Particle number to volume concentration ratios at two measurement sites in Finland

Abstract

[1] Past studies have indicated that the aerosol number to volume concentration ratio R, defined as the number concentration of particles larger than a certain cutoff diameter divided by the total particle volume concentration, shows little variability in marine conditions. In this study, aerosol number size distributions from Pallas in northern Finland and from the island of Utö in the Baltic Sea, representing more continental aerosol systems, were analyzed. The data set covered continuous measurements for 20 months at Pallas and 21 months at Utö. The average value and standard deviation of R, calculated from the measured data, were found to be inversely dependent on the cutoff diameter used, Also, the relative deviation (standard deviation divided by average value) of R(d c ) was found to be dependent on the cutoff diameter. A clear seasonal pattern of the values of R was observed at both sites, and this was found to be related to ambient temperature. The value of R was found to depend on the air mass type at Pallas, but at Utö such dependency was not observed. Besides temperature, R was also analyzed against several meteorological parameters, including global solar radiation, air pressure, relative humidity, and rain. The found dependencies between the number of potential cloud condensation nuclei and particle volume concentration provide tools for developing more accurate empirical parameterizations describing cloud formation in large-scale atmospheric models.

Pallas cloud experiment, PaCE 2012

“…Sammaltunturi station is, due to topography of the surrounding terrain, a great place for ground-based observations of orographic clouds. Thus providing an opportunity to investigate not only the cloud droplet activation of aerosol particles but also directly the cloud particle phase [3,5,6,7]. In autumn, probability the station to be inside cloud is high.…”

Parameterization of cloud droplet activation using a simplified treatment of the aerosol number size distribution

“…The main reason for using R ( d c ) in our analysis is that in all aerosol systems investigated so far, this quantity has been found to vary much less than what random size distributions would produce [ Van Dingenen et al , 2000; Hegg and Russell , 2000; Dusek et al , 2004; Eleftheriadis et al , 2006; Kivekäs et al , 2007]. In previous studies the cut‐off diameter d c has been chosen to be between 80 and 120 nm.…”

“…The variability of R depends on d c , being larger for smaller values of d c . By using a cut‐off diameter of 100 nm, the average values of R (0.1 μ m) in northern European background air were found to be approximately 200 ± 70 μ m −3 [ Kivekäs et al , 2007]. A seasonal pattern for R was found, and it was found to correlate well with the seasonal pattern of the ambient temperature.…”

“…A seasonal pattern for R was found, and it was found to correlate well with the seasonal pattern of the ambient temperature. At higher temperatures the value of R was, on average, higher and varied more [ Kivekäs et al , 2007]. In this parameterization, a value of 100 nm was used for d c .…”

“…In a paper by Kivekäs et al [2007] the magnitude of R ( d c ) was shown to depend inversely on the cut‐off diameter d c . This dependency is used in the second part of the parameterization for relating R ( D 50) to R (0.1 μ m): The constant factor of 0.10 in the fit was estimated by linear fitting so that it gives the minimum sum of weighted differences between the simulated and parameterized values.…”