Abstract. The chemical cycling and spatiotemporal distribution of mercury in the troposphere is poorly understood. We measured gaseous elemental mercury (GEM), reactive gaseous mercury (RGM) and particulate mercury (HgP) along with CO, ozone, aerosols, and meteorological variables at Storm Peak Laboratory at an elevation of 3200 m a.s.l., in Colorado, from 28 April to 1 July 2008. The mean mercury concentrations were 1.6 ng m−3 (GEM), 20 pg m−3 (RGM) and 9 pg m−3 (HgP). We observed eight events of strongly enhanced atmospheric RGM levels with maximum concentrations up to 135 pg m−3. RGM enhancement events were unrelated to daytime/nighttime patterns and lasted for long time periods of 2 to 6 days. During seven of these events, RGM was inversely correlated to GEM (RGM/GEM regression slope ~ −0.1), but did not exhibit correlations with ozone, carbon monoxide, or aerosol concentrations. Relative humidity was the dominant factor affecting RGM levels with high RGM levels always present whenever relative humidity was below 40 to 50%. We conclude that RGM enhancements observed at Storm Peak Laboratory were not induced by pollution events and were related to oxidation of tropospheric GEM, but the mechanism remain unclear. Based on backtrajectory analysis and a lack of mass balance between RGM and GEM, we propose that in situ production of RGM may have occurred in some distance allowing for scavenging and/or deposition of some RGM prior to reaching the laboratory, and that GEM oxidation is an important tropospheric Hg sink. Our observations provide evidence that the tropospheric pool of mercury is frequently enriched in divalent mercury and that high RGM levels are not limited to the upper troposphere.