Vertical Profiles of CO2 and the Latitudinal Partitioning of Carbon FluxesBritton B. Stephens*, K. R. Gurney, P. P. Tans, C. Sweeney, W. Peters, L. Bruhwiler, P. Ciais, M. Ramonet, P. Bousquet, T. Nakazawa, S. Aoki, T. Machida, G. Inoue, N. Vinnichenko, J. Lloyd, A. Jordan, M. Heimann, O. Shibistova, R. L. Langenfelds, L. P. Steele, R. J. Francey, A. S. Denning

*National Center for Atmospheric Research, Boulder, Colorado, USA, stephens@ucar.edu, see reference for co-author affiliations.

Overview: Measurements of midday vertical atmospheric CO2 distributions reveal annual-mean vertical CO2 gradients which are inconsistent with atmospheric models that estimate a large transfer of terrestrial carbon from tropical to northern latitudes. The 3 models that most closely reproduce the observed annual-mean vertical CO2 gradients estimate weaker northern uptake of -1.5 PgCyr-1 and weaker tropical emission of +0.1 PgCyr-1 compared to previous consensus estimates of -2.4 and +1.8 PgCyr-1 respectively. This suggests a smaller role for northern terrestrial uptake of industrial CO2 emissions than previously thought and, after subtracting land use emissions, that tropical ecosystems may currently be strong sinks for CO2. Adding airborne flask data to existing inverse models without first fixing their transport biases will not improve their flux estimates. Future inverse models must be validated against available metrics, including vertical profiles of CO2, SF6, and other tracers.

1. Airborne flask sampling for CO2 was first conducted five decades ago by Dave Keeling. Although funding limitations ended the program, these early measurements revealed the attenuation of the seasonal CO2 cycle with altitude and latitude.

Although considerable uncertainty remains, this result implies that northern countries may not be able to rely on their forests to offset their industrial emissions as much as hoped, but also that tropical countries can help to mitigate CO2 increases by preserving their forests.

4. Observed northern-hemisphere average profiles reveal stronger vertical mixing in summer than winter. We extracted and averaged model-predicted profiles to match the observations. After optimization to marine-boundary-layer surface data, these models show large variations with generally overestimated summer mixing and both over and underestimated winter mixing.

6. Reference: Stephens, B.B., K.R. Gurney, P.P. Tans, C. Sweeney, W. Peters, L. Bruhwiler, P. Ciais, M. Ramonet, P. Bousquet, T. Nakazawa, S. Aoki, T. Machida, G. Inoue, N. Vinnichenko, J. Lloyd, A. Jordan, M. Heimann, O. Shibistova, R.L. Langenfelds, L.P. Steele, R.J. Francey, A.S. Denning, Weak northern and strong tropical land carbon uptake from vertical profiles of atmospheric CO2, Science, 316, 1732-1735, 2007.

20

-15

10

-10

10

-10

0

-5

20

-15

10

-10

10

-10

0

-5 Northern Hemisphere sites include Briggsdale, Colorado, USA (CAR); EstevanPoint, British Columbia, Canada (ESP); Molokai Island, Hawaii, USA (HAA); Harvard Forest, Massachusetts, USA (HFM); Park Falls, Wisconsin, USA (LEF); Poker Flat, Alaska, USA (PFA); Orleans, France (ORL); Sendai/Fukuoka, Japan (SEN); Surgut, Russia (SUR); and Zotino, Russia (ZOT). Southern Hemisphere sites include Rarotonga, Cook Islands (RTA) and Bass Strait/Cape Grim, Australia (AIA).

Map of airborne flask sampling locations

3. Existing airborne samplingincludes these 12 sites with records over 5 years long and several with records over 20 years long, with reasonable global and continental coverage. However, the use of these data in global inverse models has been limited.

Altitude-time CO2 contour plots for all sampling locations

Observations

Model Predictions

J ul-Aug-S ep

C O 2 (ppm)

-12 -8 -4 0

Pre

ssu

re (

mb

ar)

250

400

550

700

850

1000

Annual means

C O 2 (ppm)

-2 0 2 4

J an-F eb-Mar

C O 2 (ppm)

-4 0 4 8 12

Alti

tud

e (

km)

0.0

1.0

2.5

4.0

6.0

9.0A B C ARE S PHAAHF MLE FP F AR T AAIAOR LS E NS URZOTNH mean

C

J ul-Aug-S ep

C O 2 (ppm)

-8 -6 -4 -2 0

Pre

ssu

re (

mb

ar)

250

400

550

700

850

1000

Annual means

C O 2 (ppm)

-1 0 1 2 3 4

J an-F eb-Mar

C O 2 (ppm)

2 4 6 8

Alti

tud

e (

km)

0.0

1.0

2.5

4.0

6.0

9.0A B

Modeled NH mean

Observed NH mean

C

J ul-Aug-S ep

1 km - 4 km C O 2 (ppm)

-3.0 -2.5 -2.0 -1.5 -1.0 -0.5

Po

st-i

nve

rsio

n f

lux

(Pg

Cyr

-1)

-14

-12

-10

-8

-6

-4

-2

1 2

34 56 78

9A BC

1 2 34 56 789

A BC

Annual means

1 km - 4 km C O 2 (ppm)

0.0 0.5 1.0 1.5 2.0 2.5

-6

-4

-2

0

2

4

6

12

3

4

67

89

A

B

C

1

23

456

789

A

B

C

J an-F eb-Mar

1 km - 4 km C O 2 (ppm)

1.0 1.5 2.0 2.5 3.0 3.5 4.0

0

2

4

6

8

10

12

23

4568

9A

BC

12 3

4568 9A

BC

A B

r2 = 0.11

r2 = 0.54

r2 = 0.54

r2 = 0.27 r

2 = 0.76

r2 = 0.67

1 - C Northern land

1 - C T ropical land

Observed

C

5 177

5. Seasonal and annual-mean comparisons show that no model accurately reproduces the observations at all times of year. We use the annual-mean 1 - 4 km gradient as a metric to select three models, which we then average to produce the new flux estimates.

Observed profiels at all sites (colors) and averaged across northern hemisphere sites (black).

Northern-hemisphere average modeled (red) and observed profiles (gray).

Model estimated fluxes plotted versus predicted vertical profiles. The observed vertical profiles and uncertainties are indicated by gray bars.

2. TransCom 3 Level 2 estimated fluxes for northern and tropical aggregated regions vary across models by as much as 5 billion tons of carbon per year.

These variations are systematically related to how the models transport CO2 in and out of the northern-hemisphere lower troposphere. Without validation, the across-model mean is no more robust than any individual estimate.

Post-inversion Flux (PgCyr

-1)

-4

-2

0

2

4

1 2 345 6 78 AB C

Tropical Total, r2 = 0.61Northern Total, r2 = 0.78

9

-4

-2

0

2

4

Pre-inversion 53°N-Eq. Terrestrial Response (ppm)

-0.5 0.0 0.5 1.0 1.5 2.0

Tropical Land, r2 = 0.63Northern Land, r2 = 0.60

A

B

Post-inversion Flux (PgCyr

-1)

TM3C

TM2B

NIREA

NIES9

MATCH.MACCM28

MATCH.NCEP7

MATCH.CCM36

JMA5

UCI4

UCB3

GCTM2

CSU1

Model NameModel

TM3C

TM2B

NIREA

NIES9

MATCH.MACCM28

MATCH.NCEP7

MATCH.CCM36

JMA5

UCI4

UCB3

GCTM2

CSU1

Model NameModel

349.5

350.0

350.5

351.0

351.5

352.0

352.5

353.0

353.5

354.0

-90 -70 -50 -30 -10 10 30 50 70 90

Transcom 3 Neutral Biosphere Response