scholze edinburgh06

Slide1: 

Results from the Carbon Cycle Data Assimilation System (CCDAS) 3 2 Marko Scholze1, Peter Rayner2, Wolfgang Knorr1 Heinrich Widmann3, Thomas Kaminski4 & Ralf Giering4 1

Methodology sketchCCDAS – Carbon Cycle Data Assimilation System: 

Methodology sketch CCDAS – Carbon Cycle Data Assimilation System Biosphere Model: BETHY Atmospheric Transport Model: TM2 Misfit to observations Forward Modeling: Parameters –> Misfit

CCDAS set-up: 

CCDAS set-up Background fluxes: Fossil emissions (Marland et al., 2001 und Andres et al., 1996) Ocean CO2 (Takahashi et al., 1999 und Le Quéré et al., 2000) Land-use (Houghton et al., 1990) Transport Model TM2 (Heimann, 1995)

BETHY(Biosphere Energy-Transfer-Hydrology Scheme): 

BETHY (Biosphere Energy-Transfer-Hydrology Scheme) GPP: C3 photosynthesis – Farquhar et al. (1980) C4 photosynthesis – Collatz et al. (1992) stomata – Knorr (1997) Plant respiration: maintenance resp. = f(Nleaf, T) – Farquhar, Ryan (1991) growth resp. ~ NPP – Ryan (1991) Soil respiration: fast/slow pool resp., temperature (Q10 formulation) and soil moisture dependant Carbon balance: average NPP =  average soil resp. (at each grid point) b<1: source b>1: sink t=1h t=1h t=1day lat, lon = 2 deg

Methodology: 

Methodology Minimize cost function such as (Bayesian form):

Calculation of uncertainties: 

Calculation of uncertainties Error covariance of parameters

Improvements and further applications since Rayner et al. 2005: 

Fate of terrestrial C under climate change Including biomass burning Uncertainties of prognostic (2000-2004) net fluxes (still calculating) Improvements and further applications since Rayner et al. 2005 Improved carbon balance Improved spin-up of fast soil pool Weaker prior constraint on parameters

Slide8: 

Seasonal cycle of CO2 at Barrow, Alaska The red line is the simulation of R05 while the green line Is the improved simulation. Observations are shown by diamonds.

Global atmospheric growth rate: 

Global atmospheric growth rate Weighted sum of Mauna Loa (0.75) and South Pole (0.25) concentrations

Parameters I: 

Parameters I 3 PFT specific parameters (Jmax, Jmax/Vmax and ) 18 global parameters 56 parameters in all plus 1 initial value (offset)

Parameters II: 

Parameters II Relative Error Reduction

Some values of global fluxes: 

Some values of global fluxes Value Gt C/yr

Carbon Balance: 

Carbon Balance net carbon flux 1980-2000 gC / (m2 year)

Terrestrial C cycling under climate change: 

Terrestrial C cycling under climate change

Off-line model for prognostic slow pool: 

Off-line model for prognostic slow pool Some equations: P: slow pool, rF: fast resp., fS: allocation fast to slow pool : soil moisture Ta: air temperature

Initial slow pool size: 

Initial slow pool size

Decadal mean global NEP 1980-2090: 

Decadal mean global NEP 1980-2090 Red lines indicate simulations with climate change and black lines with no climate change. Solid lines indicate simulations with optimized parameters and broken lines with a priori parameters.

Including biomass burning : 

Including biomass burning

Parameters revisited: 

Parameters revisited

Global fluxes revisited: 

Global fluxes revisited Mean value 1980-1999 Gt C/yr

Global growth rate revisited: 

Global growth rate revisited Calculated as: Atmospheric CO2 growth rate observed no fire with fire

Slide22: 

blue bars CCDAS red bars v. d. Werf et al. Interannual variability in biomass burning estimate year Gt C/yr

Slide23: 

Conclusions & Outlook Prognostic future net carbon flux under climate change: more productive & more sensitive More processes: fire (‘weak constraint’ as a first step) More components: ocean (not-shown, but “free” optimization indicates no big changes, ideally also process-based) Prognostic uncertainties on net carbon flux for 2000-2004: calculations finished by now.. More data: inventories, regional inversions and budgets, satellite CO2 columns, isotopes, O2/N2