Determining Fluxes of CO 2 using Mass Constraints

Transcription

1 Determining Fluxes of CO 2 using Mass Constraints Paul O. Wennberg Gretchen Keppel-Aleks, Debra Wunch, Tapio Schneider

2 Fluxes from variations in boundary layer CO2 Annual mean surface CO2 [ppm] Mixing ratio [mol m -3 ] Transport model Denning et al., 1995 Flux [mol m -2 s -1 ] Using variations in boundary layer observations of CO2 to estimate fluxes tied to vertical mixing.

3 The vertically integrated mixing ratio, <CO2> Column [mol m -2 ] Remote sensing of <CO2> will improve flux estimates since variations in <CO2> are directly related to mass fluxes.

4

5 Total Carbon Column Observing Network (TCCON)

6 How do <CO2> and boundary layer CO2 compare? Park Falls: 4 July 2006 CO 2 [ppm] Column Eddy covariance drawdown 396 m Hour [CDT] <CO2> and boundary layer CO2 have different sensitivity to the underlying biospheric fluxes.

7 Comparison of column drawdown and local flux at Park Falls Column Drawdown [ppm] Daily Drawdown from local flux [ppm] one-to-one best fit Drawdown in <CO2> not well predicted by local flux measured on the tower. Keppel-Aleks et al., Biogeosciences, 9, , 2012

8 Summer daily median <CO2> at Park Falls 4 <CO 2 > [ppm] /10/06 08/01/06 08/20/06 Synoptic scale variations complicate the relationship between <CO2> and the underlying flux.

9 TCCON <CO2> from across the network <CO 2 > [ppm] Bialystok Orleans Park Falls Lamont Pasadena Lauder We want to understand the relationship between surface fluxes and <CO2> variations at seasonal, synoptic, and diurnal timescales.

10 Simulations Use a general circulation model to simulate variations in <CO2> Modify underlying fluxes to test the sensitivity of <CO2> to local and large scale fluxes Use these simulations and other available data to understand how best to use <CO2> data to constrain surface fluxes

11 Natural CO2 fluxes CASA Net Ecosystem Exchange Ocean Exchange 60 Latitude J F M A M J J A S O N D J F M A M J J A S O N D NEE [kgc km 2 day 1 ] Ocean Flux [kgc km 2 day 1 ] There is large seasonality in northern hemisphere biospheric fluxes. Keppel-Aleks et al., Biogeosciences, 9, , 2012

12 Gradients in <CO2> from natural fluxes <CO2>bio <CO2>ocean Latitude J F M A M J J A S O N D J F M A M J J A S O N D <CO 2 > [ppm] <CO 2 > [ppm] Variations in <CO2> owe largely to biospheric fluxes.

13 Effect remote fluxes on variations in <CO2> <CO 2 > [ppm] Simulated: Park Falls all fluxes local fluxes remote fluxes /03 06/08 06/13 06/18 06/23 06/28 Date Variations in <CO2> dominated by large scale fluxes.

14 Local vs zonally uniform fluxes a 160 o W 120 o W 80 o W 40 o W 0 o 72 o N c 160 o W 120 o W 80 o W 40 o W 0 o 60 o N 48 o N 36 o N 24 o N 12 o N b 72 o N d 60 o N 48 o N 36 o N 24 o N 12 o N NEE [g C m 2 day 1 ] <CO 2 > bio [ppm] Spatial gradients tied to north-south flux gradient, not local fluxes.

15 Summer correlation with potential temperature <CO 2 > [ppm] Park Falls 700 hpa! [K] x /10/06 08/01/06 08/20/06 Day-to-day variations in <CO2> correlate with a dynamical tracer.

16 INTEX-NA: July-August o N 120 o W 100 o W 80 o W 60 o W 40 o W 48 o N 42 N 36 o N 30 o N [CO2] [ppm] Large-scale observations of tropospheric CO2 show much variability, only weekly determined by latitude.

17 INTEX-NA: July-August o N 120 o W 100 o W 80 o W 60 o W 40 o W 48 o N 42 N 36 o N 30 o N [CO2] [ppm] Variations in tropospheric CO2 tied to variations in potential temperature.

18 Conclusions (?) For LEO (GOSAT/OCO-2/etc) I think we can essentially think of If and <CO2> as separate products (e.g. I don t see a value in the simultaneous nature of the observations. In a GEO application I believe the simultaneity could be useful (e.g. at regional scales, does d<co2>/dt (over a day, say or over a few days) follow expectation based on If (GPP)?).