LAUSR

Abstract Long-term observation of atmospheric mercury (Hg) concentration and observation-based statistical methods are important tools to quantify the impacts of anthropogenic and natural perturbation on the global atmospheric Hg reservoir. In this study, two campaigns were conducted at a suburban site in eastern China with continuous measurements of gaseous elemental mercury (GEM) during the periods of August 2014 to July 2015 (Campaign 1) and May 2018 to April 2019 (Campaign 2). The overall mean GEM concentrations were 3.77 ± 1.32 and 3.24 ± 1.26 ng m−3, respectively. The potential source contribution function (PSCF) model based on backward trajectories were used to examine the variation of the potential source regions for GEM in different seasons. Generalized additive models (GAMs) were utilized in this study to quantify the impacts of local anthropogenic emissions, regional transport, and meteorological factors on the GEM concentration. Case studies with results from GAMs and observations of other air pollutants were conducted to provide more evidence for impacting mechanisms. The reduction of model residuals and the variation of contributions from direction and distance of air parcel transport imply the alleviation of local and regional anthropogenic Hg emissions from Campaign 1 to Campaign 2, respectively. The impact of relative humidity on GEM was crucial via the reduction of Hg(II) in droplets or on particles in the atmosphere. The impact of air stagnation on GEM was embodied mainly through the contribution of wind speed and partially by day of year (DOY) in winter. One DOY-controlled case also indicates the impacts of the 2015 El Nino event. With the decrease of anthropogenic emissions, the impacts of meteorological factors on GEM are getting more and more prominent. GAMs provide a promising tool for better understanding how anthropogenic and natural perturbations affect atmospheric Hg pollution.