1. On the need for a DOAS reference spectrum:
Challenges and Opportunities for retrievals
R. Volkamer, S. Coburn, B. Dix, T. Koenig, I. Ortega, R. Sinreich, R. Thalman, and the TORERO Science team
University of Colorado at Boulder, CO, USA.
DOAS reference spectra:
BrO, IO, CHOCHO
H2O line lists
Recent comparison campaigns:
Field measurements in remote free troposphere (Volkamer et al., 2015)
EUPHORE (Thalman et al., 2015)
3. BrO vertical profiles in Pensacola, FL (Coburn et al., in prep.)
Noxon et al., Stutz and Platt, 2008

2. Role of the DOAS reference spectrum
Instrument properties
Lamp emission spectrum (e.g., variable Xe-emission lines),
Optical components that modify transmission (e.g., fibers, mirrors, grating)
Detector pixel-to-pixel variability of sensitivity, ethalon, non-linearity
Atmospheric state variables:
Fraunhofer lines,
Stratospheric gases (NO2, BrO, etc.)
Free tropospheric gases ?
Aerosols, etc
Benefit: RMS is ~10-100x lower:
External ref: d‘ ~ 10-3 to 10-2
Internal ref: d‘ ~ 10-5 to 10-4
Drawback: a relative technique!
dSCD = SCD - SCDref

3. CU-AMAX-DOAS instrument aboard NSF/NCAR GV
University of Colorado Airborne Multi-AXis
Differential Optical Absorption Spectroscopy
spectrographs/detectors
Volkamer et al., SPIE 2009
Baidar et al., AMT 2013
Telescope pylon
motion
stabilized
Sinreich et al., 2010, ACP
Coburn et al., 2011, AMT
Baidar et al., 2013, AMT
Dix et al., 2013, PNAS
Oetjen et al., 2013, JGR
Sun
elevation angle
height
concentration
solar zenith angle
I added the reference to SPIE
* 30 sec, ** 60 sec integration time
Passive remote sensing column observations
Trace gases and aerosols
Volkamer et al., 2009

4. xr = xa + A(xt-xa) SCD = dSCD + SCDRef 27 xr retrieved profile
xa a-priori profile
A averaging kernel matrix
xt true atmospheric profile
Tropospheric BrO: limb avoids complications from high O3 dSCD!
Excellent control to separate tropospheric and stratospheric gases
NO2, H2O comparison with models and in-situ
Volkamer et al., 2015, AMT

5. BrO and IO detection SH tropical troposphere
NH/SH tropics: (1.5 ± 0.3) x1013 molec cm-2
SH sub-tropics: (1.7 ± 0.3) x1013 molec cm-2
SH mid-latitudes: (1.0 ± 0.3) x1013 molec cm-2

6. Active and widespread halogen chemistry in the tropical and subtropical free troposphere
Aircraft: x1013 molec cm-2
(Volkamer et al., 2015; Wang et al., 2015, PNAS)
Satellite: x1013 molec cm-2
(Chance et al., 1998; Wagner et al., 2001; Richter et
al., 2002; Van Roozendael et al., 2002; Theys et al.,
2011)
Ground : <0.4-3 x1013 molec cm-2
(Leser et al., 2003; Hendrick et al., 2007; Theys et al.,
2007; Coburn et al., 2011)
Balloon: x1013 molec cm-2
(Pundt et al., 2002; Schofield et al., 2004, 2006;
Dorf et al., 2008)
Models: x1013 molec cm-2
(von Glasow et al., 2004; Yang et al., 2010; Theys et al.,
2011; Saiz-Lopez et al., 2012; Parrella et al., 2012;
Long et al., 2014) – in the tropics
Wang et al., 2015 PNAS

7. Atmospheric models do not predict any glyoxal over oceans
Satellites show widespread glyoxal over oceans, but disagree over the remote ocean
Wittrock et al., 2006; Myriokefalitakis et al., 2008; Sinreich et al., 2010; Lerot et al., 2010
Wittrock et al., 2006; Myriokefalitakis et al., 2008; Sinreich et al., 2010; Lerot et al., 2010
Atmospheric models do not predict any glyoxal over oceans
Continental source: ~45 TgC/yr
50% unaccounted
30% biogenic (i.e. isoprene)
14% anthropogenic
6% biomass burning
Atmospheric lifetime: ~ 2.5hrs
52% photolysis
18% OH
22% SOA in clouds/aerosols?
8% Dry/wet deposition
Stavrakou et al., 2009

8. Comparison in-situ and remote sensing (CHOCHO)
Volkamer et al., 2015

9. Effect of SCDref on VCD
Trace gases in the FT (SCDref) have a strong effect on the partical MBL and tropospheric VCD (40-90%) !!
Volkamer et al., 2015

10. BrO: Opportunities for ground MAX-DOAS ?
BrO dSCD peaks at EA=10

11. Determine SCDRef from an iterative approach
xr retrieved profile
xa a-priori profile
A averaging kernel matrix
xt true atmospheric profile.
xr = xa + A(xt-xa)
SCD = dSCD + SCDRef

12. Comparing the derived SCDRef with WACCM model
Dorf et al., 2008;
Liang et al., 2014
Determining the SCDref for different zenith spectra
Check consistency with a forward RTM calculation
Final analysis is based on dSCDs using reference spectra that produce consistent results with the atmospheric model

13. Accounting for SCDRef increases AVK in FT
xr retrieved profile
xa a-priori profile
A averaging kernel matrix
xt true atmospheric profile.
xr = xa + A(xt-xa)
SCD = dSCD + SCDRef

14. Comparison with the aircraft BrO profile (different location also in the tropics)
xr retrieved profile
xa a-priori profile
A averaging kernel matrix
xt true atmospheric profile.
xr = xa + A(xt-xa)
SCD = dSCD + SCDRef

15. Conclusions
BrO, IO and glyoxal vertical profiles by limb-sounding in the tropical FT – these gases are ubiquitous in the tropical FT (Volkamer et al., 2015 AMT)
The amount of gas in the free troposphere poses a challenge to the interpretation of ground-based MAX-DOAS (and satellites).
A retrieval was developed for BrO that accounts for SCDref (extends upon Hendrick et al., Theys et al. but avoids high SZA)
Accounting for SCDref can have a 40-80% effect on VCD! VCD primarily increases due to ~0.5 DoF in the FT.
The relevance of SCDref is expected to be most pronounced in clean air! How widespread are gases in the FT?
Funding: NSF-CAREER, NSF-AGS (TORERO), NASA, DoE, EPRI
Acknowledgements: NCAR/EOL and RAF, TORERO team

16. Comparison Ship and AMAX remote sensing (IO)
RMV – 10/30: Needs discussion with Ivan/Barbara on why IO agrees so nicely

17. Selected recent publications
O4 and radiative transfer:
Spinei et al., Direct sun and airborne MAX-DOAS measurements of the collision induced oxygen complex, O2O2, absorption with significant pressure and temperature differences. 2014, Atmos. Meas. Tech. Discuss., 7, , 2014.
Thalman, R., K. Zarzana, M.A. Tolbert and R. Volkamer, Rayleigh scattering cross-section measurements of nitrogen, argon, oxygen and air. 2014, JQSRT, 147,
Thalman, R. and R. Volkamer. Temperature Dependant Absorption Cross-Sections of O2-O2 Collision Pairs between 340 and 630 nm at Atmospherically Relevant Pressure, 2013, PCCP, 15(37),
Coburn et al., Measurements of diurnal variations and Eddy Covariance (EC) fluxes of glyoxal in the tropical marine boundary layer: description of the Fast LED-CE-DOAS instrument. 2014, Atmos. Meas. Tech., 7,
Thalman et al., Instrument inter-comparison of glyoxal, methyl glyoxal and NO2 under simulated atmospheric conditions. 2014, Atmos. Meas. Tech. Discuss.,7,  
Ortega et al., The CU Two Dimensional MAX-DOAS instrument - part 1: retrieval of NO2 in 3 dimensions and azimuth dependent OVOC ratios. 2014, Atmos. Meas. Tech. Disc., in press
Baidar et al., Combining Active and Passive Airborne Remote Sensing to Quantify NO2 and Ox Production near Bakersfield, CA, British J. for Environ. Climate Change, 3(4), 2013,
Instruments & intercomparisons:
Applications:

18. Unique information about marine organic carbon sources
Coburn et al., 2014, AMT

19. H2O, NO2, O3, OVOC interference 9 instruments, 7 techniques
Precision vs Accuracy
H2O, NO2, O3, OVOC interference
9 instruments, 7 techniques
Atmospheric concentrations (CE-DOAS, BBCEAS, LIP)
Thalman et al., 2015, AMT