1. GLOBAL CONSTRAINTS ON BIOGENIC VOC EMISSIONS FROM ATMOSPHERIC OBSERVATIONS Daniel J. Jacob with Paul I. Palmer, Dorian S. Abbot, May Fu, Brendan Field, Mat J. Evans, Yaping Xiao and Randall V. Martin (Dalhousie U. ), Kelly V. Chance (Harvard-Smithsonian), Hanwant B. Singh (NASA-Ames), Joost DeGouw (NOAA/AL), Armin Hansel (U. Innsbruck), Don Blake (UCI), Nicholas Jones (U. Woolagong)

2. PART1: MAPPING OF ISOPRENE EMISSIONS USING HCHO COLUMN MEASUREMENTS FROM SPACE

3. SPACE-BASED MEASUREMENTS OF HCHO COLUMNS AS CONSTRAINTS ON VOC EMISSIONS VOCHCHO Oxidation (OH, O 3, NO 3 ) Emissions many steps h  nm), OH lifetime of hours 340 nm

4. MEASUREMENT OF HCHO COLUMNS FROM THE GOME SATELLITE INSTRUMENT (P.I. John Burrows) HCHO column is determined from backscattered solar radiance in 340 nm absorption band Instrument is in polar sun-synchronous orbit, 10:30 a.m. observation time 320x40 km 2 field of view, three cross-track scenes Complete global coverage in 3 days Operational since 1995 Expect higher-resolution measurements soon from SCIAMACHY (30x60 km 2, launched 2002) and OMI (13x24 km 2, to be launched in June)

5. RETRIEVING SLANT COLUMNS FROM SOLAR BACKSCATTER MEASUREMENTS absorption wavelength   Slant optical depth EARTH SURFACE Scattering by Earth surface and by atmosphere Backscattered intensity I B “Slant column”

6. FITTING OF HCHO SLANT COLUMNS FROM GOME SPECTRA [Chance et al., 2000]  s = 1.0 ± 0.3 x10 16 cm -2  s = 3.0 ± 0.4 x10 16 cm -2  s = 8.4 ± 0.7 x10 16 cm -2 Fitting uncertainty of 4x10 15 molecules cm -3 corresponds to ~ 1 ppbv HCHO in lowest 2 km

7. HCHO SLANT COLUMNS MEASURED BY GOME (JULY 1996) High HCHO regions reflect VOC emissions from fires, biosphere, human activity -0.5 0 0.5 1 1.5 2 2.5x10 16 molecules cm -2 South Atlantic Anomaly (disregard)

8. AIR MASS FACTOR (AMF) CONVERTS SLANT COLUMN  S TO VERTICAL COLUMN  “Geometric AMF” (AMF G ) for non-scattering atmosphere: EARTH SURFACE 

9. IN SCATTERING ATMOSPHERE, AMF DEPENDS ON VERTICAL DISTRIBUTION OF ABSORBER Observations (Y.N. Lee) Model SOS (southeast U.S., Jul 1995) Use GEOS-CHEM chemical transport model to specify shape of vertical profile for given scene HCHO

10. AMF CALCULATION FOR SCATTERING ATMOSPHERE Geometric AMF GOME sensitivity = f (sun angle, albedo, aerosols, cloud…) RADIATIVE TRANSFER MODEL Vertical concentration profile shape factor (normalized) ATMOSPHERIC CHEMISTRY MODEL (GEOS-CHEM) Vertical column = Slant column AMF From GOME From model

11. GOME sensitivity w(z) HCHO mixing ratio profile S(z) (GEOS-CHEM) what GOME sees AMF G = 2.08 actual AMF = 0.71 ILLUSTRATIVE PROFILE FOR SCENE OVER TENNESSEE

12. FORMALDEHYDE COLUMNS FROM GOME: July 1996 means …compare to GEOS-CHEM including GEIA biogenic VOC emissions and EPA anthropogenic VOC emissions GEOS-CHEM vs. GOME: R = 0.83, bias = +14%

13. RELATING HCHO COLUMNS TO VOC EMISSION VOC i HCHO h (340 nm), OH oxn. k ~ 0.5 h -1 Emission E i smearing, displacement In absence of horizontal wind, mass balance for HCHO column  HCHO : yield y i … but wind smears this local relationship between  HCHO and E i depending on the lifetime of the parent VOC with respect to HCHO production: Local linear relationship between HCHO and E VOC source Distance downwind  HCHO Isoprene  -pinene propane 100 km

14. SEASONALITY OF GOME HCHO COLUMNS (9/96-8/97) Largely reflects seasonality of isoprene emissions; general consistency with GEIA but also some notable differences SEP AUG JUL OCT MAR JUN MAY APR GOME GEOS-CHEM (GEIA) GOME GEOS-CHEM (GEIA)

15. INTERANNUAL VARIABILITY OF GOME HCHO COLUMNS 1995 1996 1997 1998 Augusts 1995-2001: correlation with temperature anomaly explains some but not all of the HCHO column variability 1999 2000 2001 GOME HCHO Temp. anomaly

16. OZARKS “ISOPRENE VOLCANO” AS SEEN BY GOME (but not always) GOME HCHO columns over the Ozarks, July 1996: daily orbits and relationship to temperature Temperature dependence of isoprene emission (GEIA)

17. ULTIMATE YIELD OF HCHO FROM ISOPRENE Uncertainty in peroxide recycling under low-NO x conditions: OH. OO NO HCHO, MVK, MACR… HO 2 HOO h,OH ? Isoprene peroxides are recycled in GEOS-CHEM (consistent with MCM)

18. HCHO COLUMN vs. ISOPRENE EMISSION RELATIONSHIP IN GEOS-CHEM MODEL Isoprene emission [10 13 atomC cm -2 s -1 ] NWNE SESW Model HCHO column [10 16 molec cm -2 ] Results for U.S. quadrants in July 1996 simulation w/ 2 o x2.5 o horizontal resolution show: (1) dominance of isoprene emission as predictor of  HCHO variability; (2) linear relationship between the two Standard simulation HCHO from simulation w/o Isoprene emission We use this relationship to derive “top-down” isoprene emissions from the GOME HCHO column observations R 2 = 0.51 R 2 = 0.65 R 2 = 0.43 R 2 = 0.49

19. ISOPRENE EMISSION INVENTORIES, JULY1996 GEIA (7.1 Tg) BEIS2 (2.6 Tg) GOME top-down (5.7 Tg) Paui Palmer to show comparisons to MEGAN inventory Wednesday

20. MODEL vs. OBSERVED SURFACE HCHO Mean daytime HCHO observations Jun-Aug 1988-1998 GEOS-CHEM simulation with “GOME” isoprene emissions Inventoryr2r2 Bias GOME0.71-9% GEIA0.47+17% BEIS20.58-40% high outliers GOME isoprene emission inventory gives better fit to surface HCHO data than either GEIA or BEIS2

21. WHAT ABOUT THE REST OF THE WORLD? We’re starting to look at China High emissions from forests in NE China? Need to be careful about possible fire influence GOME (July 1997)GEOS-CHEM using GEIA (July 1997)

22. PART 2: GLOBAL BUDGET OF METHANOL

23. GLOBAL GEOS-CHEM BUDGET OF METHANOL (Tg yr -1 ) GLOBAL GEOS-CHEM BUDGET OF METHANOL (Tg yr -1 ) with (in parentheses) ranges of previous budgets from Singh et al. [2000], Heikes et al. [2002], Galbally and Kirstine [2003], Tie et al. [2003] Plant growth: 128 (50-312) Ocean uptake: 11 (0-50) Plant decay: 23 (13-20) Biomass burning: 9 (6-13) Biofuels: 3 Urban: 4 (3-8) CH 3 OH lifetime 10 days (5-12) VOCCH 3 O 2 CH 3 O 2 (85%) RO 2 (15%) Atmospheric production: 37(18-31) OH 130 OH(aq) - clouds <1 (5-10) Dry dep. (land) : 56 Wet dep.: 12 NPP based, x3 for young leaves

24. SIMULATED METHANOL CONCENTRATIONS IN SURFACE AIR Representative observations In ppbv [Heikes et al., 2002]: Urban: 20 (<1-47) Forests: 10 (1-37) Grasslands: 6 (4-9) cont. background: 2 (1-4) NH oceans: 0.9 (0.3-1.4) January July ppb

25. METHANOL-CO RELATIONSHIP OVER N. INDIAN OCEAN INDOEX cruise [Wisthaler et al., 2002] Positive correlation reflects outflow from India, where CO is mainly from combustion and methanol mostly from terrestrial biosphere Small dots: obs Large dots: model

26. METHANOL IN ASIAN OUTFLOW OVER PACIFIC Observed [H.B. Singh] Model Plant growth tracer Biomass burning tracer TRACE-P campaign, March-April 2001

27. METHANOL AT NORTH AMERICAN HIGH LATITUDES (TOPSE MISSION) : MODEL (red) vs. OBSERVED (black) Observations from D.R. Blake (U.C. Irvine)

28. METHANOL VERTICAL PROFILES OVER S. PACIFIC Could the atmospheric source from CH 3 O 2 + CH 3 O 2 be underestimated? Could there be a biogenic VOC “soup” driving organic and HO x chemistry in the remote troposphere? In model over S. Pacific, CH 4 OH CH 3 O 2 HO 2 CH 3 OOH NO HCHO CH 3 O 2 0.6 CH 3 OH +… ~ 70% ~ 20% 5-10% Photochemical model calculations for same data set [Olson et al., 2001] are 50% too high for CH 3 OOH, factor of 2 too low for HCHO 0 0.6 1.2 1.8 2.4 3 Methanol, ppbv model atmospheric source obs. From H.B. Singh

29. PART 3: ACETONE. ACETALDEHYDE, HCN

30. GLOBAL GEOS-CHEM BUDGET OF ACETONE (Tg yr -1 ) with photolysis update from Blitz et al. [2004] GLOBAL GEOS-CHEM BUDGET OF ACETONE (Tg yr -1 ) from Jacob et al. [2002] with photolysis update from Blitz et al. [2004] Vegetation: 33 (22-42) Ocean uptake: 14 19 Plant decay: 2 (-3 - 7) Biomass burning: 5 (3-7) Urban: 1 (1-2) (CH 3 ) 2 CO lifetime 15 days 18 days OH 46 27 Dry dep. (land) : 9 propane i-butane OH terpenes MBO OH, O 3 h microbesDOC+hv Ocean source: 27 (21-33) 21 (16-26) 7 (3-11) 28 37 12

31. OCEANIC SOURCE OF ACETONE IN MODEL NEEDED TO MATCH OBSERVATIONS OVER S. PACIFIC a priori sources/sinks;  2 = 1.3 Optimized sources/sinks (including “microbial” ocean sink, photochemical ocean source);  2 = 0.39 from Jacob et al. [2002] obs from Solberg et al. [1996] obs. From H.B. Singh

32. MORE RECENT AIRCRAFT DATA IMPLY A NET OCEANIC SINK FOR ACETONE Observed Model TRACE-P observations over tropical North Pacific in spring [Singh et al., 2003]

33. CORRELATION OF ACETONE WITH TRACERS OF SOURCES IN ASIAN OUTFLOW (TRACE-P DATA) Acetone [pptv] CO [pptv] Methanol [pptv]HCN [pptv] Ethane [pptv] Acetone [pptv] Acetone =  0 +  1 [Ethane] +  2 [HCN] +  3 [Methanol] Intercept = 200 pptv Acetone =  0 +  1 [CO] +  2 [HCN] +  3 [Methanol] Intercept = 238 pptv Multiple regression: Propane source Continental source Biomass burning source Biogenic source How to explain the pervasive 200 pptv acetone background?

34. HIGH CONCENTRATIONS OF ALDEHYDES OVER REMOTE NORTH PACIFIC Singh et al. [2003] Inconsistent with observed PAN/NO x [Staudt et al., 2003] …Also inconsistent with observed PAN/PPN ~100! HOW RELIABLE ARE THE OBSERVATIONS?

35. GLOBAL GEOS-CHEM BUDGET OF HCN (Tg N yr -1 ) GLOBAL GEOS-CHEM BUDGET OF HCN (Tg N yr -1 ) from Li et al. [2003] Vegetation: ? Ocean uptake: 0.73 Biomass burning: 0.63 HCN lifetime 5 mos. OH 0.1 Residential fuel: 0.2 HCN(aq)/CN - 3 mos.

36. FTIR SURFACE-BASED MEASUREMENTS OF HCN COLUMNS Lines are model values Japan Kitt Peak Jungfraujoch Spitzbergen

37. CONFIRMATION OF BIOMASS BURNING SOURCE, OCEAN SINK IN TRACE-P AIRCRAFT DATA Mean vertical profile over remote N. Pacific Correlation with CO Li et al. [2003]; HCN observations from H.B. SIngh

38. SIMULATED GLOBAL DISTRIBUTION OF HCN [Li et al., 2003] Lauder Neumayer

39. …BUT MODEL UNDERESTIMATES RECENT HCN COLUMN OBSERVATIONS AT NEUMAYER Obs, Neumayer (N. Jones) Obs, Lauder (C. Rinsland) Model, Neumayer Model, Lauder Need better understanding of HCN(aq)/CN - chemistry in ocean and of role of terrestrial biosphere in HCN budget Southern Ocean is not a sink for HCN; compensation point?