Title: Linking variations in sea spray aerosol particle hygroscopicity to composition during two microcosm experiments
"Keywords: sea spray aerosols, light scattering, hygroscopicity"
Timeframe for the work start of experimentation: 7/2014/ publication: 7/2016
Contact for further info regarding this dataset: Christopher Cappa (cdcappa@ucdavis.edu) or Sara Forestieri (sforest@ucdavis.edu)
README for data in Figures 2-5 of Forestieri et al. 2016
Figure 2 
"This file contains daily and average size distributions for the ""indoor"" and   outdoor   MARTs"""
Size distributions for dried particles (RH < 20%) were measured with a scanning electrical
"mobility sizer (SEMS; BMI; model 2002), and an aerodynamic particle sizer (APS; TSI Inc.;"""
Model 3321). The SEMS combines a differential mobility analyzer (DMA) and a mixing
condensation particle counter (MCPC) to characterize particles according to their mobility
"diameter (dp,m). The APS characterizes particles according to their aerodynamic diameter (dp,a)."
"The SEMS characterized particles over the range 10 nm < dp,m < 1900 nm and the APS over the"""
"range 0.7 ?m < dp,a < 20 ?m. The SEMS size distributions were corrected for the influence of"""
multiply charged particles using software provided by the manufacturers. No diffusion correction
"was performed, which has negligible influence on this study because the smallest particles, which"
"are sensitive to diffusion corrections, contribute negligibly to the observed scattering. The APS"
"had a time resolution of 1 minute, while the SEMS had a time resolution of 5 minutes and the APS"
distributions were accordingly averaged to 5 minutes to facilitate generation of a merged size
distribution. The SEMS and APS distributions were merged using the SEMS distribution up to 1 um 
"and the (dp,m equivalent) APS distribution at larger diameters. The APS dp,a values were"
converted to mobility equivalent values assuming a particle density of 1.8 g cm-3. Note that the
number of size bins varied throughout the study.

The study average integrated scattering for each MART
was calculated from Mie theory using the observed dry particle size distributions (Figure 2C). The
diameters at which 50% of the total scattering occurs were 570 nm for the outdoor MART and 530
"nm for the indoor MART and particles with dp,m > 1000 nm contributed <10% of the total scattering"""
"in both MARTs,, indicating that the derived GF(85%) values for these two experiments are most"""
"sensitive to submicron particles with dp,m values between about 400 nm and 800 nm."

Figure 3
This file contains the daily averages and standard deviations of light scattering measurements 
"made by a cavity ringdown spectrometer (CRD), growth factors derived from optical measurements and the "
relative humidity of the wet CRD channel  for the indoor MART. Also included are a time series of 
"chlorophyll-a, SSA organic volume fractions  measured by  an aerodyne aerosol mass spectrometer (AMS), "
and single particle composition quantified by an aerosol time of flight mass spectrometer ATOFMS.

The hygroscopicity of the SSA particles was characterized through simultaneous measurements of light 
"extinction coefficients (equivalent to scattering coefficents, since absorption was measured to be negligible)"
"using a home-built UC Davis Cavity-Ringdown Spectrometer (CRD) (Langridge et al., 2011; Cappa et al., 2012). "
The(bext) for particles that were either dried to RH < 20 % (“dry”) or humidified to RH = 85 % (“wet”). 
The f (RH) values (wet scattering/dry scattering) measured using the CRD instrument have been used to 
determine optically weighted physical growth factors (the algorithm is described further in 
"Zhang et al. 2013). For particles of a given size, the growth factor is defined as GF(RH) = dp(RH_high)/ dp (RH_low), "
"where dp is the geometric particle diameter, which is equivalent to dp,m for spherical particles. As the relative humidity (RH) of the "
"humidified channel was not perfectly constant during measurements, the derived individual GF(RH) values have been "
"adjusted to 85 % by using Eq. 11 in Petters and Kreidenweis 2007. For each sampling day, a combination of the average f(RH) value,"
"average RH, and the average daily size distribution provided in the Figure 2 document can be used to calculate an average"
growth factor at RH = 85%.  

"An aerosol time-of-flight mass spectrometer (ATOFMS) (Gard et al., 1998; Pratt et al., 2009) was "
used to characterize the composition of individual dried SSA particles with vacuum aerodynamic 
"diameters (dva) from ? 300 nm to 3 µm, with the highest transmission and sampling of particles with "
"dva ~ 1–2 µm (Wang et al., 2015). The ATOFMS single particle spectra have been analyzed using a "
"statistical clustering algorithm (ART-2a) that groups particles with similar spectra together (Zhao et al., 2008). "
"Five particle mass spectra categories were generated and are described as sea salt (SS), salt mixed with organic carbon (SSOC),  "
"predominately organic carbon (OC), containing a large iron peak (Fe), and containing a large magnesium peak (Mg)  "
"(Lee et al., 2015; Sultana et al., 2016; Wang et al., 2015). "

Figure 4
This file contains growth factors for both indoor and outdoor MARTs as a function of AMS volume fraction.
Measurements are described in the description for Figure 3.

Figure 5
" This file contains a time series of chlorophyll-a, SSA organic volume fractions measured by  "
" an aerodyne aerosol mass spectrometer (AMS), and single particle composition quantified by an  "
aerosol time of flight mass spectrometerATOFMS for the outdoor MART.

Table S1

First row: In this table sensitivity calculations were made using a typical f(RH) value (wet scattering/dry scattering) 
"at 85% relative humidity. This f(RH) value and the average size distribution for the ""indoor"" MART were used to calcualte "
a baseline growth factor value. The f(RH) was then adjusted by 0.05 (or 7%). This new value and the average size distribution were 
used to calculate a new growth factor. The delta growth factor is the difference between the new growth factor and the baseline 
growth factor. 
"Second row: Similarly, using the same f(RH) value in the first row, a baseline growth factor was calculated assuming RH=85%. The RH was "
nudged by estimated uncertainties of 1.2%/0.3% (see Table S2) and new growth factors were calculated using equation 11 from Petters and 
Kreidenweis 2007. 
Third row: An uncertainty of 0.04 was estimated for the refractive index of sea spray aerosols. A baseline growth factor was calulcated 
using the original f(RH) value and a refractive index of 1.55 and a new growth factor was calculated using a refractive index of 1.51. 
"Fourth row: A baseline growth factor was calculated using the average ""indoor"" MART distribution. A new growth factor was calcualted "
"using the same distribution, but with diameters nudged by 1%. "

Table S2

Uncertainty in RH was assessed by comparing optically-derived growth factors pure atomized NaCl and Amm. Sulfate (observed) to literature values
"(actual). For NaCl, the ratio f(RH) (wet extinction/dry extinction) was 6.05 +/- 0.50 at RH = 78.32  +/- 0.11 and, for amm. sulfate, "
the value for f(RH) was 2.36 +/- 0.13 at an RH = 78.56  +/- 0.67. The growth factors calculated from these f(RH) values were adjusted to 
RH = 85% using Eqn. 11 from Petters and Kreidenweis 2007.