The study of riverine biogeochemical processes has implications for our understanding of the present global carbon cycle and its ramification for the greenhouse effect. One of the most puzzling issues of the present-day carbon budget is the so called "missing sink". The problem is that about _ of the CO2 released by burning of fossil fuels and deforestation is not accounted for by the post-industrial increase of ~ 70 ppm CO2 in the atmosphere. Until very recently, it was believed that this "missing" CO2 was taken up by the oceans. The latest results show, however, that the net ingassing by the oceans is insufficient to account for the "missing" CO2. It was therefore proposed that it is the terrestrial biosphere (and soils) that acts as the mysterious sink. In particular, the temperate forest ecosystems should have been "fertilized" to such a degree that their regrowth should have exceeded deforestation of the tropics. Alternatively, or complementarily, the rivers can transport large quantities of carbon (DIC, DOC, POC) directly into the ocean where it is remineralized. In either case, transport of CO2 from soils into ground waters and rivers and the subsequent transport into the oceans plays a crucial role in quantifying the discrepancies in the global carbon budget.
My groups have studied a number of world rivers (Rhine, Danube, Elbe, Oder, Indus, St. Lawrence-Niagara-Detroit, Ottawa, Fraser and most of their larger tributaries). Carbon isotopic techniques are particularly suitable for tracing the origin of carbon and quantifying the fluxes. Carbon from respiration of organic matter has 13CDIC of about -18‰ (e.g. the Amazon), but after reacting with the soil and bedrock carbonate, it is about -10‰ (e.g. the lower Rhine). On the other hand, rivers that derive their carbon mostly from ingassing of atmospheric CO2(e.g. St. Lawrence) have 13CDIC close to 0‰. Most rivers actively degas carbon dioxide fed into them by groundwater discharges, but the CO2 overpressures - up to 2 orders of magnitude - are inversely proportional to the residence time of the water in the system. Lacustrine rivers, with water residing in the lakes for many months to decades, may be completely degassed and equilibrated with the atmosphere. The p CO2 in such rivers fluctuates seasonally by only about a factor of 2, due to photosynthetic drawdown of carbon dioxide in the warm season and its release from deeper portions of the lakes following a fall overturn (e.g. Great Lakes).
On a geological time scale, the CO2 consumed by carbonate dissolution on the continents is balanced by the CO2 released to the atmosphere by carbonate precipitation in the oceans. Then with regards to the CO2 budget in the atmosphere, it is important to consider above all, the CO2 uptake by silicate rock weathering which is balanced on a steady state by CO2 released in the atmosphere by volcanic activity, metamorphism and carbonate precipitation too.
A global continental erosion modelling (GEM-CO2) has been developped (Amiotte Suchet and Probst, 1995; Ludwig et al. 1996) to simulate the atmospheric/soil CO2 uptake by carbonate and silicate rock weathering and the subsequent river discharge of alkalinity into the oceans. For the present-day conditions, the global figure gives 320 Tg C.y-1 of alkalinity of which 70% originate from the atmospheric/soil CO2 and 30% from carbonate dissolution. The CO2 uptake only by silicate weathering amounts 140 Tg C.y-1 .This flux is maximum in the tropical equatorial regions (63%) because of high runoff intensity and in the high latitudes of northern hemisphere (35%) because of largest areas of silicate rock outcrops.
It has been proposed that increased rates of chemical weathering may have at least partly contributed to the low atmospheric CO2 concentrations during LGM (Gibbs and Kump, 1994; Munhoven and François, 1996). To test such an hypothesis, GEM-CO2 has been applied to a LGM scenario by using GCM outputs to reconstruct changes in precipitation and air temperature and a runoff empirical modelling (Ludwig et al., in press). The results show that LGM climatic conditions were 9% drier than today. Consequently the CO2 uptake by silicate weathering were 10% lower than today.
Moreover during glacial periods, due to a steeper temperature gradient towards the poles, the humid climate zones of the mid and high latitudes (non lateritic areas) were considerably smaller than during interglacial periods, and the tropical-equatorial humid climatic areas (lateritic zones) increases. Boeglin and Probst (1998) have recently shown that for the same runoff intensity, the CO2 uptake by silicate weathering is two times less important in lateritic areas than in non-lateritic ones. This difference between lateritic and non-lateric soil covers has not been considered in GEM-CO2. If we consider that lateritic soils develop mainly under tropical wet climatic conditions, the CO2 flux consumed by silicate weathering in these regions (calculated from GEM-CO2) could be divided by two. Then the decreasing of total CO2 uptake by silicate weathering during LGM would be more important than 10% (about 13-16%).
These results show that lateritic soil covers must be taken into account in such a climate change scenario and that at the time scale of glacial/interglacial periods, continental weathering is probably not the driving force of atmospheric CO2 content and climate change.
Recent studies (e.g., Gibbs and Kump, 1994, Munhoven and François, 1994, 1996) have identified continental weathering as a potentially important process for the glacial-interglacial evolution of atmospheric CO2. Gibbs and Kump (1994) constructed a model of terrestrial weathering based upon empirical relationships between the production of bicarbonate by weathering processes and the drainage intensity for five rock types. However, this model did not explicitly consider glacial weathering. A sensitivity calculation based upon hypothetical maximum meltwater discharge and bicarbonate production rates in glacial areas nevertheless revealed a possible 80% increase of the global flux. Both hypotheses were nevertheless only poorly defined and not constrained by actual observations from glacial areas. Besides this, Gibbs and Kump (1994) did not calculate the glacial-interglacial evolution of bicarbonate production rate, but only its difference between the Last Glacial Maximum (LGM) and present-day difference. The most important glacial fluxes are therefore missed, because glacial runoff was largest between 12,000 and 6000 years ago (Arnold and Sharp, 1992). Munhoven and François (1996), on the other hand, constructed a time-dependent history of bicarbonate production rates, showing a 90-250% increase during peak glacial time. Their history is obtained from the reconstruction of the riverine silica fluxes derived by inversion of the marine Ge/Si record (Froelich et al., 1992), which they interpreted in terms of CO2 consumption. Their history is hence derived from an oceanic mass balance equation, and the problem of identifying and assessing possible driving mechanisms and provenance regions of the reconstructed solute fluxes could only by addressed qualitatively.
Despite these shortcomings, both studies nevertheless indicate that glacial fluxes may have some impact on atmospheric CO2 levels. In this study, we therefore tried to devise ways in which to constrain the quantity, quality and location of glacial runoff during the last glacial cycle. Glacial runoff rates (excluding iceberg calving) for the past 130,000 years are derived from a state-of-the-art three-dimensional thermomechanical model of ice sheet growth and decay in the Northern Hemisphere, coupled with a mass-balance model (Huybrechts and T'siobbel, 1995). The chemistry of the calculated runoff is determined on the basis of high precision field data from small valley glaciers (cold-, warm- and polythermal-based) on one hand, and of theoretical considerations using known reactions in subglacial environments on the other. The reconstructed fluxes can then be used in an oceanic carbon cycle model to evaluate their impact on the glacial-interglacial evolution of CO2 in the atmosphere.
Arnold N & Sharp M, Journal of Quaternary Science, 7, 109-124, (1992).
Froelich PN, Blanc V, Mortlock RA, Chillrud SN, Dunstan W, Udomkit A & Peng T-H, Paleoceanography, 7, 739-767, (1992).
Gibbs MT & Kump LR, Paleoceanography, 9, 529-543, (1994).
Huybrechts P & T'siobbel S, Annals of Glaciology, 21, 111-116, (1995).
Munhoven G & François LM, Carbon Cycling in the Glacial Ocean: Constraints on the Ocean's Role in Global Change, Springer-Verlag, Berlin, 39-58, (1994).
Munhoven G & François LM, Journal of Geophysical Research, 101, 21423-21437, (1996).
Recent tentatives to calculate the distribution of chemical weathering over the continents and its past variations have used empirical relationships between the weathering rate of most important rock types and water runoff. Although runoff is one of the most important environmental factors controlling chemical weathering at large spatial scales, many complex processes may influence weathering and its response to climatic forcings. For instance the effect of vegetation and soil microbial activity is recognized as important in promoting mineral dissolution, through enhancement of the soil CO2 pressure and organic secretion from plant roots and fungi. It is unlikely that simple statistical relationships calibrated on the present-day system can be successful in predicting chemical weathering under past climatic conditions when the atmospheric CO2 level and the vegetation distribution were different from today. For this reason, it is necessary to build mechanistic models of the coupled vegetation-soil-rock system, which should describe in some detail the various processes involved in rock weathering, plant growth and soil microbial activity.
A preliminary step towards such a process-oriented description of rock weathering and its interactions with vegetation and soil biogeochemistry is presented here. Our approach couples a soil chemistry model with the CARAIB (Carbon Assimilation In the Biosphere) model of the land biosphere. CARAIB involves various modules dealing with soil hydrology, photosynthesis, plant growth and soil microbial oxidation of organic matter. The monthly heterotrophic respiration fluxes and the soil organic carbon content from CARAIB are used respectively to evaluate the monthly soil CO2 pressure and the dissolved organic carbon (DOC) content of soil water. The rate of mineral dissolution is composed of two terms: proton-promoted and organic ligand-promoted (oxalate) weathering. The budgets of cations are estimated on a monthly basis, as the difference between inputs from rain water and weathering, and output in drainage water. The composition of the soil water solution is calculated from a set of chemical equilibrium equations.
This biosphere-weathering model is applied to several large river basins. The drainage-weighted average of the model soil water chemical composition is compared with river data, providing a preliminary validation of the model. Several sensitivity tests are performed to investigate the role of the biosphere in promoting rock weathering, as well as to study the response of weathering to climate and CO2 changes which occurred from the last glacial maximum to the present.
The Himalayan uplift generates since Miocene a major erosion system. This potentially affects the carbon cycle throughout weathering processes, organic carbon (Corg) erosion/burial and, metamorphic CO2 degassing. In addition, climate-tectonic interactions affect the erosion regime and may have triggered large environmental changes since 20 Ma. Several hypotheses have been proposed for "tectonic forcing" of the carbon cycle by the Himalayan uplift and erosion. The study of both the modern and past erosional fluxes via the river system and the synorogenic sedimentary record provides tests for such hypotheses.
On the modern river system, a detailed study of the riverine chemistry and weathering mechanism in Himalayan watersheds and the delta of Ganges and Brahmaputra (G-B) allows to evaluate how the flux of alkalinity of these rivers is related to silicate and carbonate weathering. The flux of silicate alkalinity is around 2.7*1011 mol/yr. It is mostly related to Na silicate alteration, which tends to reduce the efficiency of silicate weathering in term of CO2 uptake. The long term CO2 uptake is likely lower than 1*1011 mol/yr. The particulate Corg flux to the ocean is around 4.2*1011 mol/yr based on an average [Corg] of 0.5wt% in the monsoonal suspended load. The rate of preservation is unknown but is likely very high because Corg content in recent sediments of the Bengal fan is very similar. Metamorphic CO2 release is more difficult to quantify. It is observed in several thermal springs in Himalaya and produce high 13C alkalinity in certain rivers. Based on this type of signal, metamorphic CO2 represents less than 5% of the total alkalinity flux of G-B.
The sedimentary records does not allow to reconstruct past fluxes because sedimentary volumes and accumulation are not sufficiently well established. However, the geochemical difference between the buried sediments and the Himalayan source rock allows to compare the uptake of CO2 by silicate weathering and Corg burial. As in the modern system, uptake of CO2 from the net Corg burial is 2-3 time that resulting from silicate weathering.
The Himalayan uplift and erosion consumes carbon over the long term essentially via the burial of Corg rather than silicate weathering. This is largely due to the combination of tectonic and climatic factors. Despite considerable generation of surface area due to various erosion processes, chemical erosion of silicates for the whole G-B is ca. 10 t/km2/yr, similar to that of flat shield under tropical climate (e.g. Orinoco). The chemical weathering is strongly limited by rapid transport. On the other hand, high particulate flux and rapid accumulation in the fan favor exportation and preservation of Corg.
The effect of geological processes (weathering, sedimentation, metamorphism, subduction, and etc.) in global carbon cycle; source and sink of atmospheric CO2 and climate change are controversial.They are generally considered as slow [1], and long-term cycling [2,3,4] in comparison with the processes of the three "mobile" carbon reservoirs of the Earth, i.e., atmosphere, oceans, and the terrestrial biosphere, consisting of plants and soils [1]. These ideas guide the International Geosphere-Biosphere Program (IGBP). In the 10 ongoing Core Projects of IGBP, no one is seriously dealing with the geological processes yet. Although in its Resolution of 4th Scientific Advisory Council (SAC-IV), there was a move to pay more attention on this direction [5], no practical action is taken until now. A hundred years ago, geological processes were considered by Thomas C. Chamberlain (1898) as having taken important part in global carbon cycle to regulate atmospheric CO2 by drawdown (weathering which is usually intensified by crust uplift, fossil fuel formation, and etc.) or emission (volcanism, carbonate rock deposition, and etc.), and thus bringing about world climate periodicity [6]. Following the general acceptance of Milankovitch Theory (1920), which considers the Earth's orbital parameters responsible for climatic cycle, the enthusiasm to pursue the relationship between geological processes and climate change was more or less declined. It has been revived in recent years when high resolution paleoenvironment proxies revealed millennial climate oscillations which can not be explained by Milankovitch Theory [7,8]. Many papers are published dealing with the CO2 source and sink of geological processes. Some of them still regard the processes as long-term cycling, and put the issue in a geological history time scale [9,10,11], but there are also papers considering geological processes as actively involved in modern carbon cycle. For instance, Martin Sharp et al estimated the CO2 drawdown by chemical denudation in a Swiss modern glacier underlain by metamorphic and igneous rock [12]. John E Mylroie [13], Chris Langdon and his Biosphere II team, and many others [14] consider the 6´105 km2 modern coral reef bank (limestone) interacts with and becomes source and sink of atmospheric CO2 through water in a dynamic CaCO3-H2O-CO2 system. The involvement of terrestrial carbonate rocks to contribute the modern atmospheric CO2 by decomposition is another focus of attention, especially in the plate margin [15, 16, 17, 18, 19]. Some of the works are based on monitoring the CO2-related geological processes in addition to geological records and modeling. The terrestrial carbonate rocks behave the same as its marine counterpart (coral reef bank) in a CaCO3-H2O-CO2 Karst Dynamic System [20], but enjoy a more extensive area of 2.2´107 km2. The 32 monitoring sites of the UNESCO/IUGS supported IGCP 379 "Karst processes and the Carbon Cycle" (1990-1999) reveals that carbonate rock, the biggest carbon reservoir on the Earth is still active in global carbon cycle. The global annual removal of carbon from the atmosphere by karst processes is estimated as 6.08´108 t/a, i.e., one-third of the missing sink in the current global carbon cycle model [21]. Many points of deep source CO2 emission were also observed, especially along the Tethys realm. In the Carbon Cycle Models, geological processes are usually given factors 1-2 order of magnitude less than those short-term cycling (photosynthesis/respiration), and medium-term cycling (organic sedimentation) [4], but in the nature, geological processes are generally not a mere inorganic slow processes. Taking the karst system as an example, the involvement of enzyme carbonic anhydrase can accelerate limestone dissolution by 1 order of magnitude [22]. The interactions between the Biosphere and Lithosphere are often overlooked in current models.It is therefore considered, that in addition to a better understanding on the behaviours of the three "mobile" carbon reservoirs, a more reasonable global carbon cycle model depends on a proper assessment of geological processes, but the latter relies on more field monitoring besides hypothesis and modeling. Some national projects were put forward, e.g., the National Research Program "Carbon Fluxes in Hydrologic and Geologic Processes" [23], and a more ambitious one "The Mississippi Basin Carbon Project" [24], but extensive international collaboration is a premise for such a gigantic issue of global scale. It is the time for research on geological processes to catch up the step of other fields in Global Change Study.
PP Tans and JWC White, Science 281, 183 (1998).
ET Sundquist, in The Carbon Cycle and Atmospheric CO2: Natural Variation, Archean to Present,
ET Sundquist and WS Broecker, Eds. (Geophysical Monograph Series 32, 5-593, (1985).
EK Berner, RA Berner, The Global Water Cycle, Geochemistry and Environment, Prentice-Hall, Inc. , 384 (1987).
FT Mackenzie, JA Mackenzie, Our Changing Planet, An Introduction to Earth System Science and Global Environmental Change, Prentice Hall, 123, 2925, (1995).
The full text of the resolutions is available on p.10 of the Global Change Newsletter No.24, published in December 19956. Roger Revelle, 1985, 1-4 of [2].
Isaac J Winograd et al, Science 258, 255, (1992).
D Oppo, Science 278, 1244, (1997).
Crayton J Yapp and Harald Poths, Nature 355, 342, (1992).
RA Berner, American Journal of Science 294, 56, (1994).
Derrill M. Kerrick, Ken Caldeira, GSA Today 4, 61, (1994).
Martin Sharp et al, Geology 23, No.1, 61, (1995).
John E. Mylroie, IGCP 379, Newsletter 32, (1996).
Elizabeth Pennisi, Science 279, 989, (1998).
BE Nesbitt, CA Mendoza, DM Kerrick, Geology 23, 99, (1995).
F Arthaud, J Dazy, J-C Grillot. Geodinamica Acta (Paris) 7, 86, (1994).
A Pentecost, Journal of Hydrology 167, 263, (1995).
PR Van Gardingen et al, Agricultural and Forest Meteorology 73, 17, (1995).
AL Mayo, AB Muller, Journal of Hydrology 194, 286, (1997).
Yuan Daoxian, Z Geomorph, NF, Suppl.-Bd. 108, 91, (1997)
ET Sundquist, Science 259, 934, (1993).
Zaihua Liu, Wolfgang Dreybrodt, Geochimica et Cosmochimica Acta 61, 2879, (1997).
Atmospheric CO2 and CH4 evolutions have been reconstructed over the last 4 climatic cycles. This extension of the previous published records on several climatic cycles allows us to place the relationship between the atmospheric composition and the climate in a more global perspective. The general relationships found on the first cycle can be extended back to 420 Kyrs, in particular during deglaciations CO2, CH4 and Antarctic temperature vary in phase (±1000 years) while during the entrance into glacial periods, CO2 generally lags the temperature. In the high frequency band, the rapid CH4 fluctuations found in the GRIP core associated with the Dansgaard-Oescgher events during the stages 2 and 3 are also suggested during the stage 6. On the opposite during these periods CO2 exhibits only small variations of about 20 ppmv. These results will be discussed in terms of global carbon cycle modification linked with high and low latitudes climatic changes.
Proxy data from the marine and terrestrial realm provide increasing evidence for climatic variability during the Holocene. The driving force behind high frequency oscillations is still uncertain. Explanations for the variability reach from primary forcing by external processes to ocean - atmosphere interactions. The available high resolution records which provide the base for the paleoclimate reconstructions comprise proxy records from lake sediments, peat deposits, tree rings, glaciers and marine sediments. Until now, the only method to involve the atmospheric [CO2] as part of the earth's climate system are measurements of gas enclosures in ice-cores. Although this method reveals direct evidence for the history of this greenhouse gas, the resolution that can be reached is often insufficient to cope with the frequency of climate shifts recently postulated for the Holocene. Since the sensitivity of atm. [CO2] is documented for the large scale glacial - interglacial cycles, one can expect that similar changes occurred as a result from perturbations in the global carbon cycle during millenium or century scale climatic oscillations. An alternative method to estimate secular variations in the atm. [CO2] is the analysis of stomatal frequencies on leaves of woody plants. The inverse relationship between stomatal densities and / or stomatal index in leaves of C3 plants and atm. [CO2] has been repeatedly demonstrated by analysis of herbarium material and experimentally by growing seedlings under changing [CO2]. Here we present a paleoatmospheric [CO2] curve based on the stomatal index record of birch leaves covering the first 600 14C years of the Holocene. The analyzed leaves are derived from a palynological and 14C dated peat section from The Netherlands with a resolution of about 30 14C years per analyzed sample. The stomatal index analysis for this interval reveals two distinct phases of change in [CO2] that can be correlated to 1) the termination of the Younger Dryas and 2) the Preboreal oscillation.
AlBIOc (ALbedo - BIOsphere - carbon) is an integrated land-surface biosphere model that simulates the potential distribution of vegetation, terrestrial carbon storage and land-surface properties, as a function of climate.The parameterisations used in the model are physically and physiologically based. This makes ALBIOc an ideal tool for the study of vegetation responses to very different climate conditions known in the past or predicted in the future. The architecture of the model facilitates a fast execution: an equilibrium is reach within a few minutes on a Pc.
ALBIOc produces a reasonable simulation of modern land-surface conditions, carbon storage and vegetation distribution.
It has been applied to estimate changes in the continental biosphere carbon storage at 21 kyr (the Last Glacial Maximum) using 17 atmospheric general circulation model simulation made within the Paleoclimate Modelling Intercomparison Project (PMIP).
Solfatara volcano is located inside the Phlegrean Fields caldera. In 1970-72 and in 1982-84 the area was affected by shallow seismicity and ground uplift. The Solfatara crater is characterised by a strong fumarolic activity which cause both single vent emissions, with a temperature up to 160°C, and diffuse degassing from the the entire crater's soil. H2O and CO2 are the main components of the fumarole discharges which originate from the boiling of an hydrothermal system partially recharged by magmatic fluids. The main elements of the hydrothermal system are: a relatively shallow magma chamber acting as heat source (zone 1), one or more aquifers located over it (zone 2) and an intensely fractured zone above the uppermost aquifer where the vapor phase is wholly separated from the liquid phase at T, P conditions close to 236°C and 31 bar (zone 3).
Strong compositional variations of fumarolic fluids were detected before and during the last crisis of om 500 (April 1996) to 850 t/d (July 1997). The estimations increase by a factor 1.7 if the anomalous areas wich border the Solfatara crater are also considered. This large quantity of gas, which is of the same order of magnitude of the mean values of the CO2 released from arc volcanoes, directly affects the highly inhabited areas of Pozzuoli and Napoli city (few kilometres south).
The data from the continuous monitoring instrument suggest a strong control of CO2 fluxes by environmental parameters during the periods characterised by very low seismic activity and no-ground uplifting.
During the last two decades, the detection and quantitative determination of gas hydrate deposits as well as the study of clathrate formation and dissociation kinetics have become central fields of marine research. Gas hydrates are ice-like solid crystals that form from gas (mainly methane) and water under conditions of low temperature and high pressure. The presence of gas hydrates in oceanic sediments along the Norwegian continental margin is documented in high-frequency near-vertical and wide-angle seismic reflection data. The base of the hydrate stability zone (HSZ) is detected in reflection seismic sections by the occurence of a strong bottom simulating reflector (BSR). The BSR mimics the shape of the sea floor, crosses the sedimentary strata and is characterised by a strong phase-reversed event. Below the BSR a low velocity layer is interpreted as a gas bearing zone. The top of gas hydrates is marked by the occurence of a second BSR. The area between both BSRs is called the hydrate existence zone (HEZ) with a high potential of gas hydrates. The inferred thickness of the HEZ allows for the estimation of the total amount of carbon trapped in gas hydrates at the present time. However, if BSRs exist without gas hydrates and free gas, this would certainly have great consequences for the calculation of the carbon budget. Modelling of the HSZ as a function of temperature and pressure shows a distinct decrease of the HSZ at the Norwegian margin from the Last Glacial Maximum (LGM) to the present time. This highly-dynamic HSZ system may provide a complex seismic expression of gas hydrate occurences which is far from being well understood.
Geographical and temporal variations of Nd isotopes in seawater have the potential to provide information on changes in erosional input and ocean circulation patterns, and their relation to tectonic and climatic events. The residence time of Nd in the oceans is sufficiently short that this isotopic system has the capability to respond, in phase, to short term fluctuations in input. Foraminifera potentially offer a high-resolution radiogenic isotope record, amenable to precise biostratigraphic or isotopic dating, that can be related directly to the stable isotope, trace metal and paleoenvironmental record from the same material. This study presents Nd isotopic variations recorded by planktic foraminifera for the past 150 Ka from ODP site 758 in the North East Indian Ocean.
P. Obliquiloculata and G. menardii were separated and organic material and metallic coatings were removed following established cleaning techniques, and for all cleaned samples Mn/Ca ratios were less than 100 µ mol/mol (Boyle & Keigwin, 1985/1986). Holocene obliquiloculata and menardii give <epsilon>Nd values of -10.12±0.16 and -10.28±0.16, respectively (where <epsilon>Nd is the measured 143Nd/144Nd ratio relative to the chondrite reservoir) these values are indistinguishable from recent direct measurements of surface seawater in this area (Amakawa & Nozaki, 1998). In contrast, the same two species in samples from the last glacial maximum give <epsilon>Nd values of -7.39±0.16 and -7.02±0.16, respectively, indicating a shift of around 3 <epsilon>Nd units. A high-resolution record of G. menardii across the same time interval confirms this glacial-interglacial shift, but also reveals a remarkable correlation between Nd and oxygen isotopes, where unradiogenic Nd is associated with depleted 18O values.
If the cleaned foraminifera preserve the chemical composition of surface seawater then these results indicate that there are significant Nd isotopic variations on climatic timescales. The correlation between Nd and oxygen isotopes suggest that there was an increase in the input of old continental material (charaterized by unradiogenic Nd) during warmer interglacial intervals. This glacial-interglacial variation may be attributed to a local increase in erosional input from the Himalayas or else global variations in deep-water circulation.
Boyle, EA & Keigwin, LD, Earth Planet. Sci. Letts, 76, 135-150, (1985/86).
Amakawa, H & Nozaki, Y, Min. Mag, 62A, 47-48, (1998).
There is currently great interest in the relative magnitudes of HCO3- and Si fluxes during the last deglaciation, since these may determine how much CO2 is sequestered from the atmosphere by terrestrial silicate weathering (Munhoven and François, 1996). It has also been inferred that glacial Si fluxes were double those of interglacials (Froelich et al., 1992), but the locus of the enhanced Si is presently unknown. Modelling studies to date have focussed on a comparison of fluxes between the last glacial maximum (LGM) and the present day, and the partition of fluxes from ice-free and glaciated terrain has been imprecise. Indeed, peak glacial fluxes are missed by this approach, since runoff from the Great Ice Sheets peaked during ~12 - ~6 ka (Arnold and Sharp, 1992).
This study determines Si and HCO3- fluxes from ice-free terrain at five times (6, 11, 14, 16 and 21 ka) during the last deglaciation. The modelling approach used by Gibbs and Kump (1994) was adapted to calculate the evolution and the distribution of the global riverine Si and HCO3- fluxes in relation to runoff intensity for different rock classes. Climatologies derived from NCAR-CCM1 GCM simulations (Kutzbach et al., 1998) are used to reconstruct the distribution of global runoff. The non-glacial fluxes are compared with recent estimates of glacial solute fluxes, determined from a 3-D thermomechanically-coupled model of ice sheet growth and decay (Huybrechts and T'siobbel, 1995).
The key results of this study are as follows. First, terrestrial fluxes of Si are relatively constant from the LGM to the present day. This suggests that interpretation of variations in marine Ge-Si ratios that rely on a doubling of terrestrial Si fluxes may be in error, and that it may be important to identify another terrestrial chemical weathering environment that produces high fluxes of Ge relative to Si. Second, fluxes of HCO3- from ice-free terrain are usually at least one order of magnitude greater than from glaciated terrain. They change by ~20% during deglaciation. However, the change in HCO3- flux from ice-free areas is often of the same order of magnitude as the higher fluxes of glacial solute during break up of the Great Ice Sheets, and there are phase differences between these changing fluxes. The HCO3- fluxes have been partitioned into those derived from silicate and carbonate sources in an attempt to assess their impact on atmospheric CO2 concentrations and the marine carbonate cycle. Much of the change in bicarbonate flux is due to changes in the amount of carbonate weathering. This can promote oceanic degassing of CO2 dependent on the mechanism of carbonate weathering. We provide illustrations of the impact on atmospheric CO2 concentrations as a result of the changing fluxes from ice-free and glaciated terrain.
Arnold N & Sharp M, Journal of Quaternary Science, 7, 109-124, (1992).
Froelich PN, Blanc V, Mortlock RA & Chillrud SN, Paleoceanography, 7, 739-767, (1992).
Gibbs MT & Kump LR, Paleoceanography, 9, 529-543, (1994).
Huybrechts P & T'siobbel S, Annals of Glaciology, 21, 111-116, (1995).
Kutzbach JE, Gallimore R, Harrison S, Behling P, Selin R & Laarif F, Quaternary Science Reviews, 17, 473-506, (1998).
Munhoven G & François LM, Journal of Geophysical Research, 101, 21423-21437, (1996).
Atmospheric CO2 is consumed both by organic matter formation and chemical rock weathering, and subsequently discharged as dissolved organic carbon, particulate organic carbon, and dissolved inorganic carbon to the oceans by rivers. In the long term, varying the ratio of the amount of atmospheric CO2 consumed by continental erosion and the amount of CO2 released during carbonate precipitation and organic matter respiration in the oceans can change the CO2 content in the atmosphere.
The purpose of this paper is to determine whether riverine organic carbon fluxes during the last glacial maximum (LGM) may have been different from today in order to assess the potential impact on atmospheric CO2. Previous studies mainly focused on the role of the river fluxes of inorganic carbon in this respect, but none of them examined possible variations in the fluxes of organic carbon, although the erosion of organic carbon actually represents the bulk of the atmospheric CO2 consumption by continental erosion. We therefore applied a organic carbon erosion model (Ludwig et al., 1996; 1998) to a LGM scenario in order to determine the riverine fluxes of organic matter during that time. The climatic conditions during the LGM were reconstructed using a computer simulation with a general circulation model (Ludwig et al., in press).
It is found that during the LGM the riverine organic carbon input into the oceans was at least ~ 10% lower than today (Ludwig and Probst, in press). Most of the reduction of the total organic matter fluxes is due to the reduction of the fluxes of dissolved organic carbon. The fluxes of particulate organic carbon remained almost unchanged. The oceanic response to the lower carbon input was also estimated on the basis of a present-day steady state budget for organic river carbon in the oceans, and implies that the reduction of the river fluxes were more than counterbalanced by lower burial rates due to the smaller shelf area during the LGM. This suggests that both the lower river carbon input and the relatively greater share of this carbon being subjected to oceanic respiration, acted as a negative feedback to the low atmospheric CO2 content during the LGM.
Ludwig, W, Probst, J-L & Kempe, S, Biogeochemical Cycles, 10, 23-41, (1996).
Ludwig, W, Amiotte-Suchet, P, Munhoven, G & Probst, J-L, Global and Planetary Change, 16-17, 95-108, (1998).
Ludwig, W, Amiotte-Suchet, P & Probst, J-L, Chemical Geology, (in press).
Ludwig, W & Probst, J-L, Tellus B, (in press).
It is of major interest to quantify the various pathways of organic carbon in order to balance the long-term carbon cycle and even if one wants to understand the fate of the anthropogenic perturbations. This work focuses on one important component in the global carbon cycle - the transport of terrestrial organic carbon from land to the ocean by rivers and its burial in marine sediments. A re-estimate of modern fluviatile organic carbon discharge and burial rates on a global scale is performed.
According to these results, about 430*1012g of terrestrial organic carbon is transported to the ocean in modern times. However, only the minor amount of 10% or about 43*1012gC yr-1is most likely buried in marine sediments. This amount is similar to the burial of marine organic carbon in the coastal ocean (55*1012gC yr-1). Adding both estimates gives about 100*1012gC yr-1, which is the value calculated by Berner (1982) for "terrestrial" deltaic-shelf sediments. The major part of the terrestrial organic carbon that enters the ocean by rivers (about 400*1012gC yr-1) seems to be (i) remineralised in the ocean (incl. at the river-ocean-boundary), whereas the mechanism by which the terrestrial organic carbon is oxidised in the ocean are unknown; or (ii) is heavily underestimated because it is dispers distributed throughout the oceans and accumulates in pelagic sediments.
Going one step further in order to calculate the global burial of OC in the ocean, estimates on OC burial for the open ocean, coastal ocean, the burial of terrestrial OC from river input and the aeolian supply are added up. This amounts to about 140 *1012g of total OC buried in marine sediments every year, which is close to the global estimate of Berner (1982), and to mean total burial rates of 82 to 157*1012gC yr-1 for whole ocean sediments calculated by Wollast (1991).
Berner RA, American Journal of Science, 282, 451-473, (1982).
Wollast R, Ocean margin processes in global change; John Wiley and Sons, 365-381, (1991).
It is generally considered that the organic carbon present in the current soils or watersheds only results from the current vegetal cover production. Detailed and punctual studies of the organic matter (OM) occuring in soils under various forest cover and on different parent-rocks (Di-Giovanni et al., in press) show however that inherited OM coming from rock weathering is added in a significant way to the OM resulting from the current production. The aim of this paper is to propose an initial estimation of the annual organic production induced by the dissolution of carbonated rocks. It was calculated by determining the residual material quantity produced by such dissolution. Calculations lean on data published by numerous authors (Probst, 1992; Amiotte-Suchet, 1995): measurements of the carbonate flows resulting from the carbonated rock dissolution in the major world watersheds, organic carbon content of the rocks. The results obtained from the watersheds have been extrapolated to a global scale. We then show that there is a significant annual organic production of sedimentary rock (0,13 Gt) which is a non-negligible element in the carbon cycle. This production obviously varies depending on the OM content in the weathered rock, but also on climatic parameters. The temperate, arctic and contrasted tropical zones, characterised by high inherited organic production oppose themselves to the dry zones. This relationship between climatic parameters and organic production suggests a temporal variation of the latter, particularly from glacial to interglacial periods. Moreover the organic production of the sedimentary rocks could contribute in a significant way to the organic stock in soils and rivers. The proposed results confirm earlier punctual studies and enhance the problem of the evaluation of organic stocks in soils and rivers only based on climatic data and current vegetal production.
Amiotte-Suchet Ph., Sci. Géol. Mém., Strasbourg, 97, 156 p., (1995).
Di-Giovanni Ch., Disnar J.R., Campy M. & Macaire J.-J., Analusis Mag, (in press).
Probst JL, Sci. Géol. Mém., Strasbourg, 94, 161 p, (1992).
There are, in Amazonia, three main reservoirs of carbon: living biomass, soil carbon and wetland sediments. There is too few data in this too large region (8. 106 km2) to quantify, without high uncertainty, these reservoirs at present and, moreover, in the past. This lack of data provoked considerable discussions, mainly related to the size of rainforest during the LGM.Vegetation records older than LGM are very scarce. Their data indicate that forest existed during glacial times, although it was characterised by montane elements, at low altitude, interpreted as a 4-6°C lower temperature. These evidences are often dated between 30.000 and 22.000 14C yrs B.P. During the LGM, the four available, sometime indirect, records of paleo-vegetation, show that rainforest was present in the northwest part of Amazonia, where precipitation is nowadays higher than 2500 mm, whereas in two other regions, with a present-day rainfall of 2000 mm or lower, the forest was replaced by savanna. In all Amazonian regions, as well as in several neighboring sites, it does exist clear evidences of a drier (and colder) climate during the LGM. A 20 ka cycle appears in the longest vegetation record (Funza-I in Bogota) and seems also present in Eastern Amazonia, indicating that the glacial climate forcing is not restricted to the LGM. One question is how large was the forest regression during these dry glacial phases. But, on the carbon storage point of view, we must also consider the effect of a drier climate (and also probably of a lower atmospheric CO2) on living biomass. The studies of modern forest biomass indicate that it varies from 80 to 180 t C ha-1. This variability is directly related to the presence of large trees. The largest, and oldest, trees representing only 3% in number of individuals, frequently correspond to 40% of the biomass. It should mean that a dryness (or low CO2) stress, shortening tree life duration, could have greatly reduced the living biomass carbon.The soil reservoir seems to be the more stable during glacial times but strong soil erosion events, observed during the transitions from dry to humid stages, could have released soil carbon.The wetland sediment reservoir is not quantified. Although it has been observed that present-day wetlands are an important source of water evaporation in the region and probably the main source of methane, the carbon accumulation in these areas is still unknown. This reservoir, as well as the associated methane flux, have certainly greatly changed during glacial times, not only in relation with the dry climate phases discussed above, but also because the low sea-level stands had provoked incisions of the valleys and a better drainage of the most part of contemporary wetlands.
The Sahara can be defined as the area of the Earth's continents that bears the smaller carbon stored on its surface (if ice caps are excluded). Because vegetation is rare and limited to small oasis or along some ouadis, the biomass and the soil organic carbon of the Sahara represent only a few gigatons (Gt) of carbon. This must be compared to the 2000 Gt of carbon stored on all other continental ecosystems. (Faure et al. 1998)
The environmental carbon deficiency of the Sahara surface is depending directly on lack of water. It is well-known that in steppic areas the graminea biomass is related to pluviometry by a logarithmic (quasi linear) relation. This water/carbon couple can be visualized (and measured) along a North-South transect on the field or on satellite imagery. On the southern limit on the Sahara there is a gradual transition between the Saharian, Sahelian, and Sudanian ecosystems that are practically following the isohyets controlled by the summer moonsoon rains. (Williams et al. 1980).
This present-day situation was very different during early to mid-Holocene. The summer monsoon was then reaching further north and rain-fall contributed to a higher water table. A multitude of lakes occurred in depressed areas all over the Sahara. The vegetation very likely followed the changing hydrology (with an unknown lag). This green Sahara lasted a few thousand years with numerous fluctuations and followed a previous dry period (during and just after the LGM) even worst than the present one. The LGM to Holocene change would have deeply modify the ecosystems of the total area of the Sahara.
It is now of utmost importance to have a detailed reconstruction history of this Holocene reforestation of Sahel and Sahara. The rate of monsoon fluctuation, the timing of reforestation, of water-table changes and lake level fluctuations must be evaluated at a decadal time scale. Seasonality changes must be quantified as well as all times constance related to this natural experiment.
The Holocene natural experiment of the Sahara/Sahel carefully studied in various geological, geomorphological and hydrological localities will be one main component of the future action to take to reforest the Sahara. The case of Mauritania shall be presented.
The Holocene reforestation may be a key for a future action, and "A one percent action on the global natural fluxes may be more feasible than to oblige mankind to reduce fuel consumption to 50%" (Faure et al, 1993)
Faure H, Faure-Denard L & Adams JM (Eds), Global and Planatary Change, Elsevier, 16-17, 199pp, (1998).
Williams MAJ, Faure H, Tha Sahara and the Nile, Balkema, 607pp, (1980).
Faure H, Faure-Denard L, Fairbridge RW, NATO ASI,Series C, Kluwer Acad, 325, 459-462, (1990).
A significant correlation between atmospheric CO2 concentrations and oxygen-isotope inferred temperature is suggested from long-term glaciochemical records from Antarctica and Greenland. During the Holocene, several century-scale climatic fluctuations took place on a global scale, such as the Medieval Warm period and Little Ice Age. Linking these temperature fluctuations to paleo-CO2 concentrations derived from the Greenland and Antarctic ice core records is difficult due to their rel. low time resolution. An excellent alternative tool for high-resolution Holocene CO2 reconstructions is the stomatal frequency analysis of leaves preserved in Holocene peat and lake sediments, as shown in a study of the stomatal frequency in Betula leaves covering the earliest Holocene (Wagner 1998). A pollen and macrofossil record spanning the last 6000 years is preserved in Jay Bath, a pond in Mount Rainier National Park (Washington, U.S.A.). The lake sediments contain high concentrations of fossil conifer needles, mainly representing species of Tsuga, Abies and Pinus (Dunwiddie, 1986). Because of their high abundances, a fine resolution can be reached for the Holocene, up to 30 years per sample for the last 3400 years. Stomatal frequency analysis of conifer needles (Pinus flexilis) has revealed their sensivity to Glacial-Interglacial CO2 fluctuations (Van de Water et al., 1994), but during the Holocene less stomatal response was measured. However, preliminary results from a herbarium study on needles from Tsuga heterophylla indicate that the number of rows of stomata on needles of this species declined consistently during the last century as a response to increasing CO2 concentrations. In this study an attempt will be made to correlate high-resolution records of stomatal frequency of Tsuga heterophylla needles to century-scale climatic changes during the (Late) Holocene.
Dunwiddie, PW, Ecology, 67, 58-68, (1986).
Van de Water, PK, Leavitt, SW & Betancourt, JL, Science, 264, 239-242, (1994).
Wagner, F, Ph.D. thesis, Utrecht University, LPP Contributions Series, 9, 1-102, (1998).
The decomposition of sedimentary organic matter with special emphasis on carbon biogeochemical cycling was studied in the Makirina Bay which is a small, shallow (0.2-1 m deep) lagoon in Central Dalmatia, Croatia. The recent sediment is represented by an up to 2 m thick layer of organic- and carbonate rich clayey silt with typical characteristics of peloid muds. The presence of framboidal pyrite indicates the existence of micro-environments with strongly reducing conditions even in the uppermost part of the sediment column, in spite of intensive bioturbation and irrigation due to macrobenthic organisms. Measurements of the Eh in the sediment column revealed a value of +60 mV at the sediment/water interface, decreasing rapidly to -300 mV in a depth of 2 cm and then stabilising around -400 mV below 10 cm. No bubble methane could be observed in-situ.
In-situ benthic fluxes were measured in a benthic chamber and calculated from concentration vs. depth profiles of dissolved inorganic carbon (DIC), Ca2+, Mg2+, and dissolved nutrients in the pore water. The modelling of sources of carbon fluxes at the sediment/water interface using concentration and 13C-DIC data from depth profiles indicated the possibility of an intensive methane formation in the sediment. An incubation experiment with sediment slurries was thus performed to study the decompositon of sedimentary organic carbon in controlled laboratory conditions and to estimate whether and how methanogenesis proceeds. To determine the sources of fluxes of dissolved inorganic carbon (carbonate dissolution, decomposition of sedimentary organic matter and methanogenesis), stable carbon isotopes were used, as well as to asses the reaction pathway of methanogenesis and concurrent methane oxidation. It was found that in the incubation experiment at in-situ temperature, methane was formed probably by CO2 reduction although sulphate was still present in the solution. Methane oxidation occurred simultaneously so that the isotopic composition of methane and DIC was changed by 13C-depleted CO2 deriving from oxidised methane.
It is well known that the atmospheric CO2 level has changed in association with the glacial-interglacial cycles (Jouzel et al., 1993). Time-series of carbon isotope ratio in seawater recorded in foraminifers may have useful information for reconstructing the carbon cycle during this period, although most of previous studies were limited in qualitative discussion. There seem to be few numerical models which can reconstruct the temporal variation of carbon cycle in this time scale. Therefore, we developed a multiple box ocean model to study the carbon cycle during the last glacial cycle. In this model, we consider the transportation of dissolved carbon and alkalinity due to advection and diffusion, biological pumps, weathering, material exchange between the ocean-atmosphere and the terrestrial biosphere, carbonate dissolution on the seafloor, and carbon isotopic fractionation in these processes. The model is time-integrated by using the carbon isotope records of planktonic and benthic foraminifers derived from the deep-sea sediment core, and the atmospheric CO2 level recorded in the ice core.
We obtain the result that the ocean circulation rate and carbonate precipitation rate should have been much weaker during the glacial period especially in the cold stage (oxygen isotope stage 2 and 4). Production rate of organic carbon in the surface ocean was reduced in the cold stages, when the size of terrestrial biosphere and weathering rate of organic carbon are assumed to be constant. However, bioproduction rate could increase during cold stages if we considervariation of the terrestrial biomass size (Crowley, 1995). We calculated the change of each carbon, which indicates that the change of ocean reservoir would have much larger than that of seafloor sediment reservoir. In other words, the release of terrestrial carbon to the atmosphere-ocean system during the glacial period would have been stocked mainly in seawater because of slower ocean circulation rate and higher pH in surface water.
The phosphate concentration in mixed-layer, which is calculated as a passive tracer in this model, indicate that the phosphate utilization must have increased in the glacial stage in any assumption of the terrestrial biomass change.The change of the calcite lithocline depth is estimated to be larger than 3,000 km, which seems to be inconsistent with the observation. These problems will also be discussed.
Jouzel J, Barkov NI, Barnola JM, Bender M, Chappellaz J, Genthon C, Kotlyakov VM, Lipenkov V, Lorius C, Petit JR, Raynaud D, Raisbeck G, Ritz C, Sowers T, Stievenard M, Yiou F & Yiou P, Nature, 364, 407-412, (1993).
Crowley TJ, Global Biogeochemical Cycle, 9, 377-389, (1995).
In the course of repeated restructuring of the vegetation cover within a glacial-interglacial climatic rhythm, the phytomass storage and, consequently, the carbon storage underwent considerable fluctuations. The changes were most pronounced at the key intervals during the last 125,000 years, those are: the Mikulino (Eemian) Interglacial optimum (125 ka BP), with the mean global temperature about 2°C above that of today; at the Last Glacial Masximum (18 ka BP), with the mean global temperature about 3 to 4°C below that of today; and that at the Holocene optimum (5.5 ka BP), the temperature about 1°C higher than at present.
The phytomass and carbon estimates were based on vegetation reconstructions by V.P. Grichuk for Mikulino (Eemian) optimum and LGM and those compiled by N.A. Khotinsky - for Holocene optimum. Reconstructions of plant formations of the past are based on paleobotanic data and ecologic chatacteristics of the plants. Specific values of phytomass storage in the reconstructed plant communities are taken to be equal to those of their modern analogues (Bazilevich, 1993).
Plant formations that existed in Northern Eurasia 5.5 ka BP were similar to those of today; the warmer and wetter climate, however, caused noticeable differences in the boundaries position and therefore in the areas of plant units from those of today. Phytomass storage within the limits of the area under study was up to 292.1 Gt, the carbon storage - about 131.4 Gt, that is about 120% of the present value calculated for the same region.
Environments during the Valdai (weichselian) Glaciation differed radically from both interglacial and modern situations. Due to a severe climate of the ice age the forest belt did not exist in the East Europe, where periglacial tundra and periglacial steppe occupied the major part of the area free from ice cover, and was narrow and discontinuous in Siberia. Consequently, the phytomass and carbon sdtorage was very low - 66.1 Gt and 29.9 Gt respectively, only 27% of the modern values.
At the last Interglacial (Mikulino, Eemian) optimum the forest limit shifted 4 to 5° of latitude to the north from its present position. Plant formations with high specific phytomass values (such as broad-leaved and coniferous-broad-leaved forests and meadow steppes) were more extensive than at the Holocene, while formations with low phytomass storage (larch forest and open woodland, dry steppe, semidesert) were reduced in the area. As the result of those changes in vegetation, total storage of the phytomass ran up to 377.1 Gt and the carbon storage was about 169.7 Gt, that is 155% of the modern storage.
As a whole, the results obtained bring out clearly that the phytomass accumulated by terrestrial plants varied considerably with climatic fluctuations, and was an important constituent of the carbon budget throughout the Pleistocene and the Holocene.
Index of EUG 10 Volume
Further EUG 10 Information
Index of the Journal of Conference Abstracts
Cambridge Publications Home Page
Last Updated on Wednesday, March 17, 1999.
© 1997 Cambridge Publications