Abstract. Multiple year-round (2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015) records of the bulk and size-segregated composition of aerosol were obtained at the inland site of Concordia located in East Antarctica. The well-marked maximum of non-seasalt sulfate (nssSO 4 ) in January (100 ± 28 ng m −3 versus 4.4 ± 2.3 ng m −3 in July) is consistent with observations made at the coast (280 ± 78 ng m −3 in January versus 16 ± 9 ng m −3 in July at Dumont d'Urville, for instance). In contrast, the well-marked maximum of MSA at the coast in January (60 ± 23 ng m −3 at Dumont d'Urville) is not observed at Concordia (5.2 ± 2.0 ng m −3 in January). Instead, the MSA level at Concordia peaks in October (5.6 ± 1.9 ng m −3 ) and March (14.9 ± 5.7 ng m −3 ). As a result, a surprisingly low MSA-to-nssSO 4 ratio (R MSA ) is observed at Concordia in mid-summer (0.05 ± 0.02 in January versus 0.25 ± 0.09 in March). We find that the low value of R MSA in mid-summer at Concordia is mainly driven by a drop of MSA levels that takes place in submicron aerosol (0.3 µm diameter). The drop of MSA coincides with periods of high photochemical activity as indicated by high ozone levels, strongly suggesting the occurrence of an efficient chemical destruction of MSA over the Antarctic plateau in mid-summer. The relationship between MSA and nssSO 4 levels is examined separately for each season and indicates that concentration of non-biogenic sulfate over the Antarctic plateau does not exceed 1 ng m −3 in fall and winter and remains close to 5 ng m −3 in spring. This weak non-biogenic sulfate level is discussed in the light of radionuclides ( 210 Pb, 10 Be, and 7 Be) also measured on bulk aerosol samples collected at Concordia. The findings highlight the complexity in using MSA in deep ice cores extracted from inland Antarctica as a proxy of past dimethyl sulfide emissions from the Southern Ocean.