  Since no direct measurements of NH₄⁺ regeneration (i.e., heterotrophy) were made in this study, 342 potential heterotrophic activity is evaluated from the abundance of heterotrophic cells 343 determined via flow cytometry and the ratio of bulk POC to PON concentrations (POC:PON). 344 The availability of organic matter to heterotrophs is estimated from the abundance of detritus 345 and the ratio of POC-to-chl-a concentrations (POC:chl-a; Holm-Hansen et al., 1989). 346 The correlations among latitude, N concentrations, inorganic carbon and N uptake rates, and 347 NH₄⁺ oxidation rates were investigated at the 5% significance level using the Pearson 348 correlation coefficient and the R packages, stats (R Core Team, 2020) and corrplot (Wei & 349 Simko, 2017). 350 (61.7°S) by ~17 °C (Fig. 1) The surface and mixed-layer concentrations of NH₄⁺ ranged from below detection to 0.70 µM 363 along legs S and N ( Fig. 2a and b). The concentrations were higher in the PFZ, OAZ, and PAZ 364 (0.42 ± 0.01 µM, 0.52 ± 0.01 µM, and 0.58 ± 0.01 µM, respectively) than in the Subtropical 365

Results
Zone (STZ) and SAZ (0.08 ± 0.03 µM and 0.06 ± 0.01 µM, respectively), with a sharp gradient 366 observed in the PFZ, just south of the SAF. South of the SAF, high NH₄⁺ concentrations 367 persisted near-homogeneously throughout the mixed layer, ranging from 0.65 ± 0.01 µM at 368 station 58.5°S to 0.27 ± 0.01 µM at station 48.0°S, with concentrations that were below 369 detection north of the SAF (Fig. 2b). Beneath the mixed layer, the NH₄⁺ concentration The proportion of chl-a in the >2.7 µm size class (hereafter, "nano+" size class) varied across 383 the region but was >50% at all stations, with higher (>80%) contributions near the fronts and at 384 many OAZ and PAZ stations (Fig. 3b). The nano+ contribution was ≤60% at only five stations 385 (three in the SAZ, two in the OAZ). 386 The concentrations of bulk POC and PON were highest north of the STF and slightly higher in 387 the OAZ than in the SAZ and PFZ ( Fig. S3a and b). The contribution of the nano+ size fraction 388 to POC and PON across the transect was 80.6 ± 31.8% and 69.8 ± 50.3%, respectively ( Fig.  389 S3c and d). The ratio of bulk POC:chl-a (weight:weight) was on average low in the STZ, SAZ, 390 and PFZ, and reached a maximum in the OAZ (Fig. 4a). Contrastingly, the ratio of POC:PON 391 (mol:mol) appeared to decrease southwards, although there was no significant difference 392 among zones (p-value > 0.05) (Fig. 4b). The 15 N-PON also decreased southwards from the 393 STZ and SAZ to the PFZ and OAZ (Fig. 4c). Despite considerable differences among zones, 394 the 15 N-PON was relatively homogenous within each zone. 395 4.4 Rates of net primary production, nitrogen uptake, and ammonium oxidation 396 The surface rates of bulk NPP were high in the STZ, and two-to six-fold higher in the SAZ and 397 PFZ than has been reported previously for the Atlantic sector in winter ( The bulk NH₄⁺ uptake rates (ρNH₄⁺) generally increased southwards from the STZ to the SAZ 403 and PFZ, and then decreased across the OAZ to reach a minimum at the southernmost station 404 (58.5°S; 3.0 ± 0.8 nM day⁻¹) (Fig. 5b). In the nano+ size fraction, ρNH₄⁺ changed little 405 latitudinally, although it was slightly lower in the PFZ than in the other zones. The contribution 406 of nanoplankton to ρNH₄⁺ ranged from 32.8% in the PFZ to 71.9% in the STZ. The bulk NO 3 -407 uptake rates (ρNO 3 -) were also low in the STZ, while the highest ρNO 3 was measured in the 408 SAZ before decreasing southwards. ρNO 3 in the nano+ size class followed the same trend as 409 total community ρNO 3 -, with the nanoplankton accounting for 71.5 ± 0.3% of bulk ρNO 3 on 410 average. The rates of bulk urea uptake (ρUrea) were highest in the STZ, with the SAZ and the 411 PFZ hosting similar rates, and the lowest rates were measured in the OAZ. ρUrea for the nano+ 412 size class followed a similar trend to bulk ρUrea, and nanoplankton accounted for 51.8% of 413 ρUrea in the SAZ to 100% in the PAZ. The uptake rates of the different N forms were not 414 significantly correlated with one another or with the ambient N concentrations (Fig. S4). 415 Surface ammonium oxidation rates (NH₄⁺ ox ) increased southwards, with higher NH₄⁺ ox in the 416 OAZ and PAZ than in the STZ, SAZ, and PFZ (Fig. 5c)  Nano-and picoeukaryotes, Synechococcus, and small heterotrophs (collectively, "small cells") 445 sampled at 13 stations along leg S were roughly 10 3 -times more abundant than the 446 microplankton (Fig. 6b) The relative contribution of heterotrophs to total small cells varied considerably (10-62%), 459 reaching a maximum south of the PF at 53.0°S and 57.8°S (62% and 50%; Fig. 7a). 460 Heterotroph abundance followed a similar pattern to that of the nanoeukaryotes, with higher 461 abundances in the SAZ than in the PFZ and OAZ. The food source available to heterotrophs, 462 represented by the small detrital particles, was highest near the southern edge of the SAF. More 463 generally, detrital particles were more abundant in the PFZ than in the SAZ and OAZ. The 464 relative contributions of detrital, photosynthetic, and heterotrophic particles are shown in Fig.  465 S5. 466 concentrations, as summarized in Fig. 8. In this study, we directly measured the rates of NH₄⁺ 479 uptake by different size fractions of the winter plankton community, as well as the rates of 480 NH₄⁺ oxidation. We infer the contribution of heterotrophic bacteria and microzooplankton to 481

Discussion
NH₄⁺ production from cell count data and the abundance of small heterotrophs relative to 482 phytoplankton and detritus. For the NH₄⁺ cycle processes in Fig. 8  ocean mixing. One implication of this suggestion is that the wintertime NH₄⁺ pool likely 491 reflects processes that occurred earlier in the season, as well as those that are ongoing. We posit 492 that the elevated NH₄⁺ concentrations in the PFZ and AZ may result from higher wintertime 493 rates of NH₄⁺ production than consumption and/or from the gradual but incomplete depletion in 494 winter of NH₄⁺ produced mainly in late summer and autumn. We evaluate both possibilities 495 throughout the discussion below. concentrations of chl-a and rates of NPP were low across our transect, they were not negligible 501 ( Fig. 3a and  including in the AZ (Fig. 5b). 519 The 15 N-PON data (Fig. 4c) suggest that this elevated reliance on recycled N persisted from 520 the late summer. In theory, PON generated in early-through mid-summer from the 521 consumption of upwelled NO₃⁻ NH₄⁺ concentrations are not elevated in the SAZ mixed layer in winter (Fig 2b.) indicates that 535 the remineralized NH₄⁺ is rapidly re-assimilated by phytoplankton and/or oxidized to NO 2 in 536 this zone. In the AZ, the 15 N-PON of -3 to -1‰ that we observe in winter surface waters 537 requires the sustained consumption of low-15 N N (i.e., recycled NH 4 + and urea) to offset a 538 remineralization-driven 15 N rise similar to that of the SAZ. We conclude that Southern Ocean 539 phytoplankton dominantly consume regenerated N from late summer until at least July (albeit 540 at low rates in winter), particularly south of the PF. 541 The fact that the NH₄⁺ concentration was high in the winter mixed layer despite NH₄⁺ being the 542 preferred phytoplankton N source in late summer through winter implies that low rates of NH₄⁺ 543 uptake contributed to the accumulation of this N form. Multiple factors may cause low rates of 544 photoautotrophic NH₄⁺ uptake, including deplete NH₄⁺ and micronutrient concentrations, light 545 limitation, and low temperatures. North of the SAF, NH₄⁺ concentrations below detection likely 546 limited ρNH₄⁺, as evidenced by the fact that in a series of experiments conducted on the same 547 cruise, ρNH₄⁺ increased with the addition of NH₄⁺ at these stations (Mdutyana, 2021). By 548 contrast, south of the SAF, NH₄⁺ concentrations were similar to or higher than the half-549 saturation constant (K m ) derived for NH₄⁺ uptake in the winter Southern Ocean (0.2 to 0.4 µM; 550 Mdutyana, 2021), suggesting that something other than NH₄⁺ availability was limiting to 551 phytoplankton at these latitudes. 552 Iron is not directly involved in NH₄⁺ assimilation but is required for electron transport during 553 photosynthesis and respiration (Raven, 1988 However, since NH₄⁺ consumption by phytoplankton is fairly energetically inexpensive 567 (Dortch, 1990), it should occur even under low light (recognizing that light remains critical for 568 coincident CO₂ fixation). Heterotrophic bacteria can also consume NH₄⁺ (Kirchman, 1994), 569 including in the dark since they derive energy from organic carbon oxidation rather than light. 570 At an ecosystem level, therefore, NH₄⁺ consumption may not be primarily limited by light, 571 although this parameter clearly strongly controls the rate of NPP (Fig. 5a). 572 Previous observations suggest that temperature influences NH₄⁺ uptake, especially in winter 573 (Glibert, 1982;Reay et al., 2001). The negative effect of temperature appears to be enhanced 574 under high-nutrient and low-light conditions, at least in the case of phytoplankton growth rates 575 (Baird et al., 2001). Additionally, Southern Ocean phytoplankton may be psychrotolerant and 576 not psychrophilic, which means that while they can function at in situ wintertime temperatures, 577 their optimal temperatures for growth and photosynthesis are higher (Reay et al., 2001;Smith 578 Jr & Harrison, 1991;Tilzer et al., 1986). Experiments conducted coincident with our sampling 579 showed that the maximum rate of NH₄⁺ uptake (V max ) achievable by the in situ community was 580 strongly negatively correlated with temperature and latitude (Mdutyana, 2021), with the latter 581 parameter indicative of the combined role of light, temperature, and possibly iron, the 582 concentration of which appears to increase from the SAZ to the AZ (Tagliabue et al., 2012). 583 We conclude that these three drivers, along with NH₄⁺ availability north of the SAF, all play a 584 role in controlling photoautotrophic NH₄⁺ uptake in the winter Southern Ocean, with complex 585 interactions among them that are difficult to disentangle. 586 In addition to physical and chemical limitations, microbial preference for other N species may 587 impact the depletion of the NH₄⁺ pool. For example, the preferential uptake of urea and other 588 DON species by some organisms (e.g., cyano-or heterotrophic bacteria) could dampen total 589 NH₄⁺ uptake rates. While large contributions of urea to total N uptake have previously been 590 observed in the Southern Ocean in summer and autumn (predominantly in the SAZ; Joubert et 591 al., 2011; Thomalla et al., 2011), we measured fairly low ρUrea (Fig. 5b), which is perhaps 592 unsurprising given the low ambient urea concentrations (Table 1). The exceptions were stations 593 37°S and 43.0°S where ρUrea was higher than ρNH₄⁺, coincident with very low ambient NH₄⁺ 594 (0.10 µM and below detection) and relatively high urea concentrations (0.36 µM and 0.15 µM). 595 Community composition can also alter the N uptake regime. Smaller phytoplankton, such as 596 the numerically-dominant nano-and picoeukaryotes, are more likely to consume NH₄⁺ and urea 597 than NO 3 - (Koike et al., 1986;Lee et al., 2012Lee et al., , 2013, especially in the Southern Ocean where 598 NO 3 assimilation is severely limited by iron and light availability (Sunda & Huntsman, 1997). 599 Across our transect, the sum of NH₄⁺ and urea uptake (i.e., reduced N uptake) exceeded NO 3 -600 uptake for both the total phytoplankton community (transect average of 12.0 ± 0.9 nM day⁻¹ for 601 reduced N versus 5.8 ± 1.0 nM day⁻¹ for NO 3 -; f-ratio of 0.36) and the 0.3-2.7 µm size fraction 602 (5.0 ± 1.2 nM day⁻¹ versus 1.9 ± 1.2 nM day⁻¹; f-ratio of 0.27 (Fig. 5b) strong correlation with NH₄⁺ concentration south of the SAF (r = 0.65). In the nano+ size class, 614 NO 3 uptake was likely driven in the SAZ by dinoflagellates and some nanoeukaryotes, and in 615 the PFZ and AZ by diatoms, which remain active in these zones in winter (Weir et al., 2020). 616 By contrast, nanoeukaryotes, which have a higher per-cell nutrient requirement than the 617 equally-abundant picoeukaryotes, may have dominated NH₄⁺ uptake in the PFZ and AZ given 618 that higher nanoeukaryote abundances corresponded with lower NH₄⁺ concentrations at a 619 number of stations (e.g., stations 50.0°S, 51.1°S, and 55.5°S; Fig. 6b). 620 The low abundances of diatoms and dinoflagellates and absence of coccolithophores (Fig. 6a pennate abundance was associated with lower NH₄⁺. Diatom success in winter may also be 632 limited by enhanced mixing, as this group is generally adapted for stratified waters 633 (Kopczynska et al., 2007). 634 In sum, NH₄⁺ uptake rates were low across our transect but not negligible, indicating that 635 phytoplankton activity in winter, which is dominated by smaller species, represents a sink for 636 NH 4 + . Hostile Southern Ocean conditions imposed limitations on NH₄⁺ uptake that varied with 637 latitude, with NH₄⁺ concentrations controlling ρNH₄⁺ north of the SAF, while light and 638 temperature were important south of the SAF, with a possible supporting role for iron. 639 Additionally, Synechococcus, nanoeukaryotes, and pennate diatoms likely dominated NH₄⁺ 640 consumption, consistent with previous observations from the Southern Ocean and elsewhere 641 (Klawonn et al., 2019;Semeneh et al., 1998). 642 Ammonium oxidation -Nitrification removes more mixed-layer NH₄⁺ than phytoplankton 643 consumption south of the PF, with NH₄⁺ oxidation rates that were two-to five-times the co-644 occurring NH₄⁺ uptake rates (Fig. 5c). The comparative success of NH₄⁺ oxidisers may be due 645 to decreased competition with phytoplankton for NH₄⁺ in winter, augmented by decreased The K m derived for NH₄⁺ oxidation in the winter Southern Ocean has recently been reported to 667 be low (0.03 to 0.14 µM), with ammonia oxidizers observed to become saturated at ambient 668 NH₄⁺ concentrations of ~0.1-0.2 µM (Mdutyana, 2021). This means that south of the SAF in 669 winter 2017, ammonia oxidizers were not substrate limited (further implied by the lack of 670 correlation between NH₄⁺ ox and NH₄⁺ concentration; Fig. S4), which raises the question of why 671 NH₄⁺ oxidation did not occur at higher rates. The answer may involve temperature, in that 672 psychrophilic organisms can be less responsive to high substrate concentrations at low NH₄⁺ production, although not measured directly in this study, must be sustained during the 682 winter to retain an NH₄⁺ pool that is high in concentration relative to the early summer. With 683 low or no NH₄⁺ production in the autumn and winter, the NH₄⁺ pool south of the SAF would be 684 depleted in 10 to 38 days (median of 21 days) given the consumption rate (ρNH₄⁺ + NH₄⁺ ox ) 685 and NH₄⁺ concentration measured at each station (Text S2). Heterotrophic NH 4 + production 686 must, therefore, be ongoing in winter despite the limited production of PON substrate. 687 winter NH₄⁺ pool includes residual NH₄⁺ produced towards the end of the growing season. At 693 the time of our sampling, heterotrophic abundances were ten-fold lower to two-fold higher than 694 total pico-and nanophytoplankton abundances (Fig. 7a). Higher ratios of heterotrophic-to-695 photosynthetic cells occurred at stations with higher NH₄⁺ concentrations (e.g., stations 48.9°S, 696 53.0°S, 54.0°S and 57.8°S), suggesting a role for the short-term balance between NH₄⁺ 697 https://doi.org/10.5194/bg-2021-149 Preprint. Discussion started: 10 June 2021 c Author(s) 2021. CC BY 4.0 License. production and consumption in controlling the ambient NH₄⁺ concentration in winter. The 698 heterotrophic bacteria were likely consuming detritus (as opposed to living cells), with the 699 relative availability of detrital substrate evident from the high detrital particle counts (Fig. 7b) 700 and the persistently high POC:chl-a ratios, particularly south of the PF (Fig. 4a; values (i.e., the proportional increase in growth rate with a 10 °C rise in temperature) between 720 phytoplankton and heterotrophs are required for heterotrophic NH₄⁺ production to exceed 721 phytoplankton NH₄⁺ uptake (Koike et al., 1986). Nonetheless, it is highly unlikely that the 722 surface NH₄⁺ pool measured in winter derived solely from wintertime bacterial production 723 given that yet higher NH₄⁺ concentrations have been observed in late summer/autumn 724 (Becquevort et al., 2000;Dennett et al., 2001); this is discussed further in section 5.2 below. 725 Heterotrophic activity by zooplankton -The microzooplankton enumerated in this study may 726 also contribute to NH₄⁺ accumulation, although they are probably less important in winter than 727 heterotrophic bacteria given their low and variable abundances (Fig. 6a) By the early spring, the NH₄⁺ concentrations south of the SAF had declined to near or below 797 the methodological detection limit (0.09 ± 0.08 µM; Fig. 9d), implicating increased 798 photosynthetic activity following the alleviation of light-limitation that results in the 799 consumption of nutrients introduced into surface waters in winter. We postulate that the 800 residual NH₄⁺ would have been consumed prior to significant NO 3 drawdown because far less 801 energy (i.e., light) is required for its assimilation (Dortch, 1990). NH₄⁺ concentrations south of 802 the SAF rose again by the late spring to an average value only slightly lower than that 803 measured in winter (0.37 ± 0.69 µM; Fig. 9e). However, late-spring NH₄⁺ concentrations were 804 only elevated in the PFZ (range of 0.11 ± 0.01 to 4.39 ± 0.03 µM, average of 0.71 ± 1.04 µM), 805 as has been observed previously , which we attribute to increased 806 heterotrophic activity in response to elevated regional springtime phytoplankton growth driven 807 by frontal upwelling (Becquevort et al., 2000;Mayzaud et al., 2002). Excluding the PFZ data 808 yields a far lower late-spring average NH₄⁺ concentration of 0.18 ± 0.14 µM, which we take as 809 broadly representative of this season. 810 Using our high-resolution NH₄⁺ concentration measurements, we propose a seasonal cycle for 811 mixed-layer NH₄⁺ south of the SAF (Fig. 9f). Our proposal is consistent with previous 812 characterizations of the early summer-to-autumn evolution of Southern Ocean NH₄⁺ 813 concentrations (i.e., from below detection due to phytoplankton uptake to elevated due to net 814 heterotrophic activity), but contradicts the hypothesis that NH₄⁺ will subsequently decline due 815 to persistent but low rates of photosynthesis that yield insufficient biomass to support late-816 summer heterotrophy, thus resulting in a coincident decrease in photosynthetic and 817 heterotrophic activity ( suggesting that this effect is not straightforward. In winter 2017, we observed little evidence of 846 NH₄⁺ inhibition of NO 3 uptakefor example, the southward decrease in ρNO₃⁻ was not 847 sharper than that of ρNH₄⁺ despite the increase in NH₄⁺ concentration, and we observed no 848 relationship between NH₄⁺ concentration and the proportion of NO₃⁻-to-total N uptake (i.e., the 849 f-ratio, r = 0.28 including urea; n=7). We conclude that NH₄⁺ inhibition of NO₃⁻ uptake is globally, however, the magnitude of the marine NH 3 source remains highly uncertain (Paulot et 881 al., 2015). Surface ocean NH 4 + concentrations play a central role in determining the sign and 882 magnitude of the air-sea NH 3 flux, along with wind speed, surface ocean temperature, and pH. 883 Therefore, the biogeochemical pathways that drive seasonality in surface ocean NH 4 + 884 concentrations are an important control on the remote Southern Ocean air-sea NH 3 flux, with 885 implications for aerosol composition, cloud formation, and climate (Altieri et al., 2021). 886

Summary 887
This study, conducted in the Southern Ocean during the infrequently-sampled winter season, 888 provides new insights into the internal cycling of N in the mixed layer of a globally-important 889 region. We used measurements of NO 3 -, NH₄⁺, and urea uptake, NH₄⁺ oxidation rates, 15 N-890 PON, and the ratio of heterotrophic-to-photosynthetic cells to investigate NH₄⁺ consumption, 891 and the ratios of POC:chl-a and POC:PON, the relationship of V Ntot to V C , and measurements 892 of plankton community composition to evaluate the potential for heterotrophic NH₄⁺ 893 production. We attribute the elevated NH₄⁺ concentrations that persist in the winter mixed layer 894 south of the SAF to sustained heterotrophic NH₄⁺ production in excess of phytoplankton-and 895 nitrifier-mediated NH₄⁺ consumption, driven by temperature-, light-, and possibly iron-896 limitation of the NH₄⁺ consumers. We further conclude that heterotrophic bacteria are the main 897 NH₄⁺ producers in winter and that the contributions of DON degradation, nitrogen fixation, 898 aerosol deposition, and sea-ice melt to the Southern Ocean's mixed-layer NH₄⁺ pool are 899 negligible. Future measurements of heterotrophic NH₄⁺ production rates are required to validate 900 our conclusions, and higher spatial resolution sampling of community composition and N 901 consumption rates may help to explain smaller-scale variability in NH₄⁺ concentrations, 902 particularly near the fronts. 903 From observations of surface NH₄⁺ concentrations made between December 2018 and 904 November 2019, we suggest that the high-concentration NH₄⁺ pool cannot be generated solely 905 during winter. Instead, we propose that NH₄⁺ initially accumulates in late summer following the 906 peak phytoplankton growing season, after which sustained heterotrophy throughout the autumn 907 and winter prevents this NH₄⁺ from being depleted until the early spring. The persistence of 908 elevated NH₄⁺ concentrations across the polar Southern Ocean between late summer and winter 909 https://doi.org/10.5194/bg-2021-149 Preprint.