We have developed a multiple element (C, N, P, Si, Ca, Fe) biogeochemical model of marine ecology that includes small, large and diazotrophic phytoplankton as well as explicit ballast-driven sinking and remineralization of detrital organic matter and cycling of dissolved organic matter. Phytoplankton growth is described through a new formulation including co-limitation by N, P, Si, Fe and light to reproduce broad observational trends.  Phytoplankton grazing is described through different power laws in the closure terms for small and large phytoplankton to reproduce observed augmentation of large phytoplankton with increasing production. Detritus production is assumed to be a temperature dependent fraction of small and large phytoplankton. This model has been imbedded in a 1-degree; global ice/ocean general circulation model (MOM4) forced by a 43-year atmospheric reanalysis forcing from the Common Ocean Reference Experiments (CORE) program to quantify the relationship between food web structure, biogeochemical cycles and the atmospheric CO₂ signature on inter-annual timescales. Novel aspects in the model structure are described, the impact of the formulation of ecosystem structure on biogeochemical cycling are discussed, and results of the atmospheric reanalysis forcing experiment presented. Of particular interest are the dynamical roles played by equatorial ENSO variability and polar sea ice dynamics on air-sea CO₂ fluxes.

Temporal and spatial distributions of the δ¹⁸O value of atmospheric CO₂ (dC_a) can be used to constrain regional ecosystem carbon exchanges and linkages between carbon and water cycling. However, our understanding of the substantial observed temporal and spatial variability in dC_a is limited. Among many contributing factors, seasonal and inter-annual variations in climate are likely to be important. In this study we investigate the impact of dry climatic conditions on the ecosystem-atmosphere C¹⁸OO isoflux.

We conducted this study in the U.S. Southern Great Plains using five-year monthly-averaged precipitation δ¹⁸O values (δ_p) from the National Atmospheric Deposition Program (NADP) network, Mesonet meteorological forcing, and MODIS-derived NDVI and land-cover characterization. These data are used to force the isotope ecosystem model ISOLSM [Riley et al., 2002; Riley et al., 2003] at 10 km resolution across the region for relatively drier (2003) and wetter (2004) years. The model has been calibrated and tested in the dominant herbaceous vegetation types in the region [Biraud et al., this issue].

The Amazon region of South America plays a significant role in global cycles of water, energy and carbon, yet it is also one of the most challenging biogeographical areas of the world to model correctly. Numerous global climate models have problems with anomalous die-back of the Amazon rain forest variously attributed to inadequate representation of rainfall, faulty soil moisture dynamics or an inability to correctly simulate the drought tolerance of the vegetation. Such misrepresentation of the Amazon in global climate models can cause larger than observed excursions of the global carbon cycle. This poster explores soil moisture and drought stress for Amazonia with the Simple Biosphere Model (SiB3) and possible reasons and solutions to the rain-forest die back problem, which should lead to more reasonable estimates of carbon fluxes at the ecosystem scale.

We have developed a Climate-Carbon coupled model based on the IPSL OAGCM and on two biogeochemical models, ORCHIDEE for the continent and PISCES for the ocean, to investigate the coupling between climate change and the global carbon cycle. We have performed four climate-carbon simulations over the 1860-2100 period in which atmospheric CO₂ is interactively calculated. They are :

§ A control coupled simulation with no anthropogenic emissions.

§ A coupled simulation with anthropogenic emissions.

§ A coupled simulation with anthropogenic emissions including non-CO₂ greenhouse and sulfate aerosols.

§ An uncoupled carbon simulation with the same anthropogenic emissions as second simulation but for which atmospheric CO₂ change has no impact on climate.

Compared to the first IPSL Climate-Carbon coupled model [Dufresne, et al., 2002], the simple carbon models have been replaced by IPSL advanced ocean and land biogeochemical models, respectively PISCES and ORCHIDEE. CO₂ is transported in the atmosphere and compared with observations. Comparison with satellite data is also done. We then analyze the coupled and uncoupled simulations, highlight the importance of the climate change both on the oceanic and biosphere sink and estimate the climate-carbon feedback. The results are also compared to the outputs of other models participating in the C4MIP inter-comparison project. Finally, off-line simulations are carried out to perform sensitivity tests (fire, dynamics of land and ocean ecosystems, soil respiration) in order to identify the key processes which govern the simulated response.

Various patterns of large-scale climate variability have exhibited trends over the past few decades. These patterns of variability are known to have contributed substantially to recent trends in, for example, surface temperatures and precipitation. However, it is less clear to what extent the climate impacts of these patterns extend to the carbon cycle. Here we summarize the observed relationships between monthly and daily mean variations in concentrations of atmospheric carbon dioxide and the dominant pattern of variability in the extratropical circulations, the so-called Northern and Southern Hemisphere Annular Modes. The observed relationships are compared with results derived from surface flux estimates from the Atmospheric Tracer Transport Model Intercomparison Project (TransCom).

The world is moving in a direction of managing the carbon cycle in order to limit the forcing of earth's climate by CO₂ as well as to limit acidification of the oceans.  We may expect limitations on emissions, sequestration of carbon and enhancements of natural sinks.  It would be important to be able to observe and quantify the impact of any such measures on the growth rate of CO₂. Until now it has been difficult to quantify changes of the growth rate of CO₂ with confidence due to the large year to year variations that are caused by climate variations.  A statistical method has been developed to predict the growth rate of CO₂ based on observed variations of climate parameters.

Monthly time series of atmospheric carbon dioxide (CO₂), the relative amount of carbon-13 in CO₂ (¹³C), hydrogen (H₂) carbon monoxide (CO), and methane (CH₄) are examined and related to each other and to an index of the status of ENSO. Making use of simple 12-month running mean and difference filters isolates the year-to-year variability in the concentrations and apparent sources of these constituents.

A coupled Biogeochemistry-Ecosystem-Circulation (BEC) ocean model is used to examine the sensitivity of ocean biogeochemical cycling and air-sea CO₂ exchange to variations in mineral dust deposition from the atmosphere.  Mineral dust deposition estimates from four different climate regimes are used to force the ocean model.  Our estimated climate-induced changes in dust deposition to the oceans significantly modify phytoplankton community composition, and global-scale rates of nitrogen fixation, export production, and air-sea CO₂ flux.  Dust driven variations in air-sea CO₂ exchange exceeding 1 PgC/yr are of similar magnitude to present net oceanic anthropogenic uptake.  Dust deposition directly modifies rates of export production and CO₂ flux over large regions where iron is the primary growth-limiting nutrient.  Dust deposition also indirectly influences these rates by modifying the rates of nitrogen fixation in the tropics and subtropics where nitrogen is the primary limiting nutrient.  Initially the direct pathway dominates the ocean biogeochemical response to dust variations, but over multi-decadal timescales the indirect response may be equally important.  Our predicted decrease in mineral dust deposition over the next century would significantly slow oceanic uptake of CO₂ and act as a positive feedback mechanism for the ongoing global warming.

Because vegetation growth in the Northern Hemisphere is typically nitrogen-limited, increased nitrogen deposition could have attenuating effect on rising atmospheric CO₂ by stimulating the accumulation of biomass. Given the high carbon to nitrogen ratios and long lifetimes of carbon in wood, a most significant effect of nitrogen fertilization is expected in forests. Forest inventories indicate that the carbon content of northern forests have increased concurrently with increased nitrogen deposition since the 1950s [Spiecker et al., 1996]. In addition, variations in atmospheric CO₂ indicate a globally significant carbon sink in northern mid-latitude forest regions [Schimel et al., 2001]. It is unclear however, whether elevated nitrogen deposition or other factors are the primary cause of carbon sequestration in northern forests. We argue that the elevated nitrogen deposition is unlikely to enhance vegetation carbon sink significantly because of its differentiating effect on the carbon sequestration capacity of uneven aged forests and climatic limitations on carbon sequestration in the Northern Hemisphere. We estimate the potential of forests with lifted nitrogen limitation to decelerate CO₂ concentrations rise in the atmosphere and therefore to mitigate climate warming. We also outline areas of the Northern Hemisphere which are most sensitive to increased nitrogen deposition.