Surface Ocean pCO2 Research Articles

A Numerical reassessment of the Gulf of Mexico carbon system in connection with the Mississippi River and global ocean

Abstract. Coupled physical–biogeochemical models can fill the spatial and temporal gap in ocean carbon observations. Challenges of applying a coupled physical–biogeochemical model in the regional ocean include the reasonable prescription of carbon model boundary conditions, lack of in situ observations, and the oversimplification of certain biogeochemical processes. In this study, we applied a coupled physical–biogeochemical model (Regional Ocean Modelling System, ROMS) to the Gulf of Mexico (GoM) and achieved an unprecedented 20-year high-resolution (5 km, 1/22∘) hindcast covering the period of 2000 to 2019. The biogeochemical model incorporated the dynamics of dissolved organic carbon (DOC) pools and the formation and dissolution of carbonate minerals. The biogeochemical boundaries were interpolated from NCAR's CESM2-WACCM-FV2 solution after evaluating the performance of 17 GCMs in the GoM waters. Model outputs included carbon system variables of wide interest, such as pCO2, pH, aragonite saturation state (ΩArag), calcite saturation state (ΩCalc), CO2 air–sea flux, and carbon burial rate. The model's robustness is evaluated via extensive model–data comparison against buoys, remote-sensing-based machine learning (ML) products, and ship-based measurements. A reassessment of air–sea CO2 flux with previous modeling and observational studies gives us confidence that our model provides a robust and updated CO2 flux estimation, and NGoM is a stronger carbon sink than previously reported. Model results reveal that the GoM water has been experiencing a ∼ 0.0016 yr−1 decrease in surface pH over the past 2 decades, accompanied by a ∼ 1.66 µatm yr−1 increase in sea surface pCO2. The air–sea CO2 exchange estimation confirms in accordance with several previous models and ocean surface pCO2 observations that the river-dominated northern GoM (NGoM) is a substantial carbon sink, and the open GoM is a carbon source during summer and a carbon sink for the rest of the year. Sensitivity experiments are conducted to evaluate the impacts of river inputs and the global ocean via model boundaries. The NGoM carbon system is directly modified by the enormous carbon inputs (∼ 15.5 Tg C yr−1 DIC and ∼ 2.3 Tg C yr−1 DOC) from the Mississippi–Atchafalaya River System (MARS). Additionally, nutrient-stimulated biological activities create a ∼ 105 times higher particulate organic matter burial rate in NGoM sediment than in the case without river-delivered nutrients. The carbon system condition of the open ocean is driven by inputs from the Caribbean Sea via the Yucatan Channel and is affected more by thermal effects than biological factors.

Variability in the Global Ocean Carbon Sink From 1959 to 2020 by Correcting Models With Observations

AbstractThe ocean reduces human impact on the climate by absorbing and sequestering CO2. From 1950s to the 1980s, observations of pCO2 and related ocean carbon variables were sparse and uncertain. Thus, global ocean biogeochemical models (GOBMs) have been the basis for quantifying the ocean carbon sink. The LDEO‐Hybrid Physics Data product (LDEO‐HPD) interpolates sparse surface ocean pCO2 data to global coverage by using GOBMs as priors, and applying machine learning to estimate full‐coverage corrections. The largest component of the GOBM corrections are climatological. This is consistent with recent findings of large seasonal discrepancies in GOBMs, but contrasts the long‐held view that interannual variability is a major source of GOBM error. This supports extension of the LDEO‐HPD pCO2 product back to 1959, using a climatology of model‐observation misfits prior to 1982. Consistent with previous studies for 1980 onward, air‐sea CO2 fluxes for 1959–2020 demonstrate response to atmospheric pCO2 growth and volcanic eruptions.

Impact of coastal upwelling dynamics on the pCO2 variability in the southeastern Arabian Sea

The southeastern Arabian Sea (SEAS) experiences moderate to weak upwelling along the southwest coast of India during the southwest monsoon (SM). The coastal upwelling initiates from mid-May along the southern-most tip of India, and with the advancement of SM, it propagates towards the northern latitude. This study examines the impact of coastal upwelling dynamics on the spatio-temporal variability of pCO2 in SEAS. It aims to identify the factors controlling the variability of pCO2 using high-resolution, regional ocean-ecosystem model-simulated outputs while available in situ and ship-based observations are used to establish the capability of the model. The cold deeper water that rises to the surface during upwelling decreases surface ocean pCO2 by 50 ± 2.4 µatm, whereas the presence of carbon-rich upwelling waters to a significantly shallower depth increases pCO2 by 52 ± 1.5 µatm of SEAS during SM. The salinity component increases surface ocean pCO2 by 2.0 ± 1.6 µatm. It has relatively less impact when compared with the individual effects of temperature and dissolved inorganic carbon components in controlling surface ocean pCO2 variability. The biological activities are profound only in the southern part of SEAS where biology-driven changes decrease surface ocean pCO2 by 4.0 ± 0.4 µatm during SM. The total biology-driven changes consist of both soft and hard tissues decrease the pCO2 level of SEAS by 2.0 ± 0.2 µatm during SM. Therefore, the upwelling-driven physical dynamics dominate the biological processes in controlling the spatio-temporal variability of surface ocean pCO2 in SEAS during SM, which contrasts the physical-biological dynamics of the sea east of Sri Lanka, where the biological processes dominate over physical dynamics in controlling surface ocean pCO2 variability.

Autonomous Wintertime Observations of Air‐Sea Exchange in the Gulf Stream Reveal a Perfect Storm for Ocean CO2 Uptake

AbstractA scarcity of wintertime observations of surface ocean carbon dioxide partial pressure (pCO2) in and near the Gulf Stream creates uncertainty in the magnitude of the regional carbon sink and its controlling mechanisms. Recent observations from an Uncrewed Surface Vehicle (USV), outfitted with a payload to measure surface ocean and lower atmosphere pCO2, revealed sharp gradients in ocean pCO2 across the Gulf Stream. Surface ocean pCO2 was lower by ∼50 μatm relative to the atmosphere in the subtropical mode water (STMW) formation region. This undersaturation combined with strong wintertime winds allowed for rapid ocean uptake of CO2, averaging −11.5 mmol m−2 day−1 during the February 2019 USV mission. The unique timing of this mission revealed active STMW formation. The USV proved to be a useful tool for CO2 flux quantification in the poorly observed, dynamic western boundary current environment.

Growth rate rather than temperature affects the B∕Ca ratio in the calcareous red alga <i>Lithothamnion corallioides</i>

Abstract. The B/Ca ratio in calcareous marine species is informative of past seawater CO32- concentrations, but scarce data exist on B/Ca in coralline algae. Recent studies suggest influences of temperature and growth rates on B/Ca, the effect of which could be critical for the reconstructions of surface ocean pH and atmospheric pCO2. In this paper, we present the first laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) analyses of Mg, Sr, Li, and B in the coralline alga Lithothamnion corallioides collected from different geographic settings and depths across the Mediterranean Sea and in the Atlantic Ocean. We produced the first data on putative temperature proxies (Mg/Ca, Li/Ca, Sr/Ca, Mg/Li) and B/Ca in a coralline algal species grown in different basins from across the photic zone (12, 40, 45, and 66 m depth). We tested the B/Ca correlation with temperature proxies and growth rates in order to evaluate their possible effect on B incorporation. Our results suggested a growth rate influence on B/Ca, which was evident in the sample with the lowest growth rate of 0.10 mm yr−1 (Pontian Isl., Italy; 66 m depth) and in Elba (Italy; 45 m depth), where the algal growth rate was the highest (0.14 mm yr−1). At these two sites, the measured B/Ca was the lowest at 462.8 ± 49.2 µmol mol−1 and the highest at 757.7 ± 75.5 µmol mol−1, respectively. A positive correlation between B/Ca and temperature proxies was found only in the shallowest sample from Morlaix (Atlantic coast of France; 12 m depth), where the amplitude of temperature variation (ΔT) was the highest (8.9 ∘C). Still, fluctuations in B/Ca did not mirror yearly seasonal temperature oscillations as for Mg/Ca, Li/Ca, and Sr/Ca. We concluded that growth rates, triggered by the different ΔT and light availability across depth, affect the B incorporation in L. corallioides.

Reconstruction of global surface ocean <i>p</i>CO<sub>2</sub> using region-specific predictors based on a stepwise FFNN regression algorithm

Abstract. Various machine learning methods were attempted in the global mapping of surface ocean partial pressure of CO2 (pCO2) to reduce the uncertainty of the global ocean CO2 sink estimate due to undersampling of pCO2. In previous research, the predictors of pCO2 were usually selected empirically based on theoretic drivers of surface ocean pCO2, and the same combination of predictors was applied in all areas except where there was a lack of coverage. However, the differences between the drivers of surface ocean pCO2 in different regions were not considered. In this work, we combined the stepwise regression algorithm and a feed-forward neural network (FFNN) to select predictors of pCO2 based on the mean absolute error in each of the 11 biogeochemical provinces defined by the self-organizing map (SOM) method. Based on the predictors selected, a monthly global 1∘ × 1∘ surface ocean pCO2 product from January 1992 to August 2019 was constructed. Validation of different combinations of predictors based on the Surface Ocean CO2 Atlas (SOCAT) dataset version 2020 and independent observations from time series stations was carried out. The prediction of pCO2 based on region-specific predictors selected by the stepwise FFNN algorithm was more precise than that based on predictors from previous research. Applying the FFNN size-improving algorithm in each province decreased the mean absolute error (MAE) of the global estimate to 11.32 µatm and the root mean square error (RMSE) to 17.99 µatm. The script file of the stepwise FFNN algorithm and pCO2 product are distributed through the Institute of Oceanology of the Chinese Academy of Sciences Marine Science Data Center (IOCAS, https://doi.org/10.12157/iocas.2021.0022, Zhong, 2021.

Improved Quantification of Ocean Carbon Uptake by Using Machine Learning to Merge Global Models and pCO2 Data

AbstractThe ocean plays a critical role in modulating climate change by sequestering CO2 from the atmosphere. Quantifying the CO2 flux across the air‐sea interface requires time‐dependent maps of surface ocean partial pressure of CO2 (pCO2), which can be estimated using global ocean biogeochemical models (GOBMs) and observational‐based data products. GOBMs are internally consistent, mechanistic representations of the ocean circulation and carbon cycle, and have long been the standard for making spatio‐temporally resolved estimates of air‐sea CO2 fluxes. However, there are concerns about the fidelity of GOBM flux estimates. Observation‐based products have the strength of being data‐based, but the underlying data are sparse and require significant extrapolation to create global full‐coverage flux estimates. The Lamont Doherty Earth Observatory‐Hybrid Physics Data (LDEO‐HPD) pCO2 product is a new approach to estimating the temporal evolution of surface ocean pCO2 and air‐sea CO2 exchange. LDEO‐HPD uses machine learning to merge high‐quality observations with state‐of‐the‐art GOBMs. We train an eXtreme Gradient Boosting (XGB) algorithm to learn a non‐linear relationship between model‐data mismatch and observed predictors. GOBM fields are then corrected with the predicted model‐data misfit to estimate real‐world pCO2 for 1982–2018. The resulting reconstruction by LDEO‐HPD is in better agreement with independent pCO2 observations than other currently available observation‐based products. Within uncertainties, LDEO‐HPD global ocean uptake of CO2 agrees with other products and the Global Carbon Budget 2020.

Seasonal Variations and Drivers of Surface Ocean pCO2 in the Seasonal Ice Zone of the Eastern Indian Sector, Southern Ocean

AbstractTo quantitatively assess the inorganic carbon cycle in the eastern Indian sector of the Southern Ocean (80–150°E, south of 60°S), we measured ocean surface temperature, salinity, total alkalinity (TA), the partial pressure of carbon dioxide (pCO2), and concentrations of chlorophyll‐a (chl a), dissolved inorganic carbon (DIC), and nutrients during the KY18 survey (December 2018–January 2019). The sea–air CO2 flux in this region was −8.3 ± 12.7 mmol m−2 day−1 (−92.1 to +10.6 mmol m−2 day−1). The ocean was therefore a weak CO2 sink. Based on the DIC and TA in the temperature minimum layer, we estimated the change of pCO2 from winter to summer (δpCO2) due to changes in water temperature, salinity, and biological activity (photosynthesis). The spatial distribution of pCO2 in the western part (80–110°E) of the study area was mainly driven by biological activity, which decreased pCO2 from December to early January, and in the eastern part (110–150°E) by temperature, which increased pCO2 from January to February. We also examined the changes in the CO2 concentrations (xCO2) over time by comparing data from 1996 with our data (2018–2019). The oceanic and atmospheric xCO2 increased by 23 and 45 ppm in 23 years, respectively. These changes of ocean xCO2 were mainly driven by an increase in CO2 uptake from the atmosphere as a result of the rise in atmospheric xCO2 and increase in biological activity associated with the change in the water‐mass distribution.

Seasonal to decadal spatiotemporal variations of the global ocean carbon sink.

The global ocean has absorbed approximately 30% of anthropogenic CO2 since the beginning of the industrial revolution. However, the spatiotemporal evolution of this important global carbon sink varies substantially on all timescales and has not yet been well evaluated. Here, based on a reconstructed observation‐based product of surface ocean pCO2 and air–sea CO2 flux (the MPI‐SOMFFN method), we investigated seasonal to decadal spatiotemporal variations of the ocean CO2 sink during the past three decades using an adaptive data analysis method. Two predominant variations are modulated annual cycles and decadal fluctuations, which account for approximately 46% and 25% of all extracted components, respectively. Although the whole summer to non‐summer seasonal difference pattern is determined by the Southern Ocean, the non‐summer CO2 sink at mid‐latitudes in both hemispheres shows an increasing trend (a total increase of approximately 1.0 PgC during the period 1982–2019), while it is relatively stable in summer. On decadal timescales for the global ocean carbon sink, unlike the weakening decade (1990–1999) and the reinvigoration decade (2000–2009) in which the Southern Ocean plays the dominant role, the reinforcement decade (2010–2019) is mainly the result from the weakening source effect in the equatorial Pacific Ocean. Our results suggest that except for the Southern Ocean's role in the global ocean carbon sink, the strengthening non‐summer's sink at mid‐latitudes in both hemispheres and the decadal or longer timescales of equatorial Pacific Ocean dynamics should be fully considered in understanding the oceanic carbon cycle on a global scale.

Seasonal cycle of surface ocean pCO2 and pH in the northern Indian Ocean and their controlling factors

This paper examines the seasonal variability of surface ocean pCO2and pH in the northern Indian Ocean. It aims to identify their controlling factors using a high-resolution, regional ocean-ecosystem model simulation and available surface ocean carbon observations. The seasonal variability of pCO2 and pH of the northern Indian Ocean is attributed to the changes in surface temperature (T), dissolved inorganic carbon (DIC), total alkalinity (ALK), and salinity (S). The western Arabian Sea has an enormous seasonal variance of pCO2 due to coastal upwelling dynamics. In contrast, the seasonal variance of pCO2 in the Bay of Bengal is governed by the upper ocean mixed layer dynamics, albeit with smaller amplitudes. The contribution of T (DIC) in the seasonal variance of pCO2 and pH at the western Arabian Sea is ±90 (∓100) μatm and ±0.18 (∓0.20) pH units, respectively. The contribution of ALK and S is complementary to each other in inducing seasonal variances of pCO2 and pH in the southeastern Arabian Sea with a magnitude of ±5∼10 μatm and ±0.02 pH units, respectively. In the northern Bay of Bengal, salinity plays a significant role in controlling seasonal variability of pCO2 and pH with amplitudes of roughly ±20 µatm and ±0.18 pH units, respectively, along the pathways of freshwater spreading. The maxima of seasonal amplitudes of pCO2 and pH occur during April-May and August-September for both basins. This study highlights the hotspots of pCO2 and pH in the northern Indian Ocean, where observational efforts to monitor pCO2 and pHbecome an essential requirement.

Observation system simulation experiments in the Atlantic Ocean for enhanced surface ocean <i>p</i>CO<sub>2</sub> reconstructions

Abstract. To derive an optimal observation system for surface ocean pCO2 in the Atlantic Ocean and the Atlantic sector of the Southern Ocean, 11 observation system simulation experiments (OSSEs) were completed. Each OSSE is a feedforward neural network (FFNN) that is based on a different data distribution and provides ocean surface pCO2 for the period 2008–2010 with a 5 d time interval. Based on the geographical and time positions from three observational platforms, volunteering observing ships, Argo floats and OceanSITES moorings, pseudo-observations were constructed using the outputs from an online-coupled physical–biogeochemical global ocean model with 0.25∘ nominal resolution. The aim of this work was to find an optimal spatial distribution of observations to supplement the widely used Surface Ocean CO2 Atlas (SOCAT) and to improve the accuracy of ocean surface pCO2 reconstructions. OSSEs showed that the additional data from mooring stations and an improved coverage of the Southern Hemisphere with biogeochemical ARGO floats corresponding to least 25 % of the density of active floats (2008–2010) (OSSE 10) would significantly improve the pCO2 reconstruction and reduce the bias of derived estimates of sea–air CO2 fluxes by 74 % compared to ocean model outputs.

Analysis of Environmental Controls on the Quasi-Ocean and Ocean CO2 Concentration by Two Intelligent Algorithms

Previous studies have demonstrated CO2 absorption by soils in arid regions, where the absorbed CO2 is conjectured to be finally sequestrated in the “subterranean ocean”—groundwater. This study compares environmental controls of ocean CO2 concentration (surface ocean pCO2) and quasi-ocean CO2 concentration (deep-soil pCO2). We aim to explore the latent relationship, both linear and nonlinear between the environmental variables, and CO2 concentration, utilizing two intelligent algorithms—the partial least linear regression (PLSR) algorithm and the artificial neural network (ANN) algorithm. For quasi-ocean CO2 concentration, RPD <1.4 and R2 <40%. While for ocean CO2 concentration, RPD >1.4 and R2 is 99.7%. Linear relationships between the considered environmental controls and ocean CO2 concentration are proved; however, there is no evident relationship between most of the considered environmental controls and quasi-ocean CO2 concentration. Groundwater level is proved to be a relatively important environmental control on the quasi-ocean CO2 concentration, suggesting groundwater discharge/recharge as a significant modulator of soil CO2 absorption in arid regions.

An observing system simulation experiment for Indian Ocean surface pCO2 measurements

An observing system simulation experiment (OSSE) is conducted to identify potential locations for making surface ocean pCO2 measurements in the Indian Ocean using the Bayesian Inversion method. As of the SOCATv3 release, the pCO2 data is limited in the Indian Ocean. To improve our modeling of this region, we need to identify where and what observation systems would produce the most good or benefit for their cost. The potential benefits of installing pCO2 sensors in the existing RAMA and OMNI moorings of the Indian Ocean, the potential of Bio-Argo floats (with pH measurements), and the implementation of the ship of opportunity program (SOOP) for underway sampling of pCO2 are evaluated. A cost function of dissolved inorganic carbon as a model state vector and CO2 flux mismatch as the source of error is minimized, and the basin-wide CO2 flux uncertainty reduction is estimated for different seasons. The maximum flux uncertainty reduction achievable by installing pCO2 sensors in the existing RAMA and OMNI moorings is limited to 30% during different seasons. One may consider that around 20 Bio-Argos are still the right choice over installing mooring based pCO2 sensors and achieve uncertainty reduction up to 50% with additional benefit of profiling the sub-surface upto 1000–2000 m. However, a single track SOOP has the potential to reduce the uncertainty by approximately 62%. This study identifies vital RAMA and OMNI moorings and SOOP tracks for observing Indian Ocean pCO2.Plain Language Summary.Surface ocean partial pressure of CO2 (pCO2) information is vital for estimating sea-to-air CO2 exchanges. This parameter is least available from the Indian Ocean as compared to other global tropical and southern oceans. There has been no effort made so far to measure surface ocean pCO2 in the Indian Ocean with routine monitoring such as by mounting instruments to moorings or by underway sampling via any ship of opportunity program. Therefore there is a considerable demand to start pCO2 observations in the Indian Ocean. However, one key question that emerges is where to deploy pCO2 instruments in the Indian Ocean to learn the most with limited resources. This study addresses this question with inverse modeling techniques. The study finds that the existing moorings of the Indian Ocean are capable of hosting pCO2 sensors, and data from those are useful to reduce the uncertainty in the surface sea-to-air CO2 flux estimation by a quarter magnitude. In contrast, the Bio-Argo floats with pH sensors, and the ship of opportunity underway sampling of pCO2 may benefit from reducing the same up to 50% and 62%, respectively.

Quantifying the Atmospheric CO2 Forcing Effect on Surface Ocean pCO2 in the North Pacific Subtropical Gyre in the Past Two Decades

Despite the well-recognized importance in understanding the long term impact of anthropogenic release of atmospheric CO2 (its partial pressure named as pCO2air) on surface seawater pCO2 (pCO2sw), it has been difficult to quantify the trends or changing rates of pCO2sw driven by increasing atmospheric CO2 forcing (pCO2swatm_forced) due to its combination with the natural variability of pCO2sw (pCO2swnat_forced) and the requirement of long time series data records. Here, using a novel satellite-based pCO2sw model with inputs of ocean color and other ancillary data between 2002 and 2019, we address this challenge for a mooring station at the Hawaii Ocean Time-series Station in the North Pacific subtropical gyre. Specifically, using the developed pCO2sw model, we differentiated and separately quantified the interannual-decadal trends of pCO2swnat_forced and pCO2swatm_forced. Between 2002 and 2019, both pCO2sw and pCO2air show significant increases at rates of 1.7 ± 0.1 μatm yr–1 and 2.2 ± 0.1 μatm yr–1, respectively. Correspondingly, the changing rate in pCO2swnat_forced is mainly driven by large scale forcing such as Pacific Decadal Oscillation, with a negative rate (-0.5 ± 0.2 μatm yr–1) and a positive rate (0.6 ± 0.3 μatm yr–1) before and after 2013. The pCO2swatm_forced shows a smaller increasing rate of 1.4 ± 0.1 μatm yr–1 than that of the modeled pCO2sw, varying in different time intervals in response to the variations in atmospheric pCO2. The findings of decoupled trends in pCO2swatm_forced and pCO2swnat_forced highlight the necessity to differentiate the two toward a better understanding of the long term oceanic absorption of anthropogenic CO2 and the anthropogenic impact on the changing surface ocean carbonic chemistry.

Role of river discharge and warming on ocean acidification and pCO2 levels in the Bay of Bengal

Shifts in surface ocean pCO2 and pH are important controls governing global climate. Based on the linear relationship of observed surface pH and pCO2 with sea surface temperature (SST), sea surface salinity (SSS) and Chlorophyll-a (Chl-a) multiple linear regression equations were developed. Based on remote sensing SST, Chl-a and model-derived SSS, pH and pCO2 data were derived from 1998 to 2015. Overall warming of BoB is noticed at the rate of 0.004° to 0.03 °C/y whereas cooling is found in the northwestern BoB during winter and spring seasons associated with an increase in atmospheric dust. Decrease in SSS is noticed during all seasons due to melting of Himalayan ice cover associated with increase in fresh water flux due to increase in atmospheric temperature. Increase in pH is observed in the eastern and southern Bay during all seasons associating with warming and decrease in salinity. In contrast, decrease in pH (−0.001 y−1) and pCO2 increase (+0.1 to +0.7 µatm y−1) is noticed in the western and head Bay during winter and spring seasons due to deposition of atmospheric pollutants. This study suggests that increase in freshwater input due to melting of Himalayan ice cover and deposition of atmospheric pollutants are dominant controlling factors on surface ocean pH and pCO2 in the BoB between 1998 and 2015 and this region is acting as a stronger sink for the atmospheric CO2 in the present than that in the past two decades. The global coastal regions are significantly influenced by river discharge and atmospheric deposition of pollutants and they are not part of the global models leading to ill-reproduction of seasonal variability in pH and pCO2. Inclusion of these processes may improve prediction of pH and pCO2 in the regions heavily influenced by discharge/deposition from land and atmosphere.

Consistency and Challenges in the Ocean Carbon Sink Estimate for the Global Carbon Budget

Based on the 2019 assessment of the Global Carbon Project, the ocean took up on average, 2.5+/-0.6PgCyr-1 or 23+/-5% of the total anthropogenic CO2 emissions over the decade 2009-2018. This sink estimate is based on global ocean biogeochemical models (GOBMs) and is compared to data-products based on surface ocean pCO2 (partial pressure of CO2) observations accounting for the outgassing of river-derived CO2. Here we evaluate the GOBM simulations by comparing the simulated pCO2 to observations. The simulations are well suited for quantifying the global ocean carbon sink on the time-scale of the annual mean and its multi-decadal trend (RMSE <20 μatm), as well as on the time-scale of multi-year variability (RMSE <10 μatm), despite the large model-data mismatch on the seasonal time-scale (RMSE of 20-80 μatm). Biases in GOBMs have a small effect on the global mean ocean sink (0.05 PgC yr−1), but need to be addressed to improve the regional budgets and model-data comparison. Accounting for non-mapped areas in the data-products reduces their spread as measured by the standard deviation by a third. There is growing evidence and consistency among methods with regard to the patterns of the multi-year variability of the ocean carbon sink, with a global stagnation in the 1990s and an extra-tropical strengthening in the 2000s. GOBMs and data-products point consistently to a shift from a tropical CO2 source to a CO2 sink in recent years. On average, the GOBMs reveal less variations in the sink than the data-based products. Despite the reasonable simulation of surface ocean pCO2 by the GOBMs, there are discrepancies between the resulting sink estimate from GOBMs and data-products. These discrepancies are within the uncertainty of the river flux adjustment, increase over time, and largely stem from the Southern Ocean. Progress in our understanding of the global ocean carbon sink necessitates significant advancement in modelling and observing the Southern Ocean including (i) a game-changing increase in high-quality pCO2 observations, and (ii) a critical re-evaluation of the regional river flux adjustment.

The ECCO‐Darwin Data‐Assimilative Global Ocean Biogeochemistry Model: Estimates of Seasonal to Multidecadal Surface Ocean pCO2 and Air‐Sea CO2 Flux

AbstractQuantifying variability in the ocean carbon sink remains problematic due to sparse observations and spatiotemporal variability in surface ocean pCO2. To address this challenge, we have updated and improved ECCO‐Darwin, a global ocean biogeochemistry model that assimilates both physical and biogeochemical observations. The model consists of an adjoint‐based ocean circulation estimate from the Estimating the Circulation and Climate of the Ocean (ECCO) consortium and an ecosystem model developed by the Massachusetts Institute of Technology Darwin Project. In addition to the data‐constrained ECCO physics, a Green's function approach is used to optimize the biogeochemistry by adjusting initial conditions and six biogeochemical parameters. Over seasonal to multidecadal timescales (1995–2017), ECCO‐Darwin exhibits broad‐scale consistency with observed surface ocean pCO2 and air‐sea CO2 flux reconstructions in most biomes, particularly in the subtropical and equatorial regions. The largest differences between CO2 uptake occur in subpolar seasonally stratified biomes, where ECCO‐Darwin results in stronger winter uptake. Compared to the Global Carbon Project OBMs, ECCO‐Darwin has a time‐mean global ocean CO2 sink (2.47 ± 0.50 Pg C year−1) and interannual variability that are more consistent with interpolation‐based products. Compared to interpolation‐based methods, ECCO‐Darwin is less sensitive to sparse and irregularly sampled observations. Thus, ECCO‐Darwin provides a basis for identifying and predicting the consequences of natural and anthropogenic perturbations to the ocean carbon cycle, as well as the climate‐related sensitivity of marine ecosystems. Our study further highlights the importance of physically consistent, property‐conserving reconstructions, as are provided by ECCO, for ocean biogeochemistry studies.

Precipitation‐Induced Reduction in Surface Ocean pCO2: Observations From the Eastern Tropical Pacific Ocean

AbstractDetermining air‐sea CO2 fluxes using pCO2 disequilibrium requires knowing the gas transfer velocity and air‐sea pCO2 difference. Most pCO2 measurements are made from ships, whose uncontaminated seawater intakes are located at >5‐m depth to prevent ingestion of air. However, there could be bias in determinations of CO2 fluxes if there is disagreement between pCO2 measurements at the surface and 5 m. Seawater measurements made at the near surface and 5 m in the eastern Equatorial Pacific Ocean show that precipitation can dilute surface seawater salinity and lower the pCO2 and dissolved inorganic carbon and raise pH of the same water, and that these changes in ocean chemistry are largely missed by measurements at 5 m. This finding implies that estimates of ocean CO2 uptake might be underestimated, since rain will lower surface ocean pCO2 in both source and sink regions and, hence, increase uptake in sink regions and decrease outgassing in source regions.

Estimating spatial and temporal variation in ocean surface pCO2 in the Gulf of Mexico using remote sensing and machine learning techniques

Research on the carbon cycle of coastal marine systems has been of wide concern recently. Accurate knowledge of the temporal and spatial distributions of sea-surface partial pressure (pCO2) can reflect the seasonal and spatial heterogeneity of CO2 flux and is, therefore, essential for quantifying the ocean's role in carbon cycling. However, it is difficult to use one model to estimate pCO2 and determine its controlling variables for an entire region due to the prominent spatiotemporal heterogeneity of pCO2 in coastal areas. Cubist is a commonly-used model for zoning; thus, it can be applied to the estimation and regional analysis of pCO2 in the Gulf of Mexico (GOM). A cubist model integrated with satellite images was used here to estimate pCO2 in the GOM, a river-dominated coastal area, using satellite products, including chlorophyll-a concentration (Chl-a), sea-surface temperature (SST) and salinity (SSS), and the diffuse attenuation coefficient at 490 nm (Kd-490). The model was based on a semi-mechanistic model and integrated the high-accuracy advantages of machine learning methods. The overall performance showed a root mean square error (RMSE) of 8.42 μatm with a coefficient of determination (R2) of 0.87. Based on the heterogeneity of environmental factors, the GOM area was divided into 6 sub-regions, consisting estuaries, near-shores, and open seas, reflecting a gradient distribution of pCO2. Factor importance and correlation analyses showed that salinity, chlorophyll-a, and temperature are the main controlling environmental variables of pCO2, corresponding to both biological and physical effects. Seasonal changes in the GOM region were also analyzed and explained by changes in the environmental variables. Therefore, considering both high accuracy and interpretability, the cubist-based model was an ideal method for pCO2 estimation and spatiotemporal heterogeneity analysis.

Ocean carbon pump decomposition and its application to CMIP5 earth system model simulations

The ocean stores 60 times as much carbon as the atmosphere, and the ocean carbon cycle has a critical role in controlling atmospheric CO2 concentration. The concept of ocean carbon pump is widely used for describing the ocean carbon cycle, but the term “ocean carbon pump” is not necessarily strictly defined and has been differently referred in previous studies. Here, using three dimensional distributions of dissolved inorganic carbon concentration, total alkalinity, phosphate, and salinity, four types of ocean carbon pump (organic matter, calcium carbonate, gas exchange, and freshwater flux pumps) were formulated. Although previously proposed decomposition emphasizes the enrichment in the deep ocean, my decomposition focuses on surface depletion which directly affects air-sea CO2 exchanges. Based on this decomposition, vector diagram for quantifying the individual roles of the pumps in controlling the ocean surface pCO2, which is in balance with atmospheric CO2 concentration, was demonstrated in this study. The method was applied to the observational climatology, and the contributions of the four carbon pump components to atmospheric CO2 were visualized in a single figure (the vector diagram); each carbon pump component was represented by one vector, and its contribution to CO2 concentration was measured from the difference in the contour values between the beginning and end of the vector. The same analysis was also applied to the CMIP5 earth system model simulations. All the models reproduced the same level of atmospheric CO2 concentration as the observation; however, the contributions from the four carbon pumps varied. The vector diagram was shown to quantify the differences in the contributions from the pumps between the models and against the observation. This study demonstrated that the proposed vector diagram analysis is a useful tool for quantifying the individual contributions of the ocean carbon pumps to atmospheric CO2 concentration and is helpful for evaluating the reproducibility of ocean carbon cycle models.