Search

A comprehensive in situ dataset of chlorophyll a (Chl a; N = 18,001), net primary production (NPP; N = 165) and net community production (NCP; N = 95), were used to evaluate the performance of Moderate Resolution Imaging Spectroradiometer on Aqua (MODIS-A) algorithms for these parameters, in the South Atlantic Ocean, to facilitate the accurate generation of satellite NCP time series. For Chl a, five algorithms were tested using MODIS-A data, and OC3-CI performed best, which was subsequently used to compute NPP. Of three NPP algorithms tested, a Wavelength Resolved Model (WRM) was the most accurate, and was therefore used to estimate NCP with an empirical relationship between NCP with NPP and sea surface temperature (SST). A perturbation analysis was deployed to quantify the range of uncertainties introduced in satellite NCP from input parameters. The largest reductions in the uncertainty of satellite NCP came from MODIS-A derived NPP using the WRM (40%) and MODIS-A Chl a using OC3-CI (22%). The most accurate NCP algorithm, was used to generate a 16 year time series (2002 to 2018) from MODIS-A to assess climate and environmental drivers of NCP across the South Atlantic basin. Positive correlations between wind speed anomalies and NCP anomalies were observed in the central South Atlantic Gyre (SATL), and the Benguela Upwelling (BENG), indicating that autotrophic conditions may be fuelled by local wind-induced nutrient inputs to the mixed layer. Sea Level Height Anomalies (SLHA), used as an indicator of mesoscale eddies, were negatively correlated with NCP anomalies offshore of the BENG upwelling fronts into the SATL, suggesting autotrophic conditions are driven by mesoscale features. The Agulhas bank and Brazil-Malvinas confluence regions also had a strong negative correlation between SLHA and NCP anomalies, similarly indicating that NCP is forced by mesoscale eddy generation in this region. Positive correlations between SST anomalies and the Multivariate ENSO Index (MEI) in the SATL, indicated the influence of El Niño events on the South Atlantic Ocean, however the plankton community response was less clear. UK Natural Environment Research Council | Ref. NERC; NE / L002434 / 1 European Space Agency | Ref. AMT4SentinelFRM (ESRIN / RFQ / 3-14457 / 16 / I-BG) European Space Agency | Ref. AMT4OceanSatFlux (4000125730/18 / NL / FF / gp) NERC International Opportunity Fund Grant Satellite estimates of marine net community production in the South Atlantic from Sentinel-3 | Ref. SemSAS; NE / P00878X / 1 P&D ANP / BRASOIL | Ref. 48610.011013 / 2014-66 Oceanographic Institute of the University of São Paulo | Ref. FAPESP 2015 / 01373-0 Oceanographic Institute of the University of São Paulo | Ref. CNPq 442926 / 2015-4 Oceanographic Institute of the University of São Paulo | Ref. FAPESP 2014 / 50820-7 Oceanographic Institute of the University of São Paulo | Ref. CNPq 565060 / 2010-4 NERC | Ref. NE / R015953 / 1

There is ongoing debate as to whether the oligotrophic ocean is predominantly net autotrophic and acts as a CO2 sink, or net heterotrophic and therefore acts as a CO2 source to the atmosphere. This quantification is challenging, both spatially and temporally, due to the sparseness of measurements. There has been a concerted effort to derive accurate estimates of phytoplankton photosynthesis and primary production from satellite data to fill these gaps; however there have been few satellite estimates of net community production (NCP). In this paper, we compare a number of empirical approaches to estimate NCP from satellite data with in vitro measurements of changes in dissolved O2 concentration at 295 stations in the N and S Atlantic Ocean (including the Antarctic), Greenland and Mediterranean Seas. Algorithms based on power laws between NCP and particulate organic carbon production (POC) derived from 14C uptake tend to overestimate NCP at negative values and underestimate at positive values. An algorithm that includes sea surface temperature (SST) in the power function of NCP and 14C POC has the lowest bias and root-mean square error compared with in vitro measured NCP and is the most accurate algorithm for the Atlantic Ocean. Nearly a 13 year time series of NCP was generated using this algorithm with SeaWiFS data to assess changes over time in different regions and in relation to climate variability. The North Atlantic subtropical and tropical Gyres (NATL) were predominantly net autotrophic from 1998 to 2010 except for boreal autumn/winter, suggesting that the northern hemisphere has remained a net sink for CO2 during this period. The South Atlantic sub-tropical Gyre (SATL) fluctuated from being net autotrophic in austral spring-summer, to net heterotrophic in austral autumn–winter. Recent decadal trends suggest that the SATL is becoming more of a CO2 source. Over the Atlantic basin, the percentage of satellite pixels with negative NCP was 27%, with the largest contributions from the NATL and SATL during boreal and austral autumn–winter, respectively. Variations in NCP in the northern and southern hemispheres were correlated with climate indices. Negative correlations between NCP and the multivariate ENSO index (MEI) occurred in the SATL, which explained up to 60% of the variability in NCP. Similarly there was a negative correlation between NCP and the North Atlantic Oscillation (NAO) in the Southern Sub-Tropical Convergence Zone (SSTC), which explained 90% of the variability. There were also positive correlations with NAO in the Canary Current Coastal Upwelling (CNRY) and Western Tropical Atlantic (WTRA) which explained 80% and 60% of the variability in each province, respectively. MEI and NAO seem to play a role in modifying phases of net autotrophy and heterotrophy in the Atlantic Ocean.

We present over 900 carbonate system observations collected over four years (2007–2010) in the Western English Channel (WEC). We determined CO2 partial pressure (pCO2), Total Alkalinity (TA) and Dissolved Inorganic Carbon (DIC) along a series of 40 km transects, including two oceanographic stations (L4 and E1) within a sustained coastal observatory. Our data follow a seasonal pattern of CO2 undersaturation from January to August, followed by supersaturation in September–October and a return to near-equilibrium thereafter. This pattern is explained by the interplay of thermal and biological sinks in winter and spring–summer, respectively, followed by the breakdown of stratification and mixing with deeper, high-CO2 water in autumn. The drawdown of DIC and inorganic N between March and June with a C:N ratio of 8.7–9.5 was consistent with carbon over-consumption during phytoplankton growth. Monthly mean surface pCO2 was strongly correlated with depth integrated chlorophyll a highlighting the importance of subsurface chlorophyll a maxima in controlling C-fluxes in shelf seas. Mixing of seawater with riverine freshwater in near-shore samples caused a reduction in TA and the saturation state of calcite minerals, particularly in winter. Our data show that the L4 and E1 oceanographic stations were small, net sinks for atmospheric CO2 over an annual cycle (−0.52±0.66 mol C m−2 y−1 and −0.62±0.49 mol C m−2 y−1, respectively).

During a Lagrangian study in the Mauritanian upwelling (April 15 to May 27 2009), we investigated the biogeochemical cycling of carbon monoxide (CO) through a series of experiments, field observations and a simple 1-D model. We carried out ten photochemical experiments in a solar simulator and a further dark experiment in order to determine the magnitude of the photochemical source and microbial sink for CO. Parallel irradiation experiments using a long-pass filter (390 nm cutoff) showed that CO photoproduction was dominated by UV-light. Assuming that the apparent quantum yield (AQY) of CO photoproduction followed an exponential decrease with increasing wavelength, we applied a non-linear fit to derive the AQY for each of 10 experiments. The average AQY at 350 nm from our irradiation experiments was 1.4 × 10− 5 mol CO produced per mol photons absorbed. One of our photochemical experiments showed a distinct dissolved organic matter absorption ‘shoulder’ in the UV-B, consistent with the presence of mycosporine-like amino acids. Though this feature was rapidly photobleached during irradiation, CO AQY did not differ from other experiments. During the microbial CO oxidation experiment we found first-order loss of CO with a rate constant of 0.17/h. We further used our experimental data to parameterise a 1-D steady state model which included photochemical and dark production of CO, microbial oxidation, mixing and sea to air loss. Our sea–air flux estimates were in the range of 4.6–9.6 μmol CO per square metre per day. The estimated annual flux of CO to the atmosphere from the Mauritanian upwelling was 0.04 Tg CO yr− 1.

Dissolved methane and nitrous oxide in seawater were measured along a 6700 km transect of the North-West Passage between the North Atlantic Ocean and Beaufort Sea in the Arctic Ocean. Over- and under-saturation with respect to atmospheric equilibrium were observed for both gases. Methane and nitrous oxide were in the range of 58–528% and 82–181% saturation, respectively. Under-saturation was attributed to melt-water with low methane and nitrous oxide, while over-saturation was found under multi-year sea-ice. Elevated methane was also found in the vicinity of the marginal ice zones and the Mackenzie River plume. Our data support both water column and sedimentary sources of methane and nitrous oxide. We found first-order methane oxidation in surface seawater with a rate constant of 3.8 × 10−3 h−1. Based on these results and a conceptual model, we suggest that future sea-ice retreat may decrease the residence times of methane and nitrous oxide in the surface Arctic Ocean and thus enhance the sea–air flux of these climatically active gases.

We carried out 12 irradiation experiments under artificial light using samples collected from two UK upland temperate rivers, the Tyne and the Tay, and from the Tyne estuary and adjacent coastal North Sea waters. During these experiments, we followed concentration changes of ammonium (NH 4 + ) and spectral changes of chromophoric dissolved organic matter (CDOM). In all experiments CDOM absorbance followed pseudo-first-order loss kinetics and photochemical NH 4 + production occurred consistently. However, the patterns of NH 4 + concentration changes with time were complex. Experiments in which irradiated waters were sampled at approximately 10 min intervals revealed an initial lag phase of ∼ 30 to 120 min during which NH 4 + concentrations did not change significantly. This lag phase was followed by a net photoproduction phase, and finally an apparent NH 4 + consumption phase. Such behaviour has not previously been reported and appeared to be independent of sample origin, initial NH 4 + concentration and initial CDOM absorption coefficient at 350 nm ( a 350 ). The observed behaviour could be described by a 3-stage irreversible reaction model: A → B → NH 4 + → D with first-order rate constants ka , kb and kc for the 3 consecutive reaction steps. Both ka and kc were significantly correlated ( p a 350 and salinity over the CDOM spectral slope ( S 290–350 ) through multiple linear regression analysis, suggesting that dissolved organic matter (DOM) properties associated with CDOM absorbance characteristics play an important role in this photochemical process. NH 4 + photoproduction rates, calculated over the lag- and production-phases, were significantly correlated with the initial a 350 for the River Tyne (including tributaries) samples, but not when samples from the estuary, coastal North Sea and River Tay were included. We also tested for possible correlations between CDOM photobleaching and spectral properties, but found no significant correlations between NH 4 + photoproduction rates and the initial spectral slope of CDOM or the CDOM photooxidation rate constants at 285, 310, 350 and 390 nm.

Methane (CH4) concentrations were measured in the water column, in sediment porewaters, and in atmospheric air, in the Ria de Vigo, NW Spain, during both the onset (April 2003) and at the end of (September 2004) seasonal upwelling. In addition, CH4 concentration and stable isotopic signatures (δ13CH4) were measured in porewaters, and sediment methanogenesis and aerobic oxidation of CH4 were determined in sediment incubations. Surface water column CH4 (2 m depth) was in the range 3–180 nmol l− 1 (110–8500% saturation) and followed a generally landward increase but with localised maxima in both the inner and middle Ria. These maxima were consistent with CH4 inputs from underlying porewaters in which CH4 concentrations were up to 3 orders of magnitude higher (maximum 350 μmol l− 1). Surface water CH4 concentrations were approximately three times higher in September than in April, consistent with a significant benthic CH4 flux driven by enhanced sediment methanogenesis following the summer productivity maximum. CH4 and δ13CH4 in sediment porewaters and in incubated sediment slurries (20 °C) revealed significant sediment CH4 oxidation, with an apparent isotopic fractionation factor (rc) of ∼ 1.004. Using turbulent diffusion models of air–sea exchange we estimate an annual emission of atmospheric CH4 from the Ria de Vigo of 18–44 × 106 g (1.1–2.7 × 106 mol). This estimate is approximately 1–2 orders of magnitude lower than a previous estimate based on a bubble transport model.

Carbon monoxide (CO) atmospheric mixing ratios and surface-water concentrations were determined during Atlantic Meridional Transect cruise number 10, April–May 2000. Atmospheric CO increased from south (mean=74±9 ppbv) to north (mean=151±19 ppbv) with a steep increase around the intertropical convergence zone. Surface-water CO (0.2–2.6 nmol L −1 ) showed pronounced diurnal variations with afternoon maxima exceeding pre-dawn minima 5–7 fold. Modest regional variations, as indicated by maximum daily CO concentrations, were also observed. Highest CO maxima occurred at ∼11.5°N, where high solar irradiance was combined with elevated coloured dissolved organic matter (CDOM) levels and modest winds, while lowest CO maxima occurred during periods of high winds and lowest solar irradiance near the western European margin at 45°N. Atlantic Ocean CO emissions were estimated to be 1.5±1.1 Tg CO-C yr −1 based on near-instantaneous atmospheric CO, sea-surface CO and windspeeds from the cruise. However, as spatial and temporal variability in both terms was considered to be unique to the timing and path of the cruise, the mean Atlantic diel cycle of sea-surface CO concentration was estimated by pooling all cruise data into 1-h sections, yielding a mean of 0.94 nmol L −1 ; and diurnal variations from 0.4 to 1.6 nmol L −1 . Using the mean diurnal cycle, the Atlantic and global open-ocean sources of CO to the atmosphere were estimated to be 0.9±0.6 and 3.7±2.6 Tg CO-C yr −1 , respectively. Therefore it is our contention that IPCC-2001 (Prather, M., Ehhalt, D., Dentener, F., Derwent, R., Dlugokencky, E., Holland, E., Isaksen, I., Katima, J., Kirchhoff, V., Matson, P., Midgley, P., Wang, M., 2001. Chapter 4: Atmospheric chemistry and greenhouse gases. In: Houghton, J.T., Ding, Y., Griggs, D.J., Noguer, M., van der Linden, P.J., Dai, X., Maskell, K., Johnson, C.A. (Eds.), Climate Change 2001: The Scientific Basis. Contribution of working group 1 to the third assessment report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, pp. 239–287; 21 Tg CO-C yr −1 ) overestimates the source of atmospheric CO from the global ocean by around 5 fold.

We report over 600 absorption spectra (250–800 nm) of unfiltered surface waters

The Eastern Mediterranean is the largest oceanic ecosystem that is phosphate-limited. To determine the impact of a transient input we executed a phosphate addition experiment in the surface waters of the Cyprus Eddy (33.31N 32.31E), and compared the ecosystem response with surrounding unperturbed water. A tracer, sulphur hexafluoride (SF6), added with the phosphate, enabled tracking of the patch when phosphate concentration declined to detection limits, and provided quantitative estimates of mixing, dilution and patch volume. The patch expanded to 4400 km 2 over 9 days with a lateral diffusion rate of 237 2m 2 /s that was consistent with previous tracer releases in eddies. Mixed layer phosphate concentration was � 110 nmol/l immediately post-release, and declined to o5 nmol/l after 6 days. A phosphate budget was developed using SF6 as a proxy to discriminate between dilution and biological pathways, with dilution resulting in loss of � 75% of added phosphate from the patch centre by day 3. Non-conservative phosphate loss was largely due to biological incorporation into particulate-P, of which 50% accumulated at the patch centre whilst the remainder was removed by lateral dilution by day 3. Non-conservative phosphate loss at the patch centre was 15–15.5 nmol/l by day 4, which was equal to the cumulative biological P uptake of 15.6 (75.6) nmol/l P and concurred with two other independent estimates of P uptake. This closure of the phosphate budget infers that the transfer of added P to mesozooplankton and higher consumers was not significant within the timescale of the experiment, despite the observed biomass increase that followed phosphate addition. Although patch dilution significantly reduced phosphate concentration and particulate accumulation, and so the apparent biological response to the added phosphate, analysis suggests that lateral mixing would not prevent bacterial biomass accumulation at the growth rates observed, suggesting that another factor such as grazing was responsible.

The south east Levantine basin of the eastern Mediterranean is a uniquely P starved system with a nitrate:phosphate ratio in the deep water of 25–28:1, a PON:POP ratio of 27–32:1 and a DON:DOPUV ratio of � 100:1 (which probably represents a DON:DOPTOTAL of � 50:1). The C:N:P ratio of nutrients accumulated in the deep water from decomposed organic matter was 106:8.5–10.8:0.34–0.43 similar to the measured ratios for dissolved and particulate organic matter and much higher than the Redfield ratio. It is concluded that the P limitation of the eastern Mediterranean is due to the lack of P within the system and not in the preferential removal of P relative to N. Results from the first extensive deployment of on-board nanomolar nutrient measurements in this low nutrient low chlorophyll system showed that free ammonia (50–80 nM) was present in the surface waters while nitrate was less than 10 nM, confirming the results obtained elsewhere in the CYCLOPS addition experiment results that grazing/nutrient recycling is a dominant process in this system. The total DIN:DIP ratio in the nutrient depleted waters above the chlorophyll maximum was predominantly greater than 16:1, suggesting that the system has not switched to N limitation. A primary nitrite maximum was observed immediately below the chlorophyll maximum at the top of the nutricline, which was similar to those found previously in oligotrophic ocean locations. Where nanomolar technology was deployed through the nutricline, it was found that the phosphocline started at the same depth as the nutricline for nitrate and silicate, a conclusion that would not have been made if only conventional micromolar technology had been available. An intercomparison of nutrient procedures suggested that freezing samples is acceptable for samples with a concentration above 20 nM (DIP) and 400 nM (nitrate and nitrite), which represent most of the previously published data from intermediate and deep waters from the Levantine basin. However for concentrations lower than this, which in practice

Ammonium, nitrate and urea uptake rates, as well as the release of dissolved organic nitrogen (DON), were measured along a latitudinal transect (49°N–33°S) in the Atlantic Ocean during the 2000 summer-to-autumn transition period and subsequently related to phytoplankton-size structure. The transect included upwelling areas, oligotrophic gyres and temperate regions and, as a result, euphotic layer integrated uptake rates of ammonium (18–355 mmol m−2 h−1), nitrate (5–2830 mmol m−2 h−1) and urea (26–1212 mmol m−2 h−1) covered a wide productivity spectrum. The percentage of extracellular DON release relative to total nitrogen uptake (PER) was highest when nitrate was used as the substrate (mean 42±SE 3.1%) whereas the uptake of ammonium and urea produced lower PER values (22±2.1 and 26±1.6%, respectively). Small cells (