SLIDE 1 Robert Portmann, Eric Ray, John Daniel, Geoff Dutton, Brad Hall, David Nance, Sean Davis, Nicholas Davis, James Elkins and Steve Montzka Cooperative Institute for Research in Environmental Sciences, University of Colorado Chemical Science Division & Global Monitoring Division, NOAA 2018.5
DIAGNOSING CFC-11’s EMISSIONS IN A CHEMISTRY-CLIMATE MODEL
PENGFEI YU
SLIDE 2 CAM5 WACCM MERRA2 MERRA1 GEOS5 Use NSF/DOE Climate Model CESM to revisit the emission from the 3-Box model
Meteorology: Wind & temp
Climate Chemistry Models (CCM)
10 20 30 40 50 60 70 80 90 100 Emission or Production (Gg/yr)
a Reported Production 1995 2000 2005 2010 2015
Emission or Production (Gg/yr)
13 ± 5 Gg/yr (25%) increase 3-Box Model Emission of CFC-11
Emission
Montzka et al. 2018
SLIDE 3 HOW RELIABLE IS 3-BOX MODEL EMISSION?
- 1.2%
- 1.0%
- 0.8%
- 0.6%
- 0.4%
- 0.2%
0.0% 2000 2005 2010 2015 Rate of Change (per year) Year
Observed Climate model using 3-Box model Emission
3-D Models: WACCM or CAM, Reanalysis met.: MERRA, MERRA2
Global concentration rate of change (per year)
CFC-11 Emission increase seems too large after 2012
SLIDE 4
HOW WE DERIVE EMISSION? CFC-11’s MASS BUDGET
Stratosphere LOSS
Total Atmosphere Budget:
Growth = Emission - LOSS
Stratosphere to troposphere exchange, STE Emission
North South
Troposphere
SLIDE 5
EMISSION, LOSS AND GROWTH RATE IN A 3-BOX MODEL
3-Box Model Emission Rate Loss Rate Growth Rate
Budget Eq. Emission = Growth - Loss
Loss is assumed constant in a 3-box model
SLIDE 6
LIFETIME VARIABILITY IS MOSTLY DRIVEN BY DYNAMICS
LOSS Bonisch et al., 2001, ACP
SLIDE 7
THE LIFETIME OF CFC-11 IS NOT CONSTANT IN A CCM BECAUSE OF VARYING DYNAMICS
Life time changes Loss Term Changes
Budget Eq. Emission = Growth - Loss
SLIDE 8
CAN WE QUANTIFY THE INFLUENCE OF DYNAMICS ON OBSERVED TRENDS AND DERIVED EMISSIONS?
Stratosphere LOSS
Total Atmosphere Budget:
Growth = Emission - LOSS
Tropospheric Budget of F11:
Growth = Emission + STE
Stratosphere to troposphere exchange, STE Emission
North South
Troposphere
SLIDE 9
Is dynamical variability (STE) derived in a CCM realistic?
Solid: STE Anomaly of F11 from CCM From Model
% per year, STE
SLIDE 10 STE CFC-11 vs. QBO: strong correlation
Solid: STE anomaly of F11 from CCM Dashed: QBO, 60 mb Singapore wind anomaly From Model
% per year, STE m/s, QBO
20
R2=0.5
SLIDE 11 Solid: STE of F11 Term from CCM Dashed: QBO, 60mb winds STE of F11 term from NOAA, SH mean
SURFACE OBSERVATIONS ALSO SHOWS CORRELATION WITH QBO AND MODELED F11
Solid: STE anomaly of F11 Term from CCM Dashed: QBO, 60 mb Singapore wind anomaly STE anomaly of F11 term from NOAA, SH mean STE anomaly of F11 term from NOAA, Global
% per year, STE m/s, QBO
To diagnose Stratospheric dynamics, require year-to-year variation at ~0.1% level from measurements INDICATION: Stratospheric Dynamics (e.g. QBO) might be diagnosed and monitored by Surface Measurements !
20
SLIDE 12
ESTIMATE CORRECTED EMISSION
Observed CFC-11 Growth Rate CCM STE of CFC-11, two model estimates
Emission = Growth - STE
OBS CCM
SLIDE 13
CORRECTED INFERRED EMISSIONS 3-Box model CFC-11 Emission CCM Corrected Emission
CCM Corrected Emission increase is about 40% lower than the 3-Box model: 2013 to 2016
Year 2017 projected
SLIDE 14 MODEL PUZZLES
- Nudging Methods change modeled dynamics significantly
(lifetime changes by 10%), why?
- Dynamics affects global growth rate, but not Hemispheric difference?
Wind only nudged Wind and Temperature nudged
SLIDE 15 SUMMARY
- An increase in CFC-11 emission is required for a Climate Model to
reproduce the slowdown of CFC-11’s decline (Montzka et al., 2018, Nature)
- Climate model suggests dynamical effects (e.g. QBO) are aliased into the
emission derived by the 3-Box model.
- The CCM corrected CFC-11 emission increase since 2012 is about 40%
lower than 3-Box. In 2017, corrected emission is higher
- Stratospheric dynamical variability including QBO seem to be reflected in
surface measurement records of CFC-11
- STE and emission estimates of chemicals require year-to-year variation at
about 0.1% level from measurements
- Model and reanalysis dataset's puzzles are revealed and need to be solved
Future Work
- How to identify the source regions of the “unexpected” emissions of CFC-
11?
SLIDE 16
3-D model
16
0.0 0.5 1.0 1.5 2.0 2010 2012 2014 2016 2018 Year [N–S] – [N–S]2010-12 (ppt)
Hemispheric difference
- Dynamics affects global growth rate, but not Hemispheric
difference, why?
MODEL PUZZLES
Solid Line: Changing Dynamics Dashed Line: Fixed Dynamics
- 1.2%
- 1.0%
- 0.8%
- 0.6%
- 0.4%
- 0.2%
0.0% 2000 2005 2010 2015 Rate of Change (per year) Year
- 1.2%
- 1.0%
- 0.8%
- 0.6%
- 0.4%
- 0.2%
0.0% 2000 2005 2010 2015 Rate of Change (per year) Year
3-D model
Growth rate (per year)
Dynamical effect NO Dynamical effect?
CFC-11
SLIDE 17
STE TERMS FOR F11, F12 AND F113
SLIDE 18
OBSERVED GROWTH RATE – FITTED EMISSION
