Available via license: CC BY 4.0
Content may be subject to copyright.
Atmos. Chem. Phys., 15, 5243–5258, 2015
www.atmos-chem-phys.net/15/5243/2015/
doi:10.5194/acp-15-5243-2015
© Author(s) 2015. CC Attribution 3.0 License.
Gas and aerosol carbon in California: comparison of measurements
and model predictions in Pasadena and Bakersfield
K. R. Baker1, A. G. Carlton2, T. E. Kleindienst3, J. H. Offenberg3, M. R. Beaver3, D. R. Gentner4, A. H. Goldstein5,
P. L. Hayes6, J. L. Jimenez7, J. B. Gilman8, J. A. de Gouw8, M. C. Woody3, H. O. T. Pye3, J. T. Kelly1,
M. Lewandowski3, M. Jaoui9, P. S. Stevens10, W. H. Brune11, Y.-H. Lin12, C. L. Rubitschun12, and J. D. Surratt12
1Office of Air Quality Planning and Standards, US Environmental Protection Agency, Research Triangle Park, NC, USA
2Dept. of Environmental Sciences, Rutgers University, New Brunswick, NJ, USA
3Office of Research and Development, US Environmental Protection Agency, Research Triangle Park, NC, USA
4Department of Chemical and Environmental Engineering, Yale University, New Haven, CT, USA
5Department of Civil and Environmental Engineering, University of California, Berkeley, CA, USA
6Département de Chimie, Université de Montréal, Montréal, Québec, Canada
7Department of Chemistry & Biochemistry, and CIRES, University of Colorado, Boulder, Colorado, USA
8Chemical Sciences Division, Earth System Research Laboratory, National Oceanic and Atmospheric Administration,
Boulder, CO, USA
9Alion Science and Technology, Inc., Research Triangle Park, NC, USA
10Center for Research in Environmental Science, School of Public and Environmental Affairs and Department of Chemistry,
Indiana University, Bloomington, IN, USA
11Department of Meteorology, Pennsylvania State University, University Park, PA, USA
12Department of Environmental Sciences and Engineering, Gillings School of Global Public Health, University of North
Carolina at Chapel Hill, Chapel Hill, NC, USA
Correspondence to: K. R. Baker (baker.kirk@epa.gov)
Received: 21 November 2014 – Published in Atmos. Chem. Phys. Discuss.: 7 January 2015
Revised: 6 April 2015 – Accepted: 23 April 2015 – Published: 12 May 2015
Abstract. Co-located measurements of fine particulate mat-
ter (PM2.5)organic carbon (OC), elemental carbon, radio-
carbon (14C), speciated volatile organic compounds (VOCs),
and OH radicals during the CalNex field campaign pro-
vide a unique opportunity to evaluate the Community Multi-
scale Air Quality (CMAQ) model’s representation of organic
species from VOCs to particles. Episode average daily 23h
average 14C analysis indicates PM2.5carbon at Pasadena and
Bakersfield during the CalNex field campaign was evenly
split between contemporary and fossil origins. CMAQ pre-
dicts a higher contemporary carbon fraction than indicated by
the 14C analysis at both locations. The model underestimates
measured PM2.5organic carbon at both sites with very little
(7% in Pasadena) of the modeled mass represented by sec-
ondary production, which contrasts with the ambient-based
SOC/ OC fraction of 63% at Pasadena.
Measurements and predictions of gas-phase anthropogenic
species, such as toluene and xylenes, are generally within
a factor of 2, but the corresponding SOC tracer (2,3-
dihydroxy-4-oxo-pentanoic acid) is systematically underpre-
dicted by more than a factor of 2. Monoterpene VOCs and
SOCs are underestimated at both sites. Isoprene is under-
estimated at Pasadena and overpredicted at Bakersfield and
isoprene SOC mass is underestimated at both sites. System-
atic model underestimates in SOC mass coupled with rea-
sonable skill (typically within a factor of 2) in predicting
hydroxyl radical and VOC gas-phase precursors suggest er-
ror(s) in the parameterization of semivolatile gases to form
SOC. Yield values (α) applied to semivolatile partitioning
species were increased by a factor of 4 in CMAQ for a sen-
sitivity simulation, taking into account recent findings of un-
derestimated yields in chamber experiments due to gas wall
losses. This sensitivity resulted in improved model perfor-
Published by Copernicus Publications on behalf of the European Geosciences Union.
5244 K. R. Baker et al.: Gas and aerosol carbon in California
mance for PM2.5organic carbon at both field study locations
and at routine monitor network sites in California. Modeled
percent secondary contribution (22% at Pasadena) becomes
closer to ambient-based estimates but still contains a higher
primary fraction than observed.
1 Introduction
Secondary organic aerosol (SOA) forms in the atmosphere
during the gas-phase photooxidation of volatile organic com-
pounds (VOCs) that produce semivolatile and water-soluble
gases that condense to form new particles or partition to pre-
existing aerosol mass (Ervens et al., 2011). SOA contributes
to the atmospheric fine particulate matter (PM2.5)burden,
with subsequent effects on air quality, visibility, and climate
(Hallquist et al., 2009). Despite its importance and abun-
dance, ambient SOA mass is not well characterized by at-
mospheric models (Wagstrom et al., 2014). For example, the
Community Multiscale Air Quality (CMAQ) model consis-
tently underpredicts surface SOA mass concentrations for a
variety of seasons and locations when compared to ambient
observational estimates (Carlton and Baker, 2011; Carlton et
al., 2010; Hayes et al., 2014; Zhang et al., 2014a).
SOA formation and the preceding gas-phase photooxi-
dation chemistry are complex and often involve multiple
oxidation steps in the gas, aqueous, and particle phase as
well as accretion reactions in the particle phase that yield
high molecular weight products. However, three-dimensional
photochemical models must represent the gas-phase chem-
istry and SOA formation in a simplified fashion for compu-
tational efficiency (Barsanti et al., 2013). Gas-phase chem-
ical mechanisms employ “lumped” VOC species, catego-
rized primarily according to reactivity (e.g., reaction rate
constants with the OH radical) (Carter, 2000; Yarwood et al.,
2005), not product volatility or solubility. Condensable SOA-
forming oxidation products are typically represented with
two products in the standard versions of publically available
and routinely applied photochemical modeling systems such
as GEOS-CHEM (Chung and Seinfeld, 2002; Henze and Se-
infeld, 2006) and WRF-CHEM (Grell et al., 2005) and those
employed in regulatory applications for rule making such as
CMAQ (Carlton et al., 2010) and the Comprehensive Air
Quality Model with extensions (CAMx) (ENVIRON, 2014).
Given the relationships between precursor VOC, OH radical
abundance, and SOA formation, it is important to simulta-
neously evaluate the model representation of all three within
the context of how organic species evolve in the atmosphere
to diagnose persistent SOA model bias.
Recent studies have shown that warm season SOA mass
concentrations are usually greater than primary organic
aerosol (POA) mass in the Los Angeles (Docherty et al.,
2008; Hersey et al., 2011; Hayes et al., 2013) and Bakers-
field (Liu et al., 2012) areas. Gas-to-particle condensation
of VOC oxidation products dominates formation of summer
SOA in Bakersfield (Liu et al., 2012; Zhao et al., 2013) and
up to one-third of nighttime organic aerosols (OA) in Bak-
ersfield are organic nitrates (Rollins et al., 2012). Sources
of warm season OA in Bakersfield include fossil fuel com-
bustion, vegetative detritus, petroleum operations, biogenic
emissions, and cooking (Liu et al., 2012; Zhao et al., 2013).
Despite numerous studies based on observations and mod-
els, less consensus exists regarding the largest sources of
warm season SOA at Pasadena. Bahreini et al. (2012) con-
cluded that SOA at Pasadena is largely derived from gaso-
line engines with minimal biogenic and diesel fuel contribu-
tion (Bahreini et al., 2012). Others concluded large contribu-
tions from gasoline fuel combustion to SOA but also found
notable contributions from diesel fuel combustion, cooking,
and other sources (Gentner et al., 2012; Hayes et al., 2013).
Zotter at al. (2014) conclude that 70% of the SOA in the ur-
ban plume in Pasadena is due to fossil sources and that at
least 25 % of the non-fossil carbon is due to cooking sources.
Lower volatility VOC measurements made at Pasadena have
been estimated to produce approximately 30 % of fresh SOA
in the afternoon with a large contribution to these low volatil-
ity VOC from petroleum sources other than on-road vehicles
(Zhao et al., 2014).
Chemical measurements of PM2.5carbon, fossil and con-
temporary aerosol carbon fraction, OC and its components,
SOC tracers, and speciated VOCs taken as part of the 2010
California Research at the Nexus of Air Quality and Climate
Change (CalNex) field study in central and southern Califor-
nia (Ryerson et al., 2013) provide a unique opportunity to
quantitatively evaluate modeled organic predictions. These
special study data combined with routine PM2.5OC mea-
surements in California are compared with model estimates
to gauge how well the modeling system captures the gas and
aerosol carbon burden using the standard CMAQ aerosol ap-
proach. The SOC mechanism in the base version of CMAQ
lends itself well to comparison with chemical tracers because
it retains chemical identity traceable to the precursor VOC
(Carlton et al., 2010). Finally, a CMAQ sensitivity simula-
tion was performed in which the yields of semivolatile gases
from VOC oxidation were increased by a factor of 4 (Zhang
et al., 2014b) to determine whether this may ameliorate the
model underprediction of secondary organic carbon (SOC)
seen here and in other studies (Ensberg et al., 2014).
2 Methods
Predictions of speciated VOC, speciated SOC, and aerosol-
phase carbon are simultaneously compared to co-located am-
bient measurements at two surface locations, one in Los An-
geles County (Pasadena) and one in the San Joaquin Valley
(Bakersfield) air basin. The CMAQ photochemical model is
applied with a fine grid resolution (4km sized grid cells) us-
ing emissions from the 2011 National Emissions Inventory
and 2010 specific point source information where available.
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5245
2.1 Model background
CMAQ version 5.0.2 (www.cmaq-model.org) was applied to
estimate air quality in California from 5 May to 1 July 2010,
coincident with the CalNex study. Gas-phase chemistry
is simulated with the SAPRC07TB condensed mechanism
(Hutzell et al., 2012) and aqueous-phase chemistry that ox-
idizes sulfur, methylglyoxal (MGLY), and glyoxal (Carl-
ton et al., 2008; Sarwar et al., 2013). The AERO6 aerosol
chemistry module includes ISORROPIAII (Fountoukis and
Nenes, 2007) inorganic chemistry and partitioning. The mod-
eling system generally does well capturing ambient inorganic
gases and PM2.5species during this time period at Pasadena
and Bakersfield (Kelly et al., 2014; Markovic et al., 2014).
Model-predicted OC species are shown in Fig. 1 by
volatility bin (log of C∗) and O: C ratio (see Supplement for
related details). Aqueous-phase species are shown with blue
circles, species largely fossil in origin are colored brown, and
those non-fossil in origin are green. A general trend of in-
creasing O: C ratio as volatility decreases is consistent with
laboratory and field measurements (Jimenez et al., 2009).
The placement of the MGLY geminal diol vertically above
gas-phase MGLY in Fig. 1 represents hydration processes.
Aqueous-phase organic chemistry represents multiple pro-
cesses, including functionalization and oligomerization, be-
cause some photooxidation products are small carboxylic
acids and others are high molecular weight species (Tan et
al., 2010; Carlton et al., 2007).
VOC precursors for SOA include isoprene, monoter-
penes, sesquiterpenes, xylenes, toluene, benzene, alkanes,
glyoxal, and methylglyoxal (Fig. 1 right panel). Benzene,
toluene, and xylene form SOA precursors with high-NOx
(RO2+NO) and low-NOx(RO2+HO2)specific yields
(Carlton et al., 2010). CMAQ converts these precursors into
multiple semivolatile products (Fig. 1 middle panel) after a
single oxidation step. These multiple products vary in terms
of assigned volatility and oxygen-to-carbon (O: C) ratio.
When semivolatile SOA mass oligomerizes in CMAQ the
SOA identity is lost and becomes classified only as anthro-
pogenic or biogenic, dependent on the VOC precursor (see
Fig. S2 in the Supplement). After oligomerization, the satu-
ration vapor pressure (C∗) and OM :OC ratio associated with
all of the two-product semivolatile SOA species change from
the individual values to the values assigned for non-volatile,
non-partitioning SOA mass (C∗≈0; OM: OC=2.1) (Carl-
ton et al., 2010).
CMAQ VOCs and SOC species are paired in time and
space with measurements (Table S2 in the Supplement).
Modeled predictions are averaged temporally to match ob-
servations and extracted from the grid cell where the mon-
itor is located. Modeled toluene and xylene SOC are ag-
gregated to match the measured SOC tracer (2,3-dihydroxy-
4-oxopentanoic acid) which is known to represent prod-
ucts from both compounds and potentially other methylated
aromatics (Kleindienst et al., 2004). Because the original
VOCs contributing to oligomerized species are not tracked
by CMAQ, biogenic oligomerized species mass is appor-
tioned to parent VOC based on the fraction each semivolatile
SOC species contributes to the total semivolatile (non-
oligomerized) biogenic SOC at that time and location. The
same technique is applied to anthropogenic SOC.
2.2 Model application
The model domain covers the state of California and part of
northwest Mexico using 4 km square sized grid cells (Fig. S1
in the Supplement). The vertical domain extends to 50mb
using 34 layers (layer 1 top ∼35m) with most resolution
in the boundary layer. Initial and boundary conditions are
from a coarser CMAQ simulation that used 3-hourly bound-
ary inflow from a GEOS-Chem (v8-03-02) global model
(http://acmg.seas.harvard.edu/geos/) simulation for the same
period (Henderson et al., 2014). The coarser continental
US CMAQ simulation was run continuously from Decem-
ber 2009 through this study period and the first week of the
finer 4km CMAQ simulation was not used to minimize the
influence of initial chemical conditions. Gridded meteorolog-
ical variables are generated using the Weather Research and
Forecasting model (WRF), Advanced Research WRF core
(ARW) version 3.1 (Skamarock et al., 2008). Surface mete-
orology including temperature, wind speed, and wind direc-
tion and daytime mixing layer height were well characterized
by WRF in central and southern California during this period
(Baker et al., 2013).
Emissions are processed to hourly gridded input for
CMAQ with the Sparse Matrix Operator Kernel Emis-
sions (SMOKE) modeling system (http://www.cmascenter.
org/smoke/). Solar radiation and temperature estimated by
the WRF model are used as input to the Biogenic Emis-
sion Inventory System (BEIS) v3.14 to generate hourly emis-
sions estimates of biogenic speciated VOC and NO (Carl-
ton and Baker, 2011). Continuous emissions monitor data
are used in the modeling to reflect 2010 emissions informa-
tion for electrical generating units and other point sources
that provide that information. Day-specific fires are repre-
sented but minimally impacted air quality during this pe-
riod (Hayes et al., 2013). Mobile source emissions were
generated using the SMOKE-MOVES integration approach
(United States Environmental Protection Agency, 2014) and
then interpolated between totals provided by the Califor-
nia Air Resources Board for 2007 and 2011. Other anthro-
pogenic emissions are based on the 2011 National Emissions
Inventory (NEI) version 1 (US Environmental Protection
Agency, 2014). Primary mass associated with carbon (non-
carbon organic mass) is estimated based on sector-specific
organic matter-to-organic carbon (OM:OC) ratios (Simon
and Bhave, 2012).
Emissions of primarily emitted PM2.5OC and the sum
of anthropogenic SOA precursors benzene, toluene, and
xylenes (BTX) are shown in Table 1 by source sector and
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5246 K. R. Baker et al.: Gas and aerosol carbon in California
Figure 1. Gas (right panel), semivolatile (middle panel), and particle-phase (left panel) CMAQ organic carbon shown by saturation vapor
pressure and O: C ratio. Compounds shown in blue exist in the aqueous phase, those in brown are generally fossil in origin, those in green
are generally contemporary in origin, and those in gray both contemporary and fossil in origin. Other known processes such as fragmentation
are not shown as they are not currently represented in the modeling system.
Table 1. Episode total anthropogenic emissions of primarily emitted PM2.5organic carbon and the sum of benzene, toluene, and xylenes by
emission sector group. The Los Angeles (LA) total includes Los Angeles and Orange counties. The southern San Joaquin Valley (SSJV) total
includes Kern, Fresno, Kings, and Tulare counties. Residential wood combustion, fugitives, and non-point area PM2.5emissions are largely
contemporary in origin.
Primarily emitted PM2.5organic carbon Benzene +toluene+xylenes
Sector SSJV (tons) SSJV (%) LA (tons) LA (%) SSJV (tons) SSJV (%) LA (tons) LA (%)
Non-point area 139.9 33.8 410.1 40.8 326.7 37.2 1229.3 35.8
On-road mobile 73.3 17.7 263.6 26.2 273.5 31.2 1190.9 34.6
Non-road mobile 23.9 5.8 161.4 16.1 170.1 19.4 822.3 23.9
Point: non-electrical generating 61.3 14.8 56.3 5.6 68.3 7.8 177.7 5.2
Point: non-electrical generating 54.1 13.1 82.7 8.2 2.0 0.2 3.2 0.1
Oil and gas exploration and related 28.5 6.9 0.0 0.0 34.2 3.9 1.1 0.0
Fugitive dust 24.9 6.0 18.1 1.8 0.0 0.0 0.0 0.0
Commercial marine and rail 3.8 0.9 11.4 1.1 2.6 0.3 12.8 0.4
Point: electrical generating 4.3 1.0 1.7 0.2 0.1 0.0 1.0 0.0
Total contemporary carbon 218.9 52.9 510.9 50.8 2.0 0.2 3.2 0.1
Total fossil carbon 195.2 47.1 494.5 49.2 875.3 99.8 3435.1 99.9
Total 414.1 1005.3 877.4 3438.3
area. Here, the southern San Joaquin Valley includes emis-
sions from Kern, Tulare, Kings, and Fresno counties, and the
Los Angeles area includes emissions from Los Angeles and
Orange counties. BTX emissions in both areas are dominated
by mobile sources (on-road and off-road) and area sources
such as solvent utilization and waste disposal (Table S1). Pri-
mary OC emissions are largely commercial cooking (non-
point area) in both locations with notable contributions from
various types of stationary point and mobile sources. BTX
emissions are almost completely fossil in origin and primar-
ily emitted OC is split fairly evenly between contemporary
and fossil origin in these areas based on the 2011 version 1
NEI (Table 1).
2.3 Sampling and analysis methods
CalNex ground-based measurements took place in Pasadena,
CA, from 15 May to 15 June 2010 and in Bakersfield, CA,
from 15 May to 30 June 2010. The Bakersfield sampling site
was located in a transition area of southeast Bakersfield be-
tween the city center and areas of agricultural activity. The
Pasadena sampling site was located on the California Insti-
tute of Technology campus with the Los Angeles metropoli-
tan area to the southwest and San Gabriel Mountains directly
north (see Fig. S3).
An ambient-based approach is used here to estimate sec-
ondary OC from individual or groups of similar hydrocar-
bons (Kleindienst et al., 2010). Concentrations of specific
compounds, tracers, are determined and used to estimate
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5247
SOC contributions from the particular source groups based
on measured laboratory tracer-to-SOC mass fractions (Klein-
dienst et al., 2007). Filter-based particulate matter sampling
conducted at each site for 23h periods starting at midnight
(PDT) on the designated sampling day was used for tracer-
based organic aerosol characterization. In total, there were
32 filter samples from Pasadena and 36 from the Bakers-
field site (Lewandowski et al., 2013). The filter sampling
protocols have been described in detail elsewhere (Kleindi-
enst et al., 2010). For the analysis of the SOC tracer com-
pounds, filters and field blanks were treated using the deriva-
tization method described by Kleindienst et al. (2007). The
mass spectral analysis for the organic compounds used as
secondary molecular tracers has been described (Edney et al.,
2003). The method detection limit (MDL) for the SOC tracer
species is 0.1 ngm−3. Additional details of this methodology
are provided in the Supplement.
OC and elemental carbon (EC) concentrations were de-
termined using the thermal–optical transmittance (TOT)
method (Birch and Cary, 1996) from 1.54cm2punches of
quartz filters collected concurrent with the filters used for
tracer analyses (hereafter referred to as UNC/EPA OC). The
outer non-loaded rings were removed from filter samples and
then sent to Woods Hole Oceanographic Institute Accelerator
Mass Spectrometry for 14C analysis. The fraction of modern
carbon is provided for each daily total PM2.5carbon sample
(Geron, 2009). The modern carbon fraction is expressed as a
percentage of an oxalic acid standard material that represents
the carbon isotopic ratio for wood growth during 1890 (Stu-
iver, 1983). To account for the atmospheric 14C enhancement
due to nuclear bomb testing in the 1950s and 1960s, a factor
of 1.044 (Zotter et al., 2014) was used to calculate the con-
temporary carbon fraction from the measured modern carbon
result (Lewis et al., 2004; Zotter et al., 2014).
Two VOC data sets (one canister based and one in situ)
from each site were used in this analysis. Three-hour inte-
grated (06:00–09:00PDT) canister samples for VOC anal-
ysis were collected at both sites. A total of 41 samples
were collected at the Bakersfield site and 31 at Pasadena.
The offline VOC analysis details are given in the Sup-
plement. In Bakersfield, online VOC mixing ratios were
collected for 30min on the hour and analyzed via gas
chromatography–flame ionization detector (GC-FID) and
gas chromatography–mass spectrometry (GC-MS) (Gentner
et al., 2012). In Pasadena, online VOC measurements were
collected for 5min every 30 min and analyzed via GC-MS
(Borbon et al., 2013; Gilman et al., 2010). Carbon monoxide
measurements at Pasadena were determined using UV fluo-
rescence (Gerbig et al., 1999).
Hydroxyl (OH) and hydroperoxyl (HO2)radical measure-
ments were made at both locations using fluorescence assay
with gas expansion (FAGE). The Bakersfield OH measure-
ments used in this analysis were collected using the OHchem
method from the Penn State ground-based FAGE instrument
(Mao et al., 2012). The Pasadena hydroperoxyl observations
were made using the Indiana University FAGE instrument
(Dusanter et al., 2009). HO2measurements from both instru-
ments could contain an interference from various RO2; there-
fore, when comparing the model output with the observations
the sum of modeled HO2and RO2has been used (Griffith et
al., 2013).
OC measurements from nearby Chemical Speciation Net-
work (CSN) sites in Pasadena and Bakersfield were also
used for comparison purposes. The Los Angeles CSN site
(60371103) was approximately 14km from the CalNex site,
and the Bakersfield CSN site (60290014) was approximately
5km from the CalNex site (see Fig. S3a and b in the Sup-
plement). The CSN network uses quartz-fiber filters and
analyzes the carbon offline using the thermal–optical re-
flectance (TOR) method. Aerodyne high-resolution time-
of-flight aerosol mass spectrometer (AMS) measurements
of PM1OC made at Pasadena are described in Hayes et
al. (2013) and online Sunset Lab Inc. PM2.5OC measure-
ments made at Bakersfield are described in Liu et al. (2012).
3 Results and discussion
The results and discussion are organized such that the con-
temporary and fossil components of PM2.5carbon at the
Pasadena and Bakersfield sites are discussed, followed by
model performance for PM2.5carbon, speciated VOC, and
SOC tracer groups. Table 2 shows episode-aggregated model
performance metrics for PM2.5organic and elemental car-
bon, SOC tracers, total VOC, and select VOC species. The
results of a sensitivity increasing semivolatile yields are pre-
sented throughout and discussed in detail before finally pro-
viding an evaluation of PM2.5carbon at all routine monitor
sites in California.
3.1 Contemporary and fossil origins of PM2.5carbon
Field campaign average total PM2.5carbon measurements in-
dicate nearly equal amounts of contemporary and fossil con-
tribution at Pasadena and Bakersfield. The field study aver-
age contemporary fraction of 23 h average PM2.5total carbon
samples is 0.51 at Bakersfield (N=35) and 0.48 at Pasadena
(N=25). The estimate for contemporary carbon fraction at
Pasadena is consistent with other 14C measurements at this
location for this period (Zotter et al., 2014) and similar to
measurements made at urban areas in the southeast USA:
52% contemporary carbon in Birmingham and 63 % in At-
lanta (Kleindienst et al., 2010).
Figure 2 shows observed daily 23h PM2.5OC shaded by
contemporary and fossil component as well as PM2.5ele-
mental carbon. The fractional contribution of contemporary
carbon to total PM2.5carbon is variable from day-to-day at
the Pasadena site and steadily increases through the study
period at the Bakersfield location (first week average of 0.44
and final week average of 0.58). Some of the contemporary
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5248 K. R. Baker et al.: Gas and aerosol carbon in California
Table 2. Episode average measured and modeled PM2.5carbon, PM2.5SOC groups, and VOC at the Pasadena and Bakersfield sites.
Species Model Location NObserved Predicted Bias Error Fractional Fractional r
run (µgCm−3) (µgC m−3) (µgC m−3) (µgCm−3) bias (%) error (%)
Elemental carbon Baseline Bakersfield 35 0.5 0.4 −0.1 0.1 −13 35 0.17
Baseline Pasadena 31 0.2 1.0 0.8 0.8 125 125 0.70
Baseline CSN/IMPROVE sites 220 0.2 0.6 0.6 0.6 77 87 0.47
Organic carbon Baseline Bakersfield 35 5.4 0.8 −4.6 4.6 −144 144 0.11
Baseline Pasadena 31 3.6 2.0 −1.6 1.6 −53 53 0.73
Baseline CSN/IMPROVE sites 220 1.9 1.3 −0.6 0.9 −34 53 0.06
Sensitivity CSN/IMPROVE sites 220 1.9 1.7 −0.2 0.8 −11 42 0.32
Species Model Location NObserved Predicted Bias Error Fractional Fractional r
run (ngCm−3) (ngC m−3) (ngC m−3) (ngCm−3) bias (%) error (%)
Isoprene SOC Baseline Bakersfield 36 96 21 −75 75 −126 128 0.36
Pasadena 32 42 27 −15 25 −60 83 0.10
Monoterpene SOC Baseline Bakersfield 35 56 21 −35 37 −75 89 0.66
Pasadena 32 82 21 −60 61 −89 93 0.55
Toluene+xylene SOC Baseline Bakersfield 35 59 15 −44 44 −114 114 0.62
Pasadena 32 125 36 −89 89 −100 100 0.82
Sesquiterpene SOC Baseline Bakersfield 41 17
Pasadena 41 7
Benzene SOC Baseline Bakersfield 41 2
Pasadena 41 2
Alkane SOC Baseline Bakersfield 41 12
Pasadena 41 22
Cloud SOC Baseline Bakersfield 41 1
Pasadena 41 5
Naphthalene SOC Baseline Bakersfield 36 43
Pasadena 32 114
Species Model Location NObserved Predicted Bias Error Fractional Fractional r
run (ppbC) (ppbC) (ppbC) (ppbC) bias (%) error (%)
Isoprene VOC 3h Baseline Bakersfield 5 0.1 0.3 0.2 0.2 79 79 0.79
Pasadena 8 0.6 0.5 −0.2 0.5 0 84 −0.21
Monoterpene VOC 3h Baseline Bakersfield 37 1.4 0.5 −0.9 1.0 −72 89 0.25
Pasadena 28 1.8 0.3 −1.5 1.6 −129 137 0.15
Toluene VOC 3 h Baseline Bakersfield 41 4.3 2.7 −1.6 1.9 −48 55 0.44
Pasadena 29 7.3 7.7 0.4 3.5 17 44 0.24
Xylene VOC 3h Baseline Bakersfield 41 4.3 1.8 −2.5 2.5 −82 83 0.34
Pasadena 29 6.7 4.5 −2.1 2.6 −33 41 0.20
Benzene VOC 3h Baseline Bakersfield 41 1.2 1.3 0.2 0.5 6 38 0.14
Pasadena 29 1.5 1.6 0.1 0.5 0 30 0.16
Total VOC 3 h Baseline Bakersfield 41 186.9 63.7 −123.2 124.2 −95 97 0.37
Pasadena 29 188.9 88.7 −100.1 100.1 −66 66 0.26
Isoprene VOC 1h Baseline Bakersfield 712 0.4 0.4 0.0 0.3 −21 83 0.15
Pasadena 718 1.6 0.8 −0.9 1.7 −32 139 −0.10
Monoterpene VOC 1h Baseline Bakersfield 605 0.8 0.3 −0.6 0.7 −63 101 0.25
Pasadena 707 0.7 0.2 −0.5 0.5 −105 111 0.05
Toluene VOC 1 h Baseline Bakersfield 737 2.5 1.7 −0.8 1.5 −25 56 0.31
Pasadena 717 4.0 6.1 2.0 2.8 36 54 0.23
Xylene VOC 1h Baseline Bakersfield 737 1.9 1.1 −0.7 1.2 −37 64 0.32
Pasadena 718 3.2 3.4 0.2 1.7 2 51 0.15
carbon fraction measurements from Pasadena were above
1.0. These samples were considered erroneous and not in-
cluded in the analysis and suggest the possibility of positive
biases due to nearby sources (e.g., medical incinerator) in the
area. It is possible some of the stronger day-to-day variability
in contemporary carbon fraction measurements at Pasadena
may be related to biases due to nearby “hot” sources. Higher
time resolution 14C measurements at Pasadena show an in-
crease in fossil fraction during the middle of the day related
to increased emissions of fossil PM2.5carbon precursors and
SOA formation in the Los Angeles area (Zotter et al., 2014).
PM2.5OC of fossil origin at Pasadena shows the strongest re-
lationship to daily average temperature (Fig. S4a) compared
with contemporary carbon, total carbon, and elemental car-
bon. At Bakersfield the relationship between daily average
temperature and fossil and contemporary carbon is similar
(Fig. S4b) and not as strong as the relationship in Pasadena.
Neither fossil nor contemporary carbon concentrations show
discernible patterns by day of the week at either location
(Fig. S5).
Modeled contemporary PM2.5carbon is estimated by sum-
ming primarily emitted PM2.5multiplied by the contempo-
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5249
Figure 2. Observed daily 23h average PM2.5elemental carbon,
PM2.5contemporary-origin organic carbon, and PM2.5fossil-origin
organic carbon at Pasadena and Bakersfield.
rary fraction of urban area emissions (see Sect. 2.1 and Ta-
ble 1) with model-estimated biogenic SOC species. The av-
erage baseline modeled contemporary fraction of PM2.5OC
in Pasadena is 0.51 and Bakersfield 0.54, both of which are
similar to average observation estimates. However, the model
shows little day-to-day variability in contemporary carbon
fraction, which does not match observed trends (Fig. S6).
Episode average modeled estimates of PM2.5OC contempo-
rary fraction are similar to the estimated contemporary frac-
tion of the urban emissions of primary PM2.5OC (Bakers-
field=0.53 and Pasadena =0.51), as noted in Table 1.
3.2 PM2.5carbon
Figure 3 shows measured (UNC/EPA data) and modeled
PM2.5OC at Bakersfield and Pasadena. Organic carbon mea-
surements from co-located instruments (AMS at Pasadena
measured PM1and Sunset at Bakersfield measured PM2.5)
and the nearest CSN monitor are also shown in Fig. 3.
The co-located AMS measurements compare well with the
UNC/EPA PM2.5organic carbon measurements at Pasadena,
while the concentrations measured at the nearby CSN site
are substantially lower. At Bakersfield, UNC/EPA measure-
ments are higher compared with the nearby CSN (episode av-
erage ∼3 times higher), and co-located daily average Sunset
(episode average 20% higher) measured PM2.5OC illustrate
possible measurement artifacts in the UNC/EPA measure-
ments at this location. These differences in measured concen-
tration at Bakersfield may be related to filter handling, vari-
ability in collected blanks, true differences in the OC concen-
trations since the CSN site is spatially distinct, differences in
the height of measurement (these CSN monitors are situated
on top of buildings), and differences in analytical methods
since CSN sites use TOR to operationally define OC and EC.
Figure 3. Model-predicted and measured PM2.5organic carbon at
Pasadena and Bakersfield. The nearby CSN measurements are in-
tended to provide additional context and are not co-located with
CalNex measurements or model estimates.
Modeled PM2.5OC is underestimated at both CalNex
locations (Fig. 3), most notably at Bakersfield. However,
given the large differences in PM2.5OC mass compared
to co-located and nearby routine measurements, it is not
clear which measurement best represents ambient PM2.5
OC concentrations and would be most appropriate for com-
parison with the model. The model generally compares
well to the CSN site nearest Pasadena and Bakersfield.
PM2.5elemental carbon is well characterized by the model
at Bakersfield (fractional bias= −13 % and fractional er-
ror=35 %) and overestimated at Pasadena (fractional bias
and error=125 %) (Fig. S7). Since the emissions are based
on TOR and UNC/EPA measurements use the TOT opera-
tional definition of total carbon, some of the model overesti-
mation may be related to the TOR method estimating higher
elemental carbon fraction of total carbon (Chow et al., 2001).
PM2.5OC is mostly primary (Pasadena 93% and Bakers-
field 88%) in the baseline model simulation. AMS measure-
ments at Pasadena suggest OC is mostly secondary in nature
with an average of 63% for the semi-volatile oxidized or-
ganic aerosol and oxygenated organic aerosol components
for this field study (Hayes et al., 2013). Model-estimated
PM2.5OC is largely from primarily emitted sources and con-
temporary in nature based on the contemporary/fossil split of
primary PM2.5emissions near both sites (Fig. S6). Primar-
ily emitted PM2.5OC emissions sources near Pasadena and
Bakersfield include mobile sources, cooking, and dust based
on emissions inventory information (Table 1). Some of these
sources of primarily emitted PM2.5OC may be semivolatile
in nature. Model treatment of POA as semivolatile may im-
prove the primary–secondary comparison with observations
but would likely exacerbate underpredictions of PM2.5OC
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5250 K. R. Baker et al.: Gas and aerosol carbon in California
Figure 4. Observed (top row) and modeled (middle and bottom rows) PM2.5organic carbon at Pasadena and Bakersfield. Mass explained by
SOA tracers shown in green (contemporary-origin tracers) and brown (fossil-origin tracers). Top row tan shading indicates mass not explained
by known observed SOC tracers. Middle and bottom row gray shading shows modeled primarily emitted PM2.5that is both contemporary
and fossil in origin. Middle row shows baseline model estimates and bottom row model sensitivity results with increased SOA yields.
unless oxidation and re-partitioning of the products is con-
sidered (Robinson et al., 2007). The underestimation of SOC
may result from underestimated precursor VOC, poorly char-
acterized oxidants, underestimated semivolatile yields, miss-
ing intermediate volatility VOC emissions (Stroud et al.,
2014; Zhao et al., 2014), other issues, or some combination
of each.
3.3 Gas-phase carbon
Model estimates are paired with hourly VOCs (Fig. S8) and
mid-morning 3h average VOC (Fig. S9) at both locations.
Compounds considered largely fossil in origin including xy-
lene, toluene, and benzene are generally well predicted at
both sites although these species tend to be slightly over-
estimated at Pasadena and slightly underestimated at Bak-
ersfield. Since emissions of these compounds near these sites
are largely from mobile sources (Table 1), this suggests emis-
sions from this sector are fairly well characterized in this ap-
plication.
Contemporary (biogenic)-origin monoterpenes are under-
estimated at both sites while isoprene is underestimated at
Pasadena and has little bias at Bakersfield based on hourly
measurements (Fig. S8; Table 2). Isoprene and monoterpene
performance may be partly related to the model not fully cap-
turing transport from nearby areas with large emitting veg-
etation to these monitor locations (Heo et al., 2015), defi-
ciencies in emissions factors, or poorly characterized vegeta-
tion. Speciated monoterpene measurements made at Bakers-
field during this field campaign suggest emissions of certain
species were elevated at the start of this time period due to
flowering (Gentner et al., 2014b), which is a process not in-
cluded in current biogenic emissions models and thus may
contribute to modeled monoterpene underestimates.
Other VOC species that are systematically underestimated
include ethane, methanol, ethanol, and acetaldehyde. Un-
derprediction of methanol and ethanol in Bakersfield may
be largely related to missing VOC emissions for confined
animal operations in the emission inventory (Gentner et
al., 2014a). Underestimates of oxygenated VOC compounds
may indirectly impact SOC formation through muted photo-
chemistry (Steiner et al., 2008). Carbon monoxide tends to
be underestimated at both locations (Fig. S8) possibly due to
boundary inflow concentrations from the global model simu-
lation being too low or underestimated regional emissions.
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5251
0 5 10 15
0 5 10 15
Xylene+Toluene
Observed (ppbC)
Predicted (ppbC)
Pasadena
012345
012345
Isoprene
Observed (ppbC)
Predicted (ppbC)
Pasadena
0.0 0.5 1.0 1.5 2.0
0.0 0.5 1.0 1.5 2.0
Monoterpenes
Observed (ppbC)
Predicted (ppbC)
Pasadena
0 100 200 300
0 100 200 300
Toluene + Xylene SOC
Observed (ngC/m3)
Baseline Predicted (ngC/m3)
Pasadena
0 20 40 60 80 100
0 20 40 60 80
Isoprene SOC
Observed (ngC/m3)
Baseline Predicted (ngC/m3)
Pasadena
0 50 100 200
0 50 150 250
Monoterpene SOC
Observed (ngC/m3)
Baseline Predicted (ngC/m3)
Pasadena
Figure 5. Comparison of CMAQ-predicted and measured VOC
(daily average of hourly samples) and corresponding SOC species
(daily 23h average samples) for Pasadena. Comparison points out-
side the gray lines indicate model predictions are greater than a fac-
tor of 2 different from the measurements.
3.4 PM2.5SOC tracers
Figure 4 shows modeled and measured total PM2.5OC mass.
Measured mass explained by fossil and contemporary SOC
tracers are shown in the top row. The unexplained observed
fraction is a mixture of primary, secondary, fossil, and con-
temporary origin. Modeled mass is colored to differentiate
primarily emitted OC and SOC. Estimates of SOC mass from
a specific or lumped VOC group (e.g., isoprene, monoter-
penes, toluene), hereafter called SOC tracer mass, comprise
little of the measured or modeled PM2.5OC at either of these
locations during this field study (Fig. 4). Total SOC tracer
estimates explain only 9% of the total measured UNC/EPA
PM2.5OC at Pasadena and 5% at Bakersfield. The percent-
age of mass explained by known secondary tracers is smaller
than urban areas in the southeast USA: 27% in Atlanta and
31% in Birmingham (Kleindienst et al., 2010).
The portion of measured and modeled PM2.5carbon
not identified with tracers may be from underestimated
adjustment factors related to previously uncharacterized
semivolatile VOC (SVOC) wall loss in chamber studies
(Zhang et al., 2014b) and unidentified SOC pathways. Ad-
ditional reasons for the low estimate of observed tracer con-
tribution to PM2.5carbon include known pathways without
an ambient tracer and tracer degradation between formation
and measurement. Based on 14C measurements, this uniden-
tified portion of the measurements is likely comprised of both
contemporary and fossil carbon in generally similar amounts.
Total modeled SOC explains only 12% of the PM2.5car-
bon at Bakersfield and 7 % at Pasadena. As noted previously,
AMS-based observations suggest most OC is SOC (63%) at
Pasadena (Hayes et al., 2013), meaning both the SOC tracer
measurements and model estimates explain little of the SOC
at this location.
Despite the relatively small component of PM2.5carbon
explained by SOC tracers, a comparison of measured and
modeled SOC and precursor VOC provides additional op-
portunity to better understand sources of PM2.5carbon in
these areas and begin to establish relationships between pre-
cursors and resulting SOC formation. Ambient and model-
estimated SOC tracers and daily average VOC precursors
are shown in Fig. 5 for Pasadena and Fig. 6 for Bakersfield.
The model underestimates toluene and xylene SOC at both
locations even though VOC gas precursors show an over-
prediction tendency at Pasadena and only a slight underes-
timation at Bakersfield. Isoprene SOC is generally under-
predicted at both sites, in particular at Bakersfield. This is
in contrast to the slight overprediction of daily 24h aver-
age isoprene at Bakersfield. One explanation may be that
isoprene SOC is formed elsewhere in the region (e.g., the
nearby foothills of the Sierra Nevada where emissions are
highest in the region), which would support the lack of rela-
tionship between isoprene SOC and isoprene concentrations
at Bakersfield (Heo et al., 2015; Shilling et al., 2013). The
lack of relationship could also be related to the reactive up-
take kinetics of isoprene-derived epoxydiols (IEPOX) (Gas-
ton et al., 2014) and methacrylic acid epoxide (MAE). Since
the model does not include the reactive uptake of IEPOX and
MAE and subsequent acid-catalyzed aqueous-phase chem-
istry, it is likely isoprene SOC would be underestimated to
some degree at both sites (Karambelas et al., 2013; Pye et
al., 2013). Of these channels the IEPOX channel is thought
to have the largest SOA production potential, but the chem-
istry in the LA basin is dominated by the high-NO channel
(Hayes et al., 2014) and thus IEPOX is not formed from iso-
prene emitted within the LA basin. Consistent with that ob-
servation, the AMS tracer of IEPOX SOA is only detected at
background level in the LA basin.
Monoterpene VOC and monoterpene SOC are underesti-
mated systematically at both locations, suggesting underpre-
dictions of the VOC precursor translates to underestimates
in SOC. As noted previously, monoterpene measurements
suggest an emissions enhancement related to flowering or
other emission events (e.g., harvest or pruning) (Gentner et
al., 2014b) that is not included in current biogenic emis-
sions model formulations. The monoterpene-measured tracer
SOC group is based on α-pinene products. Measured SOC
at these sites could be from monoterpene species other than
α-pinene. A coincident study near Bakersfield indicates α-
and β-pinene emissions represent a fairly small fraction of
total monoterpene emissions during this time period (Gen-
tner et al., 2014b). SOA yields in CMAQ for monoterpenes
are heavily weighted toward α- and β-pinene, which may be
appropriate in most places but not here where measurements
show large contributions from limonene, myrcene, and para-
cymene. This is important because yields vary among from
different monoterpenes and limonene has a much larger SOA
yield than pinenes (Carlton et al., 2010).
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5252 K. R. Baker et al.: Gas and aerosol carbon in California
●
●●
●
●●
●●●
●
●●
●
●
●
●●
●
●●
●●
●
●●
●
●
●
●●●
●
●
●
●
●
●
0 2 4 6 8
02468
Xylene+Toluene
Observed (ppbC)
Predicted (ppbC)
●Bakersfield
●
●●
●
●
●
●●
●
●
●
●●
●
●
●
●
●
●●
●
●●
●
●
●●
●
●●●
●
●
●
●●
●
0.0 0.4 0.8 1.2
0.0 0.4 0.8 1.2
Isoprene
Observed (ppbC)
Predicted (ppbC)
●Bakersfield
●●●
●
●
●
●●
●●
●
●●
●●
●●
●
●●
●
●
●●
●●●
●
●●
●
●
●●
●●●
01234
01234
Monoterpenes
Observed (ppbC)
Predicted (ppbC)
●Bakersfield
●
●
●●
●
●
●
●●●
●
●●
●
●
●
●●●
●
●
●●
●●
●●
●●
●●●
●●
●
0 50 100 150
0 50 100 150
Toluene + Xylene SOC
Observed (ngC/m3)
Baseline Predicted (ngC/m3)
●Bakersfield
●
●●
●●●
●●
●●●●
●●
●
●
●
●
●●
●
●
●●
●●●●
●●
●
●●
●●
●
0 50 150 250
0 50 150 250
Isoprene SOC
Observed (ngC/m3)
Baseline Predicted (ngC/m3)
●Bakersfield
●
●
●
●
●
●
●
●●●
●
●●
●
●●●
●●
●
●●●
●●
●
●
●
●
●●●
●●
●
0 50 100 150 200
0 50 100 150 200
Monoterpene SOC
Observed (ngC/m3)
Baseline Predicted (ngC/m3)
●Bakersfield
Figure 6. Comparison of CMAQ-predicted and measured VOC
(daily average of hourly samples) and corresponding SOC species
(daily 23h average samples) for Bakersfield. Comparison points
outside the gray lines indicate model predictions are greater than
a factor of 2 different from the measurements.
Sesquiterpene VOC and SOC tracer (β-caryophyllenic
acid) mass measurements were never above the MDL at ei-
ther site during CalNex, but the modeling system often pre-
dicts SOC from this VOC group (Table 2, Fig. S10b). The
SOC tracer measurement methodology is more uncertain
for sesquiterpene products (Offenberg et al., 2009) and gas-
phase sesquiterpenes would have oxidized before reaching
the measurement sites since sesquiterpene-emitting vegeta-
tion exists in the San Joaquin Valley (Ormeño et al., 2010).
It is also possible that SOC is forming from sesquiterpenes
other than β-caryophyllene.
One potential explanation for an underestimation of
SOC despite well-characterized precursors (e.g., toluene and
xylenes) could be the lack of available oxidants. As shown
in Fig. 7, the model tends to overestimate the hydroxyl
radical compared with measurement estimates at Pasadena.
Hydroperoxyl+peroxy radical measurements are underes-
timated at Pasadena by a factor of 2 on average. The
model overestimates preliminary measurements of both hy-
droxyl (by nearly a factor of 2 on average) and hydroper-
oxyl+peroxy radicals at Bakersfield. Model representation
of hydroxyl radical at these locations during this time period
does not seem to be limiting VOC oxidation to semivolatile
products. Better agreement between radical ambient and
modeled estimates could result in less SOC produced by the
model and exacerbate model SOC underestimates. This sug-
gests deficiencies other than radical representation by the
modeling system are more influential in SOC performance
for these areas. However, hydroperoxyl underestimates at
Pasadena could lead to muted SOA formation through low-
NOxpathways dependent on hydroperoxyl concentrations
and contribute to model underestimates of SOC.
3.5 Sensitivity simulation
OH is not underestimated in the model and biases in precur-
sor VOC do not clearly translate into similar biases in SOC
(e.g., toluene and xylene VOC are overestimated at Pasadena
but tracer SOC for this group is underestimated) for these
sites during this time period. Modeled SOC may partly be
underestimated due to the use of experimental SOC yields
that may be biased low due to chamber studies not fully ac-
counting for SVOC wall loss (Zhang et al., 2014b). Even
though Zhang et al. (2014b) showed results for one precursor
to SOA pathway, for a sensitivity study here the yield of all
semivolatile gases is increased by a factor of 4. This was done
by increasing the mass-based stoichiometric coefficients for
each VOC-to-SOA pathway in the model to provide a pre-
liminary indication about how increased yields might impact
model performance. A factor of 4 is chosen based on the up-
per limit related to SVOC wall loss in Zhang et al. (2014b).
Aside from wall loss characterization, there are a variety of
other aspects of chamber studies that could result in under-
estimated yields including particle-phase accretion, aqueous-
phase chemistry, and differences in chamber and ambient hu-
midity.
Model estimates of PM2.5OC increase in urban areas and
regionally when semivolatile yields are increased. The sen-
sitivity simulation results in episode average anthropogenic
SOC increases between a factor of 3 (benzene SOC at
Pasadena) to 4.8 (toluene and xylene SOC at Pasadena)
and biogenic SOC increases between a factor of 5.1 (iso-
prene SOC at Pasadena) to 8.9 (monoterpene SOC at Bakers-
field). Model performance improves at the CalNex locations
(Figs. 3 and 4) and at routine monitors throughout Califor-
nia (Fig. 8). Average fractional bias improves from −34 to
−11 % at routine monitor locations and fractional error is re-
duced from 53 to 42%.
The sensitivity simulation with increased semivolatile
yields results in increased model-estimated secondary contri-
bution as a percent of PM2.5carbon but still does not conform
to observation-based estimates that indicate PM2.5carbon is
largely secondary in nature at these sites (Liu et al., 2012;
Hayes et al., 2013). Modeled SOC in the sensitivity simula-
tion explains 36% of the PM2.5OC at Bakersfield and 22 %
at Pasadena, which is larger than the baseline simulation by
more than a factor of 3. The model-predicted percent con-
temporary fraction of PM2.5carbon changed very little due
to this sensitivity. The model sensitivity results are not com-
pared to SOC tracer group estimates since the conversion of
tracer concentrations to SOC concentrations would require a
similar adjustment and would result in similar relationships
between model estimates and observations.
3.6 Aqueous and other SOC processes
Measurements in Pasadena during the summer of 2009 sug-
gest aqueous processes can be important for SOC mass
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5253
Figure 7. Measured and model-estimated OH radical (top) and HO2+RO2(bottom) at Pasadena. The ratio shown on the scatterplots is the
episode average model estimate divided by the episode average measured values.
Figure 8. Episode average modeled PM2.5organic carbon and measurements from both CalNex locations and routine networks including
CSN (circles) and IMPROVE (squares). Left panel shows baseline model predictions and right panel shows model estimates with increased
SOA yields.
(Hersey et al., 2011). For the CalNex period at Pasadena,
other research showed box-model-estimated 8h average
SOC from aqueous-phase chemistry of glyoxal to be be-
tween 0.0 and 0.2µg m−3(Washenfelder et al., 2011), and
Hayes et al. (2014) showed that the observed SOA was not
different between cloudy and clear morning days. CMAQ-
predicted 24 h average SOC from glyoxal and methylglyoxal
through aqueous chemistry at Pasadena ranges from 0.0 to
0.04µg m−3. CMAQ estimates of SOC from small carbonyl
compounds via aqueous-phase processes are within the range
inferred from measurements.
Not all CMAQ SOC formation pathways can be in-
cluded in this analysis. No observational indicator exists for
SOC derived from alkanes, benzene, glyoxal, and methyl-
glyoxal since unique tracer species have not been deter-
mined. Conversely, naphthalene/polycyclic aromatic hydro-
carbons (PAHs) SOC tracers were measured but not mod-
eled in CMAQ. Measured naphthalene SOC at these sites is
minor (Hayes et al., 2014), which is consistent with other
areas (Dzepina et al., 2009). Previous CMAQ simulations
predict that PAHs contribute less than 30ng m−3of SOA
in Southern California in summer (Pye and Pouliot, 2012),
and thus including those pathways is unlikely to close the
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5254 K. R. Baker et al.: Gas and aerosol carbon in California
model–measurement gap in PM2.5OC. 2-Methyl-3-buten-2-
ol (MBO) derived SOC concentrations (3–4ngC m−3)were
low at both monitor locations throughout the campaign
(Lewandowski et al., 2013). MBO does not appear to no-
tably contribute SOC at these locations during this time pe-
riod, which is consistent with low yields estimated in labora-
tory experiments (Chan et al., 2009). Organic carbon emit-
ted from marine biological activity is not included in this
modeling assessment and may contribute to some degree at
Pasadena (Gantt et al., 2010) based on ship-based measure-
ments (Hayes et al., 2013).
3.7 Regional PM2.5organic carbon
Including routine measurement data is important to provide
broader context for PM2.5carbon in California and under-
stand how the model performs and responds to perturbations
at diverse locations beyond the two CalNex sites. The highest
average modeled PM2.5OC in California during this period
is in the Los Angeles area (Fig. 8). The Sacramento and San
Joaquin valleys also show higher concentrations of PM2.5
OC than more rural parts of the state (Fig. 8). Measure-
ments made at routine monitor networks (Fig. 8) show sim-
ilar elevated concentrations near Los Angeles, Sacramento
Valley, and San Joaquin Valley. These areas of elevated OC
generally coincide with areas of the state that experience a
build-up of pollutants due to terrain features blocking air flow
(Baker et al., 2013). The model does not tend to capture the
highest concentrations of measured PM2.5OC in the central
San Joaquin Valley, Imperial Valley, or at one CSN monitor
in the northeast Sierra Nevada that is near large residential
wood combustion emissions (Fig. S11). The model under-
estimates PM2.5OC on average across all CSN sites during
this time period (fractional bias= −34 % and fractional er-
ror=53 %). The modeling system shows an overprediction
tendency (fractional bias=77 %) across all CSN sites for
PM2.5elemental carbon in California during this period.
4 Conclusions
Total PM2.5carbon at Pasadena and Bakersfield during the
CalNex period in May and June 2010 is fairly evenly split
between contemporary and fossil origin. Total PM2.5OC is
generally underestimated at both field study locations and at
many routine measurement sites in California, and compari-
son with AMS observations suggests a large underestimation
of SOC. Semivolatile yields were increased by a factor of
4 based on recent research suggesting yields may be higher
due to updated accounting for SVOC wall loss. This sensitiv-
ity resulted in a better comparison to routine and field study
measurements. However, the model-estimated OC is still
largely primary in nature and inconsistent with observation-
based approaches at these sites. A modeling study for the
same time period using different emissions, photochemical
transport model, and SOA treatment also shows underesti-
mated OA and SOA at Pasadena and underestimated SOA
but comparable OA at the Bakersfield location (Fast et al.,
2014).
CMAQ predictions of individual VOCs are often not con-
sistent with model performance for the corresponding subse-
quent SOC species mass. Gas-phase mixing ratios of toluene
and xylene are well predicted by CMAQ, typically within
a factor of 2 of the observations at both sites. However,
measurement-based estimates of the corresponding SOC
mass are consistently greater than model-predicted mass.
Mass concentrations of the isoprene SOC are systematically
underpredicted, most noticeably at Bakersfield, while model
predictions of gas-phase isoprene are not biased in only one
direction to the same degree. Gas-phase monoterpenes and
the related SOC species are underpredicted at both CalNex
monitoring sites. The hydroxyl radical is fairly well charac-
terized at Pasadena and systematically overestimated at Bak-
ersfield, suggesting oxidants are not limiting SOC production
in the model.
Episode average CMAQ model estimates of PM2.5OC
contemporary fraction at Pasadena and Bakersfield are sim-
ilar to radiocarbon measurements but lack day-to-day vari-
ability. CMAQ PM2.5OC is predominantly primary in origin,
which is contrary to findings from other studies that indicate
PM2.5OC in these areas are largely secondary in nature dur-
ing this time period (Bahreini et al., 2012; Hayes et al., 2013;
Liu et al., 2012). Treatment of primarily emitted PM2.5OC
as semivolatile would likely result in total PM2.5OC esti-
mates that would be mostly secondary rather than primary.
However, this would likely exacerbate model underestimates
of PM2.5OC. Some model performance features, including
underestimated SOC, may be related to less volatile hydro-
carbon emissions missing from the emission inventory (Chan
et al., 2013; Gentner et al., 2012; Jathar et al., 2014; Zhao et
al., 2014) or mischaracterized when lumped into chemical
mechanism VOC species (Jathar et al., 2014). A future in-
tent is to simulate this same period using a volatility basis set
approach to treat primary OC emissions with some degree
of volatility and potential for SOC production and better ac-
count for sector-specific intermediate volatility emissions.
The Supplement related to this article is available online
at doi:10.5194/acp-15-5243-2015-supplement.
Acknowledgements. The authors would like to acknowledge
measurements taken by Scott Scheller and the contribution
from Chris Misenis, Allan Beidler, Chris Allen, James Beidler,
Heather Simon, and Rich Mason. EPA, through its Office of Re-
search and Development, funded and collaborated in the research
described here under contract EP-D-10-070 to Alion Science and
Technology. This work is supported in part through EPA’s STAR
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5255
program, grant number RD83504101. P. L. Hayes and J. L. Jimenez
were supported by CARB 11-305.
Disclaimer. Although this work was reviewed by EPA and
approved for publication, it may not necessarily reflect official
agency policy.
Edited by: R. McLaren
References
Bahreini, R., Middlebrook, A. M., de Gouw, J. A., Warneke, C.,
Trainer, M., Brock, C. A., Stark, H., Brown, S. S., Dube, W. P.,
Gilman, J. B., Hall, K., Holloway, J. S., Kuster, W. C., Perring,
A. E., Prevot, A. S. H., Schwarz, J. P., Spackman, J. R., Szi-
dat, S., Wagner, N. L., Weber, R. J., Zotter, P., and Parrish, D.
D.: Gasoline emissions dominate over diesel in formation of sec-
ondary organic aerosol mass, Geophys. Res. Lett., 39, L06805,
doi:10.1029/2011gl050718, 2012.
Baker, K. R., Misenis, C., Obland, M. D., Ferrare, R. A., Scarino, A.
J., and Kelly, J. T.: Evaluation of surface and upper air fine scale
WRF meteorological modeling of the May and June 2010 Cal-
Nex period in California, Atmos. Environ., 80, 299–309, 2013.
Barsanti, K. C., Carlton, A. G., and Chung, S. H.: Analyzing exper-
imental data and model parameters: implications for predictions
of SOA using chemical transport models, Atmos. Chem. Phys.,
13, 12073–12088, doi:10.5194/acp-13-12073-2013, 2013.
Birch, M. and Cary, R.: Elemental carbon-based method for mon-
itoring occupational exposures to particulate diesel exhaust,
Aerosol Sci. Technol., 25, 221–241, 1996.
Borbon, A., Gilman, J., Kuster, W., Grand, N., Chevaillier, S.,
Colomb, A., Dolgorouky, C., Gros, V., Lopez, M., and Sarda-
Esteve, R.: Emission ratios of anthropogenic volatile organic
compounds in northern mid-latitude megacities: Observations
versus emission inventories in Los Angeles and Paris, J. Geo-
phys. Res.-Atmos., 118, 2041–2057, 2013.
Carlton, A. G. and Baker, K. R.: Photochemical Modeling of the
Ozark Isoprene Volcano: MEGAN, BEIS, and Their Impacts on
Air Quality Predictions, Environ. Sci. Technol., 45, 4438–4445,
doi:10.1021/es200050x, 2011.
Carlton, A. G., Turpin, B. J., Altieri, K. E., Reff, A., Seitzinger,
S., Lim, H. J., and Ervens, B.: Atmospheric Oxalic Acid and
SOA Production from Glyoxal: Results of Aqueous Photooxi-
dation Experiments, Atmos. Environ., 41, 7588–7602, 2007.
Carlton, A. G., Turpin, B. J., Altieri, K. E., Seitzinger, S. P.,
Mathur, R., Roselle, S. J., and Weber, R. J.: CMAQ model perfor-
mance enhanced when in-cloud SOA is included: comparisons of
OC predictions with measurements, Environ. Sci. Technol., 42,
8798–8802, 2008.
Carlton, A. G., Bhave, P. V., Napelenok, S. L., Edney, E. O., Sarwar,
G., Pinder, R. W., Pouliot, G. A., and Houyoux, M.: Treatment of
secondary organic aerosol in CMAQv4.7, Environ. Sci. Technol.,
44, 8553–8560, 2010.
Carter, W. P. L.: Implementation of the SAPRC-99 chemical mech-
anism into the models-3 framework, U.S. Environmental Pro-
tection Agency, University of California, Riverside, CA, USA,
2000.
Chan, A. W., Isaacman, G., Wilson, K. R., Worton, D. R., Ruehl,
C. R., Nah, T., Gentner, D. R., Dallmann, T. R., Kirchstetter,
T. W., and Harley, R. A.: Detailed chemical characterization of
unresolved complex mixtures in atmospheric organics: Insights
into emission sources, atmospheric processing, and secondary
organic aerosol formation, J. Geophys. Res.-Atmos., 118, 6783–
6796, 2013.
Chan, A. W. H., Galloway, M. M., Kwan, A. J., Chhabra, P. S.,
Keutsch, F. N., Wennberg, P. O., Flagan, R. C., and Seinfeld, J.
H.: Photooxidation of 2-Methyl-3-Buten-2-ol (MBO) as a Poten-
tial Source of Secondary Organic Aerosol, Environ. Sci. Tech-
nol., 43, 4647–4652, doi:10.1021/es802560w, 2009.
Chow, J. C., Watson, J. G., Crow, D., Lowenthal, D. H., and Mer-
rifield, T.: Comparison of IMPROVE and NIOSH carbon mea-
surements, Aerosol Sci. Technol., 34, 23–34, 2001.
Chung, S. H. and Seinfeld, J. H.: Global distribution and climate
forcing of carbonaceous aerosols, J. Geophys. Res.-Atmos., 107,
AAC14.11–AAC14.33, doi:10.1029/2001jd001397, 2002.
Docherty, K. S., Stone, E. A., Ulbrich, I. M., DeCarlo, P. F., Snyder,
D. C., Schauer, J. J., Peltier, R. E., Weber, R. J., Murphy, S. N.,
Seinfeld, J. H., Grover, B. D., Eatough, D. J., and Jiimenez, J. L.:
Apportionment of Primary and Secondary Organic Aerosols in
Southern California during the 2005 Study of Organic Aerosols
in Riverside (SOAR-1), Environ. Sci. Technol., 42, 7655–7662,
doi:10.1021/es8008166, 2008.
Dusanter, S., Vimal, D., Stevens, P. S., Volkamer, R., and Molina,
L. T.: Measurements of OH and HO2concentrations during the
MCMA-2006 field campaign – Part 1: Deployment of the Indiana
University laser-induced fluorescence instrument, Atmos. Chem.
Phys., 9, 1665–1685, doi:10.5194/acp-9-1665-2009, 2009.
Dzepina, K., Volkamer, R. M., Madronich, S., Tulet, P., Ulbrich,
I. M., Zhang, Q., Cappa, C. D., Ziemann, P. J., and Jimenez, J.
L.: Evaluation of recently-proposed secondary organic aerosol
models for a case study in Mexico City, Atmos. Chem. Phys., 9,
5681–5709, doi:10.5194/acp-9-5681-2009, 2009.
Edney, E. O., Kleindienst, T. E., Conver, T. S., McIver, C. D., Corse,
E. W., and Weathers, W. S.: Polar organic oxygenates in PM2.5
at a southeastern site in the United States, Atmos. Environ., 37,
3947–3965, 2003.
Ensberg, J. J., Hayes, P. L., Jimenez, J. L., Gilman, J. B., Kuster,
W. C., de Gouw, J. A., Holloway, J. S., Gordon, T. D., Jathar, S.,
Robinson, A. L., and Seinfeld, J. H.: Emission factor ratios, SOA
mass yields, and the impact of vehicular emissions on SOA for-
mation, Atmos. Chem. Phys., 14, 2383–2397, doi:10.5194/acp-
14-2383-2014, 2014.
ENVIRON: User’s Guide Comprehensive Air Quality Model with
Extensions version 6, available at: www.camx.com (last access:
8 May 2015), ENVIRON International Corporation, Novato,
2014.
Ervens, B., Turpin, B. J., and Weber, R. J.: Secondary or-
ganic aerosol formation in cloud droplets and aqueous parti-
cles (aqSOA): a review of laboratory, field and model stud-
ies, Atmos. Chem. Phys., 11, 11069–11102, doi:10.5194/acp-11-
11069-2011, 2011.
Fast, J. D., Allan, J., Bahreini, R., Craven, J., Emmons, L., Ferrare,
R., Hayes, P. L., Hodzic, A., Holloway, J., Hostetler, C., Jimenez,
J. L., Jonsson, H., Liu, S., Liu, Y., Metcalf, A., Middlebrook, A.,
Nowak, J., Pekour, M., Perring, A., Russell, L., Sedlacek, A.,
Seinfeld, J., Setyan, A., Shilling, J., Shrivastava, M., Springston,
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5256 K. R. Baker et al.: Gas and aerosol carbon in California
S., Song, C., Subramanian, R., Taylor, J. W., Vinoj, V., Yang,
Q., Zaveri, R. A., and Zhang, Q.: Modeling regional aerosol and
aerosol precursor variability over California and its sensitivity
to emissions and long-range transport during the 2010 CalNex
and CARES campaigns, Atmos. Chem. Phys., 14, 10013–10060,
doi:10.5194/acp-14-10013-2014, 2014.
Fountoukis, C. and Nenes, A.: ISORROPIA II: a computa-
tionally efficient thermodynamic equilibrium model for K+-
Ca2+-Mg2+-NH4+-Na+-SO2−
4-NO−
3-Cl−-H2O aerosols, At-
mos. Chem. Phys., 7, 4639–4659, doi:10.5194/acp-7-4639-2007,
2007.
Gantt, B., Meskhidze, N., and Carlton, A. G.: The contribution of
marine organics to the air quality of the western United States,
Atmos. Chem. Phys., 10, 7415–7423, doi:10.5194/acp-10-7415-
2010, 2010.
Gaston, C. J., Riedel, T. P., Zhang, Z., Gold, A., Surratt, J. D., and
Thornton, J. A.: Reactive Uptake of an Isoprene-derived Epoxy-
diol to Submicron Aerosol Particles, Environ. Sci. Technol., 48,
11178–11186, doi:10.1021/es5034266, 2014.
Gentner, D. R., Isaacman, G., Worton, D. R., Chan, A. W. H., Dall-
mann, T. R., Davis, L., Liu, S., Day, D. A., Russell, L. M., Wil-
son, K. R., Weber, R., Guha, A., Harley, R. A., and Goldstein,
A. H.: Elucidating secondary organic aerosol from diesel and
gasoline vehicles through detailed characterization of organic
carbon emissions, P. Natl. Acad. Sci. USA, 109, 18318–18323,
doi:10.1073/pnas.1212272109, 2012.
Gentner, D. R., Ford, T. B., Guha, A., Boulanger, K., Brioude, J.,
Angevine, W. M., de Gouw, J. A., Warneke, C., Gilman, J. B.,
Ryerson, T. B., Peischl, J., Meinardi, S., Blake, D. R., Atlas, E.,
Lonneman, W. A., Kleindienst, T. E., Beaver, M. R., Clair, J.
M. St., Wennberg, P. O., VandenBoer, T. C., Markovic, M. Z.,
Murphy, J. G., Harley, R. A., and Goldstein, A. H.: Emissions
of organic carbon and methane from petroleum and dairy oper-
ations in California’s San Joaquin Valley, Atmos. Chem. Phys.,
14, 4955–4978, doi:10.5194/acp-14-4955-2014, 2014a.
Gentner, D. R., Ormeño, E., Fares, S., Ford, T. B., Weber, R., Park,
J.-H., Brioude, J., Angevine, W. M., Karlik, J. F., and Goldstein,
A. H.: Emissions of terpenoids, benzenoids, and other biogenic
gas-phase organic compounds from agricultural crops and their
potential implications for air quality, Atmos. Chem. Phys., 14,
5393–5413, doi:10.5194/acp-14-5393-2014, 2014b.
Gerbig, C., Schmitgen, S., Kley, D., Volz-Thomas, A., Dewey, K.,
and Haaks, D.: An improved fast-response vacuum-UV reso-
nance fluorescence CO instrument, J. Geophys. Res.-Atmos.,
104, 1699–1704, 1999.
Geron, C.: Carbonaceous aerosol over a Pinus taeda forest in Cen-
tral North Carolina, USA, Atmos. Environ., 43, 959–969, 2009.
Gilman, J. B., Burkhart, J. F., Lerner, B. M., Williams, E. J., Kuster,
W. C., Goldan, P. D., Murphy, P. C., Warneke, C., Fowler, C.,
Montzka, S. A., Miller, B. R., Miller, L., Oltmans, S. J., Ry-
erson, T. B., Cooper, O. R., Stohl, A., and de Gouw, J. A.:
Ozone variability and halogen oxidation within the Arctic and
sub-Arctic springtime boundary layer, Atmos. Chem. Phys., 10,
10223–10236, doi:10.5194/acp-10-10223-2010, 2010.
Grell, G. A., Peckham, S. E., Schmitz, R., McKeen, S. A., Frost, G.,
Skamarock, W. C., and Eder, B.: Fully coupled “online” chem-
istry within the WRF model, Atmos. Environ., 39, 6957–6975,
doi:10.1016/j.atmosenv.2005.04.027, 2005.
Griffith, S. M., Hansen, R. F., Dusanter, S., Stevens, P. S., Alagh-
mand, M., Bertman, S. B., Carroll, M. A., Erickson, M., Gal-
loway, M., Grossberg, N., Hottle, J., Hou, J., Jobson, B. T.,
Kammrath, A., Keutsch, F. N., Lefer, B. L., Mielke, L. H.,
O’Brien, A., Shepson, P. B., Thurlow, M., Wallace, W., Zhang,
N., and Zhou, X. L.: OH and HO2 radical chemistry during
PROPHET 2008 and CABINEX 2009 – Part 1: Measurements
and model comparison, Atmos. Chem. Phys., 13, 5403–5423,
doi:10.5194/acp-13-5403-2013, 2013.
Hallquist, M., Wenger, J. C., Baltensperger, U., Rudich, Y., Simp-
son, D., Claeys, M., Dommen, J., Donahue, N. M., George,
C., Goldstein, A. H., Hamilton, J. F., Herrmann, H., Hoff-
mann, T., Iinuma, Y., Jang, M., Jenkin, M. E., Jimenez, J. L.,
Kiendler-Scharr, A., Maenhaut, W., McFiggans, G., Mentel, Th.
F., Monod, A., Prévôt, A. S. H., Seinfeld, J. H., Surratt, J. D.,
Szmigielski, R., and Wildt, J.: The formation, properties and im-
pact of secondary organic aerosol: current and emerging issues,
Atmos. Chem. Phys., 9, 5155–5236, doi:10.5194/acp-9-5155-
2009, 2009.
Hayes, P. L., Ortega, A. M., Cubison, M. J., Froyd, K. D., Zhao,
Y., Cliff, S. S., Hu, W. W., Toohey, D. W., Flynn, J. H., Lefer,
B. L., Grossberg, N., Alvarez, S., Rappenglueck, B., Taylor, J.
W., Allan, J. D., Holloway, J. S., Gilman, J. B., Kuster, W. C.,
De Gouw, J. A., Massoli, P., Zhang, X., Liu, J., Weber, R. J.,
Corrigan, A. L., Russell, L. M., Isaacman, G., Worton, D. R.,
Kreisberg, N. M., Goldstein, A. H., Thalman, R., Waxman, E.
M., Volkamer, R., Lin, Y. H., Surratt, J. D., Kleindienst, T. E., Of-
fenberg, J. H., Dusanter, S., Griffith, S., Stevens, P. S., Brioude,
J., Angevine, W. M., and Jimenez, J. L.: Organic aerosol com-
position and sources in Pasadena, California, during the 2010
CalNex campaign, J. Geophys. Res.-Atmos., 118, 9233–9257,
doi:10.1002/jgrd.50530, 2013.
Hayes, P. L., Carlton, A. G., Baker, K. R., Ahmadov, R., Washen-
felder, R. A., Alvarez, S., Rappenglück, B., Gilman, J. B., Kuster,
W. C., de Gouw, J. A., Zotter, P., Prévôt, A. S. H., Szidat, S.,
Kleindienst, T. E., Offenberg, J. H., and Jimenez, J. L.: Model-
ing the formation and aging of secondary organic aerosols in Los
Angeles during CalNex 2010, Atmos. Chem. Phys. Discuss., 14,
32325–32391, doi:10.5194/acpd-14-32325-2014, 2014.
Henderson, B. H., Akhtar, F., Pye, H. O. T., Napelenok, S. L.,
and Hutzell, W. T.: A database and tool for boundary condi-
tions for regional air quality modeling: description and evalua-
tion, Geosci. Model Dev., 7, 339–360, doi:10.5194/gmd-7-339-
2014, 2014.
Henze, D. K. and Seinfeld, J. H.: Global secondary organic aerosol
from isoprene oxidation, Geophys. Res. Lett., 33, L09812,
doi:10.1029/2006gl025976, 2006.
Heo, J., de Foy, B., Olson, M. R., Pakbin, P., Sioutas, C., and
Schauer, J. J.: Impact of regional transport on the anthropogenic
and biogenic secondary organic aerosols in the Los Angeles
Basin, Atmos. Environ., 103, 171–179, 2015.
Hersey, S. P., Craven, J. S., Schilling, K. A., Metcalf, A. R.,
Sorooshian, A., Chan, M. N., Flagan, R. C., and Seinfeld, J. H.:
The Pasadena Aerosol Characterization Observatory (PACO):
chemical and physical analysis of the Western Los Angeles basin
aerosol, Atmos. Chem. Phys., 11, 7417–7443, doi:10.5194/acp-
11-7417-2011, 2011.
Hutzell, W. T., Luecken, D. J., Appel, K. W., and Carter, W. P. L.:
Interpreting predictions from the SAPRC07 mechanism based on
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
K. R. Baker et al.: Gas and aerosol carbon in California 5257
regional and continental simulations, Atmos. Environ., 46, 417–
429, doi:10.1016/j.atmosenv.2011.09.030, 2012.
Jathar, S. H., Gordon, T. D., Hennigan, C. J., Pye, H. O. T.,
Pouliot, G., Adams, P. J., Donahue, N. M., and Robinson, A.
L.: Unspeciated organic emissions from combustion sources and
their influence on the secondary organic aerosol budget in the
United States, P. Natl. Acad. Sci. USA, 111, 10473–10478,
doi:10.1073/pnas.1323740111, 2014.
Jimenez, J. L., Canagaratna, M. R., Donahue, N. M., Prevot, A. S.
H., Zhang, Q., Kroll, J. H., DeCarlo, P. F., Allan, J. D., Coe,
H., Ng, N. L., Aiken, A. C., Docherty, K. S., Ulbrich, I. M.,
Grieshop, A. P., Robinson, A. L., Duplissy, J., Smith, J. D., Wil-
son, K. R., Lanz, V. A., Hueglin, C., Sun, Y. L., Tian, J., Laak-
sonen, A., Raatikainen, T., Rautiainen, J., Vaattovaara, P., Ehn,
M., Kulmala, M., Tomlinson, J. M., Collins, D. R., Cubison, M.
J., Dunlea, E. J., Huffman, J. A., Onasch, T. B., Alfarra, M. R.,
Williams, P. I., Bower, K., Kondo, Y., Schneider, J., Drewnick,
F., Borrmann, S., Weimer, S., Demerjian, K., Salcedo, D., Cot-
trell, L., Griffin, R., Takami, A., Miyoshi, T., Hatakeyama, S.,
Shimono, A., Sun, J. Y., Zhang, Y. M., Dzepina, K., Kimmel,
J. R., Sueper, D., Jayne, J. T., Herndon, S. C., Trimborn, A.
M., Williams, L. R., Wood, E. C., Middlebrook, A. M., Kolb,
C. E., Baltensperger, U., and Worsnop, D. R.: Evolution of Or-
ganic Aerosols in the Atmosphere, Science, 326, 1525–1529,
doi:10.1126/science.1180353, 2009.
Karambelas, A., Pye, H. O. T., Budisulistiorini, S. H., Surratt, J. D.,
and Pinder, R. W.: Isoprene epoxydiol contribution to urban or-
ganic aerosol: evidence from modeling and measurements, En-
viron. Sci. Technol. Lett., 1, 278–283, doi:10.1021/ez5001353,
2013.
Kelly, J. T., Baker, K. R., Nowak, J. B., Murphy, J. G., Markovic,
M. Z., VandenBoer, T. C., Ellis, R. A., Neuman, J. A., Weber, R.
J., and Roberts, J. M.: Fine-scale simulation of ammonium and
nitrate over the South Coast Air Basin and San Joaquin Valley of
California during CalNex-2010, J. Geophys. Res.-Atmos., 119,
3600–3614, doi:10.1002/2013JD021290, 2014.
Kleindienst, T. E., Conver, T. S., McIver, C. D., and Edney, E.
O.: Determination of secondary organic aerosol products from
the photooxidation of toluene and their implications in ambient
PM2.5, J. Atmos. Chem., 47, 79–100, 2004.
Kleindienst, T. E., Jaoui, M., Lewandowski, M., Offenberg, J. H.,
Lewis, C. W., Bhave, P. V., and Edney, E. O.: Estimates of the
contributions of biogenic and anthropogenic hydrocarbons to
secondary organic aerosol at a southeastern US location, Atmos.
Environ., 41, 8288–8300, doi:10.1016/j.atmosenv.2007.06.045,
2007.
Kleindienst, T. E., Lewandowski, M., Offenberg, J. H., Edney, E.
O., Jaoui, M., Zheng, M., Ding, X. A., and Edgerton, E. S.: Con-
tribution of Primary and Secondary Sources to Organic Aerosol
and PM2.5at SEARCH Network Sites, J. Air Waste Ma., 60,
1388–1399, doi:10.3155/1047-3289.60.11.1388, 2010.
Lewandowski, M., Piletic, I. R., Kleindienst, T. E., Offenberg, J. H.,
Beaver, M. R., Jaoui, M., Docherty, K. S., and Edney, E. O.: Sec-
ondary organic aerosol characterisation at field sites across the
United States during the spring–summer period, Int. J. Environ.
An. Ch., 93, 1084–1103, 2013.
Lewis, C. W., Klouda, G. A., and Ellenson, W. D.: Radiocar-
bon measurement of the biogenic contribution to summertime
PM-2.5 ambient aerosol in Nashville, TN, Atmos. Environ., 38,
6053–6061, doi:10.1016/j.atmosenv.2004.06.011, 2004.
Liu, S., Ahlm, L., Day, D. A., Russell, L. M., Zhao, Y., Gentner,
D. R., Weber, R. J., Goldstein, A. H., Jaoui, M., Offenberg, J.
H., Kleindienst, T. E., Rubitschun, C., Surratt, J. D., Sheesley, R.
J., and Scheller, S.: Secondary organic aerosol formation from
fossil fuel sources contribute majority of summertime organic
mass at Bakersfield, J. Geophys. Res.-Atmos., 117, D00V26,
doi:10.1029/2012JD018170, 2012.
Mao, J., Ren, X., Zhang, L., Van Duin, D. M., Cohen, R. C., Park,
J.-H., Goldstein, A. H., Paulot, F., Beaver, M. R., Crounse, J.
D., Wennberg, P. O., DiGangi, J. P., Henry, S. B., Keutsch, F.
N., Park, C., Schade, G. W., Wolfe, G. M., Thornton, J. A., and
Brune, W. H.: Insights into hydroxyl measurements and atmo-
spheric oxidation in a California forest, Atmos. Chem. Phys., 12,
8009–8020, doi:10.5194/acp-12-8009-2012, 2012.
Markovic, M., VandenBoer, T., Baker, K., Kelly, J., and Murphy, J.:
Measurements and modeling of the inorganic chemical compo-
sition of fine particulate matter and associated precursor gases in
California’s San Joaquin Valley during CalNex 2010, J. Geophys.
Research-Atmos., 119, 6853–6866, doi:10.1002/2013JD021408,
2014.
Offenberg, J. H., Lewandowski, M., Edney, E. O., Kleindienst,
T. E., and Jaoui, M.: Influence of Aerosol Acidity on the
Formation of Secondary Organic Aerosol from Biogenic Pre-
cursor Hydrocarbons, Environ. Sci. Technol., 43, 7742–7747,
doi:10.1021/es901538e, 2009.
Ormeño, E., Gentner, D. R., Fares, S., Karlik, J., Park, J. H., and
Goldstein, A. H.: Sesquiterpenoid emissions from agricultural
crops: correlations to monoterpenoid emissions and leaf terpene
content, Environ. Sci. Technol., 44, 3758–3764, 2010.
Pye, H. O. T. and Pouliot, G. A.: Modeling the Role of Alkanes,
Polycyclic Aromatic Hydrocarbons, and Their Oligomers in Sec-
ondary Organic Aerosol Formation, Environ. Sci. Technol., 46,
6041–6047, doi:10.1021/es300409w, 2012.
Pye, H. O. T., Pinder, R. W., Piletic, I. R., Xie, Y., Capps, S. L., Lin,
Y.-H., Surratt, J. D., Zhang, Z., Gold, A., and Luecken, D. J.:
Epoxide pathways improve model predictions of isoprene mark-
ers and reveal key role of acidity in aerosol formation, Environ.
Sci. Technol., 47, 11056–11064, 2013.
Robinson, A. L., Donahue, N. M., Shrivastava, M. K., Weitkamp,
E. A., Sage, A. M., Grieshop, A. P., Lane, T. E., Pierce, J. R., and
Pandis, S. N.: Rethinking organic aerosols: Semivolatile emis-
sions and photochemical aging, Science, 315, 1259–1262, 2007.
Rollins, A., Browne, E., Min, K.-E., Pusede, S., Wooldridge, P.,
Gentner, D., Goldstein, A., Liu, S., Day, D., and Russell, L.: Ev-
idence for NOxcontrol over nighttime SOA formation, Science,
337, 1210–1212, 2012.
Ryerson, T. B., Andrews, A. E., Angevine, W. M., Bates, T. S.,
Brock, C. A., Cairns, B., Cohen, R. C., Cooper, O. R., de Gouw,
J. A., Fehsenfeld, F. C., Ferrare, R. A., Fischer, M. L., Flagan, R.
C., Goldstein, A. H., Hair, J. W., Hardesty, R. M., Hostetler, C.
A., Jimenez, J. L., Langford, A. O., McCauley, E., McKeen, S.
A., Molina, L. T., Nenes, A., Oltmans, S. J., Parrish, D. D., Peder-
son, J. R., Pierce, R. B., Prather, K., Quinn, P. K., Seinfeld, J. H.,
Senff, C. J., Sorooshian, A., Stutz, J., Surratt, J. D., Trainer, M.,
Volkamer, R., Williams, E. J., and Wofsy, S. C.: The 2010 Cali-
fornia Research at the Nexus of Air Quality and Climate Change
www.atmos-chem-phys.net/15/5243/2015/ Atmos. Chem. Phys., 15, 5243–5258, 2015
5258 K. R. Baker et al.: Gas and aerosol carbon in California
(CalNex) field study, J. Geophys. Res.-Atmos., 118, 5830–5866,
doi:10.1002/jgrd.50331, 2013.
Sarwar, G., Fahey, K., Kwok, R., Gilliam, R. C., Roselle, S. J.,
Mathur, R., Xue, J., Yu, J., and Carter, W. P. L.: Potential im-
pacts of two SO2oxidation pathways on regional sulfate concen-
trations: Aqueous-phase oxidation by NO2and gas-phase oxida-
tion by Stabilized Criegee Intermediates, Atmos. Environ., 68,
186–197, doi:10.1016/j.atmosenv.2012.11.036, 2013.
Shilling, J. E., Zaveri, R. A., Fast, J. D., Kleinman, L., Alexander,
M. L., Canagaratna, M. R., Fortner, E., Hubbe, J. M., Jayne, J.
T., Sedlacek, A., Setyan, A., Springston, S., Worsnop, D. R.,
and Zhang, Q.: Enhanced SOA formation from mixed anthro-
pogenic and biogenic emissions during the CARES campaign,
Atmos. Chem. Phys., 13, 2091–2113, doi:10.5194/acp-13-2091-
2013, 2013.
Simon, H. and Bhave, P. V.: Simulating the Degree of Oxidation in
Atmospheric Organic Particles, Environ. Sci. Technol., 46, 331–
339, doi:10.1021/es202361w, 2012.
Skamarock, W. C., Klemp, J. B., Dudhia, J., Gill, D. O., Barker,
D. M., Duda, M. G., Huang, X., Wang, W., and Powers, J. G.:
A description of the Advanced Reserch WRF version 3., NCAR
Technical Note NCAR/TN-475+STR, 2008.
Steiner, A. L., Cohen, R. C., Harley, R. A., Tonse, S., Millet, D. B.,
Schade, G. W., and Goldstein, A. H.: VOC reactivity in central
California: comparing an air quality model to ground-based mea-
surements, Atmos. Chem. Phys., 8, 351–368, doi:10.5194/acp-8-
351-2008, 2008.
Stroud, C. A., Liggio, J., Zhang, J., Gordon, M., Staebler, R. M.,
Makar, P. A., Zhang, J., Li, S. M., Mihele, C., and Lu, G.: Rapid
organic aerosol formation downwind of a highway: Measured
and model results from the FEVER study, J. Geophys. Res.-
Atmos., 119, 1663–1679, 2014.
Stuiver, M.: International agreements and the use of the new oxalic
acid standard, Radiocarbon, 25, 793–795, 1983.
Tan, Y., Carlton, A. G., Seitzinger, S. P., and Turpin, B. J.: SOA
from methylglyoxal in clouds and wet aerosols: measurement
and prediction of key products, Atmos. Environ., 44, 5218–5226,
doi:10.1016/j.atmosenv.2010.08.045, 2010.
US Environmental Protection Agency: 2011 National Emissions
Inventory, version 1 Technical Support Document, available
at: http://www.epa.gov/ttn/chief/net/2011nei/2011_nei_tsdv1_
draft2_june2014.pdf (last access: 8 May 2015), edited by:
Agency, U. E. P., 2014.
Wagstrom, K. M., Baker, K. R., Leinbach, A. E., and Hunt, S. W.:
Synthesizing Scientific Progress: Outcomes from US EPA’s Car-
bonaceous Aerosols and Source Apportionment STAR Grants,
Environ. Sci. Technol., 48, 10561–10570, 2014.
Washenfelder, R. A., Young, C. J., Brown, S. S., Angevine, W.
M., Atlas, E. L., Blake, D. R., Bon, D. M., Cubison, M. J.,
de Gouw, J. A., Dusanter, S., Flynn, J., Gilman, J. B., Graus,
M., Griffith, S., Grossberg, N., Hayes, P. L., Jimenez, J. L.,
Kuster, W. C., Lefer, B. L., Pollack, I. B., Ryerson, T. B., Stark,
H., Stevens, P. S., and Trainer, M. K.: The glyoxal budget and
its contribution to organic aerosol for Los Angeles, California,
during CalNex 2010, J. Geophys. Res.-Atmos., 116, D00V02,
doi:10.1029/2011jd016314, 2011.
Yarwood, G., Rao, S., Yocke, M., and Whitten, G. Z.: Updates to
the carbon bond chemical mechanism: CB05, ENVIRON Inter-
national Corporation, Novato, CA, 2005.
Zhang, H., Chen, G., Hu, J., Chen, S.-H., Wiedinmyer, C., Klee-
man, M., and Ying, Q.: Evaluation of a seven-year air qual-
ity simulation using the Weather Research and Forecasting
(WRF)/Community Multiscale Air Quality (CMAQ) models in
the eastern United States, Sci. Total Environ., 473, 275–285,
2014a.
Zhang, X., Cappa, C. D., Jathar, S. H., McVay, R. C., Ensberg, J. J.,
Kleeman, M. J., and Seinfeld, J. H.: Influence of vapor wall loss
in laboratory chambers on yields of secondary organic aerosol, P.
Natl. Acad. Sci. USA, 111, 5802–5807, 2014b.
Zhao, Y., Kreisberg, N. M., Worton, D. R., Isaacman, G., Gentner,
D. R., Chan, A. W., Weber, R. J., Liu, S., Day, D. A., and Rus-
sell, L. M.: Sources of organic aerosol investigated using organic
compounds as tracers measured during CalNex in Bakersfield, J.
Geophys. Res.-Atmos., 118, 11388–11398, 2013.
Zhao, Y., Hennigan, C. J., May, A. A., Tkacik, D. S., de Gouw, J. A.,
Gilman, J. B., Kuster, W. C., Borbon, A., and Robinson, A. L.:
Intermediate-Volatility Organic Compounds: A Large Source of
Secondary Organic Aerosol, Environ. Sci. Technol., 48, 13743–
13750, 2014.
Zotter, P., El-Haddad, I., Zhang, Y., Hayes, P. L., Zhang, X., Lin,
Y. H., Wacker, L., Schnelle-Kreis, J., Abbaszade, G., and Zim-
mermann, R.: Diurnal cycle of fossil and nonfossil carbon using
radiocarbon analyses during CalNex, J. Geophys. Res.-Atmos.,
119, 6818–6835, doi:10.1002/2013JD021114, 2014.
Atmos. Chem. Phys., 15, 5243–5258, 2015 www.atmos-chem-phys.net/15/5243/2015/
