Validation of SAGE II NO2 measurements

Abstract

Stratospheric aerosol and gas experiment (SAGE) II satellite‐borne measurements of the stratospheric profiles of NO2 at sunset have been made since October 1984. The measurements are made by solar occultation and are derived from the difference between the absorptions in narrow bandwidth channels centered at 0.448 and 0.453 μm. The precision of the profiles is approximately 5% between an upper altitude of 36 km and a latitude‐dependent lower altitude at which the mixing ratio is 4 ppbv (for example, approximately 25 km at mid‐latitudes and 29 km in the tropics). At lower altitudes the precision is approximately 0.2 ppbv. The profiles are nominally smoothed over 1 km except at altitudes where the extinction is less than 2×10−5/km. (approximately 38 km altitude), where 5 km smoothing is employed. The profile measurement noise has an autocorrelation distance of 3–5 km for 1 km smoothing and more than 10 km for 5 km smoothing. The absolute accuracy of the measurements is estimated to be 15% based on uncertainties in the absorption cross‐sections and their temperature dependence. Comparisons against two sets of balloon profiles and atmospheric trace molecules spectroscopy experiment (ATMOS) measurements show agreement within approximately 10% over the altitude range of 23 to 37 km at mid‐latitudes. SAGE II NO2 measurements are calculated to be approximately 20% smaller at the mixing ratio peak than average limb infrared monitor of the stratosphere (LIMS) measurements in the tropics in 1979. They show acceptable agreement with SAGE I sunset NO2 measurements in the tropics in 1979–1981 when the limited resolution and precision of the SAGE I measurements and the differences between the two measurement techniques are considered.