PP43E-01 INVITED
Controls on the Time Scale of Carbonate Neutralization of Carbon Dioxide Released to the Atmosphere
Once released to the atmosphere, carbon dioxide is removed on a range of time scales. On the time scale of years to centuries, carbon dioxide removal from the atmosphere is dominated by transport processes within the ocean. On the time scale of hundreds of thousands of years, carbon dioxide removal from the atmosphere is dominated by processes related to the weathering of silicate rocks on land. Between these time scales, carbon dioxide removal is dominated by interactions involving carbonate minerals both on land and in the sea. Net dissolution of carbonate minerals (on land or in the sea) increases ocean alkalinity to an extent that exceeds the amount of carbon addition; the result is a transfer of carbon from the atmosphere to the ocean and moderation of the effects of added carbon on ocean chemical parameters such as pH and carbonate mineral saturation. There has been some controversy over how fast equilibration with carbonate minerals can neutralize carbon acidity, with claims ranging from the extreme and untenable claim that this process is essentially instantaneous to more plausible claims that the equilibration time scale may approach 10 kyr. Even within the domain of informed discourse, estimates of the carbonate neutralization timescale can vary by an order-of-magnitude. Here, in an effort to understand the sources of the lack of consensus on this issue, we examine how various processes (e.g., ocean transport, sediment pore water diffusion, carbonate-mineral dissolution, and carbonate weathering on land) influence the time scale for carbonate neutralization of carbon dioxide releases to the atmosphere.
PP43E-02
Calcium Carbonate Dissolution in Sinking Particulate Matter
Atmospheric carbon dioxide is sequestered into the deep ocean via sinking particles that are produced in the euphotic zone. These particles are composed of organic carbon, calcium carbonate and silica that remineralizes as the particles sink through the water column. Calcium carbonate is a key component in ballasting the particles; according to model calculations it has the potential to determine to a large degree the strength of the biological pump. Understanding the mechanisms of calcite dissolution in the water column are important to the understanding of cycling of organic and inorganic material in the ocean. We have developed an idealized, mechanistic model of particles sinking to the deep ocean, in which settling velocity is calculated from first principles. Calcium carbonate (calcium + aragonite) dissolution is both biologically and thermodynamically mediated. Opal dissolution and organic carbon respiration are temperature dependent. Global compilations of sediment trap data are used to constrain poorly known parameters such as the CaCO3 dissolution constant and organic carbon respiration rates. Coupling our sinking particle model to a sediment diagenesis model, we are able to reproduce global burial and sea floor dissolution rates within a factor of 2. A higher export ratio has been hypothesized to lower the glacial pCO2 level. Stratification and circulation changes associated with global warming is likely to bring about a change in export production, thus possibly changing the export ratio (orgC/CaCO3) of material sinking out of the euphotic zone. We explore the response of the lysocline to changes in export ratio.
PP43E-03
Calcite Dissolution in Saline Waters
The specific effect of ionic strength on the reaction kinetics of calcite dissolution in intermediate to high ionic strength (0.5 < I < 6.0) solutions applicable to natural waters has been investigated using classical free-drift methods where all other parameters (mCa2+, PCO2, and T) have been held constant. Both phosphate-free solutions of potassium chloride (KCl) and sodium chloride (NaCl) as the dominant ionic strength determining salt were investigated where calcium concentrations were held constant in all solutions at approximately 0.010 molal. Reaction rates were found to vary significantly as a function of ionic strength of the reacting solution, which we suggest is due to the lowered activity of water with an increase in ionic strength which decreases the rate of cation hydration. When modeled with the general rate equation, R = k(1-Ømega)n, first-order kinetics (n=1) are sufficient to fit the experimental data. Furthermore, the rate constant (k) appears to be a function of the square root of the ionic strength of the reacting solution. These results may have potential applications to the response of the ocean to acidification by fossil fuel CO2 as well as CO2 sequestration in subsurface saline waters in calcium carbonate-hosted reservoirs.
PP43E-04 INVITED
A Cenozoic History of Atmospheric Carbon Dioxide
Much is assumed about atmospheric carbon dioxide and the climatic character of the Cenozoic, with long- and short-term global temperature changes commonly ascribed to parallel changes in pCO2. Recent pCO2 reconstructions, isotope, and sedimentological evidence for the early Paleogene suggest this relationship is largely valid. For example, the decline in pCO2 from the mid-Eocene to Oligocene tracks increasingly colder bottom-water and global temperatures, while the hyperthermals of the Eocene, such as the Paleocene-Eocene thermal maximum (PETM) and associated ocean acidification, undeniably resulted from massive inputs of carbon. However, there are times — during the Miocene (~24-5 Ma) and the Eocene- Oligocene climate transition (~34 Ma) — when the linkage between climate change and CO2 appears less robust, suggesting that climate sensitivity to CO2 changed over time or that ocean circulation and/or tectonics played increasingly important climatic roles. While the evidence indicates that large-scale changes in carbon dioxide occurred during the Cenozoic, parallel changes in the saturation state of the ocean with respect to calcium carbonate minerals are lacking unless the rate of CO2 change was extraordinarily rapid.
PP43E-05 INVITED
First observational evidence for long-term CO2-driven weathering feedback
Feedbacks controlling long-term carbon cycle fluxes and atmospheric CO2 are critical in stabilizing Earth's long-term climate. CO2 levels over millions of years are controlled by a CO2-driven weathering feedback that maintains a mass balance between CO2 input to the atmosphere from volcanism/metamorphism and net organic carbon oxidation, and CO2 removal from the atmosphere by silicate-rock weathering and subsequent burial of carbonate minerals. This view is frequently challenged by hypotheses avoiding the need for a mass balance and invoking other processes that affect CO2, including continental uplift and organic carbon burial. Hitherto, the prevailing conjectural argument in support of the mass balance theory was that flux imbalances would lead to untenable variations in atmospheric CO2 within a few million years. Here we provide the first observational evidence for a close mass balance of long-term carbon cycle fluxes, based on ice core CO2 data over the late Pleistocene and indicators of carbonate dissolution in the ocean. Our analysis shows that mean global atmospheric CO2 and Antarctic temperature have changed during the past 610~ky by at most ~\!22~ppmv and 1.5 to 2~K, respectively. Forcing of carbon cycle models with these data allows for a maximum imbalance of ca.\ 1-2% between CO2 supply and CO2 uptake by silicate weathering over 610~ky. This presupposes that only changes in long-term processes caused the average CO2 change - otherwise, even finer balances are possible. This shows that a tight balance exists between CO2 inputs to the atmosphere and continental weathering uptake on the time scale of several hundred thousand years, despite climate variations on shorter timescales. Our results provide strong support for a CO2-driven weathering feedback.
PP43E-06 INVITED
Effects of acidification on coccolithophore assemblages and on coccolith shapes
The coccolithophores are not only an important group of oceanic primary producers, they are also one of the main carbonate producers in the ocean. Because the calcium carbonate plates (coccoliths) secreted by these unicellular algae are so small and light (a few picograms each) they cannot be directly weighed, with the result that very little is known about the effect of primary production and dissolution on the weight of coccoliths. Using a our new method that allows a rapid estimate of the weight of discrete coccoliths, we analysed the effect of dissolution on coccoliths. We performed four acidification experiments on, (1) a surface sediment sample, (2) a laboratory culture of E. huxleyi, (3) a filtered water sample combining several depths in photic zone, (4) a sample composed of a mixture of several cultured coccolithophores species. The results of the first two acidification experiments, using cultured and fossil coccoliths, indicate that acidification has very little effect on the shape of coccoliths. Based on these experiments, we identify four morphological parameters for identifying the effects of dissolution on a coccolith assemblage. These parameters were applied in samples from two sediment trap time-series taken at the same tropical Atlantic site (EUMELI), but separated by 2000m water depth. There was no evidence of increased dissolution with water depth with from 250 and to 2500m. This is indicative of the absence of dissolution of coccoliths between the photic zone and the lysocline and of dissolution in photic zone. We tested this hypothesis by comparing coccolithophores sampled in photic zone with coccoliths settling below at the same site. We compare those results with those of the third dissolution experiment made with the same suite of samples. The last dissolution experiment, on cultured artificial assemblages permitted to strengthen the results. With this suite of experiment we are able to modelize the dissolution of a coccolithophore assemblage from the photic zone to the sea floor.
PP43E-07
Deep Pacific Carbonate Chemistry During the LGM and Deglaciation: Reconstructing ΔCO32- using foraminiferal Mg/Ca ratios.
Deep Pacific carbonate preservation reconstructions are at odds suggesting both enhanced preservation (Farell and Prell, 1989) and increased dissolution (Broecker and Clark, 2003). We present a reconstruction of the western equatorial Pacific deep-water carbonate ion profile for the Last Glacial Maximum (LGM) and a deglacial preservation event based on changes in the Mg/Ca ratio of planktonic foraminifera with increased water depth. Mg/Ca measurements were generated for three species commonly used in paleothermometry, G. ruber, G. sacculifer, and N. dutertrei, obtained from a suite of six cores spanning 1.6 to 4.4 km on the Ontong Java Plateau. Mg/Ca ratios obtained from the shallow core are used to reconstruct sea surface and thermocline temperatures. Changes in the Mg/Ca ratio with increased water depth are used to assess temporal changes in the deep-water carbonate saturation state and generate a paleo-carbonate ion profile. Our data confirm the strong dissolution effect on N. dutertrei Mg/Ca. The N. dutertrei Mg-derived LGM ΔCO32- profile suggests that preservation was enhanced in the shallowest three cores in comparison to modern. In the modern deep Pacific, there is little change in the carbonate ion concentration with increasing water depth; decreases in the calcite saturation state are primarily controlled by the pressure effect. Overall the carbonate ion concentration gradient in the deep Pacific is steeper during the LGM than today. Our reconstruction suggests that there were changes in both ocean carbonate chemistry and circulation. While the shape of our reconstructed profile is consistent with the gradient derived from shell weight data (Broecker and Clark, 2003), the carbonate ion concentrations our data imply are higher and, at least in the shallowest three cores, more consistent with reconstructions based on % carbonate (e.g. Farell and Prell, 1989). During the deglaciation at ~14 ky bp a preservation event occurred that is marked by increased Mg/Ca in N. dutertrei and the presence of pteropods in the shallowest three cores. During this interval, the data imply a whole ocean increase in ΔCO32- with enhanced preservation in all cores shallower than 4.0 km. The shape of the reconstructed gradient is similar to the modern gradient which suggests that by 14 ky bp the circulation in the Pacific was similar to modern circulation.
PP43E-08
Modern Calcium Carbonate Preservation in Equatorial Pacific Sediments in the Context of Late Pleistocene Glacial Cycles
The CaCO3 content of marine sediments in many regions of the ocean has varied systematically with climate throughout the late-Pleistocene glacial cycles. Both biological productivity and carbonate preservation have been proposed to be the master variable regulating this variability. We have evaluated the preserved flux of CaCO3 in cores from the central equatorial Pacific Ocean (140° W) using the 230Th - normalization technique. Neither barite fluxes nor 10Be /230Th ratios, both geochemical proxies for export production, correlate with CaCO3 fluxes, indicating that productivity is not the principal factor controlling CaCO3 accumulation in these sediments. Preserved fluxes of CaCO3 in central equatorial Pacific sediments correlate in time with the benchmark CaCO3 record from the Cape Basin (South Atlantic Ocean; Hodell et al., EPSL 192 (2001) 109-124), supporting the view that changes in ocean chemistry (carbonate ion concentration) have controlled the pattern of CaCO3 preservation and accumulation at these sites. Modern CaCO3 preservation in equatorial Pacific sediments has dropped to levels nearly as low as those experienced at any time in the late Pleistocene. Similar changes occurred at the end of each of the late- Pleistocene interglacial periods, from which we infer that ocean carbonate chemistry has already undergone changes that are expected to precede the transition into the next ice age. However, during the late Pleistocene, the time interval between the decrease in CaCO3 preservation and the end of the interglacial has varied substantially from one interglacial to another (from 2,000 to 15,000 years), so the late-Holocene decrease in CaCO3 preservation cannot be used to predict the end of the Holocene interglacial period.