Carley D. Fredrickson

Quantification of wildfire emissions is essential for comprehending and simulating the effects of wildfires on atmospheric chemical composition. Sub-orbital measurements of vertical column nitrous acid (HONO) and nitrogen dioxide (NO₂) were made during the Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) field campaign using the GeoCAPE Airborne Simulator (GCAS) instrument on board the NASA ER-2 aircraft. The emission rates and lifetimes of HONO and NO₂ from the Sheridan Fire were estimated by fitting exponentially modified Gaussians (EMGs) to line densities, a technique previously used to estimate urban and point source NO₂ emissions. As the EMG approach does not capture temporal changes in emissions and lifetimes due to time-varying fire behavior, we developed a Monte Carlo implementation of the Python Editable Chemical Atmospheric Numeric Solver (PECANS) model that includes diurnal fire radiative power (FRP) behavior. We assessed the validity of a range of emission rate and lifetime combinations for both HONO and NO₂ as the fire evolves by comparing the resulting line density predictions to the line density observations. We found that our method results in emissions that are lower than top-down biomass burning emissions inventories and higher than bottom-up inventories. Our approach is applicable to interpreting time-resolved remotely sensed measurements of atmospheric trace gases such as those now becoming available with instruments on board geostationary satellites such as the Tropospheric Emissions: Monitoring of Pollution (TEMPO) and the Geostationary Environment Monitoring Spectrometer (GEMS) instruments.

Quantification of wildfire emissions is essential for comprehending and simulating the effects of wildfires on atmospheric chemical composition. Sub-orbital measurements of vertical column nitrous acid (HONO) and nitrogen dioxide (NO₂) were made during the Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) field campaign using the GeoCAPE Airborne Simulator (GCAS) instrument aboard the NASA ER-2 aircraft. Emission rates and lifetimes of HONO and NO₂ from the Sheridan Fire were estimated by fitting exponentially modified Gaussians (EMGs) to line densities, a technique previously used to estimate urban and point source NO₂ emissions. As the EMG approach does not capture temporal changes in emissions and lifetimes due to time-varying fire behavior, we developed a Monte Carlo implementation of the Python Editable Chemical Atmospheric Numeric Solver (PECANS) model that includes diurnal fire radiative power (FRP) behavior. We assess the validity of a range of emission rate and lifetime combinations for both HONO and NO₂ as the fire evolves by comparing the resulting line density predictions to the observations. We find that our method results in emissions that are lower than top-down biomass burning emissions inventories and higher than bottom-up inventories. Our approach is applicable to interpreting time-resolved remotely sensed measurements of atmospheric trace gases such as those now becoming available with instruments aboard geo-stationary satellites such as the Tropospheric Emissions: Monitoring of Pollution (TEMPO) and the Geostationary Environment Monitoring Spectrometer (GEMS) instruments.

Heterogeneous chemical cycles of pyrogenic nitrogen and halides influence tropospheric ozone and affect the stratosphere during extreme Pyrocumulonimbus (PyroCB) events. We report field‐derived N₂O₅ uptake coefficients, γ(N₂O₅), and ClNO₂ yields, φ(ClNO₂), from two aircraft campaigns observing fresh smoke in the lower and mid troposphere and processed/aged smoke in the upper troposphere and lower stratosphere (UTLS). Derived φ(ClNO₂) varied across the full 0–1 range but was typically γ(N₂O₅) was low in agricultural smoke (0.2–3.6 × 10⁻³), extremely low in mid‐tropospheric wildfire smoke (0.1 × 10⁻³), but larger in PyroCB processed smoke (0.7–5.0 × 10⁻³). Aged biomass burning aerosol in the UTLS had a higher γ(N₂O₅) of 17 × 10⁻³ that increased with sulfate and liquid water, but that was 1–2 orders of magnitude lower than values for aqueous sulfuric aerosol used in stratospheric models.

Heterogeneous chemical cycles of pyrogenic nitrogen and halides influence tropospheric ozone and affect the stratosphere during extreme pyrocumulonimbus (PyroCB) events. We report field-derived N2O5 uptake coefficients, γ(N2O5), and ClNO2 yields, φ(ClNO2), from two aircraft campaigns observing fresh smoke in the lower and mid troposphere and processed/aged smoke in the upper troposphere and lower stratosphere (UTLS). Derived φ(ClNO2) varied across the full 0–1 range but was typically < 0.5 and smallest in a PyroCB (< 0.05). Derived γ(N2O5) was low in agricultural smoke (0.2–3.6 ×10-3), extremely low in mid-tropospheric wildfire smoke (0.1 × 10-3), but larger in PyroCB processed smoke (0.7–5.0 × 10–3). Aged BB aerosol in the UTLS had a higher median γ(N2O5) of 17 × 10–3 that increased with sulfate and liquid water, but that was nevertheless 1–2 orders of magnitude lower than values for aqueous sulfuric aerosol used in stratospheric models.

Wildfires are important sources of atmospheric reactive nitrogen. The reactive nitrogen species partitioning generally depends on fire characteristics. One reactive nitrogen compound, nitrous acid (HONO), is a source of hydroxyl radicals and nitric oxide, which can impact the oxidizing capacity of the atmosphere and fire plume chemistry and composition. We study the Australian wildfire season of 2019–2020, known as Black Summer, where numerous large and intense wildfires burned throughout the continent. We use HONO and nitrogen dioxide (NO₂) from the TROPOspheric Monitoring Instrument (TROPOMI) and fire radiative power (FRP) from the Visible Infrared Imaging Radiometer Suite to investigate HONO and NO₂ relationships with fire characteristics. The ratio of HONO to NO₂ increases linearly with FRP both in Australia and globally. Both Australian and global fire relationships depend strongly on land cover type. These relationships can be applied to emission inventories to improve wildfire emission representation in models.

Nitrous acid (HONO) mixing ratios measured in aged wildfire smoke plumes were higher than expected from known homogeneous chemical reactions. In a representative smoke plume, intercepted hours to days downwind of the source, the missing HONO source was highly correlated to particulate nitrate photolysis and NO₂ reactive uptake to particles. Using a multilinear regression involving these two sources, we could explain the missing HONO production in this plume (R² = 0.77). The resulting fit parameters from this plume had good explanatory power (R² = 0.64) for missing HONO production in other fire plumes. The mean enhancement factor for particulate nitrate photolysis relative to gas‐phase nitric acid photolysis was 63 and the mean NO₂ reactive uptake coefficient to submicron aerosol surface area forming HONO was 4.9 × 10⁻⁴. Given the likelihood of other neglected secondary HONO sources, these values are upper‐limits, suggesting a need to revisit HONO formation mechanisms in aged wildfire smoke.

Biomass burning is a major source of light‐absorbing organic aerosol (brown carbon), but its composition, chemical evolution, and lifetime are not well known. We measured water‐soluble brown carbon absorption from 310 to 500 nm on the National Oceanic and Atmospheric Administration Twin Otter aircraft during flights downwind of western United States wildfires in summer 2019. The sampling strategy was near‐Lagrangian and the plume ages spanned 0–5 hr. Trends in brown carbon mass absorption coefficient with plume age varied between flights, and did not show an exponential decay over these short time scales. The measured absorption spectra were smoothly varying, without identifiable contributions from individual chromophores with structured absorption. Using aerosol tracer ions and reference absorption spectra, the calculated contribution of 4‐nitrocatechol to total absorption was <22 ± 9% and <11 ± 5%, although spectral fitting showed that it may be as low as <1.1% and <0.6% at 365 and 405 nm, respectively.

Cloud condensation nuclei (CCN) are mediators of aerosol‐cloud interactions, which contribute to the largest uncertainty in climate change prediction. Here, we present a machine learning (ML)/artificial intelligence (AI) model that quantifies CCN from model‐simulated aerosol composition, atmospheric trace gas, and meteorological variables. Comprehensive multi‐campaign airborne measurements, covering varied physicochemical regimes in the troposphere, confirm the validity of and help probe the inner workings of this ML model: revealing for the first time that different ranges of atmospheric aerosol composition and mass correspond to distinct aerosol number size distributions. ML extracts this information, important for accurate quantification of CCN, additionally from both chemistry and meteorology. This can provide a physicochemically explainable, computationally efficient, robust ML pathway in global climate models that only resolve aerosol composition; potentially mitigating the uncertainty of effective radiative forcing due to aerosol‐cloud interactions (ERF_aci) and improving confidence in assessment of anthropogenic contributions and climate change projections.