Nutrient analysis
\n\u00a0
\nNutrient samples were filtered through 0.22 \u03bcm pore size Durapore\u00a0GVWP filters (Millipore Sigma) and frozen at \u221220_C immediately\u00a0after collection, then stored at \u221280_C until analysis.\u00a0Dissolved nitrate (NO3\u00a0\u2212), nitrite (NO2\u00a0\u2212), phosphate (PO4\u00a03\u2212), and\u00a0silicate (SiO4\u00a04\u2212) were measured using a Bran and Luebbe AA3\u00a0autoanalyzer as described previously (Wilkerson et al. 2015).\u00a0Ammonium and urea were measured manually using the\u00a0phenolhypochlorite method (Sol\u00f3rzano 1969) and the diacetylmonoxime\u00a0method (Rahmatullah and Boyde 1980; Mulvenna\u00a0and Savidge 1992), respectively.\u00a0
\nOxidation rate measurements\u00a0
\nWe used 15N-labeled substrates (98\u201399% 15N, Cambridge\u00a0Isotope Laboratories) to measure the oxidation of N supplied\u00a0as NH4+, urea, 1,2-diaminoethane (DAE), 1,3-diaminopropane\u00a0(DAP), 1,4-diaminobutane (putrescine, PUT), L-glutamic acid\u00a0(GLU), and L-arginine (ARG). 15N oxidation from NH4+, urea,\u00a0PUT, and GLU were measured extensively, whereas 15N oxidation\u00a0from DAE, DAP, and ARG was only measured at a subset\u00a0of stations (Supporting Information Table S1). GLU and ARG\u00a0were included as a control for remineralization, as their central\u00a0roles in microbial metabolism leads to rapid catabolism and\u00a0NH4\u00a0+ regeneration (Hollibaugh 1978; Goldman et al. 1987).\u00a0PUT was used in routine assessments of the oxidation of\u00a0polyamine-N because it is one of the most consistently detected\u00a0polyamines in seawater (Nishibori et al. 2001a, 2003; Lu\u00a0et al. 2014; Liu et al. 2015). Although spermine and spermidine\u00a0are also common in seawater, 15N-labeled stocks of these polyamines\u00a0were not commercially available. We measured the oxidation\u00a0of N from DAE and DAP to investigate the effect of\u00a0aliphatic chain length (which affects pKa) on oxidation rate.\u00a0
\nDuplicate seawater samples contained in 1-liter polycarbonate\u00a0or 250 mL high density polyethylene (HDPE) bottles\u00a0wrapped with aluminum foil (to exclude light) were\u00a0amended with 10\u201350 nM 15N-labeled substrate. Marsh\u00a0Landing samples were then placed in an incubator held at\u00a0in situ temperature in the dark. Samples taken at the Skidaway\u00a0dock were placed in a mesh bag and immersed at the\u00a0sea surface at the sampling site. Samples collected at sea\u00a0were incubated in a tank of flowing surface seawater or in an\u00a0incubator held at 18 C in the dark. Incubation bottles were sampled\u00a0for 15N analysis immediately after substrate addition and\u00a0again after a period of ~ 24 h. 15N samples were subsampled into\u00a050 mL polypropylene centrifuge tubes, frozen at \u221220_C, and\u00a0stored at \u221280_C until analysis. The 15N/14N ratios of the NO3\u00a0\u2212\u00a0plus NO2\u00a0\u2212 (NOX) pools (\u03b415NNOx) in the samples were measured\u00a0using the bacterial denitrifier method to convert NOX to nitrous\u00a0oxide (N2O; Sigman et al. 2001). The \u03b415N values of the N2O\u00a0produced were measured using a Finnigan MAT-252 isotope\u00a0ratio mass spectrometer coupled with a modified GasBench II interface (Casciotti et al. 2002; Beman et al. 2011; McIlvin\u00a0and Casciotti 2011). Oxidation rates were calculated using an\u00a0endpoint model (Beman et al. 2011; Damashek et al. 2016).\u00a0Since the substrates used were uniformly labeled with 15N, the\u00a0amount of the N added as the 15N spike (in \u03bcM) was multiplied\u00a0by the number of moles of 15N per mole of substrate, which\u00a0assumes that all of the N atoms have equal probability of being\u00a0oxidized. This is likely true for urea, DAE, DAP, and PUT, which\u00a0are symmetrical molecules, but not likely to be true for ARG,\u00a0which contains 4 N atoms (one in the \u03b1-amino position and\u00a0three in the guanidine structure of its R-group). Abiotic oxidation\u00a0of organic N was assessed by measuring 15NOX production\u00a0following 15N amendment and incubation of 0.22 \u03bcm filtered\u00a0seawater (as described above), and potential metabolism of\u00a0DON by the denitrifying bacteria used to convert NOX to N2O\u00a0was checked by adding 15N-labeled substrates into the bacterial\u00a0cultures prior to mass spectrometry.\u00a0
\nWe were unable to measure the in situ concentrations of\u00a0the individual components of DON used in oxidation experiments,\u00a0other than urea. Based on previous measurements\u00a0made in the SAB (Lu et al. 2014; Liu et al. 2015), we assumed\u00a0concentrations of 1 nM and 0.25 nM for DAE, DAP and PUT,\u00a0and 10 nM and 5 nM for GLU and ARG, at inshore and\u00a0mid-shelf/shelf-break/oceanic stations, respectively. Rates of\u00a0polyamine and amino acid oxidation reported below should\u00a0therefore be considered potential rates, as amendments as low\u00a0as 10\u201350 nM are likely to increase substrate concentrations\u00a0substantially above in situ. Initial substrate 15N activity was\u00a0calculated using isotope mass balance using the known concentration\u00a0and 15N activity of the labeled substrates added\u00a0and assuming the concentrations described above and natural\u00a0abundance 15N activity (i.e., 0.3663 atom% 15N).\u00a0
Samples were collected from four regions (inshore, midshelf,\u00a0shelf-break, and oceanic) of the SAB off the Georgia\u00a0(U.S.A.) coast (Fig. 1; Supporting Information Table S1), with\u00a0terminology modified from Liu et al. (2018) as follows.\u00a0\u201cInshore\u201d stations were within the barrier island complex.\u00a0\u201cMid-shelf\u201d stations were outside the barrier island complex\u00a0to depths < 40 m; due to limited sampling in this zone, no\u00a0demarcation between \u201cmid-shelf\u201d and \u201cnearshore\u201d stations\u00a0(as in Liu et al. 2018) was made. \u201cShelf-break\u201d stations were\u00a0between 40 m and 500 m depth. While Liu et al. (2018) did\u00a0not sample waters past the shelf-break, we included deeper\u00a0stations further offshore (bottom depth > 500 m), which are\u00a0designated \u201coceanic\u201d stations. Note that the maximum depth\u00a0sampled was \u2264 500 m due to equipment limitations.\u00a0
\nInshore samples were collected from a dock at Marsh Landing\u00a0on the Duplin River (Sapelo Island) and the dock at the\u00a0Skidaway Institute of Oceanography (Fig. 1). Both inshore sites\u00a0are salt marsh-dominated estuaries. Water from both sites was\u00a0sampled from a depth \u2264 1 m and was processed immediately\u00a0at a nearby laboratory (the University of Georgia Marine Institute\u00a0on Sapelo Island or onboard the R/V Savannah). Water\u00a0quality data for Marsh Landing samples were collected as part\u00a0of the Sapelo Island National Estuarine Research Reserve monitoring\u00a0program. Relevant data from the Lower Duplin (\u201cLD\u201d)\u00a0sonde were downloaded from NOAA/CDMO (http://cdmo.\u00a0baruch.sc.edu/aqs/, last accessed 22 May 2018).\u00a0
\nMost SAB samples were collected in August 2017 on the\u00a0R/V Savannah (cruise SAV-17-16) along transects across the\u00a0continental shelf and the Gulf Stream and into the western\u00a0Sargasso Sea, with sampling focused around the shelf-break\u00a0(Fig. 1). Water from multiple depths was collected using\u00a012-liter Niskin bottles mounted on a rosette equipped with a\u00a0Sea-Bird SBE25 CTD. Profiles of salinity, temperature, dissolved\u00a0oxygen, fluorescence, and photosynthetically active\u00a0radiation (PAR) were collected using the CTD system as\u00a0described previously (Liu et al. 2018). PAR attenuation (Kd) was\u00a0calculated from plots of ln(PAR) vs. depth as in Liu et al. (2018).\u00a0 Two additional SAB stations were sampled in October 2011\u00a0(described previously by Liu et al. 2015 and Tolar et al. 2017)\u00a0and are referred to as \u201c2011-4\u201d and \u201c2011-12\u201d (Fig. 1). Environmental\u00a0data and some of the microbial and rate data from 2011\u00a0stations are available in other publications (Liu et al. 2015; Tolar\u00a0et al. 2017; see BCO-DMO dataset qPCR_Parameters at https://www.bco-dmo.org/dataset/767141).\u00a0
BCO-DMO Processing Notes:
\n-\u00a0added conventional header with dataset name, PI name, version date
\n-\u00a0modified parameter names to conform with BCO-DMO naming conventions
\n- converted latitude and longitude values from degrees minutes seconds to degrees and appended a new column (lat, lon).
\n- converted Date column to ISO convention and appended ISO_Date column for those values.
\n- replaced \"x 10\" with \"e\" to signify power of 10.
\n- split columns containing \"\u00b1\" into two columns, one for value and one for\u00a0\u00b1 value. Appended \"_sd\" to the\u00a0\u00b1 value column.
\n- converted \"nan\" and blank cells to \"nd\".