160 likes | 300 Vues
Hemispheric transport of ozone pollution and the nonlinearities. Shiliang Wu, Harvard Bryan Duncan, NASA / GSFC Arlene Fiore, NOAA / GFDL Daniel Jacob, Harvard Oliver Wild, University of Lancaster. GMI meeting UC Irvine March 17, 2008. EU. NA. EA. SA.
E N D
Hemispheric transport of ozone pollution and the nonlinearities Shiliang Wu, Harvard Bryan Duncan, NASA / GSFC Arlene Fiore, NOAA / GFDL Daniel Jacob, Harvard Oliver Wild, University of Lancaster GMI meeting UC Irvine March 17, 2008
EU NA EA SA HTAP Multi-model approach to estimate intercontinental source-receptor (S-R) relationships for ozone Conduct base simulations with 3-D models horizontal resolution of 4°x5° or finer 2001 meteorology each group’s best estimate for emissions in 2001 methane set to a uniform value of 1760 ppb Conduct sensitivity simulations(17 total) 20% decreases in: • anthrop. emis. in HTAP regions for NOx, CO, NMVOC individually • anthrop. emis. of all O3 and aerosol precursors in HTAP regions • global CH4 (Slide from Arlene Fiore @ GFDL)
SOURCE REGIONS: S 3 foreign regions S(-20% foreign emis.) -20% domestic emis. S(-20% foreign emis.) -20% domestic emis. ≈ 85% ≈ 55% RECEPTOR = EAST ASIA EU and NA ≥ SA Larger uncertainty in role of foreign VOC ppbv Estimates of S-R relationships: Springtime surface O3 decrease in HTAP receptor regions from 20% reductions of O3 precursors RECEPTOR = NORTH AMERICA EA/EU contribute similarly EU VOC at least as important as NOx ppbv (Slide from Arlene Fiore @ GFDL)
Some questions remaining • Why so large variations across models associated with the role of VOCs? • Can 20% emission perturbations be scaled to other magnitudes? (Do the perturbations show good linearity? – this could have implications for the relative importance of different precursors in impacting the regional ozone export) • Do different models have similar/consistent linearity/nonlinearity for various perturbations?
Large variations in EU AVOC inventories across models lead to large variations in calculated ozone export from EU Annual mean NA surface O3 decreases (ppb) in response to 20% EU anthro. NMVOC reductions in different models GMI GEOS-Chem
Changes in surface O3 due to changes in EU AVOC emissions(simulation results from GMI model for JJA) (ppb) EU AVOC -20% Control EU AVOC = 0 Full perturb / (20% perturb x 5) VOC perturbations are strongly linear
VOC perturbations show very good linearity for all seasons Changes in surface O3 over North America due to perturbations in EU anthro. VOC emissions (GMI results) How about NOx?
Changes in surface O3 due to changes in EA NOx emissions(simulation results from GMI model for JJA) (ppb) Control EA NOx -20% EA NOx = 0 Full perturb / (20% perturb x 5) NOx perturbations show non-linearity
Sensitivity to NOx emissions for different intercontinental pairs Source region: EA The nonlinearities for NOx perturbations are even larger for other seasons. - - - - (effects from -20% NOx) x 5 effects from -100%NOx Source region: EU Source region: NA Distinctive seasonal pattern for the North America impacts on Europe?
20% reductions (HTAP models / MAM) Fiore et al. [2008] 100% reductions (GMI/JJA) North America East Asia South Asia Europe 100% reductions (GMI/MAM) Much weaker sensitivity to VOC compared to NOx with full reductions or in summer North America Europe East Asia Intercontinental influence on surface O3 for different precursors Decrease in surface O3 (ppb) due to foreign emission reductions Sensitivity to VOC comparable to NOx
Intercontinental influence on surface O3 for different regions (HTAP) Foreign impact on NA (GMI) EU ~ EA? Fiore et al. [2008] (HTAP) Foreign impact on EA (GMI) NA ~ EU
Consistency and discrepancy across models(Changes in NA surface O3 due to changes in EU NOx emissions) • - - ( perturbations from -20% EU anthro. NOx) x 5 • __ perturbations from zeroing out EU anthro. NOx MZ = MOZART-v2 (from Arlene) FR = FRSGCUNI (from Oliver)
Consistency and discrepancy across models(Changes in NA surface O3 due to changes in EU anthropogenic emissions) • - - ( perturbations from -20% all EU anthro. emis) x 5 • __ perturbations from zeroing out EU anthro. emis (NOx & VOC & aerosols)
On the family definition of Ox Inconsistency in the family definition of Ox in literature on ozone budgets, leading to ambiguities in model intercomparisons. Ox = O3 + O + NO2 + 2 NO3 + 3 N2O5 + PANs + HNO3 + HNO4 Ox = O3 + O + NO2 + 2 NO3 + 3 N2O5 + PANs + 2 HNO3 + HNO4 + other nitrates Ox = O3 + O + NO2 + 2 NO3 + 3 N2O5 + PANs + 2 HNO3 + 2 HNO4 + HO2 + OH …
Ox = O3 + O + NO2 + 2 NO3 + 3 N2O5 + PANs + 1.5 HNO3 + HNO4 + 0.5 HOy Transport from stratosphere A new family definition of Ox Ox family HNO4 HONO O hn OH NO3 N2O5 HNO3 O3 NO2 1 Carbonyls RNOx 0.5 1 1.5 2 3 2 HO2/RO2 H2O2/ROOH (Slide from Daniel Jacob) Emission Deposition H2O, O2