Kinnison WKWG 2016 Final v1 - CESM® · 2016. 8. 14. · Current CCMI Publications. Currently have ~30 Science publication in progress. Garcia, R. R. Anne K. Smith, D. Kinnison 1,

___________________________Status of CCMI and Evaluation of

Chemistry in CMIP6 WACCM________________________

Doug Kinnison, Chuck Bardeen, and the entire WACCM Team.

WACCM Working Group MeetingBreckenridge Colorado

June 21, 2016.

Outline

• Status of CCMI simulations.

• Preliminary Comparison of CCMI with CMIP6 (SD only)

• Halogen, GHG LBCs for CMIP6 (New Capability)

CESM1 (WACCM4) - CCMI

CCMI: CESM1 (WACCM) – CAM4 column physics – 2 degree Res.

• TSMLT Chemistry, ~180 species, >450 gas-phase rxns; 17 heterogeneous rxns on sulfate, NAT, and Water-Ice.

• Solar Variability (Lean ETF; SPEs & Auroral NOx).

• Halogen (LL, VSL); GHGs (CO2, CH4, N2O).

• Sulfate SAD time series (including Volcanoes).

• QBO nudged to observed winds.

• SD simulation using MERRA met.

CESM1-WACCM4 CCMI Simulations: Status

Scenario Period Ocean RCP Members Completed

CMOR#

REFC1 1955-2014 Data - 5 Done

REFC1-PI 1850-1960 Data - 1 -

REFC1SD 1979-2014 Data - 1 Done

REFC2 1960-2100 Interactive RCP6.0 3 Done

SENC2 2001-2100 Interactive RCP4.5 1 Done

SENC2 2001-2100 Interactive RCP8.5 3 Done

SENC2-fGHG1960 1960-2100 Interactive RCP6.0 3 Soon

SENC2-fODS1960 1960-2100 Interactive RCP6.0 3 Soon

SENC2-fODS2000 2000-2100 Interactive RCP6.0 3 Soon

SENC2-nVSL 1955-2100 Interactive RCP6.0 1 Soon

#CMOR’izing addressed by Simone Tilmes and Gary Strand.

Current CCMI Publications.

Currently have ~30 Science publication in progress.

Garcia, R. R. Anne K. Smith, D. Kinnison1, Á. de la Cámara, and D. Murphy, Modification of the gravity wave parameterization in the Whole Atmosphere Community Climate Model: Motivation and results, in review, JGR, 2016.

Solomon, S., D. J. Ivy, D. Kinnison, M. J. Mills, R. R. Neely III, A. Schmidt, Emergence of Healing in the Antarctic Ozone Layer, accepted, Science, 2016.

Tilmes, S., J.-F. Lamarque, L. K. Emmons, D. Kinnison, D. Marsh, R. R. Garcia, A. K. Smith, R. R. Neeley, A. Conley, F. Vitt, Maria Val Martin, H. Tanimoto, I. Simpson, D. R. Blake, and N. Blake, Representation of the Community Earth System Model (CESM1) CAM4-Chem within the Chemistry-Climate Model Initiative, Geosci. Model Dev., 9, 1853–1890, doi:10.5194/gmd-9-1853-2016.

Jackman, C. H., D. R. Marsh, D. E. Kinnison, C. J. Mertens, and E. L. Fleming, Atmopspheric changes caused by galactic cosmic rays over the period 1960-2010, Atmos. Chem. Phys. , doi:10.5194/acp-16-5853-2016.

Garcia, R. R., M. Lopez-Puertas, B. Funke, D. E. Kinnison, D. R. Marsh, L. Qian, On the secular trend of COx and CO2 in the lower thermosphere, J. Geophys. Res., 121, 3634-3644, doi:10.1002/2015JD024553.

Solomon, S., D. E. Kinnison, J. Bandoro, R.Garcia, Simulations of Polar Ozone Depletion: An Update, J. Geophys. Res., 120, 7958-7974, doi:10.1002/2015JD0233652015.

Sakazaki, T., T. Sasaki, M. Shiotani, Y. Tomikawa, and D. Kinnison, Zonally uniform tidal oscillations in tropical stratosphere, Geophys. Res. Lett., , 42, 9553-9560, doi:10.1002/ 2015GL066054.2015.

Randall, C, V. Harvey, L. Holt, D. Marsh, D. E. Kinnison , B. Funke, P. F. Bernath, Simulation of energetic particle precipitation effects during the 2003-2004 Arctic winter, J. Geophy. Res., Space Phys., 120, 5035-5048, doi:10.1002/2015JA021196, 2015.

Daily and Monthly Ozone will be used for input for non-interactive CMIP6 models.

TOZ [63S-90S] - October

1850 1900 1950 2000 2050 2100Year

150

200

250

300

350

400

450To

tal C

olum

n O

zone

(DU)

REFC1 [data ocean]REFC2 [interactive ocean]REFC1SD [MERRA]Observations

Thin black line ~1980 TOZSymbols are individual realizations3-point smoothing for solid lines

The SD simulation gets most of the NH depletion – but not all (issues with T-bias, SAD input).

TOZ [63N-90N] - March

1850 1900 1950 2000 2050 2100Year

300

350

400

450

500

550To

tal C

olum

n O

zone

(DU)

REFC1 [data ocean]REFC2 [interactive ocean]REFC1SD [MERRA]Observations

Thin black line ~1980 TOZ

Symbols are individual realizations

3-point smoothing for solid lines

CESM Column Physics

CAM4/WACCM4 CAM5.1 CAM5.5/WACCM5.5Horizontal Resolution 1.9˚x2.5˚ 1.9˚x2.5˚ 0.95˚x1.25˚Vertical Layers 26/66 30 32/70Boundary Layer HB UW CLUBBShallow Convection Hack UW CLUBBDeep Convection ZM ZM ZMMacrophysics R&K UW CLUBBMicrophysics R&K MG 1.0 MG 2.0Radiation CAMRT RRTMG RRTMGAerosols Bulk MAM3 MAM4

NOTE: MAM4 includes prognostic stratospheric sulfates.

NEXT: Will show how only year 2011 SD simulations (88-levels).

Evaluation of Stratospheric Chemistry (SD-WACCM / MERRA)

increases the variability of ClONO2, in terms of both high and low extremes. The high extremes are linked to therapid formation of ClONO2 from enhanced ClOwhenever NO2 is available, while the low extremes reflect effectiveheterogeneous removal that subsequently drives ClONO2 toward zero if heterogeneous chemistry is sufficientlyrapid compared to the rate of reformation of ClONO2. Variations linked to wave activity that lead to sporadiclow and high temperatures, and transport air from sunlight to darkness and vice versa, strongly affect theClONO2 extremes. Arnone et al. [2012] show MIPAS observations of high ClONO2 inside the edge of the Arcticvortex at times, similar to that presented here for Antarctica. It is important to note that the computedvariability of the reference and LIQUIDS#2 model values decreases substantially as temperatures increase in thespring, again illustrating the fact that not only the abundance but also the variability of ClONO2 is tightly linkedto the rate of heterogeneous chemistry. MIPAS data display a similar marked decrease in variability after aboutday 275 in Figure 10. The LIQUIDS#2 case yields minimum ClONO2 mixing ratios that are higher than theircounterparts in ACE and MIPAS data (particularly after about day 250 in the latter), indicating a premature netrecovery into the ClONO2 reservoir (as suggested by the behavior of HCl in Figure 8 as well). At this latitudeand altitude, the model reference case is able to match the extremes of both ACE and MIPASobservations of ClONO2, although there is a clear bias in the reference model probability distributionfunction for days 260–280, where WACCM displays too many high values and too few low ones.

Figure S5 shows similar results as those of Figure 10 but for latitudes deeper in the vortex, 75–85°S, near 30hPa, where the reference model values appear to overestimate somewhat both the high and low extremesduring days 260–280. It is noticeable that the reference model is closer to the MIPAS range than theLIQUIDS#2 case throughout the season.

In summary, Figures 3, 4, 8, and 10 illustrate how the abundance, variability, and timing of the seasonalevolution of ClONO2 provide a particularly sensitive indicator of chemical processing that is key to ozonedepletion and its timing.

4.4. Transport and ClONO2

Figure 11 examines the zonally averaged budget of ClONO2 in the reference case for Antarctica, by showingthe contributions of chemistry and advective transport to the calculated rate of change of ClONO2 in SD-WACCM

Figure 8. Evolution of zonally averaged chemical species at (left) 80°S and (right) 70°S near 30 hPa, for the (top) referenceand (bottom) LIQUIDS#2 cases. Concentrations of ozone and HCl calculated by SD-WACCM (lines) are compared to MLSsatellite data (crosses), and calculated abundances of related species are also indicated.

Journal of Geophysical Research: Atmospheres 10.1002/2015JD023365

SOLOMON ET AL. POLAR OZONE DEPLETION 7969

Amazing representation of stratospheric chemistry. Comparisons above are made with: Aura MLS (HCl, O3); MIPAS (ClONO2).

Wegner, T, D. E. Kinnison, R. R. Garcia, S. Madronich, and S. Solomon, Polar Stratospheric Clouds in SD-WACCM4, J. Geophys. Res., VOL. 118, 1-12, doi:10.1002/jgrd.50415, 2013.

Solomon, S., D. E. Kinnison, J. Bandoro, R.Garcia, Simulations of Polar Ozone Depletion: An Update, J. Geophys. Res., 120, 7958-7974, doi:10.1002/2015JD0233652015.

31hPa

Agreement not as well represented in SD-WACCM / GEOS5 (1-deg) CMIP6 (between days 270-290)?

31hPa

ClOx Activation is lower in CMIP6?

Evaluation of Stratospheric Chemistry (SD)

Net Activation is lower in CMIP6 near day 280.

31hPa, 80S

Less ICE Activation

Evaluation of Stratospheric Chemistry (SD) - lower down.61hPa, 74S

Ozone depletion is also less in CMIP6 (relative to obs). However, in much better agreement at 61hPa vs 31hPa.

ICE SAD is lower in magnitude in CMIP6 (relative to CCMI). This may be partially why ozone depletion is underestimated in CMIP6 near day 280.

CCMI: H2O, HNO3, and Temperature

Overall – pretty amazing agreement.

CMIP6: H2O, HNO3, and Temperature

XXHNO3 (higher) and H2O (lower) are not as well represented in CMIP6.

Temperature is consistent between both model systems.

Transport / Mixing Differences?

Chemistry Summary

• Specified Dynamics simulations were conducted for year 2011 using the CCMI (2-deg; WACCM4) and CMIP6 (1-deg).

• There are some difference between the high latitude polar results published in Solomon et al., JGR 2015 (CCMI) versus the what is currently derived in CMIP6. The agreement is worse at lower pressures (higher altitude).

• This might be due to changes in transport/mixing. It could alsobe partially dependent on the magnitude of SAD ICE, which is lower in CMIP6 vs CCMI.

• More work (ASAP) is needed to understand the SD CCMI / CMIP6 differences.

New Lower Boundary Conditions for CMIP6

Malte Meinshausen et al., Historical greenhouse gas concentrations for CMIP6, to be submitted to GMD, 2016

Key improvements from CMIP6:

• Latitudinal gradients• Seasonal gradients

Ground-based and aircraft observation are used.

Methyl Bromide, CH3Br (pptv)

Impressive representation of the latitudinal gradient in CH3Br. Not having this gradient makes comparisons to aircraft campaigns (in the NH) problematic. Values for present day approaching 1960s conditions (~1.5 years, lifetime controlled by reaction with OH).

Methyl Chloroform, CH3CCl3 (pptv)

The atmosphere is almost free of MCF (lifetime ~5 years; lifetime controlled with OH)!

Pretty consistent with CMIP5 historical time series, except observations are now used from 2005 forward (lifetime ~9 years; lifetime controlled with OH)!.

Methane, CH4 (ppbv)

Pretty consistent with CMIP5 historical time series, except observations are now used from 2005 forward (lifetime ~9 years; lifetime controlled with OH)!.

Carbon Dioxide, CO2 (ppmv)

Carbon Dioxide, CO2 (ppmv), global average.

Increasing at a rapid rate – near 400 ppmv in 2014!

?

Thank you for your attention.