How representative are discrete sampling campaigns of...
16
Lee T. Murray 1 ( [email protected] ) and Arlene M. Fiore 2,3 1. Department of Earth and Environmental Sciences, University of Rochester, Rochester, NY 2. Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY 3. Department of Earth and Environmental Sciences, Columbia University, New York, NY How representative are discrete sampling campaigns of reactive tropospheric composition?
Transcript of How representative are discrete sampling campaigns of...
Lee T. Murray1 ([email protected]) and Arlene M. Fiore2,31. Department of Earth and Environmental Sciences, University of Rochester, Rochester, NY2. Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY3. Department of Earth and Environmental Sciences, Columbia University, New York, NY
How representative are discrete sampling campaigns of reactive
tropospheric composition?
Over 40 yrs of 3-D in situ sampling of the troposphere
WOUDC (http://www.woudc.org)
NASA, NOAA, NSF/NCAR, NERC, DLR, et. al
IAGOS (http://iagos.org)MOZAIC/CARIBIC
Observations discretely sample a dynamic 4-D system
Sondes Commercial Field Campaigns
Can we use these observations to constrain CCMs?
Chemical-transport models (CTMs) may be evaluated by exact space-time matchingChemistry-climate models (CCMs) generate their own weather so cannot match observations exactly in space and timeCCMs are typically evaluated with observed climatologies
Are aggregated in situ observations indicative of background mean conditions?Where can these observations be used to constrain processes in CCMs?Can discrete sampling be used to constrain long-term trends?
Questions
Approach: Use GEOS-Chem to compare in situ and “CCM” output
Compare all three to assess suitability of observations to characterize mean atmospheric composition
Decadal Monthly Mean Climatology
“CTM-like sampling”
Sample location at exact time of flight/sonde track
“CCM-like sampling”
Sample location’s clim. monthly mean for each observation
Simulated global “true” background
concentrations
GEOS-Chem CTMv9_02; 2ºx2.5º; 47L; MERRA
Jan 2003-Dec 2012
Data aggregated to G-C spatial, vertical & temporal res.
WOUDC/SHADOZ
MOZAIC/CARIBIC
Field Campaigns*
Total
O3 852,733 1,398,076 116,423 2,367,232
CO 1,375,094 109,677 1,484,771
NOy 84,588 17,268 101,856
NO 27,493 57,462 84,955
NO2 223 49,041 49,264
NOx 3,832 22,695 26,527
PAN 24,371 24,371
HNO3 23,771 23,771
*Pre-AVE, TROCCINOX-2004, COBRA04, INTEX-NA, NEAQS-ITCT, ITOP-DLR, ITOP-UK, FAAM, AVE-04, Polar-AVE, TROCCINOX-2005, AVE-05, AMMA-SCOUT-F20, CR-AVE, INTEX-B-J31, MAXMex-GV, INTEX-B-DC8, INTEX-B-C130, INTEX-B-DUCHESS, INTEX-B-CESSNA, TexAQS-P3B, TC4-DC8, TC4-WB57, ARCPAC, ARCTAS-WP3B, ARCTAS-DC8, START-08, VOCALS-G1, VOCALS-C130, VOCALS-TwinOtter, VOCALS-Dornier, HIPPO-1, HIPPO-2, HIPPO-3, HIPPO-4, DISCOVER-AQ-DC-WP3B, DISCOVER-AQ-DC-UMD, HIPPO-5, DC3-DC8, DC3-GV, DC3-F20, TACTS, ESMVal
Ask me about: Planeflight.dat.YYYYMMDD and merges for 1979-present
2003-2012
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
● ● ● ●● ● ● ● ● ● ●
●
n = 1,272
● ●●
●●
● ● ● ●●
●●
n = 921
●● ●
●
●● ●
● ●
●
● ●
n = 2,507
● ● ●● ●
● ●●
●●
● ●
n = 1,388
● ● ● ● ●
● ●
●●
●
● ●
n = 1,595
●●
●●
●●
●● ●
● ● ●
n = 7,599
● ●●
●● ●
●
●● ●
● ●
n = 2,057
● ●● ●
●
●
●● ●
● ●●
n = 2,101
●●
●
●● ● ●
●
●
●● ●
n = 12,274
●●
●● ● ● ● ●
●● ● ●
n = 12,018
0
25
50
75
100
0
25
50
75
100
0
25
50
75
100
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
O3
(ppb
v)
Ozone most-sampled tropospheric trace gas2003-2012 Sondes + Passenger Programs + Field Campaigns
Ozone increases w/ latitude and altitude; large variability in FT; spring surface maxima
Observations
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●● ●● ●● ●●
●●●●
●● ●● ●● ●● ●●●●
n = 1,272
●●
●●
●●
●●
●●
●●
●●
●● ●
● ●
● ●
● ●●
n = 921
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
● ●
●
●●
●
●
n = 2,507
●
●
●
●
●●
●●
●●●
●●
●
●
● ●●
●
●
●●
●●
n = 1,388
●●
●
●
●●
●
●
●● ●
●●
●●● ●
●
●●
●●
●●
n = 1,595
●
●
●
●
●
●●
● ●●
●●
●●
●
●
●
●
●
●
●
●
●
●
n = 7,599
●
●
●
●
●
● ●
●
●● ●
●
●●●●
●● ●●●●
●
●
n = 2,057
●●
●●●● ●●
●●
●●
●
● ●
●
●● ●
●
●● ●
●
n = 2,101
●●●●
●●
●●
●●
●● ●●
●
●
●
●
●●
●●
●●
n = 12,274
●●●●
●●
●
●
●●
●● ●●
●●
●●
●● ●● ●●
n = 12,018
0
25
50
75
100
0
25
50
75
100
0
25
50
75
100
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
O3
(ppb
v)
GEOS-Chem space-time sampling captures salient features
GEOS-Chem biased high ~7%; captures 87% of meridional, vertical, seas. variability (n=10 reg. x 12 mon.)
ObservationsCTM-Sampling
2003-2012 Sondes + Passenger Programs + Field Campaigns
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●●● ●●● ●●● ●●●
●●●●●●
●●● ●●● ●●● ●●●
●●●●●●
n = 1,272
●●●
●●●
●●●
●●●
●●●
●●●
●●●
●●● ●
●● ●
●●●
●● ●●●
n = 921
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●
●● ●
●●
●●
●
●
●●
n = 2,507
●
●●
●
●●
●●●
●●●
●●●
●
●●●
●●●
●● ●●●
●
●●
●●●
●●●
n = 1,388
●●●
●
●●
●●●
●
●●
●●● ●
●●●
●●●●● ●
●●
●●●
●●●
●●●
n = 1,595
●
●
●
●
●●
●
●●
●
●●
●●●
●●●
●●●
●
●●
●
●●
●
●●
●
●●
●
●●
n = 7,599
●
●●
●
●●
●
●●
●
●●
●●● ●
●●
●●●●●●
●●● ●●●●●●
●
●●
n = 2,057
●●●
●●● ●●● ●●
●
●●●
●●●
●
●● ●
●●
●●● ●
●●
●●● ●
●●
n = 2,101
●●●●●●
●●●
●●●
●●●
●●● ●●●
●
●●
●
●●
●●●
●●●
●●●
n = 12,274
●●●●●●
●●●
●
●●●●●
●●● ●●●
●●●
●●●
●●●
●●● ●●●
n = 12,018
0
25
50
75
100
0
25
50
75
100
0
25
50
75
100
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
O3
(ppb
v)
Sampling O3 decadal climatology matches direct sampling in space and time
2003-2012 Sondes + Passenger Programs + Field Campaigns
CCM decadal mean ozone patterns can be constrained with aggregated climatological observations
ObservationsCTM-Sampling
“CCM”-Sampling
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●●● ●●● ●●● ●●●
●●●●●●
●●● ●●● ●●● ●●●
●●●●●●
● ● ● ● ● ●● ● ● ● ● ●
n = 1,272
●●●
●●●
●●●
●●●
●●●
●●●
●●●
●●● ●
●● ●
●●●
●● ●●●
● ● ●●
● ● ● ● ●●
●●
n = 921
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●
●● ●
●●
●●
●
●
●●
●●
● ●
●
● ●●
● ●●
●
n = 2,507
●
●●
●
●●
●●●
●●●
●●●
●
●●●
●●●
●● ●●●
●
●●
●●●
●●●
●●
● ●●
●●
●● ●
●
●
n = 1,388
●●●
●
●●
●●●
●
●●
●●● ●
●●●
●●●●● ●
●●
●●●
●●●
●●●
● ● ● ●●
●●
●●
●●
●
n = 1,595
●
●
●
●
●●
●
●●
●
●●
●●●
●●●
●●●
●
●●
●
●●
●
●●
●
●●
●
●●
● ● ● ● ● ●● ● ● ●
●●
n = 7,599
●
●●
●
●●
●
●●
●
●●
●●● ●
●●
●●●●●●
●●● ●●●●●●
●
●●● ● ● ● ● ● ● ● ● ● ●●
n = 2,057
●●●
●●● ●●● ●●
●
●●●
●●●
●
●● ●
●●
●●● ●
●●
●●● ●
●●● ● ● ●● ● ● ● ● ● ●
●
n = 2,101
●●●●●●
●●●
●●●
●●●
●●● ●●●
●
●●
●
●●
●●●
●●●
●●●● ●
●●
●●
● ●
●
●● ●
n = 12,274
●●●●●●
●●●
●
●●●●●
●●● ●●●
●●●
●●●
●●●
●●● ●●●● ● ● ● ● ● ● ● ●● ● ●
n = 12,018
0
25
50
75
100
0
25
50
75
100
0
25
50
75
100
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
O3
(ppb
v)
O3 clim. fairly representative of “true” background mean & seasonality
CCM/CTM sampling biased 6% higher than “true” mean; captures 84% of spat./seas. monthly variability
2003-2012 Sondes + Passenger Programs + Field Campaigns
Airports, Sonde Stations
Obs. biased toward South Atlantic ozone maximum; add’l obs. needed to fully constrain zonal mean/variability
ObservationsCTM-Sampling
“CCM”-SamplingTrue Model Mean
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●●●●●●
●●●
●●●
●●●●●●
●●●
●●●
●●●
●
●●● ● ● ● ● ● ●
●● ●
●●
n = 47
●●●
●
●●●●● ●●● ●●
● ●●●
●●●
●●
●
● ● ● ● ● ● ● ●● ●
●●
n = 32
●
●●
●
●●
●
●●
●
●●●
●●●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 3,077
●
●●
●
●●●
●●●●●
●●●
●
●●
●
●●
●
●●
●
●
●●
●● ●
●●
●
●●● ● ● ● ● ● ● ●● ●
● ●
n = 380
●●●
●●●
●
●●●●●
●●●
●●●
●●
●
●●
●
●●●
●
●
●
●
●●
●
●●● ● ● ● ●
●●
●●
●
● ●
n = 824
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●●● ●
●●
●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 7,476
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●●●
●●●
●●● ●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 988
●●●
●
●●●
●●
●
●●
●
●●
●
●●●
●●
●●●
●
●●
●
●●
●●●
●
●
●●● ●
●● ● ● ● ● ● ●
●
n = 1,031
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●●
● ● ●●
● ● ● ● ●●
●
n = 6,813
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●●
● ● ●●
●● ● ● ●
●●
n = 6,794
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
CO (p
pbv)
CO reasonably represented by climatology, except in SH
Large IAV in CO skews statistics for “CCM” method
Model biased, but representative of
seasonality
2003-2012 Passenger Programs + Field Campaigns
ObservationsCTM-Sampling
“CCM”-SamplingTrue Model Mean
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●●●●●●
●●●
●●●
●●●●●●
●●●
●●●
●●●
●
●●● ● ● ● ● ● ●
●● ●
●●
n = 47
●●●
●
●●●●● ●●● ●●
● ●●●
●●●
●●
●
● ● ● ● ● ● ● ●● ●
●●
n = 32
●
●●
●
●●
●
●●
●
●●●
●●●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 3,077
●
●●
●
●●●
●●●●●
●●●
●
●●
●
●●
●
●●
●
●
●●
●● ●
●●
●
●●● ● ● ● ● ● ● ●● ●
● ●
n = 380
●●●
●●●
●
●●●●●
●●●
●●●
●●
●
●●
●
●●●
●
●
●
●
●●
●
●●● ● ● ● ●
●●
●●
●
● ●
n = 824
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●●● ●
●●
●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 7,476
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●●●
●●●
●●● ●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 985
●●●
●
●● ●
●●
●
●●
●
●●
●
●●●
●●
●●●
●
●●
●
●●
●●●
●
●
●●● ●
●● ● ● ● ● ● ●
●
n = 1,031
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●●
● ● ●●
● ● ● ● ●●
●
n = 6,807
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●●
● ● ●●
●● ● ● ●
●●
n = 6,794
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
CO (p
pbv)
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●●● ●●●
●●●●●●
●●●
●●●
● ● ● ● ● ● ●●
● ●●
●
n = 38
●●●
●●● ●●● ●●
● ●●●
●●●
● ● ● ● ● ● ● ●● ●
●●
n = 29
●●●
●●●
● ● ● ● ● ● ● ● ● ● ● ●
n = 3
●●●
●●
● ●●● ●●●
●●● ●●●
●
●
●
●●●
●●●
● ● ● ● ● ● ● ●● ●
● ●
n = 69
●●● ●●● ●●●
●●● ●●●
●
●
●
●●●
●●●● ● ● ● ●
●●
●●
●
● ●
n = 105
●
●●●
●●
●
●●
●
●●
●
●●
●
●●
●●●
●●●
● ● ● ● ● ● ● ● ● ● ● ●
n = 30
●
●●
●
●●●●●
●
●● ●
●●
●
●● ●●●
●●●
●
●●
●●●
● ● ● ● ● ● ● ● ● ● ● ●
n = 154
●●
●
●
●● ●
●●
●
●●●
●●
●
●●
●●
● ●●● ●
●
● ●●●
●● ●
●● ● ● ● ● ● ●
●
n = 279
●
●●
●
●●
●
●●
●
●●●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●●
● ● ●●
● ● ● ● ●●
●
n = 638
●
●●
●
●●
●
●●
●
●●
●
●●●
●●
●
●●
●
●●
●
●●
●●●
●
●●●●●
●● ● ●
●
●● ● ● ●
●●
n = 1,166
0
50
100
150
200
0
50
100
150
200
0
50
100
150
200
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
CO (p
pbv)
Field campaign aggregation mitigates “plume chasing”
ObservationsCTM-Sampling
“CCM”-SamplingTrue Model Mean
Field Campaigns Only
2003-2012 Passenger Programs + Field Campaigns
Campaigns; AMMA-SCOUT-F20; ARCTAS-DC8; AVE-04; AVE-05; COBRA-04; CR-AVE; DC3-GV; DISCOVER-AQ-DC-WP3B; FAAM; HIPPO; INTEX-B-C130;
INTEX-B-DUCHESS; INTEX-B-DC8; INTEX-NA; ITOP-UK; MAXMex-GV; NEAQS-ITCT; Polar-AVE; Pre-AVE; START-08; VOCALS-C130; VOCALS-G1
Short-lived, infrequently sampled species poorly characterized
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ● ● ● ● ● ●
●
●● ●●● ●●●
●●●
● ● ● ● ● ● ● ● ● ● ● ●
n = 75
● ● ● ● ● ● ● ● ● ● ● ●
● ● ● ● ● ● ●● ● ● ● ●
●●●●●● ●●●
●●●
●●● ●●● ●●● ●●●●●●
●●● ●●●●●●● ● ● ● ● ● ● ● ● ● ● ●
n = 411
●
●● ●●● ●●●
●●●
●
●● ●●
●
●
●●
●
●●
●
●● ●●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 33
●●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●●
●
●
● ●●●
● ● ● ● ● ● ● ● ● ● ● ●
n = 31
●●●
●
●●
●
●●●●●
●●● ●
●● ●●●
●
●● ●●●
●●● ●●● ●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 311
●
●●
●
●●
●
●●
●
●●
●
●●●●●
●●●
●
●●
●
●●
●
●●
●
●●
●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 481
0
2
4
6
0
2
4
6
0
2
4
6
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
NOy
(ppb
v)
ObservationsCTM-Sampling
“CCM”-SamplingTrue Model Mean
2003-2012 Passenger Programs + Field Campaigns
Additional observations critical to characterize reactive nitrogen budgets
Can observations constrain processes in CCMs?
90ºS−30ºS 30ºS−EQ EQ−30ºN 30ºN−90ºN
●●●● ●● ●● ●● ●● ●● ●● ●● ●● ●● ●●
● ● ● ● ● ● ● ● ● ● ● ●
n = 1,128
●● ●● ●● ●● ●● ●● ●● ●● ●● ●● ●● ●●● ● ● ● ● ● ● ● ● ● ● ●
n = 853
●●●● ●●
●●●●
●●
●●●●
●
●
●●
●●●●
● ● ● ● ● ●● ●
●● ●
●
n = 4,255
●● ●● ●● ●●●●
●● ●●●●
●● ●
●●●
●●
● ● ● ● ●● ● ●
● ● ●●
n = 1,373
●● ●● ●● ●● ●● ●● ●● ●●●● ●● ●
● ●●● ● ● ● ● ● ● ● ● ● ● ●
n = 1,774
●● ●●
●● ●● ●●
●●
●●●●
●●●●
●● ●●● ● ● ● ●
● ● ● ● ●● ●
n = 9,892
●● ●● ●●●● ●●
●●●●
●●●● ●● ●●
●●● ● ● ● ● ● ● ● ● ● ● ●
n = 1,998
●● ●● ●● ●● ●● ●● ●● ●● ●● ●● ●● ●●● ● ● ● ● ● ● ● ● ● ● ●
n = 2,118
●● ●● ●● ●●●●
●●●● ●●
●●
●●●● ●●● ● ● ●
●
●● ●
●● ● ●
n = 12,996
●● ●● ●● ●● ●●●●
●● ●● ●● ●● ●● ●●● ● ● ● ●●
● ●● ● ● ●
n = 12,656
0
10
20
30
40
50
0
10
20
30
40
50
0
10
20
30
40
50
250 mb
500 mb
750 mb
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun Jul
Aug
Sep
Oct
Nov
Dec
Month
∆O
3 (p
pbv)
Lightning NOx contribution to ozone at in situ locations (2004-2012)∆O
3 fr
om L
NO
x (pp
bv) CTM-Sampling
“CCM”-SamplingTrue Model Mean
Both CTM and CCM sampling capture mean ∆O3/∆LNOx response; implying
utility for top-down constraints
Only local sens.
NH in situ clim. evenly sample zonal emission-ozone sensitivities (incl. Soil, FF, BB, BVOC); SH does not
Ongoing Work: Assessing Long-Term FT Trends
GEOS-Chem v9.01.03; 4ºx5º; MERRA + MACCity; Jan 1980-Dec 2010
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●●
●
●
●
●●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●●
●
●
●
●
●
●●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●●●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●●
●
●
●●
●●
●
●●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●●●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●
●●●
●
●●
●
●
●
●
●
●
●
●
●●●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●●
●●
●
●●
●
●●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●●
●
●
●
●
●
●●●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●●●●●
●
●●●
●●●●
●●
●
●
●
●●●
●●
●
●
●●
●
●
●
●
●
●
●●●●
●
●
●
●
●
●
●●
●
●●●
●●
●●
●●●
●
●
●
●
●
●
●
●●
●●
●
●●
●●●
●
●
●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●●
●●
●
●●
●
●
●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●●
●●
●●
●
●●
●
●●
●●●
●●●
●●●
●
●●●
●
●
●●●●●●●
●
●
●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●●
●●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●●●
●
●
●
●
●
●
●●●●●
●●●
●
●
●●●
●
●
●
●
●●
●
●
●
●●●●●●
●
●●
●
●
●
●
●
●●
●
●
●
●●
●
●
●
●●●
●
●
●
●
●●
●
●
●
●
●●
●
●●
●●
●
●
●
●
●
●
●
●
●
●
●
●
●
●●
●
●●●●●●●
●●
●
●
●
●●
●
●●●●●
●●
●●●
●
●●
●●●●
●
●
●●
●●
●
●
●
●
●●●
●
●
●
●
●
●●●●●
●●
●
∆O3 = 4.24e−16 ppb/10−yr, p = 1∆O3 = 5.08e−16 ppb/10−yr, p = 1
∆O3 = 0.936 ppb/10−yr, p = 3.73e−11−20
−10
0
10
20
1980 1985 1990 1995 2000 2005 2010Time
Des
easo
naliz
ed O
zone
(ppb
)
Deseasonalized Ozone @ 500 mb above Hohenpeißenberg, Germany
Currently assessing whether aggregated sonde + aircraft data may constrain multi-decadal trends in vertical structure
Statistically significant trend in simulated monthly mean ozone…
..but not in observations or model sampled at observations
Summary
Northern hemispheric sampling mostly indicative of background mean O3 and CO conditions; some biases toward polluted regionsSouthern hemisphere needs additional constraints on zonal asymmetries in O3 and CO and/or longer averaging intervalsReactive nitrogen species poorly characterizedSampling dense enough in northern hemisphere to constrain zonal emission-ozone/CO relationships; less so in the southern hemisphereOngoing work will assess the suitability of the aggregated in situ data to characterize long-term trends
AcknowledgementsThe many individuals that collected, archived and made public data through the yearsWMO/GAW/WOUDC, SHADOZ, MOZAIC, CARIBIC, IAGOS, Pre-AVE, TROCCINOX-2004, COBRA04, INTEX-NA, NEAQS-ITCT, ITOP-DLR, ITOP-UK, FAAM, AVE-04, Polar-AVE, TROCCINOX-2005, AVE-05, AMMA-SCOUT-F20, CR-AVE, INTEX-B-J31, MAXMex-GV, INTEX-B-DC8, INTEX-B-C130, INTEX-B-DUCHESS, INTEX-B-CESSNA, TexAQS-P3B, TC4-DC8, TC4-WB57, ARCPAC, ARCTAS-WP3B, ARCTAS-DC8, START-08, VOCALS-G1, VOCALS-C130, VOCALS-TwinOtter, VOCALS-Dornier, HIPPO-1, HIPPO-2, HIPPO-3, HIPPO-4, DISCOVER-AQ-DC-WP3B, DISCOVER-AQ-DC-UMD, HIPPO-5, DC3-DC8, DC3-GV, DC3-F20, TACTS, ESMVal
