Coupled Carbon Cycle Climate Model Intercomparison Project (C4MIP)

1
Coupled Carbon Cycle ClimateModel
Intercomparison Project(C4MIP)

• Inez Fung
• Scott Doney, Keith Lindsay, Jasmin John
• CCSM Biogeochemistry Working Group
• PCMDI Curt Covey, Karl E. Taylor, Dave Bader,
  Charles Doutriaux (LSCE)
• SciDAC/ORNL Forrest Hoffman, John Drake
• Additional thanks to NSF, DOE

2
Natural Carbon Cycle is perturbed
3
Coupled Carbon Cycle-Climate Experiments

• Specify emissions from FF combustion and landuse
  modification
• 19th-20th century historical emissions
• 21st century SRES A2 and A1B
• Prognostic CO2 in atm
• Model Expts
• Coupled radiatively active CO2 prognostic CO2
• Uncoupled radiatively active CO2 282 ppmv
  (control climate)

4
The Ocean Carbon Cycle from the Atms Perspective
100 Pg C/yr
DIC 2 mole C/m3 NO3 0.02 mole
N/m3 Redfield CN 6.6 1
5
Terrestrial Carbon Cycle
6
Stable 1000-year Control Run no adjustments

• Climate variability ? land C (t 101 yr) ? atm
  C ? ocn C (t 102-103 yr) ? atm C

Doney et al. Natural variability in a stable,
1000 yr global coupled climate-carbon cycle
simulation. J Climate, in press 2006
7
Coupled Climate Carbon Cycle Model
Intercomparison Project (C4MIP)

• phase 1 19th 20th century
• Atm-Land only
• historical SST FFLandUse
• phase 2 19th to 21st century (IPCC AR4)
• Atm-Land-Ocean carbon-climate
• CO2 fertilization (max uptake)
• SRES A2 emission for 21st C (large forcing)
• A2_ROL Radiative coupling (coupled)
• A2_OL No radiative coupling (uncoupled)
• Climate fdbk A2_ROL minus A2_OL

8
C4MIP BYOModel
9
Outline

• Off-line assessment of C modules
• Constraints by contemporary obs
• COU Projections for 2100
• Carbon-climate feedback
• COU (600-300ppmv)-UNC (600-300ppmv)
• Outlook

10
Off-line intercomparisons of Terrestrial Carbon
Models

• VEMAP
• NPP (Cramer et al. 1999)
• Dynamic Vegetation (Cramer et al. 2001)
• Interannual carbon fluxes (e.g. McGuire et al. )

Land carbon No equivalent of Oort and
Rasmusson, NCEP/ECMWF reanalysis Levitus data set
11
Evaluating global ocean carbon modelsThe
importance of realistic physics (Doney et al.
2004)
Matsumoto 2004
12
Constraints on C4MIP Simulations from
Contemporary Observations
PgC/yr
(Cumulative)
13
Carbon Budget, 1990s
Apparent Airborne fraction Atm incr / FF
emission 50-60
PgC/yr
6 models OK
14
Fossil Fuel Expts Historical SRES A2 FF
Emission
Carbon-climate coupling A2_ROL Land and Ocn Uptake
Carbon-climate uncoupled A2_OL Land and Ocn Uptake
750
Climate feedback COU-UNC
15
21stC 20thC

• Carbon-climate feedback accelerates CO2 increase
  and accelerates global warming
• Feedback magnitude not proportional to actual
  increase

16
Climate Fdbk on Ocean C Storage
atlantic
pacific
Fung et al. (PNAS 2005)
17
Climate Feedback on Ocean Sink (COU-UNC)
Opposing carbon and climate effects
18
Land Carbon Storage

• Increases with
• Rapid NPP increase
• Longer lag between NPP and Respiration

19
Climate Feedback on Land C storage
20
NPP in the tropics (30S-Equator)

• Carbon-climate feedbacks
• ? NPP
• Decreases in 6 models
• Little change in 3 models
• Increases in 2 models

unc
cou
No obvious relationship with DT
21
Hydrologic Cycle
With warming, EVAP(T)gtPrecip Soil moisture
decreases even though precip increasing
22
Carbon-Water Coupling

• Models differ in degrees of
• Soil water-rainfall coupling
• Soil water-evap coupling
• Evap-rainfall coupling
• (Koster et al. 2004)

wet
Dry
Hypothesis magnitude of carbon-climate feedback
depends on bias in precip and soil moisture in
control runs
23
Summary C4MIP Model Uncertainties

• N.B. Uncertainties due to
• representation of BGC processes as well as
• propagation of biases in control climate and
  uncertainties in climate projection

24
Summary C4MIP Robust Result
Carbon-climate feedback accelerates warming
25
Anticipating AR5 BGC WG C-LAMP

• Common physical climate model CAM3-CLM3, forced
  by observed SST and sea ice boundary conditions
• Terrestrial models CASA, C-N, IBIS
• Runs being carried out at ORNL
• Standard set of output archived at PCMDI
• PCMDI is developing new software to facilitate
  remote access to data w/o downloading everything
• Comparison with in-situ, satellite and all
  available observations

26
Example NPP
CSM1.4-CASA
CAM3-CLM3-CASA
Hoffman
27
Example Comparison of time series atindividual
obs sites
Harvard Forest data CSM1.4-CASA-OCMIP simulation
at Harvard Forest
Covey
28
Taylor Diagram
29
Comparison of time series data at a site
Covey
30
Comparison of Global Distributions
precip
TSA
Covey / John
31
Posters

• Thornton Effects of terrestrial C-N coupling on
  carbon-climate feedbacks
• Hoffman Terrestrial BGC Intercomparison Expt
  (movie)
• Lu Evaluating CLM3-simulated vegetation
  phenology with satellite and ground based
  observations

32
Lorenz The general circulation of the
atmosphere, an evolving problem. Tellus 43AB,
8-15, 1991
If we can define a time-dependent scalar quantity
U (standing for Unexplained), measuring the
extent to which we have not yet logically
explained those features of the general
circulation of which we are aware, a graph of U
ending in 1970 might look something like the
early and middle parts of Fig.1.
Each cycle ends at a time when a dynamically
consistent explanation for the observed
circulation has finally attained general
acceptance, but, almost concurrently, new
observations are contradicting the explanation,
and the next cycle begins.
There follows a period when the new observations
are rejected or ignored,
and then one when they are accepted and new
explanations are sought.
33
(No Transcript)
34
(No Transcript)
35
(No Transcript)
36
(No Transcript)
37
Figure 7.3.9. Panels (a) Three main ocean carbon
pumps govern the regulation of natural
atmospheric CO2 changes by the ocean (Source
Heinze et al., 1991) (b) The oceanic uptake of
anthropogenic CO2 is dominated by inorganic
carbon uptake at the ocean surface and physical
transport of anthropogenic carbon from the
surface to deeper layers (oceanic bottleneck).
For a constant ocean circulation, to first order,
the biological carbon pumps remain unaffected
because the nutrient cycling is not changed
significantly by these processes and (c) If the
ocean circulation slows down, anthropogenic
carbon uptake is dominated by inorganic buffering
and physical transport as before. During the
slowing down, the marine particle flux can shift
somewhat to greater depths if the sinking
velocity of the particles does not change. This
leads to a biologically-induced negative
feedback. This, however, is expected to be
smaller than the positive feedback associated
with a slower physical downward mixing of
anthropogenic carbon.
38
WCRP-IGBP Coupled Carbon Cycle Climate (C4MIP)
FFhistorical SRES A2, BYOM
Cox et al (2001)
No danger of concensus
Fung et al. (2005)
1850
2100
Friedlingstein et al. (J Clim, in press, 2005)
39
(No Transcript)
40
Feedback Analysis

• Carbon Budget
• ?CO2 EMI - ? Fao - ? Fab
• Climate response (a) from coupled run
• ?Tcou ? ?CO2cou
• Sink Potential (b) estimate from uncoupled run
• ?Faxunc bax ?CO2unc
• Climate Feedback on Carbon Cycle (g)
• DFaxcou bax ?CO2cou gax ?Tcou
• DFaxcou - ? Faxunc bax (?CO2cou - ?CO2unc )
  gax (?Tcou-0)

(Friedlingstein et al., Tellus 2003)
41
Sink Potential b PgC/ppmA2_ROL
Climate Response a K/ppm
A2_ROL

• ?Tcou ? ?CO2cou
• ?Faounc bao ?CO2unc
• DFaocou bao ?CO2cou gao?Tcou

C-Climate Fdbk g PgC/K A2_ROL minus A2_ROL
42
(No Transcript)
43
(No Transcript)