ACCENT Experiment 2: ACCENT Experiment 2 25 different models perform same experiments 15 Europe: 4 UK (STOCHEM x2, UM_CAM, TOMCAT) 3 Germany (MATCH-MPIC x2, MOZECH) 2 France (LMDzINCA x2) 2 Italy (TM5, ULAQ) 1 Switzerland (GEOS-CHEM) 1 Norway (UIO_CTM2) 1 Netherlands (TM4) 1 Belgium (IASB) 7 US: GMI (x3), NCAR (MOZART4), GFDL (MOZART2), LLNL, GISS 3 Japan: JAMSTEC – CHASER (x2), FRSGC/UCI Large ensemble reduces uncertainties, and allows them to be quantified


ACCENT Expt 2: ACCENT Expt 2 Consider 2030 – ‘the next generation’ – of direct interest for policymakers 3 Emissions scenarios ‘Likely’: IIASA CLE (‘Current Legislation’) ‘Low’: IIASA MFR (‘Maximum technically Feasible Reductions’) ‘High’: IPCC SRES A2 Also assess climate feedbacks expected surface warming of ~0.7K by 2030 Target IPCC-AR4


People & Organisation: People & Organisation Co-ordination; N+S-deposition, Tropospheric O3 F. Dentener, D. Stevenson Surface O3 - impacts on health/vegetation; web-site K. Ellingsen NO2 columns – comparison of models and satellite data T. van Noije, H. Eskes Emissions M. Amann, J. Cofala, L. Bouwman, B. Eickhout Data handling and storage (SRB; ~1 TB of model output) J. Sundet Other modellers and contributors: C.S. Atherton, N. Bell, D.J. Bergmann, I. Bey, T. Butler, W.J. Collins, R.G. Derwent, R.M. Doherty, J. Drevet, A. Fiore, M. Gauss, D. Hauglustaine, L. Horowitz, I. Isaksen, M. Krol, J.-F. Lamarque, M. Lawrence, V. Montanaro, J.-F. Müller, G. Pitari, M.J. Prather, J. Pyle, S. Rast, J. Rodriguez, M. Sanderson, N. Savage, M. Schultz, D. Shindell, S. Strahan, K. Sudo, S. Szopa, O. Wild, G. Zeng


Slide4: IPCC-AR4-ACCENT ‘High’ Ship Emission Scenario Scenario S4: IPCC A2, but with ship emissions of the year 2000 Scenario S4s: "Worst" case ship emission scenario in conjunction with S4.


Slide5: SO2 High ship emissions: A2s "2030" NOx High ship emissions: A2s "2030" SO2 emissions: A2 "2000" NOx emissions: A2 "2000"


Slide6: IPCC-AR4-ACCENT ‘High’ Ship Emission Scenario Characteristics: The idea of comparing A2 to A2s: What is the influence of ship emissions on tropospheric chemistry in 2030 if they were unabated? Does an ensemble of models give approximately the same answer regarding the influence of ship emissions? Status: Data analysis recently started Thanks to everybody who sent data so far (FRSGC_UCI, LMDz/INCA, MATCH-MPIC, TM4) We invite all other model groups to join in the inter-comparison If you are interested, please contact Veronika.Eyring@dlr.de and Axel.Lauer@dlr.de


Slide7: Year 2000 Anthropogenic NOx Emissions EDGAR database: Jos Olivier et al., RIVM Plot: Martin Schultz, MPI


Year 2000 tropospheric NO2 columns: Year 2000 tropospheric NO2 columns Model (ensemble mean) Observed (GOME) (mean of 3 methods) Courtesy Twan van Noije, Henke Eskes – figure from Dentener et al, submitted (10:30am local sampling in both cases)


Slide9: Courtesy Twan van Noije Modelled column NO2 vs GOME retrievals over Europe


Slide10: NOy wet deposition zoom over Europe Courtesy Frank Dentener


Global NOx emission scenarios: Global NOx emission scenarios Figure 1. Projected development of IIASA anthropogenic NOx emissions by SRES world region (Tg NO2 yr-1). CLE SRES A2 MFR


Slide12: Figure 4. Regional emissions separated for sources categories in 1990, 2000, 2030-CLE and 2030-MFR for NOx [Tg NO2 yr-1] Regional NOx emissions Europe: falling Asia: rising USA: ~flat Ships/aircraft: unregulated; may become larger than any regional source by 2030


Slide13: Emission Changes 2030 CLE - 2000 Plots: Martin Schultz, MPI IIASA RAINS model: Markus Amann et al.


Slide14: Year 2000 Annual Zonal Mean Ozone (24 models)


Slide15: Year 2000 Ensemble mean of 25 models Annual Zonal Mean Annual Tropospheric Column


Slide16: Ensemble mean of 25 models Absolute Standard Deviation of 25 models % Standard Deviation of 25 models Year 2000 Annual Mean O3


Slide17: Year 2000 Inter-model standard deviation (%) Annual Zonal Mean Annual Tropospheric Column


Comparison of ensemble mean model with O3 sonde measurements: Comparison of ensemble mean model with O3 sonde measurements J F M A M J J A S O N D Observed ±1SD Model ±1SD 90-30°S 30°S-Eq 30°N-Eq 90-30°N UT 250 hPa MT 500 hPa LT 750 hPa


Slide19: 2030 CLE - 2000 2030 MRF - 2000 2030 A2 - 2000


Slide20: Tropospheric O3 scales ~linearly with NOx emissions


Radiative forcing implications: Radiative forcing implications Forcings (mW m-2) 2000-2030 for the 3 scenarios: -23% +37% CO2 CH4 O3


Slide22: Impact of Climate Change on Ozone by 2030 (ensemble of 9 models) Mean Mean - 1SD Mean + 1SD Positive and negative feedbacks – no clear consensus


Budgets ofmethaneandtropospheric ozone: Budgets of methane and tropospheric ozone


Slide24: 19 Models reported O3 budgets


Slide26: Highest H2O +High Lightning NOx (8 TgN/yr) More complicated - other factors CH4 lifetime / years O3 chemical loss / Tg-O3 yr-1


Slide27: 90S Eq 90N Tropospheric H2O column / g(H2O) m-2 Tropospheric water vapour in 6 GCMs Differences of ± 10% in tropics


Slide28: AOT40, May-June-July, mean model, ppb*hours Courtesy Kjerstin Ellingsen


Change in AOT40 (CLE): Change in AOT40 (CLE)


Change in AOT40 (MFR): Change in AOT40 (MFR)


Change in AOT40 (A2): Change in AOT40 (A2)


Conclusions: Conclusions Logistics: Large group participation – partly due to IPCC-AR4 Lot of work involved – relies on funding ‘goodwill’ Need well defined experiments and diagnostics Central database and strict data format Assume mistakes will be made in first attempts Enforce deadlines if possible Science: Multi-model ensemble allows uncertainties to be assessed Sample large model parameter space Get hints about the controls on internal model processes Future work: Water vapour, convection, lightning NOx, isoprene schemes STE, biomass burning Global HOx/NOx/NOy budgets, as well as O3 and CH4