Fast nitrogen oxide photochemistry in Summit, Greenland snow

During the 1999 summer field season at Summit, Greenland, we conducted several series of experiments to follow up on our 1998 discovery that NOx is released from the sunlit snowpack. The 1999 experiments included measurements of HONO in addition to NO and NO2, and were designed to confirm, for Greenland snow, that the processes producing reactive nitrogen oxides in the snow are largely photochemical. Long duration experiments (up to 48 h) in a flow-through chamber and in the natural snowpack revealed sun-synchronous diurnal variations of all three reactive nitrogen oxides. In a second set of experiments we alternately shaded or exposed snow (again in the natural snowpack and in the chamber) to ambient sunlight for short periods to reduce any temperature changes during variations in light intensity. All three N oxides increased (decreased) very rapidly when sunlit (shaded). In all experiments NO2 was approximately 3-fold more abundant than NO and HONO (which were at similar levels). Higher concentrations of NO3− in the snow resulted in higher mixing ratios of HONO, NO and NO2 in the snow pore air, consistent with our hypothesis that photolysis of NO3− is the source of the reactive N oxides.