acp-16-13561-2016.pdf

Atmos. Chem. Phys., 16, 13561–13577, 2016

www.atmos-chem-phys.net/16/13561/2016/

doi:10.5194/acp-16-13561-2016

Why do models overestimate surface ozone in the

Southeast United States?

Katherine R. Travis

, Daniel J. Jacob

1,2

, Jenny A. Fisher

3,4

, Patrick S. Kim

, Eloise A. Marais

, Lei Zhu

, Karen Yu

Christopher C. Miller

, Robert M. Yantosca

, Melissa P. Sulprizio

, Anne M. Thompson

, Paul O. Wennberg

6,7

John D. Crounse

, Jason M. St. Clair

, Ronald C. Cohen

, Joshua L. Laughner

, Jack E. Dibb

, Samuel R. Hall

Kirk Ullmann

, Glenn M. Wolfe

11,12

, Illana B. Pollack

, Jeff Peischl

14,15

, Jonathan A. Neuman

14,15

, and

Xianliang Zhou

16,17

Harvard John A. Paulson School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USA

Department of Earth and Planetary Sciences, Harvard University, Cambridge, MA, USA

Centre for Atmospheric Chemistry, School of Chemistry, University of Wollongong, Wollongong, NSW, Australia

School of Earth and Environmental Sciences, University of Wollongong, Wollongong, NSW, Australia

NASA Goddard Space Flight Center, Greenbelt, Maryland, USA

Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA, USA

Division of Engineering and Applied Science, California Institute of Technology, Pasadena, CA, USA

Department of Chemistry, University of California, Berkeley, CA, USA

Earth System Research Center, University of New Hampshire, Durham, NH, USA

Atmospheric Chemistry Division, National Center for Atmospheric Research, Boulder, CO, USA

Atmospheric Chemistry and Dynamics Laboratory, NASA Goddard Space Flight Center, Greenbelt, MD, USA

Joint Center for Earth Systems Technology, University of Maryland Baltimore County, Baltimore, MD, USA

Department of Atmospheric Science, Colorado State University, Colorado, USA

University of Colorado, Cooperative Institute for Research in Environmental Sciences, Boulder, CO, USA

NOAA Earth System Research Lab, Boulder, CO, USA

Department of Environmental Health Sciences, State University of New York, Albany, New York 12201, USA

Wadsworth Center, New York State Department of Health, Albany, New York, USA

Correspondence to:

Katherine R. Travis (ktravis@fas.harvard.edu)

Received: 3 February 2016 – Published in Atmos. Chem. Phys. Discuss.: 16 March 2016

Revised: 5 October 2016 – Accepted: 6 October 2016 – Published: 1 November 2016

Abstract.

Ozone pollution in the Southeast US involves

complex chemistry driven by emissions of anthropogenic

nitrogen oxide radicals (NO

≡

)

and biogenic

isoprene. Model estimates of surface ozone concentrations

tend to be biased high in the region and this is of concern

for designing effective emission control strategies to meet

air quality standards. We use detailed chemical observations

from the SEAC

RS aircraft campaign in August and Septem-

ber 2013, interpreted with the GEOS-Chem chemical trans-

port model at 0.25

◦

0.3125

◦

horizontal resolution, to bet-

ter understand the factors controlling surface ozone in the

Southeast US. We find that the National Emission Inven-

tory (NEI) for NO

from the US Environmental Protection

Agency (EPA) is too high. This finding is based on SEAC

observations of NO

and its oxidation products, surface net-

work observations of nitrate wet deposition fluxes, and OMI

satellite observations of tropospheric NO

columns. Our re-

sults indicate that NEI NO

emissions from mobile and in-

dustrial sources must be reduced by 30–60 %, dependent on

the assumption of the contribution by soil NO

emissions.

Upper-tropospheric NO

from lightning makes a large con-

tribution to satellite observations of tropospheric NO

that

must be accounted for when using these data to estimate sur-

face NO

emissions. We find that only half of isoprene oxi-

dation proceeds by the high-NO

pathway to produce ozone;

this fraction is only moderately sensitive to changes in NO

Published by Copernicus Publications on behalf of the European Geosciences Union.

13562

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

emissions because isoprene and NO

emissions are spatially

segregated. GEOS-Chem with reduced NO

emissions pro-

vides an unbiased simulation of ozone observations from the

aircraft and reproduces the observed ozone production effi-

ciency in the boundary layer as derived from a regression

of ozone and NO

oxidation products. However, the model

is still biased high by 6

14 ppb relative to observed sur-

face ozone in the Southeast US. Ozonesondes launched dur-

ing midday hours show a 7 ppb ozone decrease from 1.5 km

to the surface that GEOS-Chem does not capture. This bias

may reflect a combination of excessive vertical mixing and

net ozone production in the model boundary layer.

1 Introduction

Ozone in surface air is harmful to human health and veg-

etation. Ozone is produced when volatile organic com-

pounds (VOCs) and carbon monoxide (CO) are photochem-

ically oxidized in the presence of nitrogen oxide radicals

(NO

≡

)

. The mechanism for producing ozone is

complicated, involving hundreds of chemical species inter-

acting with transport on all scales. In October 2015, the US

Environmental Protection Agency (EPA) set a new National

Ambient Air Quality Standard (NAAQS) for surface ozone

as a maximum daily 8 h average (MDA8) of 0.070 ppm not

to be exceeded more than three times per year. This is the lat-

est in a succession of gradual tightening of the NAAQS from

0.12 ppm (1 h average) to 0.08 ppm in 1997 and to 0.075 ppm

in 2008, responding to accumulating evidence that ozone is

detrimental to public health even at low concentrations (EPA,

2013). Chemical transport models (CTMs) tend to signifi-

cantly overestimate surface ozone in the Southeast US (Lin et

al., 2008; Fiore et al., 2009; Reidmiller et al., 2009; Brown-

Steiner et al., 2015; Canty et al., 2015), and this is an is-

sue for the design of pollution control strategies (McDonald-

Buller et al., 2011). Here we examine the causes of this

overestimate by using the GEOS-Chem CTM to simulate

NASA SEAC

RS aircraft observations of ozone and its pre-

cursors over the region in August–September 2013 (Toon et

al., 2016), together with additional observations from surface

networks and satellite.

A number of explanations have been proposed for the

ozone model overestimates in the Southeast US. Fiore et

al. (2003) suggested excessive modeled ozone inflow from

the Gulf of Mexico. Lin et al. (2008) proposed that the ozone

dry deposition velocity could be underestimated. McDonald-

Buller et al. (2011) pointed out the potential role of halogen

chemistry as a sink of ozone. Isoprene emitted from vegeta-

tion is the principal VOC precursor of ozone in the Southeast

US in summer, and Fiore et al. (2005) found that uncertain-

ties in isoprene emissions and in the loss of NO

from forma-

tion of isoprene nitrates could also affect the ozone simula-

tion. Horowitz et al. (2007) found a large sensitivity of ozone

to the fate of isoprene nitrates and the extent to which they

release NO

when oxidized. Squire et al. (2015) found that

the choice of isoprene oxidation mechanism can alter both

the sign and magnitude of the response of ozone to isoprene

and NO

emissions.

The

SEAC

aircraft

campaign

August–

September 2013 provides an outstanding opportunity

to improve our understanding of ozone chemistry over

the Southeast US. The SEAC

RS DC-8 aircraft hosted an

unprecedented chemical payload including isoprene and

its oxidation products, NO

and its oxidation products,

and ozone. The flights featured extensive boundary layer

mapping of the southeast as well as vertical profiling to the

free troposphere (Toon et al., 2016). We use the GEOS-

Chem global CTM with high horizontal resolution over

North America (0.25

◦

0.3125

◦

) to simulate and interpret

the SEAC

RS observations. We integrate into our analysis

additional Southeast US observations during the summer of

2013, including from the NOMADSS aircraft campaign, the

SOAS surface site in Alabama, the SEACIONS ozonesonde

network, the EPA Clean Air Status and Trends Network

(CASTNET) ozone network, the National Acid Deposition

Program (NADP) nitrate wet deposition network, and NO

satellite data from the OMI instrument. Several compan-

ion papers apply GEOS-Chem to simulate other aspects

of SEAC

RS and concurrent data for the Southeast US,

including aerosol sources and optical depth (Kim et al.,

2015), isoprene organic aerosol (Marais et al., 2016), organic

nitrates (Fisher et al., 2016), formaldehyde and its relation

to satellite observations (Zhu et al., 2016), and sensitivity to

model resolution (Yu et al., 2016).

2 GEOS-Chem model description

We use the GEOS-Chem global 3-D CTM (Bey et al., 2001)

in version 9.02 (http://www.geos-chem.org) with modifica-

tions described below. GEOS-Chem is driven with assimi-

lated meteorological data from the Goddard Earth Observing

System (GEOS-5.11.0) of the NASA Global Modeling and

Assimilation Office (GMAO). The GEOS-5.11.0 data have a

native horizontal resolution of 0.25

◦

latitude by 0.3125

◦

lon-

gitude and a temporal resolution of 3 h (1 h for surface vari-

ables and mixing depths). We use a nested version of GEOS-

Chem (Chen et al., 2009) with native 0.25

◦

0.3125

◦

hor-

izontal resolution over North America and adjacent oceans

(130–60

◦

W, 9.75–60

◦

N) and dynamic boundary conditions

from a global simulation with 4

◦

horizontal resolution.

Turbulent boundary layer mixing follows a non-local param-

eterization based on K-theory (Holtslag and Boville, 1993)

implemented in GEOS-Chem by Lin and McElroy (2010).

Daytime mixing depths are reduced by 40 % from the GEOS-

5.11.0 data as described by Kim et al. (2015) and Zhu et

al. (2016) to match aircraft lidar observations. The GEOS-

Chem nested model simulation is conducted for August–

Atmos. Chem. Phys., 16, 13561–13577, 2016

www.atmos-chem-phys.net/16/13561/2016/

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

13563

September 2013, following 6 months of initialization at

◦

resolution.

2.1 Chemistry

The chemical mechanism in GEOS-Chem version 9.02 is de-

scribed by Mao et al. (2010, 2013). We modified aerosol re-

active uptake of HO

to produce H

instead of H

O in

order to better match H

observations in SEAC

RS. We

also include a number of updates to isoprene chemistry, listed

comprehensively in the Supplement (Tables S1 and S2) and

described here more specifically for the low-NO

pathways.

Companion papers describe the isoprene chemistry updates

relevant to isoprene nitrates (Fisher et al., 2016) and organic

aerosol formation (Marais et al., 2016). Oxidation of bio-

genic monoterpenes is also added to the GEOS-Chem mech-

anism (Fisher et al., 2016) but does not significantly affect

ozone.

A critical issue in isoprene chemistry is the fate of the iso-

prene peroxy radicals (ISOPO

)

produced from the oxida-

tion of isoprene by OH (the dominant isoprene sink). When

is sufficiently high, ISOPO

reacts mainly with NO

to produce ozone (high-NO

pathway). At lower NO

lev-

els, ISOPO

may instead react with HO

or other organic

peroxy radicals, or isomerize, in which case ozone is not

produced (low-NO

pathways). Here we increase the mo-

lar yield of isoprene hydroperoxide (ISOPOOH) from the

ISOPO

reaction to 94 % based on observations of

the minor channels of this reaction (Liu et al., 2013). Ox-

idation of ISOPOOH by OH produces isoprene epoxides

(IEPOX) that subsequently react with OH or are taken up

by aerosol (Paulot et al., 2009b; Marais et al., 2016). We use

updated rates and products from Bates et al. (2014) for the

reaction of IEPOX with OH.

ISOPO

isomerization produces hydroperoxy-aldehydes

(HPALDs) (Peeters et al., 2009; Crounse et al., 2011; Wolfe

et al., 2012), and we explicitly include this in the GEOS-

Chem mechanism. HPALDs go on to react with OH or pho-

tolyze at roughly equal rates over the Southeast US. We

use the HPALD

OH reaction rate constant from Wolfe et

al. (2012) and the products of the reaction from Squire et

al. (2015). The HPALD photolysis rate is calculated using the

absorption cross section of MACR, with a quantum yield of

1, as recommended by Peeters and Müller (2010). The pho-

tolysis products are taken from Stavrakou et al. (2010). Self-

reaction of ISOPO

is updated following Xie et al. (2013).

A number of studies have suggested that conversion of

to nitrous acid (HONO) by gas-phase or aerosol-phase

pathways could provide a source of HO

radicals following

HONO photolysis (Li et al., 2014; Zhou et al., 2014). This

mechanism would also provide a catalytic sink for ozone

when NO

is produced by the NO

ozone reaction, viz.,

→

(R1)

→

HONO

(

by various pathways

(R2)

HONO

hυ

→

(R3)

Observations of HONO from the NOMADSS campaign

(https://www2.acom.ucar.edu/campaigns/nomadss) indicate

a mean daytime HONO concentration of 10 ppt in the South-

east US boundary layer (Zhou et al., 2014), whereas the

standard gas-phase mechanism in GEOS-Chem version 9.02

yields less than 1 ppt. We add the pathway proposed by Li

et al. (2014), in which HONO is produced by the reaction

of the HO

O complex with NO

, but with a slower rate

constant (

−

molecule

−

) to

match the observed

∼

10 ppt daytime HONO in the Southeast

US boundary layer. The resulting impact on boundary layer

ozone concentrations is negligible.

2.2 Dry deposition

The GEOS-Chem dry deposition scheme uses a resistance-

in-series model based on Wesely (1989) as implemented by

Wang et al. (1998). Underestimation of dry deposition has

been invoked as a cause for model overestimates of ozone

in the eastern US (Lin et al., 2008; Walker, 2014). Day-

time ozone deposition is determined principally by stom-

atal uptake. Here, we decrease the stomatal resistance from

200 s m

−

for both coniferous and deciduous forests (Wesely,

1989) by 20 % to match summertime measurements of the

ozone dry deposition velocity for a pine forest in North Car-

olina (Finkelstein et al., 2000) and for the Ozarks oak forest

in southeastern Missouri (Wolfe et al., 2015), both averaging

0.8 cm s

−

in the daytime. The mean ozone deposition ve-

locity in GEOS-Chem along the SEAC

RS boundary layer

flight tracks in the Southeast US averages 0.7

0.3 cm s

−

for the daytime (09:00–16:00 local) surface layer. Deposi-

tion is suppressed in the model at night due to both stom-

atal closure and near-surface stratification, consistent with

the Finkelstein et al. (2000) observations.

Deposition flux measurements for isoprene oxidation

products at the Alabama SOAS site (http://soas2013.rutgers.

edu) indicate higher deposition velocities than simulated by

the standard GEOS-Chem model (Nguyen et al., 2015). The

diurnal cycle of dry deposition in GEOS-Chem compares

well with the observations from SOAS (Nguyen et al., 2015).

As an expedient, Nguyen et al. (2015) scaled the Henry’s law

coefficients for these species in GEOS-Chem to match their

observed deposition velocities and we follow their approach

here. Other important depositing species include HNO

and

peroxyacetyl nitrate (PAN), with mean deposition velocities

along the SEAC

RS Southeast US flight tracks in daytime of

3.9 and 0.6 cm s

−

, respectively.

www.atmos-chem-phys.net/16/13561/2016/

Atmos. Chem. Phys., 16, 13561–13577, 2016

13564

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

Figure 1.

Surface NO

emissions in the Southeast US in GEOS-Chem for August and September 2013 including fuel combustion, soils,

fertilizer use, and open fires (total emissions

153 Gg N). Anthropogenic emissions from mobile sources and industry in the National Emis-

sion Inventory (NEI11v1) for 2013 have been decreased by 60 % to match atmospheric observations (see text). Lightning contributes an

additional 25 Gg N to the free troposphere (not included in the figure). The emissions are mapped on the 0.25

◦

0.3125

◦

GEOS-Chem grid.

The pie chart gives the sum of August–September 2013 emissions (Gg N) over the Southeast US domain as shown on the map (94.5–75

◦

29.5–40

◦

N).

2.3 Emissions

We use hourly US anthropogenic emissions from the 2011

EPA National Emissions Inventory (NEI11v1) at a hor-

izontal resolution of 0.1

◦

0.1

◦

and adjusted to 2013

using national annual scaling factors (EPA NEI, 2015).

The scaling factor for NO

emissions is 0.89, for a

2013 US NEI total of 3.5 Tg N a

−

. Further informa-

tion on the use of the NEI11v1 in GEOS-Chem can be

found at http://wiki.seas.harvard.edu/geos-chem/index.php/

EPA/NEI11_North_American_emissions. Soil NO

emis-

sions, including emissions from fertilizer application, are

computed according to Hudman et al. (2012), with a 50 %

reduction in the Midwestern US based on a previous com-

parison with OMI NO

observations (Vinken et al., 2014).

Open fire emissions are from the daily Quick Fire Emissions

Database (QFED) (Darmenov and da Silva, 2014) with di-

urnal variability from the Western Regional Air Partnership

(Air Sciences, 2005). We emit 40 % of open fire NO

emis-

sions as PAN and 20 % as HNO

to account for fast oxidation

taking place in the fresh plume (Alvarado et al., 2010). Fol-

lowing Fischer et al. (2014), we inject 35 % of fire emissions

above the boundary layer, evenly between 3.5 and 5.5 km al-

titude. Lightning is an additional source of NO

but is mainly

released in the upper troposphere, as described below.

Initial implementation of the above inventory in GEOS-

Chem resulted in an 60–70 % overestimation of NO

and

HNO

measured from the SEAC

RS DC-8 aircraft and a

70 % overestimation of nitrate (NO

−

)

wet deposition fluxes

measured by the NADP across the Southeast US. Correcting

this bias required a

∼

40 % decrease in surface NO

emis-

sions. Assuming strongly reduced soil and fertilizer NO

emissions (18 % of total NO

emissions in the southeast)

and open fires (2 %), also considering the large uncertainty

in these emissions, would be insufficient to correct this bias.

Emissions from power plant stacks are directly measured but

account for only 12 % of NEI NO

emissions on an annual

basis (EPA NEI, 2015). Several local studies in recent years

have found that NEI NO

emissions for mobile sources may

be too high by a factor of 2 or more (Castellanos et al, 2011;

Fujita et al., 2012; Brioude et al., 2013; Anderson et al.,

2014). We can achieve the required 40 % decrease in total

emissions by reducing NEI emissions from mobile and

industrial sources (all sources except power plants) by 60 %

or alternatively by reducing these sources by 30 % and zero-

ing out soil and fertilizer NO

emissions. Since it is apparent

that there is some minimum contribution by soil NO

emis-

sions, we assessed the impact of the approach of reducing the

non-power-plant NEI emissions by 60 %. The spatial over-

lap between anthropogenic and soil NO

emissions is such

that we cannot readily arbitrate between these two scenarios.

Comparisons with observations will be presented in the next

section.

We constrain the lightning NO

source with satel-

lite data as described by Murray et al. (2012). Light-

ning NO

is mainly released at the top of convective up-

drafts following Ott et al. (2010). The standard GEOS-

Chem model uses higher NO

yields for midlatitudes light-

ning (500 mol flash

−

) than for tropical (260 mol flash

−

)

(Huntrieser et al., 2007, 2008; Hudman et al., 2007; Ott et

al., 2010) with a fairly arbitrary boundary between the two

at 23

◦

N in North America and 35

◦

N in Eurasia. Zhang et

al. (2014) previously found that this leads GEOS-Chem to

overestimate background ozone in the southwestern US and

we find the same here for the eastern US and the Gulf of

Mexico. We treat here all lightning in the 35

◦

S–35

◦

N band

as tropical and thus remove the distinction between North

America and Eurasia.

Figure 1 gives the resulting surface NO

emissions for the

Southeast US for August and September 2013. With the orig-

inal NEI inventory, fuel combustion accounted for 81 % of

total surface NO

emissions in the Southeast US (not includ-

ing lightning). If the required reduction of non-power-plant

Atmos. Chem. Phys., 16, 13561–13577, 2016

www.atmos-chem-phys.net/16/13561/2016/

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

13565

0.0

0.2

0.4

0.6

0.8

, ppb

Altitude, km

0.0

0.5

1.0

1.5

HNO

+NO

, ppb

100

120

, ppb

100

ISOPN, ppt

Altitude, km

200

400

600

ISOPOOH,

ppt

DC8

GEOS-Chem

Original NO

Emissions

100

150

200

HPALDs, ppt

Figure 2.

Median vertical concentration profiles of NO

, total in-

organic nitrate (gas HNO

aerosol NO

−

)

, ozone, isoprene nitrate

(ISOPN), isoprene hydroperoxide (ISOPOOH), and hydroperoxy-

aldehydes (HPALD) for the SEAC

RS flights over the Southeast

US (domain of Fig. 1). Observations from the DC-8 aircraft are

compared to GEOS-Chem model results. The dashed red line shows

model results before adjustment of NO

emissions from fuel com-

bustion and lightning (see text). The 25th and 75th percentiles of the

DC-8 observations are shown as grey bars. The SEAC

RS observa-

tions have been filtered to remove open fire plumes, stratospheric

air, and urban plumes as described in the text. Model results are

sampled along the flight tracks at the time of flights and gridded to

the model resolution. Profiles are binned to the nearest 0.5 km. The

NOAA NO

four-channel chemiluminescence (CL) instrument

made measurements of ozone and NO

(Ryerson et al., 1998), NO

(Ryerson et al., 2000), and NO

(Pollack et al., 2010). Total inor-

ganic nitrate was measured by the University of New Hampshire

Soluble Acidic Gases and Aerosol (UNH SAGA) instrument (Dibb

et al., 2003) and was mainly gas-phase HNO

for the SEAC

conditions. ISOPOOH, ISOPN, and HPALDs were measured by the

Caltech single mass analyzer CIMS (Crounse et al., 2006; Paulot et

al., 2009a; Crounse et al., 2011).

NEI emissions is 60 %, the contribution from fuel combus-

tion would be 68 %.

Biogenic VOC emissions are from MEGAN v2.1, includ-

ing isoprene, acetone, acetaldehyde, monoterpenes, and

alkenes. We reduce MEGAN v2.1 isoprene emissions by

15 % to better match SEAC

RS observations of isoprene

fluxes from the Ozarks (Wolfe et al., 2015) and observed

formaldehyde (Zhu et al., 2016). Yu et al. (2016) show the

resulting isoprene emissions for the SEAC

RS period.

3 Overestimate of NO

x

emissions in the EPA NEI

inventory

Figure 2 shows simulated and observed median vertical

distributions of NO

, total inorganic nitrate (gas-phase

HNO

aerosol NO

−

)

, and ozone concentrations along

the SEAC

RS flight tracks over the Southeast US. Here

and elsewhere the data exclude urban plumes as diag-

nosed by [NO

]

4 ppb, open fire plumes as diagnosed by

[CH

CN]

200 ppt, and stratospheric air as diagnosed by

]

[CO]

1.25 mol mol

−

. These filters exclude

1, 7,

and 6 % of the data, respectively. We would not expect the

model to be able to capture these features even at native res-

olution (Yu et al., 2016).

Model results in Fig. 2 are shown both with the original

emissions (dashed line) and with non-power-plant NEI

fuel emissions decreased by 60 % (solid line). Decreasing

emissions corrects the model bias for NO

and also largely

corrects the bias for inorganic nitrate. Boundary layer ozone

is overestimated by 12 ppb with the original NO

emissions

but this bias disappears after decreasing the NO

emissions.

Results are very similar if we decrease the non-power-plant

NEI fuel emissions by only 30 % and zero out soil and fer-

tilizer emissions. Thus the required decrease of NO

emis-

sions may involve an overestimation of both anthropogenic

and soil emissions.

Further support for decreasing NO

emissions is offered

by observed nitrate wet deposition fluxes from the NADP

network (NADP, 2007). Figure 3 compares simulated and

observed fluxes for the model with decreased NO

emis-

sions. Model values have been corrected for precipitation

bias following the method of Paulot et al. (2014), in which

the monthly deposition flux is assumed to scale to the 0.6th

power of the precipitation bias. We diagnose precipitation

bias in the GEOS-5.11.0 data relative to high-resolution

PRISM observations (http://prism.oregonstate.edu). For the

Southeast US, the precipitation bias is

−

34 % in August and

−

21 % in September 2013. We see from Fig. 3 that the model

with decreased NO

emissions reproduces the spatial vari-

ability in the observations with only

8 % bias over the

Southeast US and

7 % over the contiguous US. In com-

parison, the model with original emissions had a 63 % over-

estimation of the nitrate wet deposition flux nationally and

a 71 % overestimation in the southeast. The high deposition

fluxes along the Gulf of Mexico in Fig. 3, both in the model

and in the observations, reflect particularly large precipita-

tion.

The model with decreased NO

emissions also reproduces

the spatial distribution of NO

in the Southeast US bound-

ary layer as observed in SEAC

RS. This is shown in Fig. 4

with simulated and observed concentrations of NO

along

the flight tracks below 1.5 km altitude. The spatial correla-

tion coefficient is 0.71. There are no obvious spatial patterns

of model bias that would point to specific source sectors as

responsible for the NO

emission overestimate, beyond the

www.atmos-chem-phys.net/16/13561/2016/

Atmos. Chem. Phys., 16, 13561–13577, 2016

13566

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

Nitrate wet deposition fluxes, Aug–Sep 2013

kg N h

−1

month

−1

0.1

0.3

0.4

0.2

0.0

Model

0.0

0.1

0.2

0.3

0.4

0.0

0.1

0.2

0.3

0.4

Southeast US

r = 0.71

NMB = +8

CONUS

r = 0.76

NMB = +7

Observation

Figure 3.

Nitrate wet deposition fluxes across the US in August–September 2013. Mean observations from the NADP network (circles in

the left panel) are compared to model values with decreased NO

emissions (background). Also shown is a scatterplot of simulated vs.

observed values at individual sites for the whole contiguous US (black) and for the Southeast US (green). The correlation coefficient (

) and

normalized mean bias (NMB) are shown inset, along with the 1 : 1 line.

Figure 4.

Ozone and NO

concentrations in the boundary layer (0–

1.5 km) during SEAC

RS (6 August to 23 September 2013). Obser-

vations from the aircraft and simulated values are averaged over the

0.25

◦

0.3125

◦

GEOS-Chem grid. NO

above 1 ppb is shown in

black. The spatial correlation coefficient is 0.71 for both NO

and

. The normalized mean bias is

−

11.5 % for NO

and 4.5 % for

blanket 30–60 % decrease of non-power-plant NEI emissions

needed to correct the regional emission total.

4 Using satellite NO

data to verify NO

x

emissions:

sensitivity to upper troposphere

Observations of tropospheric NO

columns by solar

backscatter from the OMI satellite instrument offer an ad-

ditional constraint on NO

emissions (Duncan et al., 2014;

Lu et al., 2015). We compare the tropospheric columns sim-

ulated by GEOS-Chem with the NASA operational retrieval

(Level 2, v2.1) (NASA, 2012; Bucsela et al., 2013) and the

Berkeley High-Resolution (BEHR) retrieval (Russell et al.,

2011). The NASA retrieval has been validated to agree with

surface measurements to within

20 % (Lamsal et al., 2014).

Both retrievals fit the observed backscattered solar spectra to

obtain a slant tropospheric NO

column,

, along the opti-

cal path of the backscattered radiation detected by the satel-

lite. The slant column is converted to the vertical column,

, by using an air mass factor (AMF) that depends on the

vertical profile of NO

and on the scattering properties of the

surface and the atmosphere (Palmer et al., 2001):

AMF

∫

(

)

(

)

(1)

In Eq. (4), AMF

is the geometric air mass factor that de-

pends on the viewing geometry of the satellite,

w(z)

is a scat-

tering weight calculated by a radiative transfer model that de-

scribes the sensitivity of the backscattered radiation to NO

as a function of altitude,

S(z)

is a shape factor describing

the normalized vertical profile of NO

number density, and

is the tropopause. Scattering weights for NO

retrievals

typically increase by a factor of 3 from the surface to the

upper troposphere (Martin et al., 2002). Here we use our

GEOS-Chem shape factors to recalculate the AMFs in the

NASA and BEHR retrievals as recommended by Lamsal et

al. (2014) for comparing model and observations. We filter

out cloudy scenes (cloud radiance fraction

0.5) and bright

surfaces (surface reflectivity

0.3).

Figure 5 shows the mean NO

tropospheric columns

from BEHR, NASA, and GEOS-Chem (with NO

emis-

sion reductions applied) over the Southeast US for August–

September 2013. The BEHR retrieval is on average 6 %

higher than the NASA retrieval. GEOS-Chem is on aver-

age 11

19 % lower than the NASA retrieval and 16

18 %

lower than the BEHR retrieval. With the original NEI NO

emissions, GEOS-Chem would be biased high against both

retrievals by 26–31 %. The low bias in the model with re-

duced NO

emissions does not appear to be caused by an

overcorrection of surface emissions but rather by the upper

troposphere. Figure 6 (top left panel) shows the mean ver-

tical profile of NO

number density as measured from the

aircraft by two independent instruments (NOAA and UC

Berkeley) and simulated by GEOS-Chem. At the surface,

the median difference is 1.8

molecules cm

−

, which is

Atmos. Chem. Phys., 16, 13561–13577, 2016

www.atmos-chem-phys.net/16/13561/2016/

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

13567

0.00

1.25

2.50

3.75

5.00

molec

Observed (BEHR)

Observed (NASA)

GEOS-chem

BEHR: -16+18 %

NASA: -11+19 %

NO tropospheric column in August and September 2013

Figure 5.

tropospheric columns over the Southeast US in

August–September 2013. GEOS-Chem (sampled at the 13:30 local

time overpass of OMI) is compared to OMI satellite observations

using the BEHR and NASA retrievals. Values are plotted on the

0.25

◦

0.3125

◦

GEOS-Chem grid. The GEOS-Chem mean bias

over the figure domain and associated spatial standard deviation are

inset in the bottom panel.

within the NOAA and UC Berkeley measurement uncertain-

ties of

0.030 ppbv

7 % and

5 %, respectively. The ob-

servations show a secondary maximum in the upper tropo-

sphere above 10 km, absent in GEOS-Chem. It has been sug-

gested that aircraft measurements of NO

in the upper tro-

posphere could be biased high due to decomposition in the

instrument inlet of thermally unstable NO

reservoirs such

as HNO

and methyl peroxy nitrate (Browne et al., 2011;

Reed et al., 2016). This would not affect the UC Berkeley

measurement (Nault et al., 2015) and could possibly account

for the difference with the NOAA measurement in Fig. 6.

The top right panel of Fig. 6 shows the cumulative con-

tributions from different altitudes to the slant NO

column

measured by the satellite, using the median vertical profiles

from the left panel and applying mean altitude-dependent

scattering weights from the NASA and BEHR retrievals. The

boundary layer below 1.5 km contributes only 19–28 % of the

column. The upper troposphere above 8 km contributes 32–

49 % in the aircraft observations and 23 % in GEOS-Chem.

Much of the observed upper-tropospheric NO

likely orig-

inates from lightning and is broadly distributed across the

southeast because of the long lifetime of NO

at that alti-

number density, 10

molecules cm

-3

Altitude, km

0.0

0.2

0.4

0.6

0.8

1.0

Fractional contribution

to NO column

NOAA

UC Berkeley

GEOS-chem

BEHR scattering weights

NASA scattering weights

1000

2000

3000

400

, ppt

Observed

GEOS-chem

[NO]/[NO

] mol mol

-1

Observed (NOAA + UC Berkeley)

NO-NO -O equilibrium (PSS)

Double HO and RO above 8 km

GEOS-chem

Altitude, km

Figure 6.

Vertical distribution of NO

over the Southeast US dur-

ing SEAC

RS (August–September 2013) and contributions to tro-

pospheric NO

columns measured from space by OMI. The top left

panel shows median vertical profiles of NO

number density mea-

sured from the SEAC

RS aircraft by the NOAA and UC Berke-

ley instruments and simulated by GEOS-Chem. The top right panel

shows the fractional contribution of NO

below a given altitude

to the total tropospheric NO

slant column measured by OMI, ac-

counting for increasing sensitivity with altitude as determined from

the retrieval scattering weights. The bottom left panel shows the

median vertical profiles of the daytime [NO]

[NO

] molar con-

centration ratio in the aircraft observations (NOAA for NO and UC

Berkeley for NO

)

and in GEOS-Chem. Also shown is the ratio

computed from NO–NO

–O

photochemical steady state (PSS) as

given by Reactions (4) and (6) (blue) and including Reaction (5)

with doubled HO

and RO

concentrations above 8 km (purple).

The bottom right panel shows the median H

profile from the

model and from the SEAC4RS flights over the Southeast US. H

was measured by the Caltech CIMS (see Fig. 2).

tude (Li et al., 2005; Bertram et al., 2007; Hudman et al.,

2007). The NO

vertical profile (shape factor) assumed in the

BEHR retrieval does not include any lightning influence, and

the Global Modeling Initiative (GMI) model vertical profile

assumed in the NASA retrieval has little contribution from

the upper troposphere (Lamsal et al., 2014). These underes-

timates of upper-tropospheric NO

in the retrieval shape fac-

tors will cause a negative bias in the AMF and therefore a

positive bias in the retrieved vertical columns.

The GEOS-Chem underestimate of observed upper-

tropospheric NO

in Fig. 6 is partly driven by NO

partitioning. The bottom left panel of Fig. 6 shows the

[NO]

[NO

] concentration ratio in GEOS-Chem and in the

observations (NOAA for NO, UC Berkeley for NO

)

. One

would expect the [NO]

[NO

] concentration ratio in the

www.atmos-chem-phys.net/16/13561/2016/

Atmos. Chem. Phys., 16, 13561–13577, 2016

13568

K. R. Travis et al.: Why do models overestimate surface ozone in the Southeast United States?

daytime upper troposphere to be controlled by photochem-

ical steady state:

→

(R4)

→

(R5)

hυ

−→

(R6)

Reaction

(R5)

plays

only

minor

role

then

[NO]

[NO

]

≈

(

]),

defining

the

NO–NO

–

photochemical steady state (PSS). The PSS plotted in

Fig. 6 agrees closely with GEOS-Chem. Such agreement

has previously been found when comparing photochemical

models with observed [NO]

[NO

] ratios from aircraft

in the marine upper troposphere (Schultz et al., 1999) and

lower stratosphere (Del Negro et al., 1999). The SEAC

observations show large departure. The NO

photolysis

frequencies

computed locally by GEOS-Chem are on

average within 10 % of the values determined in SEAC

from measured actinic fluxes (Shetter and Muller, 1999), so

this is not the problem.

A possible explanation is that the model underestimates

peroxy radical concentrations and hence the contribution of

Reaction (5) in the upper troposphere. Zhu et al. (2016) found

that GEOS-Chem underestimates the observed HCHO con-

centrations in the upper troposphere during SEAC

RS by a

factor of 3, implying that the model underestimates the HO

source from convective injection of HCHO and peroxides

(Jaeglé et al., 1997; Prather and Jacob, 1997; Müller and

Brasseur, 1999). HO

observations over the central US in

summer during the SUCCESS aircraft campaign suggest that

this convective injection increases HO

concentrations in the

upper troposphere by a factor of 2 (Jaeglé et al., 1998). The

bottom right panel of Fig. 6 shows median modeled and ob-

served vertical profiles of the HO

reservoir hydrogen perox-

ide (H

)

during SEAC

RS over the Southeast US. GEOS-

Chem underestimates observed H

by a mean factor of

1.7 above 8 km. The bottom left panel of Fig. 6 shows the

[NO]

[NO

] ratio in GEOS-Chem with HO

and RO

dou-

bled above 8 km. Such a change corrects significantly the

bias relative to observations.

The

PSS

and

GEOS-Chem

simulation

the

concentration ratio in Fig. 6 use

−

exp

[−

1500

]

molecule

−

and spec-

troscopic information for

from Sander et al. (2011). It is

possible that the strong thermal dependence of

has some

error, considering that only one direct measurement has been

published for the cold temperatures of the upper troposphere

(Borders and Birks, 1982). Cohen et al. (2000) found that

reducing the activation energy of

by 15 % improved

model agreement in the lower stratosphere. Correcting the

discrepancy between simulated and observed [NO]

[NO

]

ratios in the upper troposphere in Fig. 6 would require a

similar reduction to the activation energy of

, but this

reduction would negatively impact the surface comparison.

This inconsistency of the observed [NO]

[NO

] ratio with

basic theory needs to be resolved, as it affects the inference

of NO

emissions from satellite NO

column measurements.

Notwithstanding this inconsistency, we find that NO

in the

upper troposphere makes a significant contribution to the

tropospheric NO

column observed from space.

5 Isoprene oxidation pathways

Measurements aboard the SEAC

RS aircraft included first-

generation isoprene nitrates (ISOPN), isoprene hydroper-

oxide (ISOPOOH), and hydroperoxy-aldehydes (HPALDs)

(Crounse et al., 2006; Paulot et al., 2009a; St. Clair et al.,

2010; Crounse et al., 2011; Beaver et al., 2012; Nguyen et

al., 2015). Although measurement uncertainties are large (30,

40, and 50 %, respectively; Nguyen et al., 2015), these are

unique products of the ISOPO

NO, ISOPO

, and

ISOPO

isomerization pathways and thus track whether oxi-

dation of isoprene proceeds by the high-NO

pathway (pro-

ducing ozone) or the low-NO

pathways. Figure 2 (bottom

row) compares simulated and observed concentrations. All

three gases are restricted to the boundary layer because of

their short lifetimes. Mean model concentrations in the low-

est altitude bin (Fig. 2, approximately 400 m above ground)

differ from observations by

19 % for ISOPN,

70 % for

ISOPOOH, and

−

50 % for HPALDs. The GEOS-Chem sim-

ulation of organic nitrates including ISOPN is further dis-

cussed in Fisher et al. (2016). Our HPALD source is based

on the ISOPO

isomerization rate constant from Crounse et

al. (2011). A theoretical calculation by Peeters et al. (2014)

suggests a rate constant that is 1.8

higher, which would

reduce the model bias for HPALDs and ISOPOOH and in-

crease boundary layer OH by 8 %. St. Clair et al. (2015)

found that the reaction rate of ISOPOOH

OH to form

IEPOX is approximately 10 % faster than the rate given by

Paulot et al. (2009b), which would further reduce the model

overestimate. For both ISOPOOH and HPALDs, GEOS-

Chem captures much of the spatial variability (

0.80 and

0.79, respectively).

Figure 7 shows the model branching ratios for the fate of

the ISOPO

radical by tracking the mass of ISOPO

react-

ing via the high-NO

pathway (ISOPO

NO) and the low-

pathways over the Southeast US domain. The mean

branching ratios for the Southeast US are ISOPO

54 %, ISOPO

26 %, ISOPO

isomerization 15 %,

and ISOPO

5 %. The lack of dominance of the high-

pathway is due in part to the spatial segregation of iso-

prene and NO

emissions (Yu et al., 2016). This segregation

also buffers the effect of changing NO

emissions on the fate

of isoprene. Our original simulation with higher total NO

emissions (unadjusted NEI11v1) had a branching ratio for

the ISOPO

NO reaction of only 62 %.

Atmos. Chem. Phys., 16, 13561–13577, 2016