Darren J. Pilcher

Timescales for detection of trends in the ocean carbon sink

The ocean has absorbed 41 per cent of all anthropogenic carbon emitted as a result of fossil fuel burning and cement manufacture. The magnitude and the large-scale distribution of the ocean carbon sink is well quantified for recent decades. In contrast, temporal changes in the oceanic carbon sink remain poorly understood. It has proved difficult to distinguish between air-to-sea carbon flux trends that are due to anthropogenic climate change and those due to internal climate variability. Here we use a modelling approach that allows for this separation, revealing how the ocean carbon sink may be expected to change throughout this century in different oceanic regions. Our findings suggest that, owing to large internal climate variability, it is unlikely that changes in the rate of anthropogenic carbon uptake can be directly observed in most oceanic regions at present, but that this may become possible between 2020 and 2050 in some regions.

Natural Variability and Anthropogenic Trends in the Ocean Carbon Sink

Since preindustrial times, the ocean has removed from the atmosphere 41% of the carbon emitted by human industrial activities. Despite significant uncertainties, the balance of evidence indicates that the globally integrated rate of ocean carbon uptake is increasing in response to increasing atmospheric CO2 concentrations. The El Niño-Southern Oscillation in the equatorial Pacific dominates interannual variability of the globally integrated sink. Modes of climate variability in high latitudes are correlated with variability in regional carbon sinks, but mechanistic understanding is incomplete. Regional sink variability, combined with sparse sampling, means that the growing oceanic sink cannot yet be directly detected from available surface data. Accurate and precise shipboard observations need to be continued and increasingly complemented with autonomous observations. These data, together with a variety of mechanistic and diagnostic models, are needed for better understanding, long-term monitoring, and future projections of this critical climate regulation service.

Physical and biogeochemical mechanisms of internal carbon cycling in Lake Michigan

The lakewide seasonal carbon cycle of Lake Michigan is poorly quantified and lacks the mechanistic links necessary to determine impacts upon it from eutrophication, invasive species, and climate change. A first step toward a full appreciation of Lake Michigans carbon cycle is to quantify the dominant mechanisms of its internal carbon cycle. To achieve this, we use the MIT general circulation model configured to the bathymetry of Lake Michigan and coupled to an ecosystem model to simulate the seasonal cycle of productivity, temperature, circulation, and the partial pressure of CO2 in water (pCO2). This biogeochemistry is designed to be appropriate for the prequagga mussel state of the lake. The primary mechanism behind the seasonal cycle of primary productivity is lake physics. The offshore spring phytoplankton bloom begins following a reduction in deep vertical mixing and ends with the depletion of nutrients via thermal stratification. The exception is the western shoreline, where summer winds drive coastal upwelling, providing hypolimnetic nutrients and generating significant productivity. Surface pCO2 is controlled by the net effect from temperature on solubility, and is modulated by biological uptake of dissolved inorganic carbon (DIC) and isothermal mixing of DIC-rich water in winter. Temperature tends to have the greatest seasonal impact in nearshore regions, while local DIC has the greatest impact in offshore regions. Lakewide, the model suggests that carbon is absorbed from the atmosphere during the spring bloom and released to the atmosphere during winter mixing and when summer surface temperatures are at their maximum.

Phytoplankton size impact on export flux in the global ocean

Efficiency of the biological pump of carbon to the deep ocean depends largely on biologically mediated export of carbon from the surface ocean and its remineralization with depth. Global satellite studies have primarily focused on chlorophyll concentration and net primary production (NPP) to understand the role of phytoplankton in these processes. Recent satellite retrievals of phytoplankton composition now allow for the size of phytoplankton cells to be considered. Here we improve understanding of phytoplankton size structure impacts on particle export, remineralization, and transfer. A global compilation of particulate organic carbon (POC) flux estimated from sediment traps and 234Th are utilized. Annual climatologies of NPP, percent microplankton, and POC flux at four time series locations and within biogeochemical provinces are constructed. Parameters that characterize POC flux versus depth (export flux ratio, labile fraction, and remineralization length scale) are fit for time series locations, biogeochemical provinces, and times of the year dominated by small and large phytoplankton cells where phytoplankton cell size show enough dynamic range over the annual cycle. Considering all data together, our findings support the idea of high export flux but low transfer efficiency in productive regions and vice versa for oligotrophic regions. However, when parsing by dominant size class, we find periods dominated by small cells to have both greater export flux efficiency and lower transfer efficiency than periods when large cells comprise a greater proportion of the phytoplankton community.

Assessing the abilities of CMIP5 models to represent the seasonal cycle of surface ocean pCO2

The ability of Earth System Models to accurately simulate the seasonal cycle of the partial pressure of CO2 in surface water ( pCO2SW) has important implications for projecting future ocean carbon uptake. Here we develop objective model skill score metrics and assess the abilities of 18 CMIP5 models to simulate the seasonal mean, amplitude, and timing of pCO2SW in biogeographically defined ocean biomes. The models perform well at simulating the monthly timing of the seasonal minimum and maximum of pCO2SW, but perform somewhat worse at simulating the seasonal mean values, particularly in polar and equatorial regions. The results also illustrate that a single “best” model can be difficult to determine, despite an analysis restricted to the seasonality of a single variable. Nonetheless, groups of models tend to perform better than others, with significant regional differences. This suggests that particular models may be better suited for particular regions, though we find no evidence for model tuning. Timing and amplitude skill scores display a weak positive correlation with observational data density, while the seasonal mean scores display a weak negative correlation. Thus, additional mapped pCO2SW data may not directly increase model skill scores; however, improved knowledge of the dominant mechanisms may improve model skill. Lastly, we find skill score variability due to internal model variability to be much lower than variability within the CMIP5 intermodel spread, suggesting that mechanistic model differences are primarily responsible for differences in model skill scores.

Modeled sensitivity of Lake Michigan productivity and zooplankton to changing nutrient concentrations and quagga mussels

The recent decline in Lake Michigan productivity is often attributed to filter feeding by invasive quagga mussels, but some studies also implicate reductions in lake-wide nutrient concentrations. We use a 3D coupled hydrodynamic - biogeochemical model to evaluate the effect of changing nutrient concentrations and quagga mussel filtering on phytoplankton production and phytoplankton and zooplankton biomass. Sensitivity experiments are used to assess the net effect of each change separately and in unison. Quagga mussels are found to have the greatest impact during periods of isothermal mixing, while nutrients have the greatest impact during thermal stratification. Quagga mussels also act to enhance spatial heterogeneity, particularly between nearshore-offshore regions. This effect produces a reversal in the gradient of nearshore-offshore productivity: from relatively greater nearshore productivity in the pre-quagga lake to relatively lesser nearshore productivity after quaggas. The combined impact of both processes drives substantial reductions in phytoplankton and zooplankton biomass, as well as significant modifications to the seasonality of surface water pCO2, particularly in nearshore regions where mussel grazing continues year-round. These results support growing concern that considerable losses of phytoplankton and zooplankton will yield concurrent losses at higher trophic levels. Comparisons to observed productivity suggest that both quagga mussel filtration and lower lakewide total phosphorus are necessary to accurately simulate recent changes in primary productivity in Lake Michigan.

The Importance of Freshwater to Spatial Variability of Aragonite Saturation State in the Gulf of Alaska

High latitude and subpolar regions like the Gulf of Alaska (GOA) are more vulnerable than equatorial regions to rising carbon dioxide (CO2) levels, in part due to local processes that amplify the global signal. Recent field observations have shown that the shelf of the GOA is currently experiencing seasonal corrosive events (carbonate mineral saturation states Ω, Ω<1), including suppressed Ω in response to ocean acidification as well as local processes like increased low alkalinity glacial melt water discharge. While the glacial discharge mainly influences the inner shelf, on the outer shelf, upwelling brings corrosive waters from the deep GOA. In this work, we develop a high-resolution model for carbon dynamics in the GOA, identify regions of high variability of Ω, and test the sensitivity of those regions to changes in the chemistry of glacial melt water discharge. Results indicate the importance of this climatically sensitive and relatively unconstrained regional freshwater forcing for Ω variability in the nearshore. The increase was nearly linear at 0.002 Ω per 100 µmol/kg increase in alkalinity in the freshwater runoff. We find that the local winds, biological processes, and freshwater forcing all contribute to the spatial distribution of Ω and identify which of these three is highly correlated to the variability in Ω. Given that the timing and magnitude of these processes will likely change during the next few decades, it is critical to elucidate the effect of local processes on the background ocean acidification signal using robust models, such as the one described here.