Text S1
Aerosol size distributions between 10 – 100 nm were measured with a custom-built nano-Scanning Mobility Particle Sizer (nSMPS) consisting of a nano Differential Mobility Analyzer (TSI 3085) and a Condensation Particle Counter (TSI 3010) [Ortega et al. , 2019]. Scan times were 30s and continuous measurements were made during both ascending and descending voltage ramps. Aerosol size distributions were also measured by three optical probes. Two Ultra High Sensitivity Aerosol Spectrometers (UHSAS; Droplet Measurement Technologies) [Kupc et al. , 2018], one mounted on a wing and one mounted in a rack inside the plane, measured aerosol particles, with effective size ranges between 0.07 - 1.0 µm for the rack-mounted and 0.08 - 1.0 µm for the wing-mounted unit. For Figs. 3a and S1, nSMPS and UHSAS measurements were combined to create continuous size distributions by averaging dN/dlogDp values in the overlapping size ranges. A wing mounted Passive Cavity Aerosol Spectrometer (PCASP; Droplet Measurement Technologies) also measured particles between 0.1 and 3.0 µm.
A single column Cloud Condensation Nuclei Counter (CCNC; Droplet Measurement Technologies) was used to measure CCN number concentrations (NCCN) at set supersaturations between ≈ 0.1 – 0.6% [Roberts and Nenes , 2005] for at least 5 min at each supersaturation. Because supersaturation is a function of pressure, the CCNC sampled from a constant pressure inlet at 500 mb, and laboratory calibrations using size selected ammonium sulfate were performed at this pressure. Refractory black carbon (rBC) number and mass concentrations were measured with a Single Particle Soot Photometer (SP2; Droplet Measurement Technologies) [Schwarz et al. , 2006] which uses a 1064 nm Nd:YAG laser to heat absorbing material > 70 nm in diameter to its vaporization temperature and measures the resulting incandescence. Scattered laser light is used to determine total aerosol size and incandescence to determine black carbon mass. The SP2 sample line was diluted with HEPA-filtered ambient air to prevent signal saturation. The single scattering albedo (SSA) of the aerosol was calculated by dividing the difference of particulate absorption measured by a photoacoustic absorption spectrometer (PAS) [Foster et al. , 2019] and the extinction measured by a CAPS PMSSA [Onasch et al. , 2015] by the CAPS extinction. This approach was taken to avoid potential errors in the scattering channel of the CAPS PMSSA due to truncation errors from larger particles or nonlinearity that can occur in the optically thick plumes encountered during this study.
Submicron nonrefractory aerosol mass and composition were measured with a High-Resolution time-of-flight Aerosol Mass Spectrometer (HR-AMS; Aerodyne Inc.) with a pressure controlled inlet [Garofalo et al. , 2019]. For single particle composition, ambient particles were collected with the NCAR Solid Diffuser Inlet (SDI) (based on the University of Hawaii design of McNaughton et al. [2007]), and a counterflow virtual impactor (CVI) was used to collect evaporated cloud droplet residual particles [Noone et al. , 1988;Twohy et al. , 2003]. These individual aerosol and droplet residual particles were periodically collected onto carbon-coated electron microscope grids using a custom two-stage impactor. The impactor’s 50% cut sizes were about 0.5 and 0.1 µm dry diameter (assuming spherical particle densities of 1.5 g cm-3at typical sampling pressures of 600 mb).
The single-particle elemental composition of a subset of samples was measured via analytical Scanning Transmission Electron Microscopy (STEM) and X-ray spectroscopy at Colorado State University (JEOL JEM-2100F 200 kV microscope with an Oxford Max 80 Energy Dispersive Spectroscopy system). Particles were dividing into categories based on their elements and morphology. Organic particles were identified as those containing carbon and oxygen detectable above the background substrate, as well as sometimes potassium, nitrogen, sulfur and sodium. Because of their ubiquity in recent smoke plumes, these organic particles were presumed to be of biomass burning origin. They were also distinguishable from soot aggregates and biological entities by their quasi-round shape. A mineral dust, ash and metals category was for irregular particles with elevated levels of silicon, aluminum, iron, magnesium, zinc, titanium, often with elevated oxygen. Particles in the sulfate category contained quasi-spherical particles with sulfur and oxygen and without carbon detectable above background levels. Sea-salt based sea-spray particles were identified as those containing sodium with chlorine and/or sulfur with minor potassium, magnesium, and other elements present in seawater.
Cloud droplet number concentrations Nd were measured with a Droplet Measurement Technologies cloud droplet probe (CDP, 2-50 µm) modified with a pinpoint aperture to reduce errors due to coincidence [Lance et al. , 2010]. This modification is expected to minimize errors in droplet number concentration up to about 1850 cm-3, while measurements in clouds with higher droplet concentrations may under-report number concentrations and over-report droplet size [McFarquhar et al. , 2017]. In-cloud particles larger than 75 µm physical diameter were measured with a Particle Measuring Systems 2D-C optical array probe, modified with faster electronics for better response at high airspeeds, mounted on the left outboard wingpod location. Bulk cloud liquid water content was measured using the CSIRO/King hot-wire probe [King et al. , 1978].