RESEARCH ARTICLE Chemical Interactions Between Ship-Originated Air 10.1029/2020JD034175 Pollutants and Ocean-Emitted Halogens Key Points: Qinyi Li1 , Alba Badia2 , Rafael P. Fernandez1,3 , Anoop S. Mahajan4 , • S hipping emissions significantly Ana Isabel López-Noreña5 , Yan Zhang6 , Shanshan Wang6, Enrique Puliafito5 , perturb the level of air pollutants and ocean-emitted reactive halogen Carlos A. Cuevas1 , and Alfonso Saiz-Lopez1 species 1 • Halogens affect the abundance Department of Atmospheric Chemistry and Climate, Institute of Physical Chemistry Rocasolano, CSIC, Madrid, Spain, 2 and distribution of ship-originated Institute of Environmental Science and Technology (ICTA), Universitat Autònoma de Barcelona (UAB), Barcelona, oxidants and air pollutants Spain, 3Institute for Interdisciplinary Science (ICB), National Research Council (CONICET), FCEN-UNCuyo, Mendoza, • Chemical interactions between ship Argentina, 4Centre for Climate Change Research, Indian Institute of Tropical Meteorology, Pune, India, 5Atmospheric emissions and reactive halogens should be considered in policy and Environmental Studies Group (GEAA), National Technological University (UTN-FRM), CONICET, Mendoza, assessments of their impact on Argentina, 6Department of Environmental Science and Engineering, Shanghai Key Laboratory of Atmospheric Particle climate and coastal air quality Pollution and Prevention, Fudan University, Shanghai, China Supporting Information: Abstract Ocean-going ships supply products from one region to another and contribute to the • Supporting Information S1 world’s economy. Ship exhaust contains many air pollutants and results in significant changes in marine atmospheric composition. The role of reactive halogen species (RHS) in the troposphere has received Correspondence to: increasing recognition and oceans are the largest contributors to their atmospheric burden. However, A. Saiz-Lopez and Q. Li, a.saiz@csic.es; qli@iqfr.csic.es the impact of shipping emissions on RHS and that of RHS on ship-originated air pollutants have not been studied in detail. Here, an updated Weather Research Forecasting coupled with Chemistry model is utilized to explore the chemical interactions between ship emissions and oceanic RHS over the East Asia Citation: Li, Q., Badia, A., Fernandez, R. P., seas in summer. The emissions and resulting chemical transformations from shipping activities increase Mahajan, A. S., López-Noreña, A. the level of NO and NO2 at the surface, increase O3 in the South China Sea, but decrease O3 in the East of chlorine; ∼30% and ∼5% decrease of bromine and iodine, respectively) as well as in their partitioning. I., Zhang, Y., et al. (2021). Chemical China Sea. Such changes in pollutants result in remarkable changes in the levels of RHS (>200% increase interactions between ship-originated air The abundant RHS, in turn, reshape the loadings of air pollutants (∼20% decrease of NO and NO2; ∼15% pollutants and ocean-emitted halogens. decrease of O3) and those of the oxidants (>10% reduction of OH and HO2; ∼40% decrease of NO3) with Journal of Geophysical Research: Atmospheres, 126, e2020JD034175. https://doi.org/10.1029/2020JD034175 marked patterns along the ship tracks. We, therefore, suggest that these important chemical interactions Received 31 OCT 2020 of ship-originated emissions with RHS should be considered in the environmental policy assessments of Accepted 27 DEC 2020 the role of shipping emissions in air quality and climate. 1. Introduction Shipment is the dominant method for redistributing products from one continent to another and is critical in sustaining the world’s economy. Shipping activities result in emissions of air pollutants over natural wa- ters and contribute approximately 15% of the NOx emissions and 13% of the SOx emissions on a global scale (IMO, 2014; Kontovas, 2020). Ship-originated air pollutants (primary emissions and their secondary products) have been reported to af- fect the atmospheric composition, air quality, and human health. A pioneering study by Lawrence and Crutzen (1999) revealed that ship emissions reshape atmospheric chemistry in the marine boundary layer (MBL). Eyring et al. (2010) reported that ship emissions of air pollutants caused air quality problems with el- evated nitrogen oxides, sulfur, particulate matter, and ground-level ozone (O3) in coastal areas and harbors. Fan et al. (2016) and Lv et al. (2018) also suggested that ship emissions have a noticeable influence on air quality in East China and East Asia, respectively. The reduced air quality due to ship emissions can lead to © 2021. The Authors. This is an open access article under adverse health impacts with 14,500–37,500 worldwide premature deaths per year (H. Liu et al., 2016). Ship the terms of the Creative Commons emissions affect climate via their effects on greenhouse gases (CO2, CH4, and O3) and cloud condensation Attribution-NonCommercial-NoDerivs nuclei (Von Schneidemesser et al., 2015). Ship emissions also change marine stratocumulus regime of open License, which permits use and distribution in any medium, provided cells into closed cells inducing negative radiative forcing (Y.-C. Chen et al., 2012; Goren & Rosenfeld, 2012). the original work is properly cited, the use is non-commercial and no The budget of reactive halogen species (RHS) can also be influenced by shipping emissions. Osthoff modifications or adaptations are made. et al. (2008) reported the first ambient measurement of ClNO2, an important reservoir of reactive chlorine, LI ET AL. 1 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 and linked its production with ship plumes and coastal pollution. Ship-emitted NOx (=NO + NO2) and SO2 increase the deposition of acids (HNO3, H2SO4) to the sea surface and aerosols, decreasing their pH, which can potentially affect the debromination process of sea-salt aerosols (Badia et al., 2019; Yang et al., 2005). This emitted NOx also affects the formation of O3 and hence the subsequent emission of iodine species (I2 and HOI) from the oxidation of iodide in seawater (Carpenter et al., 2013; Macdonald et al., 2014), trans- forms reactive halogen oxides into halogen nitrates which ultimately deposit onto the sea surface (Simpson et al., 2015), and is associated with nighttime iodine chemistry (Saiz-Lopez et al., 2016), for example, via the reaction of NO3 with both I2 and HOI. Oceans are the largest source of reactive halogen (Cl, Br, and I) species and most of the impact of oceanic halogens inherently occurs in the marine atmosphere (Simpson et al., 2015). Reactive halogens change the partitioning of HO2 and OH which affects the oxidation capacity of the atmosphere (Li, Badia, Wang, et al., 2020), as well as the partitioning of NO and NO2 and hence affects the air quality (Li et al., 2019; Sherwen et al., 2017). Reactive halogens increase the loss of O3 (via reactions of halogen atoms with O3) and reduce the production of O3 (via the reactions of halogen oxides with NO2), which affects air quality and ra- diative forcing (Hossaini et al., 2015; Saiz-Lopez et al., 2012, 2014; Sherwen, et al., 2016; Sherwen, Schmidt, et al., 2016). Halogen-mediated tropospheric ozone loss is also linked to climate and is projected to be a key component of the ozone budget throughout the 21st century (Iglesias-Suarez et al., 2020). Halogen atoms (mostly chlorine but also bromine atoms) also react with volatile organic compounds (VOCs) and initiate photochemical production of O3 in polluted areas (Li et al., 2016; Sarwar et al., 2012). This work focuses on evaluating the interaction of ship-originated air pollutants and atmospheric halogens emitted from the ocean. We adopt a revised regional chemical transport model, Weather Research Fore- casting coupled with Chemistry (WRF-Chem), updated with comprehensive implementation of halogen sources and chemistry (Badia et al., 2019) and the EDGAR emission inventory for air pollutants from ship- ment activities (Crippa et al., 2018), to look into the interaction between ship-originated air pollutants and oceanic halogen emissions within the lower troposphere over the open ocean. We choose East Asia as the region of interest because (1) very few studies on reactive halogens have been conducted in this area; (2) this region contributes nearly 40% of the seaborne trade around the world (H. Liu et al., 2016); and (3) this region is one of the world’s busiest ship tracks (Crippa et al., 2018). 2. Method and Materials 2.1. WRF-Chem Model The halogen sources and chemistry incorporated in the WRF-Chem model have been documented in detail in Badia et al. (2019). Here, we briefly describe the main processes. A few heterogeneous reactions activate (relatively) inert halogen species (chloride, HCl, bromide in sea-salt aerosol, iodide in seawater, etc.) to reactive gaseous species (e.g., R1–R3). In R1, φ represents the ClNO2 production branching ratio (yield). N 2O5   Cl   ClNO 2   2    NO3 (R1) O3  Br   sea ‐ salt aerosol   Br2 (R2) O3  I  sea ‐ water   I2  HOI (R3) These halogens actively participate in atmospheric chemistry via (1) directly destroying O3 molecule (R4); (2) oxidizing VOCs which contributes to the generation of more oxidants including RO2, HO2, OH, and O3 (R5–R9); (3) perturbing the partitioning of HOx (R10 and R11) and NOx (R12); and (4) providing a sink for NOx (R13) (Saiz-Lopez & von Glasow, 2012; Simpson et al., 2015). The loss of halogen species is primarily through deposition on aerosol and ground surface with halogen nitrates (ClNO3, BrNO3, and INO3) having the highest Henry’s law constants (Badia et al., 2019). X  O3  XO  X  (R4) Cl; Br; and I  LI ET AL. 2 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 1. Emissions of NOx, CO and VOC, and SO2 (mol km-2 h-1) over the ocean within the study domain. The blue line is used for further cross-section analysis. Cl  Br   VOC  RO 2  HCl  HBr  (R5) RO 2  NO  NO 2  RO (R6) HO2  NO  NO 2  OH (R7) NO 2  hv  NO  O 3 P (R8)   O 3 P  O 2  O3 (R9)   XO  HO2  HOX (R10) (R11) HOX  hv  X  OH XO  NO  X  NO 2 (R12) XO  NO 2  XNO3 (R13) 2.2. Emission Data We follow Li, Badia, Wang, et al. (2020) to set up the reactive halogen scheme in order to calculate the reac- tive halogen emissions online. Although in the present study, we do not include the anthropogenic chlorine emission inventory over mainland China because the focus here is over the oceanic area. We use the MEIC emission inventory (www.meicmodel.org) for routine air pollutants (NOx, SO2, CO, VOC, NH3, PM2.5, and PM10) in mainland China and EDGARv4.3.2 emission inventory (Crippa et al., 2018) for air pollutants in other East Asian regions. For the ship emissions, the EDGARv4.3.2 inventory is adopted. Note that there are no halogen emissions in the EDGAR and MEIC emission inventories. The geographical distribution for ship emissions of NOx, CO and VOCs, and SO2 over the ocean is shown in Figure 1. We ac- knowledge that there have been a few emission inventories developed for East Asia with various spatial and temporal coverage, for example, Fan et al. (2016) for the year 2010, H. Liu et al. (2016) and Lv et al. (2018) for the year 2013, and D. Chen et al. (2017) for the year 2014. Please note that in the present conceptual LI ET AL. 3 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Table 1 work, we investigate the coupled chemistry of ship emissions and oceanic Simulation Scenarios halogens and we do not intend to characterize shipping emission inven- tories in East Asia. Halogen Cases Ship emission Oceanic halogen emission chemistry NO-SHIP No Yes Yes 2.3. Simulation Setup NO-HAL Yes No No The domain of interest (14°N–38°N, 108°E–132°E) is shown in Figure 1 FULL Yes Yes Yes and covers the coastlines of most East Asian regions. The simulation pe- riod is July 20, 2018 to August 31, 2018 considering the initial 11 days in July as model spin-up. Three simulation scenarios, as shown in Table 1, were conducted to identify the effects of the ship emissions on air pollutants and halogens (i.e., the chang- es between NO-SHIP and FULL cases) and those of halogens on ship-originated air pollutants (i.e., the difference of atmospheric composition between NO-HAL and FULL cases). Here, we note that we design our model simulations to quantify the sources (ship or oceanic halogen) and the corresponding subsequent chemical transformations on the atmospheric chemistry. 2.4. Model Performance Evaluation We evaluated the simulated air pollutants and halogen species from the three cases with previously reported observations within (or close to) the domain of interest (Table S1). The reported average NO2 ranges from 30 to 5,300 pptv (Lan et al., 2015; Schreier et al., 2015). The three scenarios also simulate a large concentration range (from pptv to ppbv levels) of NO2 (see Section 3 for details). WRF-Chem simulations cover the range of NO2 reported by the previous studies in this area but overestimate NO2 compared to some studies and underestimate compared to others (Table S1). The discrepancy between model and measurements could be due to (1) different period between model and field campaigns; (2) different spatial dimensions and range (a single point in ground campaigns and a 3D line for flight campaigns compared to a 27 × 27 km2 grid box in the model); (3) uncertainty in the emission inventory from shipping and the islands within the domain; and (4) missing sources and/or chemical mechanism of NOx leading to the typical underestimation of NOx levels in remote MBL by chemical transport models (e.g., Travis et al., 2020). Similarly, the present work reproduces the O3 level in the South China Sea and Taiwan strait compared to one observation study (in to ∼85 ppbv; Carmichael et al., 2003; Hatakeyama et al., 2004; Jacob et al., 2003). This could be due to (1) average 21 ppbv) (Lan et al., 2015) but underestimates O3 compared to other previous measurements (∼50 uncertainty in the emission inventories; (2) uncertainty in the initial and boundary conditions; (3) grid size and different spatial range; and (4) a lack of VOC emissions from the ocean (e.g., S. Wang, Hornbrook, et al., 2019). WRF-Chem underestimated CO compared to the observations with a range from 145 to 286 ppbv for CO (Carmichael et al., 2003; Jacob et al., 2003; Lan et al., 2015) (Table S1). As to the loading of RHS in the region, previous studies reported a range of the daytime average of BrO from 0.15 to 1.7 pptv (Koenig et al., 2017) and a daytime average of IO from 0.4 to 1.4 pptv in the West Pacific (Großmann et al., 2013; Prados-Roman et al., 2015). The NO-SHIP scenario overestimates the level of BrO (daytime average of 0.96 pptv and maximum of 7.0 pptv at the surface) and IO (daytime average of 0.76 pptv and maximum of 1.6 pptv within MBL and 1.4 pptv and 4.2 pptv at the surface). The FULL case estimates an amount closer to the observations for both halogen oxides (daytime average of 0.45 pptv and maximum of 1.4 pptv for BrO at the surface; the daytime average of 0.68 pptv and maximum of 1.3 pptv of IO in MBL; and an average of 1.2 pptv and maximum of 2.6 pptv of IO at the surface; Table S1). A recent study (S. Liu et al., 2020) reported a wide range of mixing ratios of organic bromine species (CHBr3, CH2Br2, and CHBr2Cl) in the East China Sea and West Pacific. The present study underestimates these organic bromine species but covers the lower end of the measurements (Table S1). Note that these bromine species have a lifetime of tens to more than a hundred days (Fernandez et al., 2014) and therefore have limited impact on the bromine loading at the surface and within the MBL. In general, WRF-Chem reasonably reproduces the limited observations available over the Western Pacific. LI ET AL. 4 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 3. Results and Discussion 3.1. Impact of Ship Emissions on Air Pollutants The simulated average distribution of air pollutants (NO, NO2, and O3) at the surface for the NO-SHIP and FULL cases and the relative changes from NO-SHIP to FULL (the impact of ship emissions) are presented in Figure 2. Note that in the current study, we focus on the interaction between ship-originated air pollut- ants and oceanic halogens therefore we only present the results over the marine area. The reader is referred to Li, Badia, Wang, et al. (2020) for the influence of halogens on oxidation and air pollutants over land in this region. In the NO-SHIP scenario, there are no emissions of NOx over the ocean and the levels of NO (in average 12.3 pptv) and NO2 (127.0 pptv) are low in most of the marine atmosphere (Figure 2). O3 has an average mixing ratio of 19.9 ppbv and ranges between 10.0 and 50.0 ppbv over the ocean and peaks in the northern part of the domain which is affected by the outflow from China, Korea, and Japan (Figure 2). In the FULL scenario, shipping activities add substantial (compared to the natural state of marine atmos- phere) emissions of NOx on top of this otherwise clean environment, which result in higher levels of NO (in average 38.3 pptv or 210.0% higher) and NO2 (289.8 pptv or 128.0% higher) compared to the NO-SHIP scenario. Note that in the marine areas away from the coast, the NOx enhancement near the ship tracks is as large as a factor of 10 (less than 10.0 pptv in NO-SHIP case as compared over 100.0 pptv in FULL case). The enhancement of NOx levels results in an opposite impact on background ozone levels depending on loca- tion: O3 is reduced in the northern part of the domain (East China Sea) by as much as -15.8% and enhanced in the southern part (South China Sea) up to 22.8% compared to the NO-SHIP scenario. This is because as a secondary product of photochemistry, O3 has a nonlinear relationship with its precursors, NOx and CO + VOCs (Seinfeld & Pandis, 2006). Thus, an increase of NOx leads to (1) an increase of O3 if the levels of CO in the NO-SHIP scenario are sufficient (e.g., >70 ppbv in the South China Sea) or (2) a decrease in O3 if the levels of CO are lower (e.g., <60 ppbv in the East China Sea, Figure S1). Most of the previous studies on the role of shipping activities in East Asia have focused on their impact on air quality over land (e.g., H. Liu et al., 2016; Z. Liu et al., 2017; Lv et al., 2018; Zhang et al., 2017). A model study by R. Wang, Tie, et al. (2019) in the East China Sea reported a range of –5 to 30 μg m-3 (∼–2.5 to 15 ppbv) difference in O3 over the ocean at a specific hour (14:00 on September 21, 2015) after the addition of ship emissions. Our simulations have a similar range of impacts (Figure 2) but a different spatial pattern, possibly due to the different simulation period, different emission inventory, and halogen chemistry used here but not in previous works (see Section 2 for Methods and Materials). Overall, it can be seen that the ship emissions substantially change the simulated level of air pollutants over the ocean. 3.2. Impact of Ship Emissions on the Loading of Halogens The total inorganic chlorine (Cly), bromine (Bry), and iodine (Iy) in the NO-SHIP and FULL cases and the relative changes due to ship emissions are presented in Figure 3. In the NO-SHIP case, the average levels of Cly (38.7 pptv), Bry (2.8 pptv), and Iy (9.0 pptv) present a spatial distribution that generally correlates with the level of O3, NOx, and chloride (as an indication of sea-salt aerosol; Figure S2) because the activation of halogens from sea-salt aerosol and sea surface to the gaseous phase is strongly mediated by O3, NOx, and sea-salt aerosol (Saiz-Lopez & von Glasow, 2012; Simpson et al., 2015). The large changes in air pollutants due to the addition of ship emissions induce a significant difference in the simulated levels of halogen species. The average Cly over the ocean within our domain increases to 139.3 pptv (or by 259.8%) in the FULL scenario. Note that shipping activities substantially increase the level of NOx (Figure 2) and OH (Figure S3) which activates chlorine from (relatively) inert chloride in sea- salt aerosol to reactive chlorine gases, for example, ClNO2, ClNO3, HOCl, and Cl2. For Bry, the average is reduced to 1.9 pptv (or by 32.8%) in the FULL scenario. The large enhancement of NOx transforms BrO into BrNO3 which deposits onto the sea surface and hence reduces the total gaseous bromine. Regarding Iy, the mean level is decreased to 8.5 pptv (or by 6.1%) in the FULL case. The NOx enhancement also contributes to changes in Iy by favoring the formation of INO3, instead of HOI which dominates as iodine reservoir species LI ET AL. 5 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 2. Simulated monthly average of NO (pptv, top), NO2 (pptv, middle), and O3 (ppbv, bottom) at the surface for the NO-SHIP (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. The relative change of one species (here using NO as an  NOFULL  NO NO ‐SHIP  / NO NO ‐SHIP  100. example) between two cases in this paper is calculated as: NO relativechange  in low NOx environments. The reduction of O3, in turn, decreases the emission of I2 and HOI from the ocean (Carpenter et al., 2013; Macdonald et al., 2014). Simulated levels of halogens in the FULL scenario are in line with the results in Li, Badia, Wang, et al. (2020) except that the present study does not include anthro- pogenic chlorine emissions over the land and hence simulates lower levels of chlorine. LI ET AL. 6 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 3. Simulated monthly average of Cly (pptv, top), Bry (pptv, middle), and Iy (pptv, bottom) at the surface for the NO-SHIP (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. We now look at the effects of ship emissions on the featured species of reactive halogens, ClNO2, BrO, and IO (Figure 4). Shipping activities increase the levels of NOx and hence the production of ClNO2. But for BrO, the large NOx levels along the ship tracks rapidly transform BrO to BrNO3, which is subsequently taken up on aerosols, therefore, decreasing the levels of BrO. As for IO, the ship-emitted NO2 transforms IO into INO3 and subsequently deposits on aerosol surfaces, therefore, decreasing IO in most of the domain, especially along the ship tracks. Both BrNO3 and INO3 undergo photolysis and the photolysis rate of BrNO3 LI ET AL. 7 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 4. Simulated monthly average of ClNO2 (pptv, top), BrO (pptv, middle), and IO (pptv, bottom) at surface for the NO-SHIP (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. (Figure S4) is slower (hence regenerating less halogen atom or halogen oxide) than that of INO3 (Figure S4). Such difference of photolysis rate between BrNO3 and INO3 induces a larger decrease in BrO than in IO (Figure 4). ClNO2 is produced at night and can be transported to adjacent areas before being photolyzed at sunrise, therefore, the changes in ClNO2 do not follow the ship tracks pattern as strictly as those in other halogen species. Note that the uptake of ClNO2 on aerosols (when pH < 2) to form Cl2 (Roberts et al., 2008) is not considered in the present work and is not expected to have noticeable influence because the simulated LI ET AL. 8 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 pH is rarely lower than 2 in the bins with considerable sea-salt aerosol within our domain (Figure S5). On the contrary, BrO and IO are formed during the day and rapidly transform to other species (e.g., BrNO3 and INO3), therefore the suppressing effect of ship emissions is evident along the ship tracks. 3.3. Impact of Ship Emissions on the Halogens Partitioning In this section, we investigate how ship emissions change the partitioning of halogen species. Figure 5 pre- sents the monthly averaged percentage contribution of individual gaseous species to the total Cly, Bry, and Iy in the NO-SHIP and FULL scenarios. For chlorine, ship emissions induce a shift from HOCl to Cl2, mostly driven by heterogeneous recycling (Keene et al., 2009; Pechtl & von Glasow, 2007), and result in a less effec- tive washout efficiency that increases the chlorine burden. In particular, including ship emissions reduces percentage contribution of individual species to Cly, that is, HOCl (from 71.8% to 50.3%), ClNO2 (from 3.4% to 2.7%), BrCl (2.2% to 0.4%), and ICl (0.7% to 0.2%) and significantly increase that of Cl2 (13.3% to 32.5%) and ClNO3 (2.7% to 7.5%). Only small changes are induced to ClO and Cl. Note that the total abundance of chlorine species is increased with the addition of ship emissions (Figure 3) and such enhancement effect is stronger for some species than for others. Shipping activities also affect the bromine partitioning, with the largest change in the BrNO3 fraction from 5.2% to 16.2% and in BrO from 16.1% to 8.8%, due to the large emissions of NOx transforming BrO into BrNO3. BrCl is the dominant bromine fraction for both NO-SHIP and FULL schemes. Note that including the additional NOx and SO2 sources does not introduce a significant change in the sea-salt recycling of bro- mine (mainly driven by BrNO3 and HOBr) as for the case of chlorine, where HOCl is the dominant fraction for the NO-SHIP case. The changes in iodine partitioning are similar to those of bromine species. The average concentration and partitioning of halogen species during the daytime (8:00–17:00, local time) and nighttime (18:00–07:00, local time) in the NO-SHIP and FULL scenarios are depicted in Figures S6 and S7. The total chlorine and bromine abundances in the daytime are slightly higher than those in the nighttime, while total iodine during the day is lower compared to that at night in both NO-SHIP and FULL scenarios (Figures S6 and S7). The key changes of halogen partitioning between the NO-SHIP and FULL cases in the daytime (as well as the changes in the nighttime) are similar to those in daily average (Figure 5), for example, the decrease of HOCl, the increase of BrNO3 and INO3. The partitioning of halogen species in the daytime shows distinct patterns compared to that at night (Fig- ure S6 for the FULL scenario and Figure S7 for the NO-SHIP case). Here, we take the FULL case as an example. For daytime chlorine, the most abundant species is HOCl (59.7%) followed by Cl2 (13.5%), ClNO3 (13.7%), ClO (11.7%), etc., while at night, Cl2 (48.2%) and HOCl (43.4%) contribute to most of the loading. For bromine species, HOBr (41.4%), BrNO3 (23.8%), BrO (16.5%), and BrCl (9.8%) are the dominant daytime species, while at night the partitioning is dominated by BrCl (52.5%), IBr (21.9%), Br2 (9.8%), and BrNO3 (7.4%). In a global model study by Fernandez et al. (2014), HBr and HOBr are the species with elevated sur- face concentrations during the daytime in the Western Pacific (eastern part of the domain in present work), while Br2, HBr, and BrCl are high at night. The present study does not show large levels of HBr because there are higher levels of pollutants, which activate the bromine from HBr (relatively more inert) to other more reactive species (e.g., via the reactions of ClNO3 + HBr  BrCl, HOCl + HBr  BrCl, HBr + N2O5  BrNO2, and HOBr + HBr  Br2). For the partitioning of iodine species in the daytime, HOI (66.2%), INO3 (11.2%), and IO (15.0%) contribute more than 90% of total iodine, and HOI (59.5%), INO3 (8.1%), I2O2 (9.6%), and I2 (8.4%) take up most of the iodine at night. Saiz-Lopez et al. (2014) reported HOI as the most abundant species at noon at the surface in the tropics and dihalogen molecules (IBr + ICl + I2) and HOI are the most abundant at midnight. Note that the reaction of HOI with NO3 (Saiz-Lopez et al., 2016) is not included in the present work. 3.4. Impact of Halogens on Ship-Originated Air Pollutants Figures 6 and 7 show the simulation of NO, NO2, and O3 in NO-HAL and the relative changes from NO- HAL to FULL. Halogens reduce both surface NO (on average by 9.6 pptv or 20.1%) and NO2 (on average LI ET AL. 9 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 5. Partitioning of monthly average Cly (top), Bry (middle), and Iy (bottom) for the NO-SHIP (left) and FULL (right) scenarios. by 66.8 pptv or 18.7%) in most of the regions due to the transformation of NOx to halogen nitrates, which ultimately form HNO3 and deposit to the sea surface and aerosols. Note that along the ship tracks and close to the coast, NO is enhanced and NO2 is reduced mostly because the loss of O3 decreases the formation of NO2 (from the reaction of NO + O3  NO2) resulting in an enhancement of NO. Li et al. (2019) applied the CMAQ model with comprehensive halogen chemistry in Europe and reported a decrease of NO2 (as much as 1.7 ppbv) in the North Sea, the Mediterranean sea, and the polluted coastal areas and an increase of NO2 (up to 0.20 ppbv) in other oceanic areas due to the same reason. LI ET AL. 10 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 6. Simulated monthly average of NO (pptv, top), NO2 (pptv, middle), and O3 (ppbv, bottom) at the surface for the NO-HAL (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. Halogens consistently reduce the levels of O3 (on average by 4.2 ppbv or 17.4%) throughout the domain be- cause (1) halogen atoms directly destroy O3, (2) halogen chemistry reduces the level of NO2 and hence the production of O3, and (3) the production of O3 via the added reactions of Cl and Br atoms with VOCs is not able to compensate the enhanced loss of O3 driven by (1) and the reduced production of O3 driven by (2). Figure 7 shows a vertical perspective of the air pollutants for the NO-HAL and FULL cases along the latitu- dinal transect shown in Figure 1. Halogens reduce the NOx and O3 throughout the MBL and even in the free LI ET AL. 11 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 7. Simulated monthly average of NO (pptv, top), NO2 (pptv, middle), and O3 (ppbv, bottom) over the cross-section plane for the NO-HAL (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. The black shared area corresponds to the location of the Hainan Island (see Figure 1). troposphere, affecting the vertical transport of air pollutants (emitted over the continent and transported to the coastal and open ocean areas) from the surface to the MBL and the free troposphere. A hemispheric modeling study with CMAQ in the summer of 2006 (Sarwar et al., 2015) reported a reduction in O3 of more than 6.0 ppbv in coastal areas because of halogens and >4.0 ppbv within the domain of the current study. gested that halogens reduced surface O3 by ∼10% (∼a few ppbv) in the year of 2005 within the domain of Sherwen et al. (2016) and Sherwen, Schmidt, et al. (2016) utilized a global model (GEOS-Chem) and sug- the present work. A follow-up work of GEOS-Chem (Sherwen et al., 2017) reported a larger reduction of O3 (as much as 28.9 ppbv) in July 2015 within the European domain. Muñiz-Unamunzaga et al. (2018) applied CMAQ in Los Angeles and found a decrease of 2.5–5.0 ppbv in O3 along the coast in September 2006 due to the inclusion of full halogen chemistry. Sarwar et al. (2019) conducted a year-long simulation in 2006 of CMAQ and reported a decrease of 25% over oceanic waters and >6 ppbv in the present study domain. Li, Badia, Wang, et al. (2020) employed the WRF-Chem model to report 4%–10% reduction in surface O3 along the coast in summer 2018 due to the addition of halogen sources and chemistry in China. LI ET AL. 12 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 8. Simulated monthly average of OH (pptv, top), HO2 (pptv, middle), and NO3 (pptv, bottom) at the surface for the NO-HAL (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. 3.5. Impact of Halogens on Free Radicals The elevated levels of halogens in the presence of ship emissions (results in the FULL case as shown in Fig- ure 3) in turn change the atmospheric composition with distinguished features along and around the ship tracks. Figure 8 illustrates the average spatial distribution of the main free radicals controlling atmospheric oxidation in the NO-HAL scenario, in the FULL scenario, and the relative changes. LI ET AL. 13 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 In the NO-HAL case, the average levels of OH and HO2 over the ocean are 0.32 pptv and 8.5 pptv, respec- tively. OH is higher along the ship tracks, while HO2 is lower due to the emissions of NO from ships and the subsequent reaction of NO with HO2 that transforms HO2 into OH. The average NO3 over the ocean is 7.1 pptv, with higher NO3 outside the ship tracks compared to within the nearby tracks because the high levels of NO and NO2 react with NO3, reducing the levels of this radical. Halogen chemistry decreases OH (on average by 0.05 pptv or 16.0%) and HO2 (on average by 0.9 pptv or 10.8%) in most of the domain (Figure 8) because of significant O3 destruction (Figure 6). Noticeably, in the grids with intensive ship emissions (Figure 1), OH is only slightly reduced and HO2 is increased since halogens (mostly chlorine) increase the formation of HOx via reactions of halogen atoms with ship-emitted VOCs (R5). NO3 is formed through the reaction of NO2 and O3, and halogens decrease the NO3 mixing ratios (on average by 2.8 pptv or 39.3%) because (1) halogens reduce the levels of NO2; (2) halogens decrease O3; (3) halogens directly react with NO3 radical through the reactions of I2 and I with NO3. NO3 concentration could be further decreased if the reaction of HOI with NO3 (Saiz-Lopez et al., 2016) is included. The relative change of NO3 due to halogens is not marked along the ship tracks (Figure 8) while the absolute change of NO3 is evident along the tracks (Figure S8) because halogens induce the same decreasing effects on NO3 (in contrast to the complex effects of halogens on HOx) on and near the ship tracks. We also isolate the results in the daytime (for OH and HO2) and nighttime (for NO3) and the results show the same pattern (Figure S9). Figure 9 shows the vertical cross-section results for the simulation of radicals in the NO-HAL and FULL scenarios and the corresponding changes. The effect of halogens on OH and HO2 is higher below 400 m above sea level (a.s.l.), but a noticeable influence can be seen in the upper boundary layer and the free trop- osphere mostly due to the reduction of O3. For NO3, the decreasing effect due to halogens is more intense and reaches to more than 1,000 m a.s.l. because NOx levels in addition to O3 are significantly reduced. Sarwar et al. (2015) used the CMAQ model to estimate that halogens led to an average decrease of OH (∼1%) and HO2 (∼11%) in summer 2006 in the Northern Hemisphere. A global study with GEOS-Chem for the year of 2005 (Sherwen et al., 2016; Sherwen, Schmidt, et al., 2016) suggested a change of HOx (OH + HO2) in the range of -5% to 5% due to halogens within our domain. A follow-up work by Stone et al. (2018) found an increase of OH (∼2%) and a decrease of HO2 (-5% to -2%) for the year 2009 in East Asia. Li, Badia, Wang, et al. (2020) utilized the WRF-Chem model in China and reported changes of OH and HO2 in between -10% and 20% in summer 2018 along the coast. Note that Li, Badia, Wang, et al. (2020) included anthropogenic chlorine emission over land for their simulation, which led to an increase of both OH and HO2 in polluted coastal regions. Muñiz-Unamunzaga et al. (2018) used CMAQ model in Los Angeles and found a large de- crease (-50% to -20%) in the NO3 radical in September 2006 due to halogens and dimethyl sulfide chemistry. Li et al. (2019) suggested that halogen chemistry also significantly reduced the NO3 levels by as much as 20 pptv in July 2016 in the Mediterranean Sea. Li, Badia, Wang, et al. (2020) estimated as much as a 30% decrease in NO3 along the coast of China in summer 2018 due to halogen chemistry. 4. Conclusions and Final Remarks This conceptual work exemplifies, on one hand, the impacts of shipping emissions on the atmospheric burden of RHS (increasing Cly by 260%, decreasing Bry by 33%, and decreasing Iy by 6%) as well as on air pollutants and climate-relevant species (increasing NOx by >100% and inducing a wide range of changes in O3 from -16% to 23%) over the open ocean environment. On the other hand, our results account for the than 10% and decreasing NO3 by ∼40%) and air pollutants (decreasing NOx by∼20% and decreasing O3 by effects of RHS on the abundance and distribution of ship-originated free radicals (decreasing HOx by more ∼17%) in East Asia over the open ocean in summer. All of these highlight the importance of considering the long-range transport of oceanic natural sources interacting with anthropogenic pollutants not only in the vicinity of the ship tracks but also in the surrounding areas and vertically into the free troposphere. Such chemical interaction between shipping emissions and RHS is also expected in other parts of the world with marked shipping activities, for example, the North Sea (Bencs et al., 2020), southern Europe (Nunes et al., 2020), northern Europe (Tang et al., 2020), and North Pacific and North Atlantic (Crippa et al., 2018). LI ET AL. 14 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Figure 9. Simulated monthly average of OH (pptv, top), HO2 (pptv, middle), and NO3 (pptv, bottom) on the cross-section plane for the NO-HAL (left) and FULL (center) scenarios. The relative change (in percentage) between scenarios is shown in the right column. While the chemical scheme and shipping emissions used in the present work are subject to some uncer- tainties as discussed in Sections 2 and 3, our analysis suggests that (1) ship emissions should be taken into account when studying reactive halogens from the ocean; (2) halogens should be considered when assessing the impact of shipping emission on air quality; and (3) further laboratory, field, and numerical modeling works are required to fully characterize the chemical interaction between ship emissions and oceanic halogens. A particular interest is to consider the synergistic interaction between shipping emissions and halogens on the regulation of air pollution for coastal cities with large ports, for example, Shanghai, Shenzhen, and Hong Kong. On a broader aspect, we recommend that the formulation of air quality and en- vironmental policies should consider the strong chemical interaction between ship emissions and reactive halogen species. Data Availability Statement Data that support the finding of this study can be found at https://doi.org/10.17632/yd3bdrcwtb.1 (Li, Ba- dia, Fernandez, et al., 2020). LI ET AL. 15 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Acknowledgments References This study received funding from the European Research Council Executive Badia, A., Reeves, C. E., Baker, A. R., Saiz-Lopez, A., Volkamer, R., Koenig, T. K., et al. (2019). Importance of reactive halogens in the trop- Agency under the European Union’s ical marine atmosphere: A regional modelling study using WRF-Chem. Atmospheric Chemistry and Physics, 19(5), 3161–3189. https:// Horizon 2020 Research and Inno- doi.org/10.5194/acp-19-3161-2019 vation Program (Project ERC-2016- Bencs, L., Horemans, B., Buczyńska, A. J., & Deutsch, F. (2020). Seasonality of ship emission related atmospheric pollution over coastal COG 726349 CLIMAHAL) and was and open waters of the North Sea. Atmospheric Environment X, 7, 100077. https://doi.org/10.1016/j.aeaoa.2020.100077 supported by the Consejo Superior de Carmichael, G. R., Tang, Y., Kurata, G., Uno, I., Streets, D. G., Thongboonchoo, N., et al. (2003). Evaluating regional emission estimates Investigaciones Científicas (CSIC) of using the TRACE-P observations. Journal of Geophysical Research, 108(D21), 8810. https://doi.org/10.1029/2002JD003116 Spain and The Program for Professor of Carpenter, L. J., MacDonald, S. M., Shaw, M. D., Kumar, R., Saunders, R. W., Parthipan, R., et al. (2013). Atmospheric iodine levels influ- Special Appointment (Eastern Scholar) enced by sea surface emissions of inorganic iodine. Nature Geoscience, 6(2), 108–111. at Shanghai Institutions of Higher Chen, D., Wang, X., Li, Y., Lang, J., Zhou, Y., Guo, X., & Zhao, Y. (2017). High-spatiotemporal-resolution ship emission inventory of China Learning and Shanghai Thousand based on AIS data in 2014. The Science of the Total Environment, 609, 776–787. Talents Program. The development Chen, Y.-C., Christensen, M. W., Xue, L., Sorooshian, A., Stephens, G. L., Rasmussen, R. M., et al. (2012). Occurrence of lower cloud albedo and maintenance of the WRF-Chem in ship tracks. Atmospheric Chemistry and Physics, 12, 8223–8235. https://doi.org/10.5194/acp-12-8223-2012 model are conducted by NOAA/ESRL/ Crippa, M., Guizzardi, D., Muntean, M., Schaaf, E., Dentener, F., van Aardenne, J. A., et al. (2018). Gridded emissions of air pollutants for GSD in active collaboration with the period 1970–2012 within EDGAR v4.3.2. Earth System Science Data, 10(4), 1987–2013. https://doi.org/10.5194/essd-10-1987-2018 other institutes. Computing resourc- Eyring, V., Isaksen, I. S. A., Berntsen, T., Collins, W. J., Corbett, J. J., Endresen, O., et al. (2010). Transport impacts on atmosphere and es, support, and data storage were climate: Shipping. Atmospheric Environment, 44(37), 4735–4771. https://doi.org/10.1016/j.atmosenv.2009.04.059 provided by the Climate Simulation Fan, Q., Zhang, Y., Ma, W., Ma, H., Feng, J., Yu, Q., et al. (2016). Spatial and seasonal dynamics of ship emissions over the Yangtze River Laboratory at NCAR’s Computational Delta and East China Sea and their potential environmental influence. Environmental Science & Technology, 50(3), 1322–1329. https:// and Information Systems Laboratory doi.org/10.1021/acs.est.5b03965 (CISL), sponsored by the NSF. Inter- Fernandez, R. P., Salawitch, R. J., Kinnison, D. E., Lamarque, J.-F. F., & Saiz-Lopez, A. (2014). Bromine partitioning in the tropical tropo- national cooperation between CSIC pause layer: Implications for stratospheric injection. Atmospheric Chemistry and Physics, 14(24), 13391–13410. https://doi.org/10.5194/ (Spain) and CONICET (Argentina) acp-14-13391-2014 was supported by the i-COOP program Goren, T., & Rosenfeld, D. (2012). Satellite observations of ship emission induced transitions from broken to closed cell marine stratocu- (B20331). R.P.F. would like to thank mulus over large areas. Journal of Geophysical Research, 117, D17206. https://doi.org/10.1029/2012JD017981 ANPCyT (PICT 2015-0714), UNCuyo Großmann, K., Frieß, U., Peters, E., Wittrock, F., Lampel, J., Yilmaz, S., et al. (2013). Iodine monoxide in the Western Pacific marine bound- (SeCTyP M032/3853), and UTN (PID ary layer. Atmospheric Chemistry and Physics, 13(6), 3363–3378. https://doi.org/10.5194/acp-13-3363-2013 4920-194/2018) for financial support. Hatakeyama, S., Takami, A., Sakamaki, F., Mukai, H., Sugimoto, N., Shimizu, A., & Bandow, H. (2004). Aerial measurement of air pol- IITM is funded by the Ministry of Earth lutants and aerosols during 20–22 March 2001 over the East China Sea. Journal of Geophysical Research, 109, D13304. https://doi. Sciences, Government of India. org/10.1029/2003JD004271 Hossaini, R., Chipperfield, M. P., Montzka, S. A., Rap, A., Dhomse, S., & Feng, W. (2015). Efficiency of short-lived halogens at influencing climate through depletion of stratospheric ozone. Nature Geoscience, 8(3), 186–190. https://doi.org/10.1038/ngeo2363 Iglesias-Suarez, F., Badia, A., Fernandez, R. P., Cuevas, C. A., Kinnison, D. E., Tilmes, S., et al. (2020). Natural halogens buffer tropospheric ozone in a changing climate. Nature Climate Change, 10, 147–154. https://doi.org/10.1038/s41558-019-0675-6 IMO. (2014). Third IMO GHG study 2014. London, UK: International Maritime Organization (IMO). http://www.imo.org/en/OurWork/ Environment/PollutionPrevention/AirPollution/Pages/Greenhouse-Gas-Studies-014.aspx Jacob, D. J., Crawford, J. H., Kleb, M. M., Connors, V. S., Bendura, R. J., Raper, J. L., et al. (2003). Transport and chemical evolution over the Pacific (TRACE-P) aircraft mission: Design, execution, and first results. Journal of Geophysical Research, 108(20), 9000. https://doi. org/10.1029/2002JD003276 Keene, W. C., Long, M. S., Pszenny, A. A. P., Sander, R., Maben, J. R., Wall, A. J., et al. (2009). Latitudinal variation in the multiphase chem- ical processing of inorganic halogens and related species over the eastern North and South Atlantic Oceans. Atmospheric Chemistry and Physics, 9, 7361–7385. https://doi.org/10.5194/acp-9-7361-2009 Koenig, T. K., Volkamer, R., Baidar, S., Dix, B., Wang, S., Anderson, D. C., et al. (2017). BrO and inferred Bry profiles over the western Pacific: Relevance of inorganic bromine sources and a Bry minimum in the aged tropical tropopause layer. Atmospheric Chemistry and Physics, 17(24), 15245–15270. https://doi.org/10.5194/acp-17-15245-2017 Kontovas, C. A. (2020). Integration of air quality and climate change policies in shipping: The case of sulphur emissions regulation. Marine Policy, 113(December 2019), 103815. https://doi.org/10.1016/j.marpol.2020.103815 Lan, Y. Y., Tsuang, B. J., Lin, N. H., Hsu, H. H., Yu, C. C., & Chen, Y. T. (2015). Distribution of ozone and related compounds in the marine boundary layer of the Northern South China Sea in 2010. Aerosol and Air Quality Research, 15(5), 1990–2008. https://doi.org/10.4209/ aaqr.2014.10.0242 Lawrence, M. G., & Crutzen, P. J. (1999). Influence of NOx emissions from ships on tropospheric photochemistry and climate, Nature, 402, 167–170. Li, Q., Badia, A., Fernandez, R. P., Mahajan, A. S., López-Noreña, A. I., Zhang, Y., et al. (2020). Chemical interactions between ship-origi- nated air pollutants and ocean-emitted halogens. Mendeley Data, V1. https://doi.org/10.17632/yd3bdrcwtb.1 Li, Q., Badia, A., Wang, T., Sarwar, G., Fu, X., Zhang, L., et al. (2020). Potential effect of halogens on atmospheric oxidation and air quality in China. Journal of Geophysical Research: Atmospheres, 125, e2019JD032058. https://doi.org/10.1029/2019JD032058 Li, Q., Borge, R., Sarwar, G., de la Paz, D., Gantt, B., Domingo, J., et al. (2019). Impact of halogen chemistry on summertime air quality in coastal and continental Europe: Application of the CMAQ model and implications for regulation. Atmospheric Chemistry and Physics, 19(24), 15321–15337. https://doi.org/10.5194/acp-19-15321-2019 Li, Q., Zhang, L., Wang, T., Tham, Y. J., Ahmadov, R., Xue, L., et al. (2016). Impacts of heterogeneous uptake of dinitrogen pentoxide and chlorine activation on ozone and reactive nitrogen partitioning: Improvement and application of the WRF-Chem model in southern China. Atmospheric Chemistry and Physics, 16(23), 14875–14890. https://doi.org/10.5194/acp-16-14875-2016 Liu, H., Fu, M., Jin, X., Shang, Y., Shindell, D., Faluvegi, G., et al. (2016). Health and climate impacts of ocean-going vessels in East Asia. Nature Climate Change, 6(11), 1037–1041. https://doi.org/10.1038/nclimate3083 Liu, S., He, Z., & Yang, G. (2020). Bromoform, dibromochloromethane, and dibromomethane over the East China Sea and the western Pacific Ocean: Oceanic emission and spatial variation. Chemosphere, 257, 127151. https://doi.org/10.1016/j.chemosphere.2020.127151 Liu, Z., Lu, X., Feng, J., Fan, Q., Zhang, Y., & Yang, X. (2017). Influence of ship emissions on urban air quality: A comprehensive study using highly time-resolved online measurements and numerical simulation in Shanghai. Environmental Science & Technology, 51(1), 202–211. https://doi.org/10.1021/acs.est.6b03834 LI ET AL. 16 of 17 Journal of Geophysical Research: Atmospheres 10.1029/2020JD034175 Lv, Z., Liu, H., Ying, Q., Fu, M., Meng, Z., Wang, Y., et al. (2018). Impacts of shipping emissions on PM 2.5 air pollution in China. Atmos- pheric Chemistry and Physics, 18(21), 15811–15824. https://doi.org/10.5194/acp-18-15811-2018 Macdonald, S. M., Gómez Martín, J. C., Chance, R., Warriner, S., Saiz-Lopez, A., Carpenter, L. J., & Plane, J. M. C. (2014). A laboratory characterisation of inorganic iodine emissions from the sea surface: Dependence on oceanic variables and parameterisation for global modeling. Atmospheric Chemistry and Physics, 14(11), 5841–5852. https://doi.org/10.5194/acp-14-5841-2014 Muñiz-Unamunzaga, M., Borge, R., Sarwar, G., Gantt, B., de la Paz, D., Cuevas, C. A., & Saiz-Lopez, A. (2018). The influence of ocean halogen and sulfur emissions in the air quality of a coastal megacity: The case of Los Angeles. The Science of the Total Environment, 610–611, 1536–1545. https://doi.org/10.1016/j.scitotenv.2017.06.098 Nunes, R. A. O., Alvim-ferraz, M. C. M., Martins, F. G., Calderay-cayetano, F., Durán-grados, V., Moreno-gutiérrez, J., et al. (2020). Shipping emissions in the Iberian Peninsula and the impacts on air quality. Atmospheric Chemistry and Physics Discussion, 20, 9473–9489.https:// doi.org/10.5194/acp-2019-1186 Osthoff, H. D., Roberts, J. M., Ravishankara, A. R., Williams, E. J., Lerner, B. M., Sommariva, R., et al. (2008). High levels of nitryl chloride in the polluted subtropical marine boundary layer. Nature Geoscience, 1(5), 324–328. https://doi.org/10.1038/ngeo177 Pechtl, S., & von Glasow, R. (2007). Reactive chlorine in the marine boundary layer in the outflow of polluted continental air: A model study. Geophysical Research Letters, 34, L11813. https://doi.org/10.1029/2007GL029761 Prados-Roman, C., Cuevas, C. A., Hay, T., Fernandez, R. P., Mahajan, A. S., Royer, S. J., et al. (2015). Iodine oxide in the global marine boundary layer. Atmospheric Chemistry and Physics, 15(2), 583–593. https://doi.org/10.5194/acp-15-583-2015 Roberts, J. M., Osthoff, H. D., Brown, S. S., & Ravishankara, A. R. (2008). N2O5 oxidizes chloride to Cl2 in acidic atmospheric aerosol. Science, 321(5892), 1059. Saiz-Lopez, A., Fernandez, R. P., Ordóñez, C., Kinnison, D. E., Martín, J. C. G., Lamarque, J. F., & Tilmes, S. (2014). Iodine chemistry in the troposphere and its effect on ozone. Atmospheric Chemistry and Physics, 14(23), 13119–13143. https://doi.org/10.1177/0964663912467814 Saiz-Lopez, A., Lamarque, J. F., Kinnison, D. E., Tilmes, S., Ordóñez, C., Orlando, J. J., et al. (2012). Estimating the climate significance of halogen-driven ozone loss in the tropical marine troposphere. Atmospheric Chemistry and Physics, 12(9), 3939–3949. https://doi. org/10.5194/acp-12-3939-2012 Saiz-Lopez, A., Plane, J. M. C., Cuevas, C. A., Mahajan, A. S., Lamarque, J. F., & Kinnison, D. E. (2016). Nighttime atmospheric chemistry of iodine. Atmospheric Chemistry and Physics, 16(24), 15593–15604. https://doi.org/10.5194/acp-16-15593-2016 Saiz-Lopez, A., & von Glasow, R. (2012). Reactive halogen chemistry in the troposphere. Chemical Society Reviews, 41(19), 6448–6472. https://doi.org/10.1039/c2cs35208g Sarwar, G., Gantt, B., Foley, K., Fahey, K., Spero, T. L., Kang, D., et al. (2019). Influence of bromine and iodine chemistry on annual, sea- sonal, diurnal, and background ozone: CMAQ simulations over the Northern Hemisphere. Atmospheric Environment, 231, 395–404. https://doi.org/10.1016/j.atmosenv.2019.06.020 Sarwar, G., Gantt, B., Schwede, D., Foley, K., Mathur, R., & Saiz-Lopez, A. (2015). Impact of enhanced ozone deposition and halogen chemistry on tropospheric ozone over the Northern Hemisphere. Environmental Science and Technology, 49(15), 9203–9211. https:// doi.org/10.1021/acs.est.5b01657 Sarwar, G., Simon, H., Bhave, P., & Yarwood, G. (2012). Examining the impact of heterogeneous nitryl chloride production on air quality across the United States. Atmospheric Chemistry and Physics, 12(14), 6455–6473. https://doi.org/10.5194/acp-12-6455-2012 Schreier, S. F., Peters, E., Richter, A., Lampel, J., Wittrock, F., & Burrows, J. P. (2015). Ship-based MAX-DOAS measurements of trop- ospheric NO2 and SO2 in the South China and Sulu Sea. Atmospheric Environment, 102(2), 331–343. https://doi.org/10.1016/j. atmosenv.2014.12.015 Sherwen, T., Evans, M. J., Carpenter, L. J., Andrews, S. J., Lidster, R. T., Dix, B., et al. (2016). Iodine’s impact on tropospheric oxidants: A global model study in GEOS-Chem. Atmospheric Chemistry and Physics, 16(2), 1161–1186. https://doi.org/10.5194/acp-16-1161-2016 Sherwen, T., Evans, M. J., Sommariva, R., Hollis, L. D. J., Ball, S. M., Monks, P. S., et al. (2017). Effects of halogens on European air-quality. Faraday Discussions, 200, 75–100. https://doi.org/10.1039/c7fd00026j Sherwen, T., Schmidt, J. A., Evans, M. J., Carpenter, L. J., Großmann, K., Eastham, S. D., et al. (2016). Global impacts of tropospheric halogens (Cl, Br, I) on oxidants and composition in GEOS-Chem. Atmospheric Chemistry and Physics, 16(18), 12239–12271. https://doi. org/10.5194/acp-16-12239-2016 Simpson, W. R., Brown, S. S., Saiz-Lopez, A., Thornton, J. A., & Von Glasow, R. (2015). Tropospheric halogen chemistry: Sources, cycling, and impacts. Chemical Reviews, 115(10), 4035–4062. https://doi.org/10.1021/cr5006638 Stone, D., Sherwen, T., Evans, M. J., Vaughan, S., Ingham, T., Whalley, L. K., et al. (2018). Impacts of bromine and iodine chemistry on tropospheric OH and HO2: Comparing observations with box and global model perspectives. Atmospheric Chemistry and Physics, 18(5), 3541–3561. https://doi.org/10.5194/acp-18-3541-2018 Tang, L., Ramacher, M. O., Moldanová, J., Matthias, V., Karl, M., Johansson, L., et al. (2020). The impact of ship emissions on air quality and human health in the Gothenburg area—Part I: 2012 emissions. Atmospheric Chemistry and Physics Discussions(2), 1–36. https:// doi.org/10.5194/acp-2020-94 Travis, K. R., Heald, C. L., Allen, H. M., Apel, E. C., Arnold, S.R., Blake, D. R., et al. (2020). Constraining remote oxidation capacity with ATom observations. Atmospheric Chemistry and Physics, 20(13), 7753–7781. https://doi.org/10.5194/acp-20-7753-2020 Von Schneidemesser, E., Monks, P. S., Allan, J. D., Bruhwiler, L., Forster, P., Fowler, D., et al. (2015). Chemistry and the linkages between air quality and climate change. Chemical Reviews, 115(10), 3856–3897. https://doi.org/10.1021/acs.chemrev.5b00089 Wang, R., Tie, X., Li, G., Zhao, S., Long, X., Johansson, L., & An, Z. (2019). Effect of ship emissions on O3 in the Yangtze River Del- ta region of China: Analysis of WRF-Chem modeling. The Science of the Total Environment, 683, 360–370. https://doi.org/10.1016/j. scitotenv.2019.04.240 Wang, S., Hornbrook, R. S., Hills, A., Emmons, L. K., Tilmes, S., Lamarque, J. F., et al. (2019). Atmospheric acetaldehyde: Impor- tance of air–sea exchange and a missing source in the remote troposphere. Geophysical Research Letters, 46, 5601–5613. https://doi. org/10.1029/2019GL082034 Yang, X., Cox, R. A., Warwick, N. J., Pyle, J. A., Carver, G. D., O’Connor, F. M., & Savage, N. H. (2005). Tropospheric bromine chemistry and its impacts on ozone: A model study. Journal of Geophysical Research, 110, D23311. https://doi.org/10.1029/2005JD006244 Zhang, Y., Yang, X., Brown, R., Yang, L., Morawska, L., Ristovski, Z., et al. (2017). Shipping emissions and their impacts on air quality in China. The Science of the Total Environment, 581–582, 186–198. https://doi.org/10.1016/j.scitotenv.2016.12.098 LI ET AL. 17 of 17