B. R. Carter

Detecting the Unexpected: A Research Framework for Ocean Acidification

The threat that ocean acidification (OA) poses to marine ecosystems is now recognized and U.S. funding agencies have designated specific funding for the study of OA. We present a research framework for studying OA that describes it as a biogeochemical event that impacts individual species and ecosystems in potentially unexpected ways. We draw upon specific lessons learned about ecosystem responses from research on acid rain, carbon dioxide enrichment in terrestrial plant communities, and nitrogen deposition. We further characterize the links between carbon chemistry changes and effects on individuals and ecosystems, and enumerate key hypotheses for testing. Finally, we quantify how U.S. research funding has been distributed among these linkages, concluding that there is an urgent need for research programs designed to anticipate how the effects of OA will reverberate throughout assemblages of species.

Climatological distribution of aragonite saturation state in the global oceans

Aragonite saturation state (Ωarag) in surface and subsurface waters of the global oceans was calculated from up-to-date (through the year of 2012) ocean station dissolved inorganic carbon (DIC) and total alkalinity (TA) data. Surface Ωarag in the open ocean was always supersaturated (Ωu2009>u20091), ranging between 1.1 and 4.2. It was above 2.0 (2.0–4.2) between 40°N and 40°S but decreased toward higher latitude to below 1.5 in polar areas. The influences of water temperature on the TA/DIC ratio, combined with the temperature effects on inorganic carbon equilibrium and apparent solubility product (K′sp), explain the latitudinal differences in surface Ωarag. Vertically, Ωarag was highest in the surface mixed layer. Higher hydrostatic pressure, lower water temperature, and more CO2 buildup from biological activity in the absence of air-sea gas exchange helped maintain lower Ωarag in the deep ocean. Below the thermocline, aerobic decomposition of organic matter along the pathway of global thermohaline circulation played an important role in controlling Ωarag distributions. Seasonally, surface Ωarag above 30° latitudes was about 0.06 to 0.55 higher during warmer months than during colder months in the open-ocean waters of both hemispheres. Decadal changes of Ωarag in the Atlantic and Pacific Oceans showed that Ωarag in waters shallower than 100u2009m depth decreased by 0.10u2009±u20090.09 (−0.40u2009±u20090.37%u2009yr−1) on average from the decade spanning 1989–1998 to the decade spanning 1998–2010.

An observing system simulation for Southern Ocean carbon dioxide uptake.

The Southern Ocean is critically important to the oceanic uptake of anthropogenic CO2. Up to half of the excess CO2 currently in the ocean entered through the Southern Ocean. That uptake helps to maintain the global carbon balance and buffers transient climate change from fossil fuel emissions. However, the future evolution of the uptake is uncertain, because our understanding of the dynamics that govern the Southern Ocean CO2 uptake is incomplete. Sparse observations and incomplete model formulations limit our ability to constrain the monthly and annual uptake, interannual variability and long-term trends. Float-based sampling of ocean biogeochemistry provides an opportunity for transforming our understanding of the Southern Ocean CO2 flux. In this work, we review current estimates of the CO2 uptake in the Southern Ocean and projections of its response to climate change. We then show, via an observational system simulation experiment, that float-based sampling provides a significant opportunity for measuring the mean fluxes and monitoring the mean uptake over decadal scales.

Enzyme-level interconversion of nitrate and nitrite in the fall mixed layer of the Antarctic Ocean

In the Southern Ocean, the nitrogen (N) isotopes of organic matter and the N and oxygen (O) isotopes of nitrate (NO3−) have been used to investigate NO3− assimilation and N cycling in the summertime period of phytoplankton growth, both today and in the past. However, recent studies indicate the significance of processes in other seasons for producing the annual cycle of N isotope changes. This study explores the impact of fall conditions on the 15N/14N (δ15N) and 18O/16O (δ18O) of NO3− and nitrite (NO2−) in the Pacific Antarctic Zone using depth profiles from late summer/fall of 2014. In the mixed layer, the δ15N and δ18O of NO3−u2009+u2009NO2− increase roughly equally, as expected for NO3− assimilation; however, the δ15N of NO3−-only (measured after NO2− removal) increases more than does NO3−-only δ18O. Differencing indicates that NO2− has an extremely low δ15N, oftenu2009<u2009−70‰ versus air. These observations are consistent with the expression of an equilibrium N isotope effect between NO3− and NO2−, likely due to enzymatic NO3−-NO2− interconversion. Specifically, we propose reversibility of the nitrite oxidoreductase (NXR) enzyme of nitrite oxidizers that, having been entrained from the subsurface during late summer mixed layer deepening, are inhibited by light. Our interpretation suggests a role for NO3−-NO2− interconversion where nitrifiers are transported into environments that discourage NO2− oxidation. This may apply to surface regions with upwelling, such as the summertime Antarctic. It may also apply to oxygen-deficient zones, where NXR-catalyzed interconversion may explain previously reported evidence of NO2− oxidation.

Mixing and remineralization in waters detrained from the surface into Subantarctic Mode Water and Antarctic Intermediate Water in the southeastern Pacific

A hydrographic data set collected in the region and season of Subantarctic Mode Water and Antarctic Intermediate Water (SAMW and AAIW) formation in the southeastern Pacific allows us to estimate the preformed properties of surface water detrained into these water masses from deep mixed layers north of the Subantarctic Front and Antarctic Surface Water south of the front. Using 10 measured seawater properties, we estimate: the fractions of SAMW/AAIW that originate as surface source waters, as well as fractions that mix into these water masses from subtropical thermocline water above and Upper Circumpolar Deep Water below the subducted SAMW/AAIW; ages associated with the detrained surface water; and remineralization and dissolution rates and ratios. The mixing patterns imply that cabbeling can account for ∼0.005–0.03 kg m−3 of additional density in AAIW, and ∼0–0.02 kg m−3 in SAMW. We estimate a shallow depth (∼300–700 m, above the aragonite saturation horizon) calcium carbonate dissolution rate of 0.4u2009±u20090.2 µmol CaCO3 kg−1 yr−1, a phosphate remineralization rate of 0.031u2009±u20090.009 µmol P kg−1 yr−1, and remineralization ratios of P:N:–O2:Corg of 1:(15.5u2009±u20090.6):(143u2009±u200910):(104u2009±u200922) for SAMW/AAIW. Our shallow depth calcium carbonate dissolution rate is comparable to previous estimates for our region. Our –O2:P ratio is smaller than many global averages. Our model suggests neglecting diapycnal mixing of preformed phosphate has likely biased previous estimates of –O2:P and Corg:P high, but that the Corg:P ratio bias may have been counteracted by a second bias in previous studies from neglecting anthropogenic carbon gradients.

Two decades of Pacific anthropogenic carbon storage and ocean acidification along Global Ocean Ship-based Hydrographic Investigations Program sections P16 and P02

A modified version of the extended multiple linear regression (eMLR) method is used to estimate anthropogenic carbon concentration (Canth) changes along the Pacific P02 and P16 hydrographic sections over the past two decades. P02 is a zonal section crossing the North Pacific at 30°N, and P16 is a meridional section crossing the North and South Pacific at ~150°W. The eMLR modifications allow the uncertainties associated with choices of regression parameters to be both resolved and reduced. Canth is found to have increased throughout the water column from the surface to ~1000u2009m depth along both lines in both decades. Mean column Canth inventory increased consistently during the earlier (1990s–2000s) and recent (2000s–2010s) decades along P02, at rates of 0.53u2009±u20090.11 and 0.46u2009±u20090.11u2009molu2009Cu2009m−2u2009a−1, respectively. By contrast, Canth storage accelerated from 0.29u2009±u20090.10 to 0.45u2009±u20090.11u2009molu2009Cu2009m−2u2009a−1 along P16. Shifts in water mass distributions are ruled out as a potential cause of this increase, which is instead attributed to recent increases in the ventilation of the South Pacific Subtropical Cell. Decadal changes along P16 are extrapolated across the gyre to estimate a Pacific Basin average storage between 60°S and 60°N of 6.1u2009±u20091.5u2009PgCu2009decade−1 in the earlier decade and 8.8u2009±u20092.2u2009PgCu2009decade−1 in the recent decade. This storage estimate is large despite the shallow Pacific Canth penetration due to the large volume of the Pacific Ocean. By 2014, Canth storage had changed Pacific surface seawater pH by −0.08 to −0.14 and aragonite saturation state by −0.57 to −0.82.A modified version of the extended multiple linear regression (eMLR) method is used to estimate anthropogenic carbon concentration (Canth) changes along the Pacific P02 and P16 hydrographic sections over the past two decades. P02 is a zonal section crossing the North Pacific at 30°N and P16 is a meridional section crossing the North and South Pacific at ~150°W. The eMLR modifications allow the uncertainties associated with choices of regression parameters to be both resolved and reduced. Canth is found to have increased throughout the water column from the surface to ~1000u2009m depth along both lines in both decades. Mean column Canth inventory increased consistently during the earlier (1990s-2000s) and recent (2000s-2010s) decades along P02, at rates of 0.53u2009±u20090.11 and 0.46u2009±u20090.11u2009mol C m-2 a˗1, respectively. By contrast, Canth storage accelerated from 0.29u2009±u20090.10 to 0.45u2009±u20090.11u2009mol C m˗2 a˗1 along P16. Shifts in water mass distributions are ruled out as a potential cause of this increase, which is instead attributed to recent increases in the ventilation of the South Pacific Subtropical Cell. Decadal changes along P16 are extrapolated across the gyre to estimate a Pacific Basin average storage between 60°S and 60°N of 6.1u2009±u20091.5 PgC decade˗1 in the earlier decade and 8.8u2009±u20092.2 PgC decade˗1 in the recent decade. This storage estimate is large despite the shallow Pacific Canth penetration due to the large volume of the Pacific Ocean. By 2014, Canth storage had changed Pacific surface seawater pH by ˗0.08 to ˗0.14 and aragonite saturation state by ˗0.57 to ˗0.82.

When can ocean acidification impacts be detected from decadal alkalinity measurements

We use a large initial condition suite of simulations (30 runs) with an Earth system model to assess the detectability of biogeochemical impacts of ocean acidification (OA) on the marine alkalinity distribution from decadally repeated hydrographic measurements such as those produced by the Global Ship-Based Hydrographic Investigations Program (GO-SHIP). Detection of these impacts is complicated by alkalinity changes from variability and long-term trends in freshwater and organic matter cycling and ocean circulation. In our ensemble simulation, variability in freshwater cycling generates large changes in alkalinity that obscure the changes of interest and prevent the attribution of observed alkalinity redistribution to OA. These complications from freshwater cycling can be mostly avoided through salinity normalization of alkalinity. With the salinity-normalized alkalinity, modeled OA impacts are broadly detectable in the surface of the subtropical gyres by 2030. Discrepancies between this finding and the finding of an earlier analysis suggest that these estimates are strongly sensitive to the patterns of calcium carbonate export simulated by the model. OA impacts are detectable later in the subpolar and equatorial regions due to slower responses of alkalinity to OA in these regions and greater seasonal equatorial alkalinity variability. OA impacts are detectable later at depth despite lower variability due to smaller rates of change and consistent measurement uncertainty.