Search for: All records

Abstract. The hydroxyl (OH), hydroperoxy (HO2), and organic peroxy (RO2)radicals play important roles in atmospheric chemistry. In the presence ofnitrogen oxides (NOx), reactions between OH and volatile organiccompounds (VOCs) can initiate a radical propagation cycle that leads to theproduction of ozone and secondary organic aerosols. Previous measurements ofthese radicals under low-NOx conditions in forested environmentscharacterized by emissions of biogenic VOCs, including isoprene andmonoterpenes, have shown discrepancies with modeled concentrations. During the summer of 2016, OH, HO2, and RO2 radical concentrationswere measured as part of the Program for Research on Oxidants:Photochemistry, Emissions, and Transport – Atmospheric Measurements ofOxidants in Summer (PROPHET-AMOS) campaign in a midlatitude deciduousbroadleaf forest. Measurements of OH and HO2 were made by laser-inducedfluorescence–fluorescence assay by gas expansion (LIF-FAGE) techniques,and total peroxy radical (XO2) mixing ratios were measured by the Ethane CHemical AMPlifier (ECHAMP) instrument. Supporting measurements ofphotolysis frequencies, VOCs, NOx, O3, and meteorological datawere used to constrain a zero-dimensional box model utilizing either theRegional Atmospheric Chemical Mechanism (RACM2) or the Master ChemicalMechanism (MCM). Model simulations tested the influence of HOxregeneration reactions within the isoprene oxidation scheme from the LeuvenIsoprene Mechanism (LIM1). On average, the LIM1 models overestimated daytimemaximum measurements by approximately 40 % for OH, 65 % for HO2,and more than a factor of 2 for XO2. Modeled XO2 mixing ratioswere also significantly higher than measured at night. Addition of RO2 + RO2 accretion reactions for terpene-derived RO2 radicals tothe model can partially explain the discrepancy between measurements andmodeled peroxy radical concentrations at night but cannot explain thedaytime discrepancies when OH reactivity is dominated by isoprene. Themodels also overestimated measured concentrations of isoprene-derivedhydroxyhydroperoxides (ISOPOOH) by a factor of 10 during the daytime,consistent with the model overestimation of peroxy radical concentrations.Constraining the model to the measured concentration of peroxy radicalsimproves the agreement with the measured ISOPOOH concentrations, suggestingthat the measured radical concentrations are more consistent with themeasured ISOPOOH concentrations. These results suggest that the models maybe missing an important daytime radical sink and could be overestimating therate of ozone and secondary product formation in this forest. 

Abstract. Accurate representation of aerosol optical properties is essential for the modeling and remote sensing of atmospheric aerosols. Although aerosol optical properties are strongly dependent upon the aerosol size distribution, the use of detailed aerosol microphysics schemes in global atmospheric models is inhibited by associated computational demands. Computationally efficient parameterizations for aerosol size are needed. Inthis study, airborne measurements over the United States (DISCOVER-AQ) andSouth Korea (KORUS-AQ) are interpreted with a global chemical transport model (GEOS-Chem) to investigate the variation in aerosol size when organicmatter (OM) and sulfate–nitrate–ammonium (SNA) are the dominant aerosol components. The airborne measurements exhibit a strong correlation (r=0.83) between dry aerosol size and the sum of OM and SNA mass concentration (MSNAOM). A global microphysical simulation(GEOS-Chem-TOMAS) indicates that MSNAOM and theratio between the two components (OM/SNA) are the major indicators for SNA and OM dry aerosol size. A parameterization of the dry effective radius (Reff) for SNA and OM aerosol is designed to represent the airborne measurements (R2=0.74; slope = 1.00) and the GEOS-Chem-TOMAS simulation (R2=0.72; slope = 0.81). When applied in the GEOS-Chem high-performance model, this parameterization improves the agreement between the simulated aerosol optical depth (AOD) and the ground-measured AOD from the Aerosol Robotic Network (AERONET; R2 from 0.68 to 0.73 and slope from 0.75 to 0.96). Thus, this parameterization offers a computationally efficient method to represent aerosol size dynamically. 

Abstract Wildfire emissions affect downwind air quality and human health. Predictions of these impacts using models are limited by uncertainties in emissions and chemical evolution of smoke plumes. Using high‐time‐resolution aircraft measurements, we illustrate spatial variations that can exist within a plume due to differences in the photochemical environment. Horizontal and vertical crosswind gradients of dilution‐corrected mixing ratios were observed in midday plumes for reactive compounds and their oxidation products, such as nitrous acid, catechol, and ozone, likely due to faster photochemistry in optically thinner plume edges relative to darker plume cores. Gradients in plumes emitted close to sunset are characterized by titration of O₃in the plume and reduced or no gradient formation. We show how crosswind gradients can lead to underestimated emission ratios for reactive compounds and overestimated emission ratios for oxidation products. These observations will lead to improved predictions of wildfire emissions, evolution, and impacts across daytime and nighttime. 

Abstract Reactive nitrogen (N_r) within smoke plumes plays important roles in the production of ozone, the formation of secondary aerosols, and deposition of fixed N to ecosystems. The Western Wildfire Experiment for Cloud Chemistry, Aerosol Absorption, and Nitrogen (WE‐CAN) field campaign sampled smoke from 23 wildfires throughout the western U.S. during summer 2018 using the NSF/NCAR C‐130 research aircraft. We empirically estimateN_rnormalized excess mixing ratios and emission factors from fires sampled within 80 min of estimated emission and explore variability in the dominant forms ofN_rbetween these fires. We find that reduced N compounds comprise a majority (39%–80%; median = 66%) of total measured reactive nitrogen (ΣN_r) emissions. The smoke plumes sampled during WE‐CAN feature rapid chemical transformations after emission. As a result, within minutes after emission total measured oxidized nitrogen (ΣNO_y) and measured totalΣNH_x(NH₃ + pNH₄) are more robustly correlated with modified combustion efficiency (MCE) than NO_xand NH₃by themselves. The ratio of ΣNH_x/ΣNO_ydisplays a negative relationship with MCE, consistent with previous studies. A positive relationship with total measuredΣN_rsuggests that both burn conditions and fuel N content/volatilization differences contribute to the observed variability in the distribution of reduced and oxidizedN_r. Additionally, we compare our in situ field estimates ofN_rEFs to previous lab and field studies. For similar fuel types, we findΣNH_xEFs are of the same magnitude or larger than lab‐based NH₃EF estimates, andΣNO_yEFs are smaller than lab NO_xEFs. 

Abstract Wildfires are a major source of gas‐phase ammonia (NH₃) to the atmosphere. Quantifying the evolution and fate of this NH₃is important to understanding the formation of secondary aerosol in smoke and its accompanying effects on radiative balance and nitrogen deposition. Here, we use data from the Western Wildfire Experiment for Cloud Chemistry, Aerosol Absorption, and Nitrogen (WE‐CAN) to add new empirical constraints on the e‐folding loss timescale of NH₃and its relationship with particulate ammonium (pNH₄) within wildfire smoke plumes in the western U.S. during summer 2018. We show that the e‐folding loss timescale of NH₃with respect to particle‐phase partitioning ranges from ∼24 to ∼4000 min (median of 55 min). Within these same plumes, oxidation of nitrogen oxides is observed concurrent with increases in the fraction ofpNH₄in each plume sampled, suggesting that formation of ammonium nitrate (NH₄NO₃) is likely. We find wide variability in how close ourin situmeasurements of NH₄NO₃are to those expected in a dry thermodynamic equilibrium, and find that NH₄NO₃is most likely to form in fresh, dense smoke plumes injected at higher altitudes and colder temperatures. In chemically older smoke we observe correlations between both the fraction ofpNH₄and the fraction of particulate nitrate (pNO₃) in the aerosol with temperature, providing additional evidence of the presence of NH₄NO₃and the influence of injection height on gas‐particle partitioning of NH₃. 

Abstract The Western Wildfire Experiment for Cloud Chemistry, Aerosol Absorption, and Nitrogen (WE‐CAN) deployed the NSF/NCAR C‐130 aircraft in summer 2018 across the western U.S. to sample wildfire smoke during its first days of atmospheric evolution. We present a summary of a subset of reactive oxidized nitrogen species (NO_y) in plumes sampled in a pseudo‐Lagrangian fashion. Emissions of nitrogen oxides (NO_x = NO + NO₂) and nitrous acid (HONO) are rapidly converted to more oxidized forms. Within 4 h, ∼86% of the ΣNO_yis in the form of peroxy acyl nitrates (PANs) (∼37%), particulate nitrate (pNO₃) (∼27%), and gas‐phase organic nitrates (Org N_(g)) (∼23%). The averagee‐folding time and distance for NO_xare ∼90 min and ∼40 km, respectively. Nearly no enhancements in nitric acid (HNO₃) were observed in plumes sampled in a pseudo‐Lagrangian fashion, implying HNO₃‐limited ammonium nitrate (NH₄NO₃) formation, with one notable exception that we highlight as a case study. We also summarize the observed partitioning of NO_yin all the smoke samples intercepted during WE‐CAN. In smoke samples intercepted above 3 km above sea level (ASL), the contributions of PANs andpNO₃to ΣNO_yincrease with altitude. WE‐CAN also sampled smoke from multiple fires mixed with anthropogenic emissions over the California Central Valley. We distinguish samples where anthropogenic NO_xemissions appear to lead to an increase in NO_xabundances by a factor of four and contribute to additional PAN formation. 

Abstract We present a comparison of instruments measuring nitrogen oxide species from an aircraft during the 2015 Wintertime INvestigation of Transport, Emissions, and Reactivity (WINTER) campaign over the northeast United States. Instrument techniques compared here include chemiluminescence (CL), thermal dissociation laser‐induced fluorescence (TD‐LIF), cavity ring‐down spectroscopy (CRDS), high‐resolution time of flight, iodide‐adduct chemical ionization mass spectrometry (I^‐CIMS), and aerosol mass spectrometry. Species investigated include NO₂, NO, total nitrogen oxides (NO_y), N₂O₅, ClNO₂, and HNO₃. Particulate‐phase nitrate is also included for comparisons of HNO₃and NO_y. Instruments generally agreed within reported uncertainties, with individual flights sometimes showing much better agreement than the data set taken as a whole, due to flight‐to‐flight slope changes. NO measured by CRDS and CL showed an average relative slope of 1.16 ± 0.01 across all flights, which is outside of combined uncertainties. The source of the error was not identified. For NO₂measured by CRDS and TD‐LIF the average was 1.02 ± 0.00; for NO_ymeasured by CRDS and CL the average was 1.01 ± 0.00; and for N₂O₅measured by CRDS and I^‐CIMS the average was 0.89 ± 0.01. NO_ybudget closure to within 20% is demonstrated. We observe nonlinearity in NO₂and NO_ycorrelations at concentrations above ~30 ppbv that may be related to the NO discrepancy noted above. For ClNO₂there were significant differences between I^‐CIMS and TD‐LIF, potentially due in part to the temperature used for thermal dissociation. Although the fraction of particulate nitrate measured by the TD‐LIF is not well characterized, it improves comparisons to include particulate measurements.