Presentation is loading. Please wait.

Presentation is loading. Please wait.

Atmospheric chemistry analysis and modeling at LDEO: Connecting global composition, climate, and air quality Atmos. Chem at CU Group Meeting 9-19-13 Arlene.

Similar presentations


Presentation on theme: "Atmospheric chemistry analysis and modeling at LDEO: Connecting global composition, climate, and air quality Atmos. Chem at CU Group Meeting 9-19-13 Arlene."— Presentation transcript:

1 Atmospheric chemistry analysis and modeling at LDEO: Connecting global composition, climate, and air quality Atmos. Chem at CU Group Meeting Arlene M. Fiore Group members: Olivia Clifton (CU), Gus Correa (LDEO), Jean Guo (CU), Nora Mascioli (CU), Keren Mezuman (CU), Lee Murray (LDEO), Luke Valin (LDEO) Close Collaborators: Elizabeth Barnes (now at CSU), Larry Horowitz (GFDL), Meiyun Lin (Princeton/GFDL), Vaishali Naik (UCAR/GFDL), Jacob Oberman (former NOAA Hollings intern), Harald Rieder (now at U Graz, Austria), Lorenzo Polvani (CU), Mike Previdi (LDEO)

2 Today’s Focus How and why might extreme air pollution events change? Mean shifts Variability increases Symmetry changes Figure SPM.3, IPCC SREX  Need to understand how different processes influence the distribution pollutant sources Degree of mixing Meteorology (e.g., stagnation vs. ventilation)  Shift in mean?  Change in symmetry? Changing global emissions (baseline) Changing regional emissions (episodes) T Fires NO x OH PAN H2OH2O VOCs Deposition Feedbacks (Emis, Chem, Dep)

3 Extreme Value Theory (EVT) methods describe the high tail of the observed ozone distribution (not true for Gaussian) JJA MDA8 O at CASTNet Penn State site Gaussian (ppb) Observed MDA8 O 3 (ppb) Generalized Pareto Distribution Model (ppb) EVT Approach: (Peak-over-threshold) for MDA8 O 3 >75 ppb Gaussian: Poor fit for extremes Rieder et al., ERL 2013

4 EVT methods enable derivation of “return levels” for JJA MDA8 O 3 within a given time period CASTNet site: Penn State, PA Return level: describes probability of observing value x (level) within time window T (period) Rieder et al., ERL  Sharp decline in return levels between early and later periods (NO x SIP call)  Consistent with prior work [e.g., Frost et al., 2006; Bloomer et al., 2009, 2010]  Translates air pollution changes into probabilistic language Applied methods to all EUS CASTNet sites; projections for future changes with GFDL CM3 chemistry-climate model

5 The GFDL CM3/AM3 chemistry-climate model years of CM3 CMIP5 simulations [John et al., ACP, 2012 Turner et al., ACP, 2012 Levy et al., JGR, 2013 Barnes & Fiore, GRL, 2013] Options to nudge AM3 to reanalysis; also global high-res (~0.5°x0.5°) [Lin et al., JGR, 2012ab] Donner et al., J. Climate, 2011; Golaz et al., J. Climate, 2011 Naik et al., JGR, 2013 cubed sphere grid ~2°x2°; 48 levels Atmospheric Chemistry 86 km 0 km Atmospheric Chemistry 86 km 0 km Atmospheric Dynamics & Physics Radiation, Convection (includes wet deposition of tropospheric species), Clouds, Vertical diffusion, and Gravity wave Atmospheric Dynamics & Physics Radiation, Convection (includes wet deposition of tropospheric species), Clouds, Vertical diffusion, and Gravity wave Chemistry of gaseous species (O 3, CO, NO x, hydrocarbons) and aerosols (sulfate, carbonaceous, mineral dust, sea salt, secondary organic) Dry Deposition Aerosol-Cloud Interactions Chemistry of O x, HO y, NO y, Cly, Br y, and Polar Clouds in the Stratosphere Forcing Solar Radiation Well-mixed Greenhouse Gas Concentrations Volcanic Emissions Forcing Solar Radiation Well-mixed Greenhouse Gas Concentrations Volcanic Emissions Ozone–Depleting Substances (ODS) Modular Ocean Model version 4 (MOM4) & Sea Ice Model Modular Ocean Model version 4 (MOM4) & Sea Ice Model Pollutant Emissions (anthropogenic, ships, biomass burning, natural, & aircraft) Land Model version 3 (soil physics, canopy physics, vegetation dynamics, disturbance and land use) Land Model version 3 (soil physics, canopy physics, vegetation dynamics, disturbance and land use) SSTs/SIC from observations or CM3 CMIP5 Simulations GFDL-CM3 GFDL-AM3

6 Generating policy-relevant statistics from biased models  Applying correction to projected simulations assumes model captures adequately response to emission and climate changes Rieder et al., in prep CM3 Historical simulation (3 ensemble-member mean) Corrected CM3 simulation (regional remapping) CASTNet sites (observations) Average ( ) number of summer days with O 3 > 75 ppb

7 GFDL CM3 generally captures NE US JJA surface O 3 decrease following NO x emission controls (-25% early 1990s to mid-2000s) Rieder et al., in prep  Implies bias correction based on present-day observations can be applied to future scenarios with NO x changes (RCPs)  Caveat: Model response on low-O 3 days smaller than observed Observed (CASTNet) Model (CM3) JJA MDA8 O 3 (ppb)

8 Climate and Emission Scenarios for the 21 st Century: Representative Concentration Pathways (RCPs) RCP8.5 RCP4.5 RCP4.5_WMGG Percentage changes from 2005 to 2100 Global CO 2 Global CH 4 Global NO x NE USA NO x Enables us to isolate role of changing climate Change in Global T (°C) 2006–2025 to 2081–2100 (>500 hPa) in GFDL CM3 chemistry-climate model [John et al., ACP, 2012] How will surface O 3 distributions over the NE US evolve with future changes in emissions and climate?

9 How does climate warming change extreme O 3 events over the NE USA? Base Years ( ) Change: ( ) – ( ) 1-year MDA8 O 3 Return Levels in GFDL CM3 RCP4.5_WMGG (corrected) Much of region sees ‘penalty’ (Wu et al., 2008) but some ‘benefit’, up to 4 ppb  Signal from regional ‘climate noise’ with only 3 ensemble members?  Robust across models? Change: ( ) – ( ) Rieder et al., in prep

10 Large NO x reductions offset climate penalty on O 3 extremes 1-year Return Levels in CM3 chemistry-climate model (corrected) Summer (JJA) MDA8 Surface O 3 [Rieder et al., in prep] ppb RCP4.5_WMGG: Pollutant emissions held constant (2005) + climate warming RCP4.5: Large NO x decreases + climate warming All at or below 60 ppb Nearly all at or below 70 ppb Rieder et al., in prep  Ongoing work examining relationship between NO x reductions and 1-year return levels

11 Does aerosol forcing influence extreme weather events? N. Mascioli “Aerosol only” forcing scenario in GFDL CM3 chemistry-climate model Warm Spell Duration (consecutive warm days) over Eastern North America Consecutive warm days ( ) – ( )  Decrease in warm spell duration over ENA following increase in anthropogenic aerosol (preindustrial conditions except for aerosol)

12 Identifying key factors contributing to OH variability Murray et al., 2013  Can this explain large inter-model differences in OH changes (e.g., Naik et al., 2013; Voulgarakis et al. 2013)?

13 Linearity with photochemical steady-state relationship holds for some models but not others: what factors are missing? Transient (CMIP5) and Time-slice (ACCMIP) simulations Annual mean tropospheric [OH] (10 6 molec cm -3 ) GFDL CM3 model  Incorporating additional term for sulfate achieves linearity: aerosol influence on HO x sink OR correlated causal mechanism?

14 Space-based CH 2 O: often used to constrain biogenic emissions; what can we learn from background variability? Atmospheric Column CH 2 O as retrieved by OMI on Aura satellite Jun-Aug   Dec-Feb L. Valin, NOAA/LDEO Constraints on variability in background (CH 4 ) oxidation?

15 Over the Sahara, model formaldehyde columns correlate strongly with year-to-year variations in CH 4 loss rates Dec-Feb L. Valin, NOAA/LDEO  Exploring whether this is detectable with current space-based measurements GFDL AM3 model nudged to NCEP meteorology ( ) Regional averages over 20-28N, E

16 satellites suborbital platforms models AQAST Pollution monitoring Exposure assessment AQ forecasting Source attribution Quantifying emissions Natural & foreign influences AQ processes Climate-AQ interactions AQAST

17 All AQAST projects connect Earth Science and air quality management:  active partnerships with air quality managers with deliverables/outcomes  self-organizing to respond quickly to demands  flexibility in how it allocates its resources  INVESTIGATOR PROJECTS (IPs): members adjust work plans each year to meet evolving AQ needs  “TIGER TEAM” PROJECTS (TTs): multi-member efforts to address emerging, pressing problems requiring coordinated activity Our projects for 2013 Stratospheric and Asian influence on inter-annual WUS surface O 3 variations (M. Lin) climate versus emissions impacts on 21 st C U.S. baseline O 3 (seasonality) Analysis of background O 3 influence on O 3 measured at new NV sites (M. Lin). TTs (pending selection) (PIs: Fiore, Holloway) Source attribution for high-O 3 events over EUS (PI: Streets) Satellite-derived NO x emissions and trends In progress: Multi-model W126 analysis (vegetation exposure) In progress: Estimates of background O 3 from 2 models click on “projects” for brief descriptions + link to pdf describing each project What makes AQAST unique? How is our group contributing?

18 Surface O 3 seasonal cycle over NE USA reverses – cold season increases- under extreme warming scenario (RCP8.5) in the GFDL CM3 model NO x decreases ? Monthly mean surface O 3 (land only) over the NE USA (36-46N, 70-80W) ppb month 3 ensemble members for each scenario O. Clifton

19 Why does surface O 3 increase in winter/spring over NE USA under RCP8.5? Change in monthly mean surface O 3 (land only) over the NE USA (36-46N, 70-80W) ( ) – ( ) ppb NO x decreases RCP8.5 (3 ensemble members) ? O. Clifton

20 Doubling methane raises surface O 3 over NE USA ~5-10 ppb Change in monthly mean surface O 3 (land only) over the NE USA (36-46N, 70-80W) ( ) – ( ) ppb RCP8.5 but chemistry sees 2005 CH 4 RCP8.5 (3 ensemble members) Doubling CH 4 does not fully explain wintertime increase O. Clifton Will the NE USA resemble present-day remote, high-altitude W US sites by 2100?

21 Setting achievable standards requires accurate knowledge of background levels 120 ppb hr avg 84 ppb hr 75 ppb hr O 3 (ppbv) 20 U.S. National Ambient Air Quality Standard for O 3 has evolved over time Future? (proposed) typical U.S. “background” (model estimates) [ Fiore et al., 2003; Wang et al., 2009; Zhang et al., 2011] Allowable O 3 produced from U.S. anthrop. sources (“cushion”) Lowering thresholds for U.S. O 3 NAAQS implies thinning cushion between regionally produced O 3 and background background events over WUS [Lin et al., 2012ab] 2008 O 3 NAAQS revision relied on background estimates from one model  Compare N. Amer Background in GFDL AM3 vs. GEOS-Chem

22 AM3 (~2°x2°) GEOS-Chem (½°x⅔°) Fourth-highest North American background MDA8 O 3 in model surface layer between Mar 1 and Aug 31, 2006 ppb Estimates of North American background in 2 models (simulations with N. American anth. emissions set to zero)  Models robustly agree N. American background is higher at altitude in WUS  Multi-model enables error estimates, in context of observational constraints  Need to account for biases to estimate useful policy-relevant statistics Higher background (spring): More exchange with surface? Larger stratospheric influence? High AM3 bias in EUS; caution on N. Amer. Background here! Excessive lightning NO x in summer J. Oberman Fiore et al., in prep

23 Connecting atmospheric composition, climate, and air quality: from background oxidation to extreme pollution events Variability in background CH 4 oxidation (OH)  Space-based constraints from formaldehyde columns?  Determine key driving factors (multi-model CMIP5/ACCMIP)  Extend to other regions, seasons, with a focus on extremes  Develop robust relationships with changes in meteorology + feedbacks from the biosphere “Climate penalty” can be offset by NO x reductions which preferentially decrease the highest O 3 events Background ozone and its specific sources over the U.S.  Future shifts in regional vs. transported ozone (incl. CH 4 )?  Harnessing power of multiple models + observations  Variability on scales from daily to multi-decadal  Process-oriented analysis (Asian polln, strat, storms, jets) New Directions  Role of aerosol forcing on extreme weather events  Process studies (lightning NO x, deposition)  Climate variability and composition Change in 1-yr O 3 RL: ( ) – ( ) Change in monthly mean NE USA O 3 : ( ) – ( ) Sahara region Change in WSDI (days) ( ) – ( )


Download ppt "Atmospheric chemistry analysis and modeling at LDEO: Connecting global composition, climate, and air quality Atmos. Chem at CU Group Meeting 9-19-13 Arlene."

Similar presentations


Ads by Google