PP53D-01 INVITED 13:40h
Cenozoic Climate Change and Carbon Cycling Processes
Model results and proxy data are presented to constrain the evolution of climate and pCO2 over the Cenozoic. Major controls on the partial pressure of CO2 in the atmosphere (pCO2) are discussed considering also the contribution from terrestrial and marine productivity and organic carbon burial. Moreover, feed-backs between carbon cycling and climate evolution are evaluated using different modeling approaches. Finally, the importance of greenhouse gases (CO2, H2O, CH4) versus albedo effects (ice cover, vegetation, cloud formation) is discussed in order to identify the major controls on Cenozoic climate change.
PP53D-02 INVITED 13:55h
The evolution of atmospheric carbon dioxide since the middle Eocene: a biomarker perspective
The carbon isotopic fractionation that occurs during marine photosynthetic carbon fixation ($\epsilon$$_{p}$) is primarily a function of surface-water [CO$_{2aq}$], growth rate, and cell geometry. Therefore, if temporal variations in growth rate and cell geometry are minimal or constrained, $\epsilon$$_{p}$ values provide the potential to evaluate paleo-pCO$_{2}$ concentrations and trends. Alkenone-based $\epsilon$$_{p}$ ($\epsilon$$_{p}$$_{37:2}$) records have been used to estimate atmospheric carbon dioxide concentrations during the Miocene ($\sim$24-5 Ma). These records provide evidence that pCO$_{2}$ was substantially lower than previously anticipated and suggest that changes in CO$_{2}$ played a secondary role in forcing global climate change during the Neogene. Alkenone $\delta$$^{13}$C values ($\delta$$_{37:2}$) for the middle Eocene to the late Oligocene were measured from six ocean locations encompassing a range of growth environments (DSDP sites 511, 513, 516, 612, and ODP site 803). Data from sites 511, 513, and 803 are limited but overlap other records. Site 516 (Southwest Atlantic Ocean) represents the most continuous record ranging from the middle Miocene to the earliest Oligocene. The Eocene portion of the record is derived largely from site 612 (Northwest Atlantic Ocean). Our results reveal a clear secular trend in $\delta$$_{37:2}$ from the early Miocene to the latest Eocene ($\sim$45 Ma), with $\delta$$_{37:2}$ values leveling off to $\sim$-32 to -33.4$\permil$ by the latest Eocene. By $\sim$25-30 Ma and older, $\delta$$_{37:2}$ values are more negative than the lowest values recorded from similar environments in the modern ocean. $\epsilon$$_{p}$$_{37:2}$ values were calculated using the $\delta$$^{13}$C values of aqueous CO$_{2}$ estimated from the $\delta$$^{13}$C of coeval planktonic foraminifera or modeled from the isotopic composition of the $<$60 um fraction. Surface temperatures were estimated from the $\delta$$^{18}$O compositions of shallow-dwelling foraminifera and/or modeled from the isotopic composition of the $<$60 um fraction. All carbonates were assumed to be influenced by secondary carbonate and thus estimated temperatures were increased by $3\deg$ to $6\deg$C to account for diagenesis. In general, $\epsilon$$_{p}$$_{37:2}$ values track $\delta$$_{37:2}$ reaching maximum values of $\sim$23$\permil$ by the late Eocene. Atmospheric carbon dioxide concentrations can be estimated if we apply the modern calibration for $\epsilon$$_{p}$$_{37:2}$ as a function of surface-water [PO$_{4}$$^{3-}$] and [CO$_{2aq}$] and assume that the range of paleo-[PO$_{4}$$^{3-}$] for each site was similar to modern distributions. These criteria lead to minimum estimates of pCO$_{2}$ that are highly dependent on our assumed ocean temperatures. Our results suggest that middle Eocene atmospheric carbon dioxide concentrations were $\sim$1000 to 1500 ppmv. Atmospheric carbon dioxide concentrations appear to rapidly decline following the Eocene/Oligocene boundary reaching modern values near the end of the Oligocene.
PP53D-03 INVITED 14:10h
A Cenozoic terrestrial isotope record and the evolution of C$_{4}$ photosynthesis
Our understanding of C$_{4}$ plant evolution and expansion has predominantly relied on site-specific fossil teeth, paleosols, and pedogenic carbonates carbon-isotope records and suggests a global dominance between 15-6 Ma. However, more recent techniques using bulk and compound-specific carbon-isotope ratios from terrestrial organic matter and other biomarker evidence suggest C$_{4}$ plants may have evolved multiple times. Furthermore, C$_{4}$ plants may have been present in terrestrial environments much earlier than the late Miocene expansion, but owing to their environmental preference and low preservation potential may not have been preserved in the terrestrial sedimentary record, and/or such latitudinal sites have not been fully explored. An additional implication is that the carbon-isotope composition of CO$_{2}$ (\delta $^{13}$C$_{CO2}$) has changed through time and paleoecologic reconstructions based on teeth and carbonate isotopic signatures may not reflect accurate floral contributions. Thus, terrestrial and atmospheric carbon-isotope signatures must be integrated in order to assess the Cenozoic history of C$_{4}$ photosynthesis. Presently, we are constructing a carbon-isotope record of long-chain {\it n}-alkanes with high carbon preference indices (indicative of higher plant input) from a globally distributed set of oligotropic and marginal DSDP/ODP marine sediments. As mentioned above, an estimate of the C$_{4}$ plant proportion of total land-plant biomass requires an understanding of changes in \delta $^{13}$C$_{CO2}$ through time. Accordingly, we have constrained this parameter by establishing the carbon-isotope composition of C$_{3}$ plant organic matter from Paleogene-age shallow marine shelf and lagoonal sediments from the Isle of Wight, UK, by assuming constant carbon-isotope discrimination between CO$_{2}$ and C$_{3}$ photosynthesis. Such \delta $^{13}$C$_{CO2}$ records can be directly compared with alkenone-based {\it p}CO$_{2}$ and {\it n}-alkane based floral contribution estimates. Using integrated isotopic proxies, our preliminary data suggest that the C$_{4}$ photosynthetic pathway evolved prior to the late Miocene and was possibly driven by paleoenvironmental pressures.
PP53D-04 INVITED 14:25h
How Reliable are Plants as CO$_{2}$ Proxies: Lessons From the Late Cenozoic
Stomatal frequency analysis is a biological proxy for palaeo-atmospheric CO$_{2}$ reconstructions. As a biological proxy it depends on the careful calibration of the plant response by growth experiments, field observations and comparative studies with other independent CO$_{2}$ archives (ice core and geochemical proxy records). The present contribution summarizes recent advancements in the development and validation of the method. Quantification techniques, the reproducibility of Holocene stomatal frequency records and the comparison with available ice core records are evaluated: (1) In order to address the potential influence of local environmental changes or methodological insufficiencies on the CO$_{2}$ reconstructions, multiple stomatal frequency records are compared for three climatic key periods during the Holocene. It appears that stomatal frequency records from different continents and plant species using different calibration techniques are reproducible. (2) Differences with Holocene ice-based CO$_{2}$ records are evaluated by applying a firn gas diffusion model to a stomatal frequency based CO$_{2}$ record. It allows to test the potential influence of smoothing during gas enclosure on the temporal resolution as well as the amplitude of CO$_{2}$ fluctuations. Apparently, the large discrepancies between ice core and stomatal frequency records diminish when the effect of natural smoothing of the ice core CO$_{2}$ record is simulated with the raw data obtained from stomatal frequency record. These results indicate that the differences derived by the two methods may be less significant than previously thought. (3) Having discussed some critical issues of the Holocene stomatal frequency records we will move to the Middle Miocene, a key period of Neogene climate evolution. In the marine record a marked positive \delta$^{13}$C shift between about 17.5 Ma and 13.5 Ma indicates enhanced biological productivity and burial of organic carbon, which in turn may have resulted into a drastic depletion in atmospheric CO$_{2}$ concentration and finally into global cooling (Monterey hypothesis). Multiple stomatal frequency records from Europe and N-America corroborate the predicted CO$_{2}$ drawdown during the period of the Monterey C isotope excursion.
PP53D-05 14:45h
Effects of Organic Carbon/Carbonate Burial Ratios and Biological Carbon Fixation on the Global Carbon Cycle Over the Past ~200 myr
The isotopic composition of the global carbon reservoir integrates large kinetic fractionations from photosynthesis with small thermodynamic fractionations from carbonate precipitation. We present concordant $\delta$$^{13}$C records of carbonates ($\delta$$^{13}$C$_{carb}$) and organic matter ($\delta$$^{13}$C$_{org}$), along with new carbonate (C$_{carb}$) and organic carbonate (C$_{org}$) fluxes for the past $\sim$205 myrs (Jurassic-Cenozoic) generated from bulk sediment samples from the Atlantic. The new $\delta$$^{13}$C$_{org}$ record greatly refines previous compilations (Hayes et al., 1999) by providing a sample resolution of $\sim$100-300 kyrs. Model simulations using these $\delta$$^{13}$C$_{carb}$ and $\delta$$^{13}$C$_{org}$ data provide constraints on carbon sources (mantle and weathering) and sinks (carbonate and organic carbon sedimentation); comparisons with the flux records provide insight on the components of the geological carbon cycle. Stable isotope records indicate that long-term net depletion of $^{12}$C from mobile carbon reservoirs was a consequence of an organic carbon burial fraction increase of $\sim$0.05-0.1 that began in the Jurassic ($\sim$200 Ma). Superimposed on the long-term trend are higher-order variations (5-10s of myrs) in $\delta$$^{13}$C$_{carb}$ and $\delta$$^{13}$C$_{org}$ that show episodic intervals of elevated values. In contrast to paleoceanographic convention, organic carbon burial is often decoupled from global $\delta$13C variations on the 5-10s of myrs scale. Brief episodes of elevated C$_{org}$ flux tend to occur near the onset and cessation of these intervals of elevated $\delta$$^{13}$C$_{carb}$ and $\delta$$^{13}$C$_{org}$ values; prolonged episodes of elevated C$_{carb}$ flux tend to correspond to the cessation of extended intervals of elevated $\delta$$^{13}$C values. In the latter part of the Cenozoic, the development of $\beta$ carboxylation and C$_{4}$ photosynthetic pathways in phytoplankton and terrestrial plants increasingly influenced $\delta$$^{13}$C$_{org}$, ultimately contributing to the reversal of the long-term trend in $\delta$$^{13}$C$_{carb}$. Thus, the geologic record of the global carbon cycle over the past 205 myr has been influenced by a combination of changes in carbonate burial, organic carbon burial, and biological fixation.
PP53D-06 15:00h
Isotopic Evolution of Soil Organic Matter Affects Paleo-vegetation and Paleo-pCO$_{2}$ Reconstructions
The stable carbon isotope ratio (\delta$^{13}$C) of fossil terrestrial organic matter is used to study several aspects of biosphere/atmosphere coupling in the geologic past. These range from vegetation response to climatic and pCO$_{2}$ shifts to reconstruction of paleo-pCO$_{2}$ levels. Although screening for diagenesis is typical in these studies, few have taken into account the ubiquitous but poorly understood phenomenon of progressive $^{13}$C-enrichment of soil organic matter during its decay, which is observed in modern soils worldwide. We present a simple model that describes this phenomenon and the interaction of soil organic carbon and CO$_{2}$ concentrations, fluxes and \delta$^{13}$C values. At its most basic level, the model suggests that bulk organic matter from sub-surface soil horizons will be variably enriched in $^{13}$C relative to the vegetation living on the soil surface. This complicates interpretation of paleo-isotopic records used in C3/C4 vegetation reconstructions, and may account for anomalously heavy fossil organic carbon isotope values measured in some paleosols pre-dating the end-Miocene expansion of C4 floras. The model also demonstrates that the \delta$^{13}$C evolution of soil organic carbon during its decay generates 2 types of biases that may affect soil mineral paleo-pCO$_{2}$ proxies. The first type of bias results from a steady-state inequality between the \delta$^{13}$C of organic carbon at a single depth within the soil and that of respired CO$_{2}$ in the soil. This bias is present when fossil organic matter is used to reconstruct the \delta$^{13}$C of soil-respired carbon, and can be minimized with appropriate sampling methods. The second type of bias results from a dynamic, seasonal imbalance in respiration, which may cause the soil \delta$^{13}$CO$_{2}$ flux during times of soil mineral formation to deviate from that of the annually integrated flux. At present, this bias can not be fully described or corrected for due to inadequacies in our knowledge of soil \delta$^{13}$C dynamics and the timing of soil mineral formation. Given the strong dependence of paleo-pCO$_{2}$ reconstructions on data from soil mineral isotopes, further work on these topics is warranted.
PP53D-07 15:15h
Modelling Climate-Vegetation Interactions With a Paleoclimate Earth System Model
A complex coupled earth system model for paleoclimate studies, consisting of an atmosphere and ocean general circulation model, a dynamic terrestrial vegetation and ocean biogeochemistry model and a thermodynamic icesheet model, is being developed. The model will enable studying the feedbacks within as well as between the various components of the earth system. The atmosphere and ocean GCM, together with the dynamic vegetation model, were applied for time slice experiments for 125 ky B.P. and 115 ky B.P. To investigate the effect of biophysical climate-vegetation interactions, the experiments were performed with a dynamic vegetation as well as with a prescribed (present-day potential) vegetation. A comparison of the results shows that changes in the vegetation can cause large changes in climate in places where positive feedbacks between the climate and the vegetation can enhance the insolation effects, e.g. in the Sahara region (vegetation-albedo feedback) and in the boreal forest regions in high latitudes (forest-snow-albedo feedback).
http://www.mpimet.mpg.de/en/depts/dep3/oph/deklim.html
PP53D-08 15:30h
A 10 Million Year, High-Resolution Record of C4 and C3 Plant Evolution from Arabian Sea ODP Site 722
The Siwalik paleosol sequence in Pakistan and India and the Bengal Fan sediments indicate a major expansion of C4 plants during the late Miocene, approximately 8 to 5 Ma. However, the depositional environments of paleosol sequences and deep-sea fans result in uncertainties in chronology and sediment discontinuities. The paleosol isotopic data are also local in nature. Here, we report new high-resolution carbon and hydrogen isotopic measurements of higher plant biomarkers from sediments in the Arabian Sea (ODP 722B). This site is situated at 2000 m water depth on the Owen Ridge, isolated from turbidite deposition on the adjacent Indus Fan. Continuous deposition allows more accurate age control, based on oxygen isotope and nannofossil stratigraphy. The principal source of terrestrial input to the site is from aeolian sources associated with monsoon circulation, which results in the transport of plant leaf waxes from adjacent continental regions, especially the Arabian Peninsula, the Middle East, and the Indian subcontinent. Therefore, our data represent the first continuous integrated large scale records of C3 and C4 plants for these continental regions. We analyzed ca 200 samples over the last 10 My. Our results show that C4 plants were already present by 10 Ma. Assuming end member 13C values for C3 (-34 %) and C4 (-20 %) plant leaf waxes, C4 plant percentage input increased from 25 to 45 percent from 10 to 7.5 Ma. Surprisingly, however, the C4 input decreased to 15 percent from 7.5 to 6.8 Ma. The major rise of C4 plant inputs occurred between 6.8 to 5 Ma, when the C4 percentage input increased from 15 to 65 percent. C4 percentage continued to rise slowly from 5 Ma, reaching 75 percent at 0.7 Ma. A C4 decrease from 57 percent took place from 0.7 Ma to present. Our hydrogen isotopic ratios of leaf waxes suggest a major increase in continental aridity between 8 to 6 Ma, which is followed by a slower rate of aridity increases from 6 to 0.7 Ma. Both carbon and hydrogen isotopic ratios become more variable on shorter time scales after 6 Ma than before 8 Ma. Notably, a decrease in C4 plants by ca. 15 percent, just prior to the major expansion of C4 plants, was also observed in published Siwalik paleosol carbonate 13C record. This pattern of C4/C3 plant evolution has not yet been observed in other continents, suggesting important regional climatic control, rather than a global change in atmospheric pCO2 may have played a key role on the rise of C4 plant during the late Miocene.