Abstract. A quality-controlled, 5-year dataset of aerosol number
size distributions (particles with diameters (Dp) from 7 nm through 14 µm) was developed using observations from a scanning mobility particle
sizer, aerodynamic particle sizer, and a condensation particle counter at
the Department of Energy's Southern Great Plains (SGP) site. This dataset
was used for two purposes. First, typical characteristics of the aerosol
size distribution (number, surface area, and volume) were calculated for the
SGP site, both for the entire dataset and on a seasonal basis, and size
distribution lognormal fit parameters are provided. While the median size
distributions generally had similar shapes (four lognormal modes) in all the
seasons, there were some significant differences between seasons. These
differences were most significant in the smallest particles
(Dp<30 nm) and largest particles (Dp>800 nm).
Second, power spectral analysis was conducted on this long-term dataset to
determine key temporal cycles of total aerosol concentrations, as well as
aerosol concentrations in specified size ranges. The strongest cyclic signal
was associated with a diurnal cycle in total aerosol number concentrations
that was driven by the number concentrations of the smallest particles
(Dp<30 nm). This diurnal cycle in the smallest particles
occurred in all seasons in ∼50 % of the observations,
suggesting a persistent influence of new particle formation events on the
number concentrations observed at the SGP site. This finding is in contrast with earlier
studies that suggest new particle formation is observed primarily in the
springtime at this site. The timing of peak concentrations associated with
this diurnal cycle was shifted by several hours depending on the season,
which was consistent with seasonal differences in insolation and boundary
layer processes. Significant diurnal cycles in number concentrations were
also found for particles with Dp between 140 and 800 nm, with peak
concentrations occurring in the overnight hours, which were primarily
associated with both nitrate and organic aerosol cycles. Weaker cyclic
signals were observed for longer timescales (days to weeks) and are
hypothesized to be related to the timescales of synoptic weather
variability. The strongest periodic signals (3.5–5 and 7 d cycles) for
these longer timescales varied depending on the season, with no cyclic
signals and the lowest variability in the summer.