-0.1

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

310 320 330 340 350 360 370 380 390

Atmospheric CO2 concentration (ppm)

Cumulative ocean and land fraction carbon uptake

Atmosphere Ocean Terrestrial

-1.0

0.0

1.0

2.0

3.0

4.0

1840 1860 1880 1900 1920 1940 1960 1980 2000

Year

Carbon uptake (pg C/yr)

BERN CCSM

CLIMBER FRCGC

HADLEY IPSL

LLNL IPSL-L

MPI UMD

UVIC BCM

-1.0

0.0

1.0

2.0

3.0

4.0

5.0

1840 1860 1880 1900 1920 1940 1960 1980 2000Year

Terrestrial carbon uptake (pg C/yr)

BERN CCSM

CLIMBER FRCGC

HADLEY IPSL

LLNL IPSL-L

MPI UMD

UVIC BCM

285

286

287

288

289

290

291

292

1840 1860 1880 1900 1920 1940 1960 1980 2000

Year

Global atmospheric temperature (K)

BERN CCSM CLIMBERFRCGC HADLEY IPSLLLNL IPSL-L MPIUMD UVIC CRU-dataBCM-C

280

300

320

340

360

380

400

1840 1860 1880 1900 1920 1940 1960 1980 2000

Year

Atmospheric CO2 concentration (ppm)

Law Dome dataMauna Loa dataBERNCCSMCLIMBERFRCGCHADLEYIPSLLLNLIPSL-LMPIUMDUVICBCM-C

I. AbstractA first version of Bergen earth system model (BCM-C) was developed to address the relative importance of the terrestrial and ocean carbon cycles in the global carbon cycle climate feedback. The physical model, which is based on the Bergen Climate Model (BCM: Arpege-MICOM; Furevik et al., 2003) is coupled to oceanic (HAMOCC5) and terrestrial carbon cycle model (LPJ). The model is run for the historical period (1850-2000) forced by observed anthropogenic CO2 forcing (fossil fuel emissions and land-use change) Preliminary model simulations looks promising. The model is able to reproduce the observed climate variability reasonably well. Both oceanic and terrestrial carbon fluxes also lies within reasonable range to other modelsʼ trends and variability. During this period, both the land and the ocean take up a total of 61 and 142 Pg C from the atmosphere, respectively, which represent 47% of the total emitted anthropogenic carbon (435 Pg C). In general, the efficiency of global terrestrial carbon uptake has reached a setady state, while the efficiency of carbon uptake by the ocean have slowly declined. The global averaged SST and atmospheric temperature have increased by approximately 1.2 degree Celcius from the preindustrial period. This warming significantly reduced the sea-ice extend in both poles, which increase the carbon flux into the ocean, mainly contributed by increase in net primary, hence export production. Warming in the high latitude also reduce winter MLD and carbon outgassing. Over the land, warming increase net primary pruction in the northern high latitudes.

Analysis of global and regional carbon uptake by land and ocean using an earth system modeling approach

Jerry Tjiputra, Karen Assmann, Ingo Bethke, Christoph Heinze, and Christophe Sturm Corresponding authors: jerry.tjiputra@bjerknes.uib.no, Bjerknes Center for Climate Research, Allegaten 70, 5007 Bergen, Norway

VII. ReferencesAumont, O., E. Maier-Reimer, S. Blain, and P. Monfray, An ecosystem model of teh global ocean including Fe, Si, P

colimitations, Global Biogeochemical Cycles, 7(2), 1060- doi:10.1029/2001GB001745, 2003.Bleck, R. and L. T. Smith, A wind drivern isopycnic coordinate model of the North and Equatorial Atlantic Ocean, Model

development and supporting experiments, Journal of Geophysical Research, 95, 3273-3285, 1990.Bleck, R., C. Rooth, D. Hu, and L. T. Smith, Salinity-driven thermocline transient in a wind and thermohaline forced

isopycnic coordinate model of teh North Atlantic, Journal of physical Oceanography, 22, 1486-1505, 1992.Courtier, P., C. Freydier, J. F. Geleyn, F. Rabier, and M. Rochas, The ARPEGE project at Meteo-France. in Proc ECMWF

workshop on numerical methodsin atmospheric modeling 2: 193-231, 1991.Friedlingstein et al., Climate-carbon cycle feedback analysis: Results from C4MIP model intercomparison, J. Climate,

19, 3337-3353, 2006.Furevik, T., M. Bentsen, H. Drange, I. K. T., Kindem, N. G. Kvamsto, and A. Sorteberg, Description and evaluation of the

bergen climate model: ARPEGE coupled with MICOM, Climate Dynamics, 21, 27-51, 2003.Heinze, C. and E. MAier-Reimer, The hamburg oceanic carbon cycle circulation model version "HAMOCC2s" for long

time integrations. Technical Report 20, Deutsches Klimarechenzentrum, Modellberatungsgrupper, Hamburg, 1999.Maier-Reimer, E., Geochemical cycles in an ocean general circulation model. Preindustrial tracer distributions, Global

Biogeochemical Cycles, 7, 645-677, 1993.Six, K., and E. Maier-Reimer, Effects or plankton dynamics on seasonal carbon fluxes in an ocean general circulation

model, Global Biogeochemical Cycles, 10, 559-583, 1996.Sitch, S., B. Smith, I. C. Prentice, A. Arneth, A. Bondeau, W. Cramer, J. O. Kaplan, S. Levis, W. Lucht, M. T. Sykes, K.

Thonicke, and S. Venevsky, Evaluation of ecosystem dynamics, plant geography and terrestrial carbon cycling in the LPJ dynamic global vegetationmodel, Global Change Biology, 9(2), 161-185, doi:1046/j.1365-2486.2003.00569, 2003.

Takahashi et al., Global sea-air CO2 flux based on climatological surface ocean pCO2, and seasonal biological and temperatureeffects, Deep-Sea Res. II, 49, 1601-1622, 2002.

Wanninkhof, R., Relationship between wind speed and gas exchange over the ocean. Journal of Geophysical Research, 97, 7373-7382, 1992.

Wetzel, P., A. Winguth, E. Maier-Reimer, Sea-to-air CO2 flux from 1948 to 2003: A model study, Global Biogeochemical Cycles, 19, doi:10.1029/2004GB002339, 2005.

Acknowledgement, This research is supported by the NorClim (NFR, NORKLIMA

programme, ES418361) and the EU FP6 Integrated Projecrt CARBOOCEAN (grant 511176 GOCE).

VI. Future Outlook1. The model simulation wil be integrated until year 2100 adopting the CO2

emission from the IPCC SRES-A2 scenario.

2. A second model simulation based on the pre-industrial CO2 concentration will be applied to estimate the regional and temporal variability of anthropogenic carbon uptake by both the land biosphere and the ocean.

3. A third model simulation where the atmospheric CO2 is treated as a non-radiative gas will be conducted to analyze and quantify the regional strength of climate-mediated feedback on the global carbon cycle.

4. The current model simulation does not include the feedback from terrestrial vegetation, thus future experiments evaluating the land-cover feedback to the atmosphere and climate will be conducted.

5. Development of an improved version of the earth system model with improved atmospheric and terrestrial carbon cycle components is currently underway.

III. Comparisons of simulated carbon sinks

An essential aspect in climate-carbon cycle modeling is evaluation of model output with available data and other models. Here, the simulated global atmospheric temperature, CO2 concentration, and carbon uptake by land and ocean are compared with other models and available data. Other model simulations are based on earlier C4MIP study (Friedlingstein et al., 2006). During this period (1850-2000) the surface temperature increased by approximately 1.2 degree Celcius. The simulated atmospheric CO2 concentration follows the trends of the observational data and other models. The ocean dominates the carbon uptake with close to 3 Pg C/year carbon uptake in the present day simulation. Figure 2e shows that only 14% of the emitted anthropogenic carbon has been absorbed by the land whereas approximately 33% has ended up in the ocean. However, this uptake efficiency is slowly decreasing in the future due to decreasing buffer capacity with rising surface pCO2 and other factors.

V. Regional estimate of terrestrial carbon uptake

The global accumulated carbon uptake by the land for the year 1850-2000 is approximately 61 Pg C. Figure 5 shows that most of the carbon uptake by the terrestrial biosphere occurrs in the northern high latitudes (e.g. northern Asia, northern North America, and western Europe) and the amazon rainforest. This high carbon uptake is due to significant increase in net primary productivity, which is attributed to the increase in surface temperature in these regions. For the simulated period, there is no significant change caused in the vegetation type in all regions.

Atm

osph

eric

CO

2 co

ncen

trat

ion

(ppm

)

Glo

bal A

tmos

pher

ic s

urfa

ce te

mpe

ratu

re (K

evin

)

II. Model description

Atmospheric general circulation modelThe ARPEGE is a spectral model developed by the France weather servide and ECMWF (Courtier et al., 1991). It implements semi-Lagrangian two-time level integration with hydrostatic Navier-Stokes equations to regulate the atmospheric physical flows by conserving mass, momentum, and energy.

Ocean general circulation modelThe OGCM model is the Miami Isopycnic Coordinate Ocean Model (MICOM, Bleck and Smith, 1990; Bleck et al., 1992). It has approximately 2.4˚ horizontal resolution. The vertical dimension consists of 35 layers with temporal and spatial varying densitiy in the top most layer.

Ocean carbon cycle modelThe Hamburg Oceanic Carbon Cycle (HAMOCC5, Maier-Reimer, 1993) model includes NPZD-type ecosystem model (Six and Maier-Reimer, 1996; Aumont et al., 2003), marine carbon chemistry (Heinze and Maier-Reimer, 1999), air-sea gas exchange formulation (Wanninkhof, 1992), and processes, such as denitrification, N-fixation, calcium carbonate parameterization, DMS production, and dust deposition (Wetzel et al., 2005).

Terrestrial carbon cycle modelThe Lund Postdam Jena (LPJ) dynamic global terrestrial vegetation model is a large-scale terrestrial dynamics vegetation model with land-atmosphere carbon and water fluxes in a modular framework (Sitch et al., 2003). It implements terrestrial photosynthesis, respiration, resource competition, tissue turnover, and litter dynamics as well as fire forcing.

Figure 1. Schematic diagram of the current version of Bergen earth system model (BCM-C) and the interactions between its components.

ARPEGE

MICOM HAMOCC

LPJ

heat, freshwater, wind stress

physicalforcing

irradiationtemperatureprecipitation

CO2

Bergen Climate Model

CO2

Figure 2. Evolution of (a) atmospheric CO2 concentration simulated by the BCM-C (solid blue line) and compared with the observations and other coupled climate-carbon cycle models (Friedlingstein et al., 2006), (b) atmospheric surface temperature, (c) global carbon uptake by the terrestrial biosphere, (d) the ocean, and (e) the efficiencies of land and ocean in taking up emitted anthropogenic CO2 as a response to increased atmospheric CO2 concentration.

Tere

stria

l car

bon

upta

ke (p

g C/

yr)

Oce

an c

arbo

n up

take

(pg

C/yr

)

Frac

tion

of c

umul

ated

car

bon

upta

ke b

y la

nd a

nd o

cean

Figure 5. Regional (a) accumulated carbon uptake (NPP-respiration) for the period 1850-2000 and (b) averaged anomaly surface temperature (1991-2000) simulated by the BCM-C model.

Tem

pera

ture

ano

mal

y (K

elvi

n)

Accu

mul

ated

NEP

(g C

/m2)

a) b)

a) b)

c) d)

e)

Figure 3. Regional (a) Mean and (b) anomaly of sea-to-air CO2 flux for the period 1991-2000 (average(1991-2000) less average(1850-2000)). (c) Simulated SST anomaly and (d) net primary production for the same period.

Figure 4. Mean (a) summer and (b) winter Sea-ice extend in the arctic (1991-2000) simulated by the BCM-C model. Solid blue lines represent the observation.

IV. Regional carbon uptake by the ocean

The BCM-C model is capable of reproducing the general features of global sea-air CO2 flux as observed by Takahashi et al. (2002). Warming of SST significantly reduces the mixed layer depth (MLD, up to 200m reduction in the North Atlantic) and sea-ice extent in the high latitudes. Reduction in the MLD lowers the flux of nutrients and DIC to the surface, hence the flux of CO2 to the atmosphere. In addition, the reduction in sea-ice increases the surface area for marine photosynthesis, export production, and biological pump.

a) b)

c) d)

a) b)