Publications

The Arabian Peninsula experiences severe air pollution, the extent and sources of which are poorly documented. Each year in Saudi Arabia this situation is intensified during Hajj, the Holy Pilgrimage of Islam that draws millions of pilgrims to Mecca. An initial study of air quality in Mecca and surrounding holy sites during the 2012 Hajj (October 24-27) revealed strongly elevated levels of the combustion tracer carbon monoxide (CO, up to 57 ppmv) and volatile organic compounds (VOCs) along the pilgrimage route-especially in the tunnels of Mecca-that are a concern for human health. The most abundant VOC was the gasoline evaporation tracer i-pentane, which exceeded 1200 ppbv in the tunnels. Even though VOC concentrations were generally lower during a follow-up non-Hajj sampling period (April 2013), many were still comparable to other large cities suffering from poor air quality. Major VOC sources during the 2012 Hajj study included vehicular exhaust, gasoline evaporation, liquefied petroleum gas, and air conditioners. Of the measured compounds, reactive alkenes and CO showed the strongest potential to form ground-level ozone. Because the number of pilgrims is expected to increase in the future, we present emission reduction strategies to target both combustive and evaporative fossil fuel sources.

Airborne measurements of particles and gases from a 1000‐ha savanna fire in South Africa are presented. These measurements represent the most extensive data set reported on the aging of biomass smoke. The measurements include total concentrations of particles (CN), particle sizes, particulate organic carbon and black carbon, light‐scattering coefficients, downwelling UV fluxes, and mixing ratios for 42 trace gases and 7 particulate species. The ratios of excess nitrate, ozone, and gaseous acetic acid to excess CO increased significantly as the smoke aged over ∼40–45 min, indicating that these species were formed by photochemistry in the plume. For 17 other species, the excess mixing ratio normalized by the excess mixing ratio of CO decreased significantly with smoke age. The relative rates of decrease for a number of chemical species imply that the average OH concentration in the plume was ∼1.7 × 10 7 molecules cm −3 . Excess CN, normalized by excess CO, decreased rapidly during the first ∼5 min of aging, probably due to coagulation, and then increased, probably due to gas‐to‐particle conversion. The CO‐normalized concentrations of particles <1.5 μm in diameter decreased, and particles >1.5 μm diameter increased, with smoke age. The spectral depletion of solar radiation by the smoke is depicted. The downwelling UV flux near the vertical center of the plume was about two‐thirds of that near the top of the plume.

Eight inorganic ions in fine aerosol particles (D p < 1.3 μm) were measured on board the NCAR C130 and NASA P‐3B aircraft during the 2001 Aerosol Characterization Experiment (ACE)‐Asia and the Transport and Chemical Evolution over the Pacific (TRACE‐P) experiments, respectively. Concentrations of NH 4 + , SO 4 2− , NO 3 − , Ca 2+ , K + , Mg 2+ , Na + , and Cl − were determined using a particle‐into‐liquid sampler coupled to ion chromatography (PILS‐IC) technique at a 4‐min resolution and a limit of detection <0.05 μg m −3 . The maximum total ion concentrations observed on the C130 and the P‐3B were 27 μg m −3 and 84 μg m −3 , respectively. During ACE‐Asia, NH 4 + and SO 4 2− dominated, with the dust‐derived Ca 2+ contributing nearly equally as SO 4 2− in mixing ratios. The sea‐salt‐derived Na + and Cl − were comparable to biomass‐burning tracer K + , showing >1 ppbv only in the top 1% sample population. During TRACE‐P, NH 4 + dominated, followed by SO 4 2− , Cl − , Na + , NO 3 − , Ca 2+ , and K + , in decreasing order of importance. In addition to a sea‐salt origin, Cl − showed a source in urban emissions possibly related to biofuel combustion. Both sea salt and dust contributed to Mg 2+ . In both experiments, NH 4 + , SO 4 2− , NO 3 − , and CO were strongly correlated, indicating that combustion was the dominant source of these species and that NH 3 and other alkaline materials were in sufficient supply to neutralize H 2 SO 4 . The [NH 4 + ] to ([NO 3 − ] + 2[SO 4 2− ]) ratio was ∼0.70 in the two campaigns, with deviations found only in volcano plumes, whereby SO 4 2− was found to correlate with SO 2 . Charge balance of the ions showed both positive and negative deviations whose magnitudes, ∼30%, provide estimates of the lower limits of unmeasured ions. Elevated NO 3 − and Ca 2+ coexist mainly under polluted conditions, suggesting the importance of sequestering HNO 3 by mineral dust.

peer reviewed

Abstract. The Community Earth System Model (CESM1) CAM4-chem has been used to perform the Chemistry Climate Model Initiative (CCMI) reference and sensitivity simulations. In this model, the Community Atmospheric Model version 4 (CAM4) is fully coupled to tropospheric and stratospheric chemistry. Details and specifics of each configuration, including new developments and improvements are described. CESM1 CAM4-chem is a low-top model that reaches up to approximately 40 km and uses a horizontal resolution of 1.9° latitude and 2.5° longitude. For the specified dynamics experiments, the model is nudged to Modern-Era Retrospective Analysis for Research and Applications (MERRA) reanalysis. We summarize the performance of the three reference simulations suggested by CCMI, with a focus on the last 15 years of the simulation when most observations are available. Comparisons with selected data sets are employed to demonstrate the general performance of the model. We highlight new data sets that are suited for multi-model evaluation studies. Most important improvements of the model are the treatment of stratospheric aerosols and the corresponding adjustments for radiation and optics, the updated chemistry scheme including improved polar chemistry and stratospheric dynamics and improved dry deposition rates. These updates lead to a very good representation of tropospheric ozone within 20 % of values from available observations for most regions. In particular, the trend and magnitude of surface ozone is much improved compared to earlier versions of the model. Furthermore, stratospheric column ozone of the Southern Hemisphere in winter and spring is reasonably well represented. All experiments still underestimate CO most significantly in Northern Hemisphere spring and show a significant underestimation of hydrocarbons based on surface observations.

The distribution of reactive nitrogen species over the South Pacific during austral springtime appears to be dominated by biomass burning emissions and possibly lightning and stratospheric inputs. The absence of robust correlations of reactive nitrogen species with source‐specific tracers (e.g., C 2 H 2 [combustion], CH 3 Cl [biomass burning], C 2 Cl 4 [industrial], 210 Pb [continental], and 7 Be [stratospheric]) suggests significant aging and processing of the sampled air parcels due to losses by surface deposition, OH attack, and dilution processes. Classification of the air parcels based on CO enhancements indicates that the greatest influence was found in plumes at 3–8 km altitude in the distributions of HNO 3 and peroxyacetyl nitrate (PAN). Here mixing ratios of these species reached 600 parts per trillion by volume (pptv), values surprisingly large for a location several thousand kilometers removed from the nearest continental areas. The mixing ratio of total reactive nitrogen (the NO y sum), operationally defined in this paper as measured (NO + HNO 3 + PAN + CH 3 ONO 2 + C 2 H 5 ONO 2 ) + modeled (NO 2 ), had a median value of 285 pptv within these plumes compared with 120 pptv in nonplume air parcels. Particle NO − 3 was not included in this analysis of the NO y sum due to its 10‐ to 15‐min sampling time resolution, but, in general, it was <10% of the NO y sum. Comparison of the two air parcel classifications for NO y and alkyl nitrate distributions showed no perceivable plume influence, but recycling of reactive nitrogen may have masked this direct effect. In the marine boundary layer, the NO y sum averaged 50 pptv in both air parcel classifications, being somewhat isolated from the polluted conditions above it by the trade wind inversion. In this region, however, alkyl nitrates appear to have an important marine source where they comprise 20–80% of the NO y sum in equatorial and high‐latitude regions over the South Pacific.

Emissions and secondary photochemical products from the Daesan petrochemical complex (DPCC), on the west coast of South Korea, were measured from the NASA DC-8 research aircraft during the Korea-United States Air Quality campaign in 2016. The chemical evolution of petrochemical emissions was examined utilizing near-source and downwind plume transects. Small alkenes, such as ethene (C2H4), propene (C3H6), and 1,3-butadiene (C4H6), dominated the hydroxyl (OH) radical reactivity near the source region. The oxidation of these alkenes in the petrochemical plumes led to efficient conversion of nitrogen oxides (NOx) to nitric acid (HNO3), peroxycarboxylic nitric anhydrides (PANs), and alkyl nitrates (ANs), where the sum of the speciated reactive nitrogen contributes more than 80% of NOy within a few hours. Large enhancements of short-lived NOx oxidation products, such as hydroxy nitrates (HNs) and peroxyacrylic nitric anhydride, were observed, in conjunction with high ozone levels of up to 250 ppb, which are attributed to oxidation of alkenes such as 1,3-butadiene. Instantaneous ozone production rates, P(O3), near and downwind of the DPCC ranged from 9 to 24 ppb h−1, which were higher than those over Seoul. Ozone production efficiencies ranged from 6 to 10 downwind of the DPCC and were lower than 10 over Seoul. The contributions of alkenes to the instantaneous secondary organic aerosol (SOA) production rate, P(SOA), were estimated to be comparable to those of more common SOA precursors such as aromatics at intermediate distances from the DPCC. A model case study constrained to an extensive set of observations provided a diagnostic of petrochemical plume chemistry. The simulated plume chemistry reproduced the observed evolution of ozone and short-lived reactive nitrogen compounds, such as PANs and HNs as well as the rate and efficiency of ozone production. The simulated peroxy nitrates (PNs) budget included large contributions (approximately 30%) from unmeasured PNs including peroxyhydroxyacetic nitric anhydride and peroxybenzoic nitric anhydride. The large, predicted levels of these PAN compounds suggest their potential importance in chemical evolution of petrochemical plumes. One unique feature of the DPCC plumes is the substantial contribution of 1,3-butadiene to ozone and potentially SOA production. This work suggests that reductions in small alkene, especially 1,3-butadiene, emissions from the DPCC should be a priority for reducing downwind ozone.

Abstract. Reactive halogens catalytically destroy O3 and therefore affect (1) stratospheric O3 depletion and (2) the oxidative capacity of the troposphere. Reactive halogens also partition into the aerosol phase, but what governs halogen-aerosol partitioning is poorly constrained in models. In this work, we present global-scale measurements of non-sea-salt aerosol (nSSA) bromine and iodine taken during the NASA Atmospheric Tomography Mission (ATom). Using the Particle Analysis by Laser Mass Spectrometry instrument, we found that bromine and iodine are present in 8 %–26 % (interquartile range, IQR) and 12 %–44 % (IQR) of accumulation-mode nSSA, respectively. Despite being commonly found in nSSA, the concentrations of bromine and iodine in nSSA were low but potentially important, at 0.11–0.57 pmol mol−1 (IQR) and 0.04–0.24 pmol mol−1 (IQR), respectively. In the troposphere, we find two distinct sources of bromine and iodine for nSSA: (1) a primary source from biomass burning and (2) a pervasive secondary source. In the stratosphere, nSSA bromine and iodine concentrations increased with increasing O3 concentrations; however, higher concentrations of stratospheric nSSA bromine and iodine were found in organic-rich particles that originated in the troposphere. Finally, we compared our ATom nSSA iodine measurements to the global chemical transport model GEOS-Chem (Goddard Earth Observing System); nSSA bromine concentrations could not be compared because they were not tracked in the model. We found that the model compared well to our ATom nSSA iodine measurements in the background atmosphere but not in the marine boundary layer, biomass burning plumes, or stratosphere.

The concentration of nitrogen oxides (NO _x) plays a central role in controlling air quality. On a global scale, the primary sink of NO _x is oxidation to form HNO₃. Gas-phase HNO₃ photolyses slowly with a lifetime in the troposphere of 10 days or more. However, several recent studies examining HONO chemistry have proposed that particle-phase HNO₃ undergoes photolysis 10-300 times more rapidly than gas-phase HNO₃. We present here constraints on the rate of particle-phase HNO₃ photolysis based on observations of NO _x and HNO₃ collected over the Yellow Sea during the KORUS-AQ study in summer 2016. The fastest proposed photolysis rates are inconsistent with the observed NO _x to HNO₃ ratios. Negligible to moderate enhancements of the HNO₃ photolysis rate in particles, 1-30 times faster than in the gas phase, are most consistent with the observations. Small or moderate enhancement of particle-phase HNO₃ photolysis would not significantly affect the HNO₃ budget but could help explain observations of HONO and NO _x in highly aged air.

Carbon tetrachloride (CCl4) is an ozone-depleting substance, accounting for about 10% of the chlorine in the troposphere. Under the terms of the Montreal Protocol, its production for dispersive uses was banned from 2010. In this work we show that, despite the controls on production being introduced, CCl4 emissions from the eastern part of China did not decline between 2009 and 2016. This finding is in contrast to a recent bottom-up estimate, which predicted a significant decrease in emissions after the introduction of production controls. We find eastern Asian emissions of CCl4 to be 16 (9-24) Gg/year on average between 2009 and 2016, with the primary source regions being in eastern China. The spatial distribution of emissions that we derive suggests that the source distribution of CCl4 in China changed during the 8-year study period, indicating a new source or sources of emissions from China's Shandong province after 2012.

Air samples were collected concurrently at 05:00 A.M. and 10:00 A.M. local Beijing time (geomagnetic time + 8) at 84 sites during two grid-study campaigns on 29 September 2008 and 1 March 2009 in the Pearl River Delta region, in order to offer snapshots of ambient CFCs and hydrochlorofluorocarbons (HCFCs) in different seasons and to indicate the presence of local emission sources. Compared to the subtropical northern hemisphere background levels, mean mixing ratios of CFC-11, CFC-12, CFC-113, CFC-114, and HCFC-22 were enhanced by 7%–11%, 8%–11%, 5%–6%, 8%–9%, and 71%–135%, respectively. When data from this tudy were pooled together with previous observations in the region, ambient CFC-11, CFC-12, and CFC-113 unambiguously showed declines in mixing ratios, while HCFC-22 showed an increase. Spatial variations revealed potential emission hot spots in the region, and levels of CFCs and HCFC-22 were higher in September than in March due to many more refrigeration and air-conditioning activities during summer. Source apportioning by positive matrix factorization revealed that new input of CFCs and HCFC-22 into the ambient air was largely attributed to emission from air-conditioning and refrigerating activities instead of industry activities. Average emissions in the region estimated by the CO-tracer method were 0.8 ± 0.2, 1.4 ± 0.6, 0.2 ± 0.1, 0.1 ± 0.02, and 4.4 ± 1.0 Gg/yr for CFC-11, CFC-12, CFC-113, CFC-114, and HCFC-22, respectively, and they accounted for 5.5%–25.5% of the total estimated CFC and HCFC-22 emissions in China.