Chapter 3: Observations

1. Chapter 3: Observations 3rd Meeting of Task Force on Hemispheric Transport of Air Pollution Coordinating Lead Authors: David Parrish, Hajime Akimoto (contributions from many) Presented by: Randall Martin, Dalhousie University Rich Scheffe, EPA

2. Acknowledge Contributors Hajime Akimoto, John Burrows, Russ Dickerson, David Edwards, Mat Evans, Shiro Hatakeyama, Tony Hollingsworth, Dan Jaffe, Gerard Jennings, Randall Martin, David Parrish, Joe Prospero, Lorraine Remer, Stuart Penkett, Ulrich Platt, Rich Scheffe, Akinori Takami, Kjetil Tørseth

3. Topics • Background on Observations • Satellite Perspective • Transport • Emissions • In Situ • Ozone and its precursors • Aerosols • Surface Impacts • Trends • Recommendations

4. Column abundance (through Satellites) Modeling Field Campaigns Surface Monitors Complementary Observations • Independent evidence of transport • Trends of background concentrations • Evaluating and improving models

5. Major Nadir-viewing Space-based Measurements of Tropospheric Composition (Not Exhaustive) Solar Backscatter & Thermal Infrared

6. Satellite Perspective • Long-Range Transport • Emissions

7. Dobson Units Transport Evidence from Satellites: Ozone, CO and NO2 Tropospheric O3 from GOME for summer 1997 Liu et al., 2006 CO from MOPITT for July 2004 Pfister et al., 2006 Tropospheric NO2 from SCIAMACHY for summer 2004 Martin et al., 2006

8. AIRS Observations of an Asian Plume May 5 AIRS Total CO Column for 2006 May 6 May 7 May 8 Lin Zhang Daniel Jacob Wallace McMillan

9. April 2001 Dust Transport Event Observed from TOMS April 11/10 April 10/9 April 9/8 April 14/13 April 13/12 April 12/11

10. 2006 Dust Transport Event Observed from CALIPSO Aug 17 Aug 18 Aug 19 Aug 20 Aug 21 Aug 25 Aug 22 5 km Aug 28 3 km Aug 23 David Winker

11. W e s t P a c i f i c E a s t P a c i f i c G O C A R T B C + O M + S u l f a t e M O D I S P o l 5 0 - 6 0 N 4 0 - 5 0 N 3 0 - 4 0 N 2 0 - 3 0 N 1 0 - 2 0 N - 2 . 0 0 . 0 2 . 0 4 . 0 6 . 0 8 . 0 - 4 - 3 - 2 - 1 F l u x ( T g / y e a r ) Attempting Quantitative Flux Estimates • Combines: • MODIS Aerosol Optical Depth (AOD) • Transport height from field experiments • AOD/mass from field experiments • GEOS-4 winds G O C A R T B C + O M + S u l f a t e l u t i o n M O D I S P o l l u t i o n 5 0 - 6 0 N 4 0 - 5 0 N 3 0 - 4 0 N Latitude 2 0 - 3 0 N 1 0 - 2 0 N 0 1 2 3 4 FLUX (Tg/yr) FLUX (Tg/yr) F l u x ( T g / y e a r ) Yu et al., submitted

12. Satellite Perspective • Long-Range Transport • Emissions

13. Close Relationship of NOx and VOC Emissions With Satellite Tropospheric NO2 and HCHO Columns Satellite Tropospheric NO2 column ~ ENOx Tropospheric HCHO column ~ EVOC BOUNDARY LAYER hv NO2 NO/NO2  W ALTITUDE NO HCHO CO OH hours O3 hours VOC lifetime <day HNO3 Emission Emission Nitrogen Oxides (NOx) Volatile Organic Compound (VOC)

14. Cloud-filtered Tropospheric NO2 Columns Retrieved from SCIAMACHY 2004 –2005 Martin et al., 2006

15. GOME Satellite NO2 Trends (1995-2002) Richter et al., 2005

16. Day-to-day and Intra-day Variation in NO2 Columns Day of Week (GOME) SCIAMACHY (10 AM) – OMI (1:30 PM) for August 2006 Diurnal variation driven by diurnal variation in emissions and photochemistry Beirle et al., 2003 Boersma et al., submitted

17. Top-Down Constraints on NOx Emissions Inverse Modeling 2004-2005 SCIAMACHY Tropospheric NO2 (1015 molec cm-2) NOx emissions (1011 atoms N cm-2 s-1) Martin et al., 2006 NOx Emissions Reduction in NOx Emissions During Traffic Restrictions Observed by OMI Sino-African Summit in Beijing Wang et al., 2007 Date

18. Fuel Combustion 1. Spatial location of FF-dominated regions in a priori (>90%) 1 2 Biomass Burning 2. Spatiotemporal distribution of fires used to separate BB/soil VIRS/ATSR fire counts Soils No fires + background Algorithm for partitioning top-down NOx inventory GOME NOx emissions (2000) Algorithm tested using synthetic retrieval Jaeglé et al., 2005

19. Top-Down Constraints on Isoprene Emissions Inverse Modeling OMI HCHO Columns for June-Aug 2006 Isoprene Emissions Millet et al., submitted Isoprene dominant source when ΩHCHO is high Other VOCs give rise to a relatively stable background ΩHCHO  Not to variability detectable from space ΩHCHO variability over N. America driven by isoprene Millet et al., 2006

20. Source Strengths Inferred from MOPITT CO Observations Adjustments in CO inventories Correction factors to a priori Asian CO sources for February – April 2001 Pétron et al., 2004 Kopacz et al., submitted

21. Aerosol Source Strengths from MODIS AOD MODIS AOD for August 19-30, 2000 Inverse Model AOD Fine Aerosol Source (107 kg/day) Dubovik et al., 2007

22. Satellite Summary Important role in observing long-range transport and constraining emission estimates Near Term: (Considerable Data Mining Remains) • Evaluate and improve satellite retrievals • Additional in situ vertical profiles • Additional development of inverse modeling capability • Application of top-down information to inform bottom-up inventory development • Increasing data use with efficient interfaces Long Term • Few planned future launches • Geostationary platforms for temporal variation

23. In situ observations

24. Ozone and Precursors • View major Westerly transport corridors • Trans Pacific, Atlantic, Europe-Asia • Recognize complications • Background vs transport vs production vs stratospheric intrusion vs shifting transport patterns • Adequacy of measurement locations, frequencies and parameters

25. In situ evidence for O3 and precursors: Asia to NA Observations at the Mt. Bachelor Observatory in central Oregon: April 25th, 2004 Strong correlations between precursors Back-trajectories, GEOS-Chem and chemical signature (Hg/CO) confirmed the Asian industrial source. (Jaffe et al., 2005; Weiss-Penzias et al 2006)

26. In situ evidence for O3 and precursor transport: Asia to NA Observations through airplanes (IGAC-ITCT-2K2 study) • Elevated CO on May 5, 10, 17 mark Asian emission plumes. • Each CO plume accompanied by elevated oxidized nitrogen species at similar CO/NOy ratio • Different stories for O3: • - May 5 high altitude plume with little O3 enhancement; NOy in form of PAN. • May 10 high altitude plume with NOy in form of PAN; O3 likely from stratospheric intrusion. • May 17 plume descended from high altitude; PAN has been converted to NOx and then to HNO3; the NOx likely lead to in situ formation of O3, although stratospheric O3 contribution cannot be excluded Adapted from Cooper et al., 2004; Nowak et al., 2004

27. In situ evidence for O3 and precursor transport: NA to Europe Observations through airplanes (ICARTT) and modeling (Pink and Blue lines are linear regressions to 20 and 23 July measurement data.) • A forest fire plume was sequentially sampled by three aircraft during transport across the North Atlantic Ocean • Increasing O3 versus CO slope during trans-Atlantic transport indicates in situ O3 formation. • Lagrangian model calculations (Pink and Blue symbols), initiated with black data points, give excellent reproduction of the measurements (indicated by Pink and Blue lines), 23 July measurements near French Coast 20 July measurements north of the Azores 18 July measurements near North American East Coast Real et al., 2004;

28. In situ evidence for O3 and precursor transport: Europe to Asia Mondy surface station, East Siberia: air masses delineated by origin Over land transport difficult to directly discern; occurs predominately within the continental boundary layer over Eurasia Pochanart et al., 2004;

29. Trends in Background Ozone

30. Ozone Summary • Clear observational evidence of direct transport of O3 and precursors, as well as production across transport corridors • Remaining questions • What is driving decadal increase in background O3 (and what can we say about very recent patterns?) • Can we delineate transport flux from “background” air? • Role of tracers (e.g., CO, Hg) in delineating source origins: anthropogenic, biomass nurning, start. intrusion • How well suited is our in situ observation system? • Role of satellites in tracking ozone?

31. Aerosols • Primary sources: dust, burning, sea spray, anthropogenic; and, • Secondary formation processes (natural and anthropogenic) • ? Defining “natural” and “anthropogenic” • Dust as carriers of anthropogenic sources • Behavioral impacts on natural events

32. Aerosol outflow from continentsIn Situ and Lidar

33. North American Outflow: Impact on Bermuda IGAC AEROCE The Atmosphere/Ocean Chemistry Experiment: AEROCEContributed by J. M. Prospero, University of Miami (Florida), USA and S.G. Jennings, University College Galway, IrelandReprinted from IGACtivities Newsletter No. 7, December 1996. Westerly flows dominant in Winter, Spring associated with relative peak levels (Prospero, Jennings, Savoi)

34. Outflow form China observed at Cape Hedo, Okinawa Original dominant anthropogenic signal remains throughout dust event (Takami et al., 2006)

35. Aerosol Continental Inflow examples Jaffe et al., 2005 Adopted from Brock et al., 2004: ITCT-2K2 Barrie et al., 2001 Observations at the Mt. Bachelor Observatory in central Oregon: April 25th, 2004 Biomass burning Delineating plume types through combined size distribution and CO profiles-May, 2002 event Anthropogenic gases dust

36. Aerosol summary • Sufficient qualitative observational evidence of aerosol transport, and insights into secondary processes along transport corridors • Major questions include: • Quantifying flux • Organic properties • Black C: source regions and removal processes • Developing an optimal observational approach • How do we integrate satellites, in situ measurements and global models

37. Impacts on surface air quality How does transport impact threshold and background levels of interest? Note: recent changes in U.S. NAAQS have reduced the daily PM2.5 standard from 65 to 35 μg/m3 and there is an expectation that the ozone standard (80ppbv 8hr-average) will be reduced. Such changes will elevate the relative contribution (to local) of transport. The 60 ppbv (1 hr) European ozone standard clearly is subjected to strong transport/background influences on a frequent basis. April, 2001 June 6, 2003 Washington State, U.S. Jaffe et al., 2004 and 2003

38. African Dust impacts on Barbados and U.S. (Prospero and Lamb, 2003) Note: Limited examples of “anthropogenic” aerosol transport (e.g., Jaffe et al., 2005) Shown earlier….implications for new US daily PM2.5 standard 35µg/m3 Conclusion: Attribution to surface impacts requires advances in modeling systems and coupling of observations

39. Long term monitoring • What can we say about decadal and greater time series • Recognize the gaps in adequately sited stations and chemical species • Often requires screening of coherent air masses not confounded by local sources Figure 3.6.1: Linear trends in median, normalized ozone concentrations at Kårvatn, mid-Norway, based on data filtered by back trajectory for only the background transport sectors. The boxes mark the annual medians, and the error bars the 25- and 75-percentiles. [EMEP/CCC 2005]

40. How confident are we in (the interpretation) these trends? (modified from Oltmans et al., 2006) Trends of N.H. surface ozone data

41. Aerosol Precursor Trends Bermuda BCO: Annual (Daily) Means 1989 - 1998 4.0 35 3.5 30 3.0 25 2.5 nss-Sulfate Aerosol (ug/m3) US Emissions SOx (Streets) 20 2.0 15 1.5 10 1.0 5 0.5 0.0 0 1991 1993 1994 1995 1996 1998 1989 1990 1992 1997 nss-SO4 Streets US SOx Figure 3.6.2: Anthropogenic sulfate concentrations on Midway compared to the emissions of SO2 from China [Prospero et al., 2003]. Bermuda BCO: Annual (Daily) Means 1989 - 1998 2.5 50 45 2.0 40 35 1.5 30 US Emissions NOx (Steets) Nitrate Aerosol (ug/m3) 25 1.0 20 15 0.5 10 5 0.0 0 1992 1993 1994 1995 1996 1997 1998 1989 1990 1991 NO3 Streets US NOx Note, uncertainty in NO3 trends in Barbados, Savoi et al., 2002

42. Summary comments on Trends • Available trends data for O3, aerosols and precursors; but • Disagreement? over interpretation of ozone trends • Very limited number of sites • Only two decade record for ozone • Interannual variability compromising interpretation of aerosol trends • Reasonable interpretations of precursor (NOx, SOx) trends for aerosols • Scarcity of sites, data compromises ability to evaluate models • University of Miami aerosol network discontinued

43. Conclusions • Observations demonstrate profound impact of NH transport on distribution of ozone, aerosols and precursors • Models are needed to conduct adequate attribution and projection assessments • Confidence in attribution and projection will require marrying of observations and modeling • Improvements in observational systems include:

44. Recommendations • Long term surface stations • Vertical profiling systems • Satellites • Intensive filed campaigns

45. Sustain and establish long term surface stations in critical transport paths - multi purpose: transport, climate change, trends, evaluation - adequate mix of precursor, indicator and “final” species - build on existing WMO/GAW structure

46. http://www.fz-juelich.de/icg/icg-ii/mozaic/home http://www.fz-juelich.de/icg/icg-ii/iagos/ • Vertical Profiling (sondes, lidar, aircraft) • Support expansion of MOZAIC/CARIBA • 20 more flights + NOx, CO2, aerosols • Proposed sonde network ~ $20M/yr (Cooper) • ground based lidar systems (aerosols and ozone) Proposed sonde network