Source Receptor Parameterization and Impact Assessment

Transcription

1 Source Receptor Parameterization and Impact Assessment Oliver Wild Lancaster Environment Centre, Lancaster University, UK European studies: Frank Dentener, Kathy Law, Jan Eiof Jonson, Valiyaveetil Semeena, David Simpson, Camilla Andersson, David Parrish, Markus Amann Applications: Ruth Doherty, David Stevenson, all contributors to intercomparison HTAP Meeting, Geneva, Switzerland, 22 nd March 2013

2 HTAP SR studies for 2001 NA EU Model Runs Control run (SR1), 20% reduced CH 4 run (SR2) 20% reduced NO x emissions over each region (SR3) 20% reduced VOC emissions over each region (SR4) 20% reduced CO emissions over each region (SR5) 20% reduced NO x /VOC/CO/aerosol emissions (SR6) SA HTAP focus: quantify current source receptor relationships Parameterization: use relationships to explore O 3 trends EA HTAP: Quantify impacts of major anthropogenic source regions on surface O 3 under 2001 conditions using 20% precursor emission changes, ~30 models contributed. Fiore et al., 2009

3 Sources ΔO 3 (ppbv) N. American Emissions Receptors Ozone Source Receptor Matrix European Emissions S. Asian Emissions E. Asian Emissions Rest of World Emissions 24 (+8) models contribute results; 14 models complete full set This example: FRSGC/UCI CTM Scale of grey boxes 5 times larger

4 Parameterizing Surface O 3 Changes Simple linear combination of emission perturbations Decompose emission scenario by region and precursor Scale contributions to ΔO 3 for each region and precursor Sum to estimate ΔO 3 at any location or region Determine anthropogenic contribution to O 3 changes ΔO 3 due to 20% reduction in E NOx over N. America f NA,NOx = ΔE NOx /0.2E NOx ΔO 3 due to 20% reduction in E CO over N. America f NA,CO = ΔE CO /0.2E CO ΔO 3 due to 20% reduction in E VOC over N. America f NA,VOC = ΔE VOC /0.2E VOC

5 Parameterizing Surface O 3 Changes Simple linear combination of emission perturbations Decompose emission scenario by region and precursor Scale contributions to ΔO 3 for each region and precursor Sum to estimate ΔO 3 at any location or region Determine anthropogenic contribution to O 3 changes O 3 response at location k 3 precursors NOx, CO, VOC 5 emission regions NA, EU, SA, EA, RoW scale factor HTAP O 3 response O 3 Response to CH 4 abundance f = 0.95f f 2

6 Apply to a Range of HTAP Models FE Simulations Surface O 3 changes from 2000 to 2030 under RCP 8.5 from four HTAP models Full simulations: solid lines Linearization: dashed lines South Asia East Asia Europe N. America Global 14 Models from HTAP contributing to parameterization CAM CHEM GEM AQ INCA LMDz MOZECH TM5 EMEP GISS PUCCINI LLNL IMPACT STOC HadAM3 UM CAM FRSGC/UCI GMI MOZART GFDL STOCHEM

7 How has surface O 3 changed? How well does this match observations? trend: Obs 0.15 ppb/yr, Mod 0.13 ppb/yr trend: Obs 0.17 ppb/yr, Mod 0.03 ppb/yr

8 Attributing surface O 3 changes Summer: 1.1 ppb Winter: +1.1 ppb

9 Attributing surface O 3 changes European emissions Summer: 1.9 ppb Winter: +0.3 ppb External emissions Summer: 0.0 ppb Winter: +0.3 ppb Methane Summer: +0.8 ppb Winter: +0.5 ppb ppb

10 Future changes in European surface O 3 SRES A2 RCP 8.5 RCP 2.6 RCP: SRES: GEA: Representative Concentration Pathways IPCC Special Report on Emission Scenarios Global Energy Assessment Focus on three scenarios: consider impacts of O 3 from outside Europe

11 Relating Global and Regional Scales ppb Seasonal mean O 3 changes between 2005 and 2030 with RCP 2.6 emissions from the HTAP parameterization (top row) and the EMEP regional model (bottom row) Global models are not ideal tools for air quality applications (resolution 2 3 ) EMEP model widely used for policy making in Europe (resolution 0.2 here)

12 Max 1 hr O 3 from Regional Models Focus on maximum O 3 with EMEP model (20 km resolution) Apply RCP 2.6, RCP 8.5 and SRES A2 emission scenarios Run 2005, 2030 and 2030 with 2005 boundary conditions Use 3 D boundary conditions for O 3, CO, NOy, CH 4 from HTAP study Compare climate effects with MATCH model (50 km resolution) Two climate projections: SRES A1b in ECHAM5 and HadCM3 Anthropogenic emissions held constant at 2000 level

13 Max 1 hr O 3 changes RCP 2.6 emissions Summer: 7.1 ppb Winter: +0.4 ppb RCP 8.5 emissions Summer: 3.9 ppb Winter: +0.4 ppb SRES A2 emissions Summer: +4.5 ppb Winter: +2.0 ppb

14 Max 1 hr O 3 : Imported O 3 RCP 2.6 emissions Summer: 1.7 ppb Winter: 1.2 ppb RCP 8.5 emissions Summer: +0.2 ppb Winter: +0.6 ppb SRES A2 emissions Summer: +1.9 ppb Winter: +1.8 ppb

15 Max 1 hr O 3 : Impacts of Climate Summer: +1.3 ppb Winter: 0.3 ppb (1σ = ±1.3 ppb) (1σ = ±0.5 ppb) Summary of impacts on Max 1 hr O 3 (in ppb) Scenario Summer Winter EU External CH 4 Climate EU External CH 4 Climate RCP RCP SRES A Note: effect of emissions > effects of climate change on this time scale

16 Effects on Summer Max O 3 Distributions All Europe Impacts of CH 4 and external sources Impacts of A1b climate N. Europe (Ruhr) Exceedence of 60 ppb S. Europe (Po Valley) RCP % RCP % 4.5% SRES A2 37.6%

17 Informing Emissions Policy in 2050 How would controlling CH 4 emissions affect surface O 3? RCPs have O 3 precursor controls but big differences in CH 4 abundance This accounts for most of the difference in O 3 between scenarios 75% for N. America and Europe Under future precursor emissions controls, limiting atmospheric CH 4 will become increasingly important!

18 Contributions to O 3 RF 1850s 2000s O 3 Radiative Forcing /mw m 2 Total O 3 RF 378 mwm 2 Δ Emissions w.r.t Interpretation of O 3 RF from ACCMIP studies (Stevenson et al., 2013) Discrepancy in attribution disappears when nonlinearity is accounted for

19 Tropospheric O 3 Burden Changes FRSGC/UCI CTM Szopa et al., 2012

20 FC Simulations Impacts of Climate Change on Surface O s 2090s changes: ΔT = C, ΔQ = %, ΔE isop = % GISS PUCCINI STOC HadAM3 UM CAM Reduced surface O 3 in unpolluted regions (higher H 2 O) Increased O 3 over parts of most polluted regions But large differences across models Changes in SR relationships Greater impact in source region, less impact downwind Doherty et al., 2013

21 Regional emission reductions needed to offset climate change Red: reduction in 2000 emissions needed to offset increased O 3 from climate Blue: climate change leads to reduced O 3 (so emissions can increase) White: emission reductions would lead to increased O 3 Doherty et al., 2013

22 Changes in Transport Patterns Annual average surface CO tracer (50 day lifetime: TP1x) emitted from NA region, and difference due to 2000s 2090s climate changes Complex patterns Lower CO over some of NA source region (enhanced ventilation) Higher CO over some NA regions (more frontal outflow?) Lower CO downwind (transport changes, reduced downwelling?) Could be important for changes in O 3 extremes?

23 Applications of Parameterization Regional mean surface O 3 trends (and attribution) Background/boundary conditions for regional models Climate impacts Tropospheric O 3 burden and RF Health impacts Surface O 3 metrics (mda8, max O 3, etc) Need to address sub grid scale heterogeneity Ecosystem impacts Explore O 3 deposition fluxes (by veg type?) Need to address sub grid scale heterogeneity Need to include aerosol responses! Model runs needed, then characterization of behaviour