The effect of marine isoprene emissions on secondary organic aerosol and ozone formation in the coastal United States

Atmospheric Environment 44 (2010) 115–121

Contents lists available at ScienceDirect

Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv

The effect of marine isoprene emissions on secondary organic aerosol and ozone formation in the coastal United States Brett Gantt*, Nicholas Meskhidze, Yang Zhang, Jun Xu Department of Marine, Earth, and Atmospheric Sciences, North Carolina State University, Raleigh, NC, USA

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 May 2009 Received in revised form 7 August 2009 Accepted 18 August 2009

The impact of marine isoprene emissions on summertime surface concentrations of isoprene, secondary organic aerosols (SOA), and ozone (O3) in the coastal areas of the continental United States is studied using the U.S. Environmental Protection Agency regional-scale Community Multiscale Air Quality (CMAQ) modeling system. Marine isoprene emission rates are based on the following five parameters: laboratory measurements of isoprene production from phytoplankton under a range of light conditions, remotelysensed chlorophyll-a concentration ([Chl–a]), incoming solar radiation, surface wind speed, and sea-water optical properties. Model simulations show that marine isoprene emissions are sensitive to meteorology and ocean ecosystem productivity, with the highest rates simulated over the Gulf of Mexico. Simulated offshore surface layer marine isoprene concentration is less than 10 ppt and significantly dwarfed by terrestrial emissions over the continental United States. With the isoprene reactions included in this study, the average contribution of marine isoprene to SOA and O3 concentrations is predicted to be small, up to 0.004 mg m3 for SOA and 0.2 ppb for O3 in coastal urban areas. The light-sensitivity of isoprene production from phytoplankton results in a midday maximum for marine isoprene emissions and a corresponding daytime increase in isoprene and O3 concentrations in coastal locations. The potential impact of the daily variability in [Chl-a] on O3 and SOA concentrations is simulated in a sensitivity study with [Chl-a] increased and decreased by a factor of five. Our results indicate that marine emissions of isoprene cause minor changes to coastal SOA and O3 concentrations. Comparison of model simulations with few available measurements shows that the model underestimates marine boundary layer isoprene concentration. This underestimation is likely due to the limitations in current treatment of marine isoprene emission and a coarse spatial resolution used in the model simulations. Ó 2009 Elsevier Ltd. All rights reserved.

Keywords: Marine isoprene Secondary organic aerosol Ozone CMAQ

1. Introduction It is has been well established that photosynthetic organisms can emit gases that affect the atmospheric chemistry of the area. These gases, collectively known as biogenic volatile organic compounds (BVOCs), play a role the formation of ozone (O3) (Trainer et al., 1987; Chameides et al., 1988) and help extend the lifetime of important atmospheric gases such as methane and carbon monoxide (Poisson et al., 2000). Isoprene (C5H8) is the most ubiquitous BVOC with annual global emissions estimated at 500–750 Tg of carbon (Guenther et al., 2006). While forests typically have the highest isoprene emission rates, it has been shown that productive areas of remote ocean, coastal upwelling regions, and wetlands (Bonsang et al., 1992; Broadgate et al., 1997; Holst et al., 2008) can all emit isoprene (hereafter referred to as marine isoprene) at rates that can

* Corresponding author. E-mail address: [email protected] (B. Gantt). 1352-2310/$ – see front matter Ó 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.atmosenv.2009.08.027

considerably influence atmospheric chemistry in remote marine and coastal regions (Liakakou et al., 2007). In addition to its importance in atmospheric chemistry, isoprene can also affect air quality due to the formation of secondary organic aerosols (SOA) during oxidation (Claeys et al., 2004; Edney et al., 2005). Compared to other BVOCs, isoprene has a relatively low SOA yield (defined as the ratio of the mass of SOA formed to the mass of isoprene reacted); yields up to 3% have been measured under low NOx (nitric oxide þ nitrogen dioxide) conditions (Kroll et al., 2006). However, higher SOA yields from isoprene are possible during nighttime reaction with nitrate (NO3) (Ng et al., 2008), and aqueousphase processing (Carlton et al., 2006; Ervens et al., 2008). Modeling studies estimate that isoprene photo-oxidation contributes about 50% of SOA in the United States (U.S.) (Liao et al., 2007) and 47–58% globally (Henze and Seinfeld, 2006; Liao et al., 2007). In the marine boundary layer, water soluble organic carbon (WSOC), which are mostly secondary in origin (Ceburnis et al., 2008), make up 18% of the total fine mode aerosol (0.06–0.5 mm in diameter) mass during periods of high ocean productivity (O’Dowd et al., 2004).

116

B. Gantt et al. / Atmospheric Environment 44 (2010) 115–121

With only sporadic measurements, emission inventories of marine isoprene and its contribution to the global burdens of SOA and O3 remain poorly defined. Previous global estimates of marine isoprene emissions have ranged widely, with values from 0.11 to 1.9 Tg yr1 (Palmer and Shaw, 2005; Arnold et al., 2009). Recently, Gantt et al. (2009) developed an advanced, physically-based global marine isoprene emission inventory that uses dynamic mixed layer depth and modeled solar radiation received by phytoplankton at different heights of the water column. The high temporal resolution and ability to consider different environmental parameters makes this emission inventory ideal for the incorporation into 3-D air quality models. In this study, we apply the Gantt et al. (2009) methodology to a regional domain encompassing the continental United States (CONUS). The effect of marine emissions and the subsequent contributions on the surface SOA and O3 concentrations are also examined at several large cities on the coast of the U. S. 2. Method 2.1. CMAQ model description To determine the contribution of marine isoprene emissions to SOA and O3 formation, we have conducted a month-long simulation of the U.S. Environmental Protection Agency (US EPA) Models-3/ Community Multiscale Air Quality (CMAQ) model (Version 4.4) using a 36  36 km2 spatial resolution, 1 h temporal resolution, and a domain comprising the CONUS and parts of southern Canada and northern Mexico. The model simulations are conducted using meteorological data generated offline by the Pennsylvania State University/National Center for Atmospheric Research Mesoscale Modeling System Generation 5 (MM5) Version 3.6.1 with fourdimensional data assimilation. The model runs start on July 1st, 2001 using the June 30th output from the one year CMAQ simulation of Zhang et al. (2007) as the initial condition. The boundary conditions are set every 3 h from a global chemical transport model, the Goddard Earth Observing System (GEOS)-Chem (Park et al., 2004). Emissions of anthropogenic gaseous and aerosol species are based on the 2001 National Emissions Inventory (NEI), while biogenic emissions are based on the Biogenic Emissions Inventory System (BEIS) version 3.12. As described in Zhang et al. (2007), isoprene chemistry is modeled in CMAQ by using the Carbon-Bond Mechanism version IV (CBM-IV) reactions, with SOA formation from isoprene determined by a revision of the default SOA module (Schell et al., 2001). A detailed description of model mechanisms and configurations can be found in Zhang et al. (2007). The modified SOA module of Zhang et al. (2007) uses parameterizations based on laboratory measurements of SOA formation from the oxidation of isoprene by OH (Kroll et al., 2006), but does not account for SOA formation from isoprene oxidation by nitrate, aqueous-phase processing, oligomerization, and acid catalyzation, all of which could lead to higher SOA yields (Kalberer et al., 2004; Carlton et al., 2006; Surratt et al., 2007; Ng et al., 2008; Ervens et al., 2008). However, recent laboratory measurements (Offenberg et al., 2006) also indicate that the enthalpy of vaporization for SOA from isoprene could be lower than that used in baseline simulations of Zhang et al. (2007), possibly leading to an overestimation of SOA in this work. 2.2. Marine isoprene emissions inventory An emissions inventory for marine isoprene was created based on laboratory measurements of isoprene production by several phytoplankton species under a range of light conditions. A detailed discussion of the data and the methods used for deriving the physically-based parameterization for the emission of marine isoprene can be found in Gantt et al. (2009). The isoprene production rate is

found using the following equation: P ¼ 0:042*ln ðIPAR Þ2 , where P is the isoprene production rate for diatoms (mmole isoprene (gram chlorophyll-a)1 h1) and IPAR is the ambient photosynthetically active radiation (PAR) (mE m2 s1). This rate was chosen in this study because diatoms are one of the dominant summer phytoplankton classes for both the Atlantic and Pacific coasts of the U.S. (Marshall, 1978; Chavez et al., 1991), despite changes in species dominance with latitude during the summer months. To calculate total isoprene production from a given model grid cell, production rates (derived per unit mass of chlorophyll-a) are multiplied by the mass of chlorophyll-a in each grid cell. This mass is estimated as the product of remotely-sensed chlorophyll-a concentration ([Chl-a]) and water volume of the euphotic zone. Monthlyaveraged, Level 3 surface [Chl-a] was obtained from the Sea-viewing Wide Field-of-view Sensor (SeaWiFS) (O’Reilly et al., 1998) for July 2001, regridded to a 36  36 km2, and geographically projected into the Lambert Conformal Conic coordinates of the CMAQ model domain (Supplemental Fig. S1). For most of the modeling region, statistical analyses of the SeaWiFS daily data for July, 2001 showed little daily variation in the [Chl-a]. However, the isolated areas in the Gulf of Mexico and Pacific coast showed 95% confidence limits that were about factor of five higher/lower compared to the monthly mean. To capture such fluctuations in coastal ocean productivity in Section 4 sensitivity studies are conducted with five times higher/ lower monthly-averaged [Chl-a]. The amount of solar radiation received by phytoplankton at a particular depth in the water column is calculated by applying the simulated surface solar radiation Io output from MM5 to Beer–Lambert’s Law, I ¼ Io ek490 h , to account for light attenuation at water depth. Here I is ambient solar radiation at a specific depth, k490 is the diffuse attenuation coefficient at 490 nm wavelength, and h is the depth in the water. The simulated solar radiation from MM5, originally in units of W m2, is converted to PAR using the approximate conversion 1 W m2 z 2 mE m2 s1 (Jacovides et al., 2004). Level 3, monthly-averaged values of k490 for July 2001 were downloaded from SeaWiFS (O’Reilly et al., 1998) and regridded to the same 36  36 km2, Lambert Conformal Conicprojected grids used for [Chl-a]. The total column isoprene emission is found by integrating the production equation for all depths in the euphotic zone. The maximum depth of the euphotic zone Hmax is set to the depth at which the ambient light intensity is 5 mE m2 s1 which is the lower limit of light needed for phytoplankton photosynthesis (Shaw et al., 2003). The fraction Fiso of the isoprene produced in the water column that is lost via sea-air gas exchange is calculated by the method used in Palmer and Shaw (2005), which includes chemical, biological, and mixed layer losses in addition to the sea-air gas exchange. The isoprene emission rate (Eiso) in mole s1 is given by the following equation:

Eiso ¼ SA*CF*Hmax *½Chl  a*Fiso *

H Zmax

p dh

(1)

0

where SA is the surface area of each grid cell (1.296  109 m2) and CF is the conversion factor between mmole hr1 and mole s1 ((3.6  109)1). Three separate groups of monthly (for July 2001) simulations are carried out using different isoprene emission schemes: 1) terrestrial emissions only, 2) terrestrial and marine emissions, and 3) terrestrial and marine emissions, with marine emissions increased and decreased five-fold over the calculated values. 3. Results Four model variables of particular interest to coastal atmospheric chemistry examined here are the rates of marine isoprene emissions, and the changes in surface concentrations of isoprene,

B. Gantt et al. / Atmospheric Environment 44 (2010) 115–121

117

Fig. 1. The modeling domain used in this study. Selected grids for three coastal cities denote the regions where marine isoprene emissions (filled with blue colors) and concentrations (highlighted in green) are examined. Lake Pontchartrain near New Orleans is also included as a potential source of marine isoprene.

SOA and O3 as a result of marine isoprene. For each variable, both domain-wide analysis and diurnal variations are explored. For the domain-wide analysis, we examine the spatial distribution of the model variables to determine the extent of areas impacted by marine isoprene and identify the locations with the maximum impact. Diurnal variations of marine isoprene emissions and surface concentrations are explored in three oceanic regions just offshore of major coastal cities: New York, NY (NY), New Orleans, LA (NOL), and Los Angeles, CA (LA). The resulting changes in SOA and O3 concentrations for these cities are determined by selecting model grid cells containing the majority of metropolitan area. Fig. 1 shows the model domain with the marine grid cells (in blue) selected for the analysis of marine isoprene and the urban grid cells (in green) selected for the analysis of SOA and O3 concentration changes associated with marine isoprene. These metropolitan areas are currently in (LA and NY) or near (NOL) nonattainment for the U.S. National Ambient Air Quality Standard (NAAQS) for 8-h O3 and two (LA and NY) are in nonattainment for PM2.5 (US EPA, 2004; US EPA, 2005; http://www.epa.gov/oar/oaqps/greenbk/mappm25o3. html). As these areas have lower terrestrial biogenic emissions and higher oxidant concentrations they could experience a large regulatory and environmental impact from marine isoprene. Despite being in attainment for O3 and PM2.5, we focus on the NOL area because the eutrophication of the Louisiana coast as a result of discharge from the Mississippi River that creates an area of unnaturally high [Chl-a].

3.1. Marine isoprene emissions Fig. 2a shows the simulated monthly average of the daily maximum marine isoprene emission rates for the model domain, with values ranging from <1  108 molecules cm2 s1 in much of the remote ocean to > 4  109 molecules cm2 s1 in near-coastal areas. The region with the highest emission rates is the Northern Gulf of Mexico, an area characterized by high [Chl-a] and strong solar radiation. Daily maximum (i.e., midday) emission rates were chosen for the analysis due to the importance to the photochemistry for SOA and O3 formation. Fig. 2b shows that the light-sensitivity of isoprene production results in an emission pattern that

follows the variation in incoming solar radiation, with the midday emission rates in all three areas over 7.5  108 molecules cm2 s1 and the highest rates located near NOL due to the high [Chl-a] (17.0 mg m3) as compared to NY (3.4 mg m3) and LA (0.5 mg m3). This figure also reveals large ranges (denoted by the error bars) in hourly emissions of marine isoprene. Such significant variation in the emissions rates (for a given [Chl-a]) is primarily controlled by the variation in the incoming solar radiation and to lesser extent by the changes in surface wind speed. In general, comparison of marine isoprene emission rates reported in Fig. 2b with those measured (see Table 1) over the surface waters with comparable [Chl-a] shows that for the three study regions our model calculated rates are higher than previously reported values. However, our rates are somewhat lower than model-predicted values of 6  109 molecules cm2 s1 at the Mediterranean Sea (Liakakou et al., 2007).

3.2. Coastal isoprene concentrations due to marine isoprene emissions The contribution of marine isoprene emissions to the marine boundary layer (MBL) concentration of isoprene were determined by subtracting the isoprene concentrations of the simulations with only terrestrial emissions from the simulations with terrestrial and marine sources combined. Fig. 2c shows the monthly mean of the daily-average marine isoprene concentrations in the lowest CMAQ layer (surface to 35 m) for the CONUS domain. The highest concentrations in the U.S. are located in the Pacific Northwest (NW) because of the high emissions (Fig. 2a) and low oxidant concentrations characteristic of clean marine air. Due to the short lifetime of isoprene (w0.6–2 h) (Atkinson and Arey, 2003; Liakakou et al., 2007), the spatial distribution of isoprene concentrations is similar to that of emissions. The diurnal profiles of marine isoprene concentrations in Fig. 2d show midday concentrations averaging 10 parts per trillion (ppt) for the NY and NOL coasts, and 2.5 ppt for the LA coast. Because marine isoprene emissions are not included in the default model simulations, the percentage changes in the MBL isoprene concentrations is fairly high with values up to 100%, 25%, and 7% for the LA, NY, and NOL coasts, respectively. Overall,

118

B. Gantt et al. / Atmospheric Environment 44 (2010) 115–121

Fig. 2. Model predicted domain-wide monthly average marine isoprene (a) emissions and (c) concentrations in the MBL. Monthly average hourly values for (b) emissions and (d) concentrations are shown for New York, New Orleans, and Los Angeles coasts for July, 2001. Error bars on Fig. 2b and 2d show the highest and the lowest simulated values during the monthly time period.

comparison of model-predicted surface layer marine isoprene concentrations with the observations (see Table 1) shows, that model simulated values are toward the low end of the reported MBL isoprene concentrations. The possible reasons for the lower isoprene concentrations predicted by the model include the uniform mixing of isoprene throughout the 35 m of the surface layer, analysis of coastal urban areas with high summertime oxidant levels, and the averaging of isoprene concentrations over the entire 36  36 km2 grid. Note that the diurnal variation in the model-predicted surface marine isoprene concentration, shown in Fig. 2d, is different than that of isoprene emissions. This figure shows the presence of two maxima before and after the emission maximum (13–14 local time). The mid-afternoon drop or plateau in

marine isoprene concentrations is likely due to the increased rate of photochemical oxidation of isoprene. This plateau/drop of daytime isoprene concentrations was not present in remote marine areas of the Pacific Ocean, in agreement with observational studies of remote marine air masses (Lewis et al., 1997, 2001). The difference in the diurnal cycle of isoprene concentration between remote and urban marine areas is likely the result of higher oxidation rates in urban air masses. 3.3. Coastal SOA concentrations due to marine isoprene emissions The monthly average changes in SOA surface concentration from the inclusion of marine isoprene displayed in Fig. 3a shows

Table 1 In-situ measurements of average coastal marine isoprene emissions and MBL concentrations. Location

Date

Emission rate (108 molec. cm2 s1)a

MBL conc. (pptv)a

Ref.

South Pacific Florida Coast North Sea Ireland Coast Ireland Coast Southern Ocean North Atlantic Southern Ocean Ireland Coast Ireland Coast Norway Coast Mediterranean Southern Ocean

5, 1987 9, 1993 5, 1994 7–8, 1996 4–5, 1997 12, 1997–3, 1998 5, 1997 1–2, 1999 9–10, 1998 9–10, 2003 6, 2005 2004 1–2, 2007

1.1 0.34 (0.06–0.71) 0.17 (0.01–0.47) NA NA NA 0.38 (0.08–0.6) NA 2.3 (0–6.7) 5.9 1.1 (0–8.8) (1–60)b NA

12 (<2–36) <5 NA 6.2 (0–24.3) 2.7 (0–37.1) (<0.1–250)b NA 5.7 (1.8–7.9) NA 63 (7–210) 180 60 (2–~300)b 99 (60–138)b

Bonsang et al. (1992) Milne et al. (1995) Broadgate et al. (1997) Lewis et al. (1997) Lewis et al. (1999) Yokouchi et al. (1999) Baker et al. (2000) Lewis et al. (2001) Broadgate et al. (2004) Greenberg et al. (2005) Sinha et al. (2007) Liakakou et al. (2007) Yassaa et al. (2008)

a b

Maximum and minimum measurements in parenthesis. Data is reported for MBL isoprene concentration least influenced by the terrestrial sources.

B. Gantt et al. / Atmospheric Environment 44 (2010) 115–121

119

Fig. 3. Model predicted domain-wide monthly average marine isoprene-derived (a) SOA and (c) 8-h average O3 concentrations in the MBL. Monthly average hourly values for (b) SOA and (d) O3 are shown for New York, New Orleans, and Los Angeles for July, 2001. Error bars on Fig. 3b and 3d show the highest and lowest simulated values during the monthly time period.

the greatest increases in concentration for the Pacific NW with values over 0.005 mg m3. These values are very small compared to monthly average model-predicted SOA concentrations due to terrestrial sources, that range from w0 mg m3 over the ocean to as high as 5 mg m3 in the western U.S. Additionally, an increase in PM2.5 concentrations on the order of 2–3 ng m3 is insignificant compared to the regulatory standard of 15 mg m3. Fig. 3a shows that coastal areas generally experience the greatest contribution of marine isoprene-derived SOA, though locations hundreds of kilometers inland can potentially be affected. The longer lifetime of SOA (weeks), as compared to the relatively-short lifetime of isoprene (hours), results in the enlarged footprint of regions affected by marine isoprene-derived SOA. According to Fig. 3a some areas in the Pacific NW, Eastern Canada, and Baja California experience minor decreases in SOA concentration from marine isoprene. We hypothesize that marine isoprene is acting as a sink for the oxidants that otherwise will be used for the SOA formation from other BVOC with higher SOA yields. Fig. 3b shows that the average daily maximum concentrations are 0.002 mg m3 in LA and 0.001 mg m3 in NY and NOL. Despite having higher marine isoprene concentration, NOL has relatively low SOA concentrations from marine isoprene; the high air temperatures in the Gulf of Mexico keep more of the isoprene oxidation products in the gas phase (Pankow, 1994; Odum et al., 1996) preventing them from forming SOA. The average percentage change in SOA concentration when marine isoprene emissions are included is 0.5% for LA, 0.35% for NOL, and 0.15% for NY. The diurnal pattern of SOA variation due to marine isoprene is different for each city: LA experiencing a nighttime maximum, NY experiencing higher midday levels, and the greatest increase of SOA occurs in NOL during the evening hours.

These different responses are likely due to the micrometeorology of the three regions, as the longer lifetime of SOA causes its surface concentrations to be sensitive to variability in the PBL height. 3.4. Coastal O3 concentrations due to marine isoprene emissions Complex relationships between NOx and volatile organic compounds (VOC) for surface O3 formation (Haagen-Smit, 1952) create the heterogeneous spatial distribution of 8-h surface O3 concentration changes attributed to marine isoprene. Fig. 3c shows that O3 enhancements occur in near every major urban coastal area, with values up to 0.2 ppb. Like SOA concentrations, the contribution of marine isoprene to O3 levels is quite small (<0.2%) compared to the ambient 8-h average NAAQS regulatory standard of 75 ppb. It is also important to note that these increases do not necessarily coincide with the peak O3 concentrations, as marine air masses change the chemical composition of the atmosphere. Because urban regions have greater concentrations of NOx and are often VOC-limited, the additional isoprene participates in the NOx/hydrocarbon chemistry of the area and leads to increased O3 formation (Chameides et al., 1988; Zhang et al., in press). In remote coastal areas, the additional isoprene makes little difference in O3 formation because these areas are likely to be NOx-limited (Trainer et al., 1987; Zhang et al., in press). The areas with very low NOx concentrations, like those in the Pacific Ocean off the California coast, experience a slight decrease in average O3 concentration (see Supplemental Fig. S2) due to O3 destruction by additional isoprene (Ridley et al., 1989). If the decrease in oxidant levels characteristic to this region is representative of the remote Pacific Ocean, marine isoprene

120

B. Gantt et al. / Atmospheric Environment 44 (2010) 115–121

emission may lead to the longer lifetime of pollutants (Poisson et al., 2000) and possible enhancement of trans-Pacific pollution transport. Changes in the monthly-average surface O3 concentration in the three selected regions shown in Fig. 3d are similar, with each city experiencing a maximum increase of >0.06 ppb during the afternoon hours. The average percentage change in O3 concentration when marine isoprene emissions are included is 0.1% for NOL, 0.03% for NY, and 0.01% for LA. 4. Sensitivity analysis To account for the daily variability in [Chl-a] and the associated uncertainties in phytoplankton isoprene production, a series of simulations are conducted with marine isoprene emissions increased and decreased by a factor of 5. This range was chosen based on statistical analysis of daily remotely-sensed [Chl-a], revealing 95% confidence limits to be about factor of five higher/lower compared to the monthly mean. All other emission parameters remained unchanged during sensitivity simulations. Because the effect of marine isoprene on SOA and O3 using default emissions were small, the impacts using the decreased emissions were even more insignificant (<0.1%). The sensitivity test with elevated emissions showed maximum changes in the concentrations of SOA and O3 to be 0.03 mg m3 and 4 ppb respectively. These values correspond to a percentage change of w3% for SOA and w4% for O3 in coastal urban areas, confirming the small effect of marine isoprene on coastal air quality. 5. Conclusion In this study, a physically-based marine isoprene emission inventory is applied to the coastal areas of the CONUS to quantify its effects on coastal concentrations of isoprene, SOA, and O3. Laboratory measurements of phytoplankton isoprene production under a range of light conditions are combined with remotely-sensed data and modeled meteorological variables to create marine isoprene emissions comparable in magnitude and diurnal cycle to in-situ observations. In three of the major U.S. coastal cities, NY, NOL and LA, the inclusion of marine isoprene causes minor increases in concentrations of SOA and O3, with changes up to 0.004 mg m3 (<0.5%), and 0.07 ppb (<0.2%), respectively. Nationally, the simulated changes in SOA are greatest over the Pacific NW, while O3 concentrations are affected the most in urban coastal cities such as NY and LA. Despite predicted minor role on coastal air quality, the importance of marine isoprene on a local scale cannot be discarded. It is likely that the smoothing of coastal emissions over the 36  36 km2 grid results in the loss of locally high emission sources like estuaries. While the average coastal [Chl-a] is w1–5 mg m3, estuaries have been shown to have very high [Chl-a] up to 100 mg m3 (Buzzelli et al., 2003). In addition to isoprene, there are several other sources of secondary aerosols with a marine origin not simulated in this study, including dimethyl sulfide (Shaw, 1983), iodine compounds (O’Dowd et al., 2002) and monoterpenes (Yassaa et al., 2008), and amines (Facchini et al., 2008). Marine primary organic aerosols, the emissions of which are related to phytoplankton abundance, have also been shown to comprise a substantial portion of the fine mode aerosol mass in the MBL (O’Dowd et al., 2004). In order to better constrain the model emission rates, concomitant in-situ observational studies for marine fluxes of BVOC, and phytoplankton abundances and speciation are needed. This study highlights the great need for future modeling studies conducted at a finer spatial resolution, with more complete representation of SOA formation from marine BVOC, and the inclusion of primary sources of marine organic aerosols to better quantify the effect of ocean trace gas and aerosol emissions on coastal air quality.

Acknowledgements This research was supported by the Office of Science (BER), U.S. Department of Energy, Grant No. DE-FG02-08ER64508. The baseline CMAQ simulation without marine isoprene emissions was conducted with support from NASA Award No. NNG04GJ90G and NSF Career Award No. Atm-0348819. Thanks are due to Warren Peters, George Pouliot, Kenneth L. Schere, and Tom Pierce, the U.S. EPA, for providing meteorological fields, emission inventories, initial and boundary conditions for the July 2001 baseline simulations. We also would like to thank the anonymous reviewers for their helpful suggestions. Appendix. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.atmosenv.2009.08.027. References Arnold, S.R., Spracklen, D.V., Williams, J., Yassaa, N., Sciare, J., Bonsang, B., Gros, V., Peeken, I., Lewis, A.C., Alvain, S., Moulin, C., 2009. Evaluation of the global oceanic isoprene source and its impacts on marine organic carbon aerosol. Atmos. Chem. Phys. 9, 1253–1262. Atkinson, R., Arey, J., 2003. Atmospheric degradation of volatile organic compounds. Chemical Reviews 103 (12), 4605–4638. Baker, A.R., Turner, S.M., Broadgate, W.J., Thompson, A., McFiggans, G.B., Vesperini, O., Nightingale, P.D., Liss, P.S., Jickells, T.D., 2000. Distribution and sea-air fluxes of biogenic trace gases in the eastern Atlantic Ocean. Global Biogeochem. Cy. 14, 871–886. Bonsang, B., Polle, C., Lambert, G., 1992. Evidence for marine production of isoprene. Geophys. Res. Lett. 19, 1129–1132. Broadgate, W.J., Liss, P.S., Penkett, S.A., 1997. Seasonal emissions of isoprene and other reactive hydrocarbon gases from the ocean. Geophys. Res. Lett. 24, 2675–2678. Broadgate, W.J., Malin, G., Kupper, F.C., Thompson, A., Liss, P.S., 2004. Isoprene and other non-methane hydrocarbons from seaweeds: a source of reactive hydrocarbons to the atmosphere. Marine Chem. 88, 61–73. Buzzelli, C.P., Ramus, J.A., Paerl, H.W., 2003. Ferry-based monitoring of surface water quality in North Carolina estuaries. Estuaries 26 (4A), 975–984. Carlton, A.G., Turpin, B.J., Lim, H.J., Altieri, K.E., Seitzinger, S., 2006. Link between isoprene and secondary organic aerosol (SOA): pyruvic acid oxidation yields low volatility organic acids in clouds. Geophys. Res. Lett. 33 (6), L06822. Ceburnis, D., O’Dowd, C.D., Jennings, G.S., Facchini, M.C., Emblico, L., Decesari, S., Fuzzi, S., Sakalys, J., 2008. Marine aerosol chemistry gradients: elucidating primary and secondary processes and fluxes. Geophys. Res. Lett. 35, L07804. doi:10.1029/2008GL033462. Chameides, W.L., Lindsay, R.W., Richardson, J., Kiang, C.S., 1988. The role of biogenic hydrocarbons in urban photochemical smog: Atlanta as a case study. Science 241, 1473–1475. Chavez, F.P., Barber, R.T., Kosro, P.M., Huyer, A., Ramp, S.R., Stanton, T.P., Rojas de Mendiola, B.,1991. Horizontal transport and the distribution of nutrients in the coastal transition zone off Northern California: effects on primary production, phytoplankton biomass and species composition. J. Geophys. Res. 96 (C8), 14833–14848. Claeys, M., Graham, B., Vas, G., Wang, W., Vermeylen, R., Pashynska, V., Cafmeyer, J., Guyon, P., Andreae, M.O., Artaxo, P., Maenhaut, W., 2004. Formation of secondary organic aerosols through photooxidation of isoprene. Science 303, 1173–1176. Edney, E.O., Kleindienst, T.E., Jaoui, M., Lewandowski, M., Offenberg, J.H., Wang, W., Claeys, M., 2005. Formation of 2-methyl tetrols and 2-methylglyceric acid in secondary organic aerosol from laboratory irradiated isoprene/NOx/SO2/air mixtures and their detection in ambient PM2.5 samples collected in the eastern United States. Atmos. Environ. 39, 5281–5289. Ervens, B., Carlton, A.G., Turpin, B.J., Altieri, K.E., Kreidenweis, S.M., Feingold, G., 2008. Secondary organic aerosol yield from cloud-processing of isoprene oxidation products. Geophys. Res. Lett. 35, L02816. doi:10.1029/2007GL031828. Facchini, M.C., Decesari, S., Rinaldi, M., Carbone, C., Finessi, E., Mircea, M., Fuzzi, S., Moretti, F., Tagliavini, E., Ceburnis, D., O’Dowd, C., 2008. Important source of marine secondary organic aerosol from biogenic amines. Environ. Sci. Technol. 42 (24), 9116–9121. doi:10.1021/es8018385. Gantt, B., Meskhidze, N., Kamykowki, D., 2009. A new physically-based quantification of isoprene and primary organic aerosol emissions from the world’s oceans. Atmos. Chem. Phys. 9, 4915–4927. Greenberg, J.P., Guenther, A.B., Turnipseed, A., 2005. Marine organic halide and isoprene emissions near mace head. Ireland. Environ. Chem. 2, 291–294. doi:10.1071/EN05072. Guenther, A., Karl, T., Harley, P., Wiedinmyer, C., Palmer, P.I., Geron, C., 2006. Estimates of global terrestrial isoprene emissions using MEGAN (Model of Emissions of Gases and Aerosols from Nature). Atmos. Chem. Phys. 6, 3181–3210. Haagen-Smit, A.J., 1952. Chemistry and physiology of Los Angeles smog. Industrial Engineering Chemistry 44, 1342–1346. doi:10.1021/ie50510a045.

B. Gantt et al. / Atmospheric Environment 44 (2010) 115–121 Henze, D.K., Seinfeld, J.H., 2006. Global secondary organic aerosol formation from isoprene oxidation. Geophys. Res. Lett. 33, L09812. doi:1029/2006GL025976. Holst, T., Arneth, A., Hayward, S., Ekberg, A., Mastepanov, M., Jackowicz-Korczynski, M., Friborg, T., Crill, P.M., Backstrand, K., 2008. BVOC ecosystem flux measurements at a high latitude wetland site. Atmos. Chem. Phys. Discuss. 8, 21129–21169. Jacovides, C.P., Timvios, F.S., Papaioannou, G., Asimakopoulos, D.N., Theofilou, C.M., 2004. Ratio of PAR to broadband solar radiation measured in Cyprus. Agricultural Forest Met. 121, 135–140. Kalberer, M., Paulsen, D., Sax, M., Steinbacher, M., Dommen, J., Prevot, A.S.H., Fisseha, R., Weingartner, E., Frankevich, V., Zenobi, R., Baltensperger, U., 2004. Identification of polymers as major components of atmospheric organic aerosols. Science 303, 1659–1662. Kroll, J.H., Ng, N.L., Murphy, S.M., Flagan, R.C., Seinfeld, J.H., 2006. Secondary organic aerosol formation from isoprene photooxidation. Environ. Sci. Technol. 40, 1869–1877. doi:10.1021/es0524301. Lewis, A.C., Bartle, K.D., Heard, D.E., McQuaid, J.B., Pilling, M.J., Seakins, P.W., 1997. In situ, gas chromatographic measurements of non-methane hydrocarbons and dimethyl sulfide at a remote coastal location (Mace Head, Eire) July–August 1996. Journal of the Chemical Society, Faraday Transactions 93, 2921–2927. Lewis, A.C., McQuaid, J.B., Carslaw, N., Pilling, M.J., 1999. Diurnal cycles of short-lived tropospheric alkenes at a north Atlantic coastal site. Atmos. Environ. 33, 2417–2422. Lewis, A.C., Carpenter, L.J., Pilling, M.J., 2001. Nonmethane hydrocarbons in Southern Ocean boundary layer air. J. Geophys. Res. 106, 4987–4994. Liakakou, E., Vrekoussis, M., Bonsang, B., Donousis, C., Kanakidou, M., Mihalopoulos, N., 2007. Isoprene above the Eastern Mediterranean: seasonal variation and contribution to the oxidation capacity of the atmosphere. Atmos. Environ. 41, 1002–1010. Liao, H., Henze, D.K., Seinfeld, J.H., Wu, S., Mickley, L.J., 2007. Biogenic secondary organic aerosol over the United States: comparison of climatological simulations with observations. J. Geophys. Res. 112, D06201. doi:10.1029/2006JD007813. Marshall, H.G., 1978. Phytoplankton distribution along the east coast of the USA, II, seasonal assemblages north of Cape Hatteras, North Carolina. Marine Biol. 45, 203–208. Milne, P., Riemer, D., Zika, R., Brand, L., 1995. Measurement of vertical distribution of isoprene in surface seawater, it chemical fate, and its emission from several phytoplankton monocultures. Marine Chem. 48, 237–244. Ng, N.L., Kwan, A.J., Surratt, J.D., Chan, A.W.H., Chhabra, P.S., Sorooshian, A., Pye, H.O.T., Crounse, J.D., Wennberg, P.O., Flagan, R.C., Seinfeld, J.H., 2008. Secondary organic aerosol (SOA) formation from reaction of isoprene with nitrate radicals (NO3). Atmos. Chem. Phys. 8, 4117–4140. O’Dowd, C.D., Jimenez, J.L., Bahreini, R., Flagan, R.C., Seinfeld, J.H., Pirjola, L., Kulmala, M., Jennings, S.G., Hoffmann, T., 2002. Marine particle formation from biogenic iodine emissions. Nature 417, 632–636. O’Dowd, C.D., Facchini, M.C., Cavalli, F., Ceburnis, D., Mircea, M., Dececari, S., Fuzzi, S., Yoon, Y.J., Putaud, J.-P., 2004. Biogenically driven organic contribution to marine aerosol. Nature 431, 676–680. Odum, J.R., Hoffmann, T., Bowman, F., Collins, D., Flagan, R.C., Seinfeld, J.H., 1996. Gas/particle partitioning and secondary organic aerosol yields. Environ. Sci. Technol. 30, 2580–2585. Offenberg, J.H., Kleindienst, T.E., Jaoui, M., Lewandowski, M., Edney, E.O., 2006. Thermal properties of secondary organic aerosols. Geophys. Res. Lett. 33, L03816. doi:10.1029/2005GL024623. O’Reilly, J.E., Maritorena, S., Mitchell, B.G., Siegel, D.A., Carder, K.L., Garver, S.A., Kahru, M., McClain, C., 1998. Ocean color chlorophyll algorithms for SeaWiFS. J. Geophys. Res. 103 (C11), 24937–24953.

121

Palmer, P.I., Shaw, S.L., 2005. Quantifying global marine isoprene fluxes using MODIS chlorophyll observations. J. Geophys. Res. 32, L09805. doi:10.1029/ 2005GL022592. Pankow, J., 1994. An absorption model of the gas/aerosol partitioning involved in the formation of secondary organic aerosols. Atmos. Environ. 28 (2), 189–193. Park, R.J., Jacob, D.J., Field, B.D., Yantosca, R.M., Chin, M., 2004. Natural and transboundary pollution influences on sulfate-nitrate-ammonium aerosols in the United States: implications for policy. J. Geophys. Res. 109, D15204. doi:10.1029/ 2003JD004473. Poisson, N., Kanakidou, M., Crutzen, P.J., 2000. Impact of nonmethane hydrocarbons on tropospheric chemistry and the oxidizing power of the global troposphere: 3-dimensional modeling results. J. Atmos. Chem. 36, 157–230. doi:10.1023/ A:1006300616544. Ridley, B.A., Carroll, M.A., Dunlap, D.D., Trainer, M., Sachse, G.W., Gregory, G.L., Condon, E.P., 1989. Measurements of NOx over the Eastern Pacific Ocean and Southwestern United States during the spring 1984 NASA GTE aircraft program. J. Geophys. Res. 94 (D4), 5043–5067. Schell, B., Ackermann, I.J., Hass, H., Binkowski, F.S., Ebel, A., 2001. Modeling the formation of secondary organic aerosol within a comprehensive air quality model system. J. Geophys. Res. 106 (D22), 28, 275–28, 293. Shaw, G.E., 1983. Bio-controlled thermostasis involving the sulfur cycle. Climate Change 5, 297–303. Shaw, S.L., Chisholm, S.W., Prinn, R.G., 2003. Isoprene production by Prochlorococcus, a marine cyanobacterium, and other phytoplankton. Marine Chem. 80, 227–245. Sinha, V., Williams, J., Meyhofer, M., Riebesell, U., Paulino, A.I., Larsen, A., 2007. Air-sea fluxes of methanol, acetone, acetaldehyde, isoprene and DMS from a Norwegian fjord following a phytoplankton bloom in a mesocosm experiment. Atmos. Chem. Phys. 7, 739–755. Surratt, J.D., Kleindienst, T.E., Edney, E.O., Lewandowski, M., Offenberg, J.H., Jaoui, M., Seinfeld, J.H., 2007. Effect of acidity on secondary organic aerosol formation from isoprene. Environ. Sci. Technol. 41 (15), 5363–5369. doi:10.1021/es0704176. Trainer, M., Williams, E.J., Parrish, D.D., Buhr, M.P., Allwine, E.J., Westberg, H.H., Fehsenfeld, F.C., Liu, S.C., 1987. Models and observations of the impact of natural hydrocarbons on rural ozone. Nature 329 (6141), 705–707. United State Environmental Protection Agency, 2004. Air quality designations and classifications for the 8-hour ozone national ambient air quality standards; early action compact areas with deferred effective dates. Federal Register 69 (No. 84), 23857–23951. United States Environmental Protection Agency, 2005. Air quality designations and classifications for the fine particles (PM2.5) National Ambient Air Quality Standards. Federal Register 70 (No. 3), 943–1019. Yassaa, N., Peeken, I., Zo¨llner, E., Bluhm, K., Arnold, S., Spracklen, D., Williams, J., 2008. Evidence for marine production of monoterpenes. Environ. Chem. 5, 391–401. doi:10.1071/EN08047. Yokouchi, Y., Li, H.-J., Machida, T., Aoki, S., Akimoto, H., 1999. Isoprene in the marine boundary layer (Southeast Asian Sea, eastern Indian Ocean, and Southern Ocean): comparison with dimethyl sulfide and bromoform. J. Geophys. Res. 104, 8067–8076. Zhang, Y., Huang, J., Henze, D.K., Seinfeld, J., 2007. Role of isoprene in secondary organic aerosol formation on a regional scale. J. Geophys. Res. 112, D20207. doi:10.1029/2007/JD008675. Zhang, Y., Wen, X.-Y., Wang, K., Vijayaraghavan, K., Jacobson, M.Z., 2009. Probing into Regional O3 and PM Pollution in the U.S., Part II. An Examination of Formation Mechanisms through a Process Analysis Technique and Sensitivity Study. J. Geophys. Res., in press.