| Within the Earth's mesosphere and lower thermosphere (MALT) (50--130 km) lies an important structured layer of atomic oxygen (O), created by the photo-dissociation of diatomic oxygen. The abundance and high reactivity of O plays a critical role in the heat budget, and therefore the dynamics of the region, and is a primary contributor of chemical excitation causing night and day airglow. Because of uncertainties in O concentrations, our quantitative understanding of the thermal and chemical processes within the MALT has been limited.; Over the past two decades, the technique most utilized for the determination of O concentrations has been the measurement of resonant fluorescence of the O(3S- 3P) triplet. U&barbelow;tah S&barbelow;tate U&barbelow;niversity (USU) is one of several groups that have embraced the technique, developing and flying the ATomic OXygen (ATOX) sensor system on sounding rockets. This dissertation reports on a set of observations from two sounding rocket flights launched from the University of Alaska's Poker Flat Research Range, CODA I and CODA II, which flew instruments based on this technique.; While the resonant fluorescence technique appears well suited for the measurement of O in the MALT, the dynamics of the sounding rockets which carry the sensor system create challenges that the system and data analysts must overcome. Primary challenges include density flow field disturbances, Doppler shift of the source lamp emission, and contamination. It has been well documented that these challenges, especially the flow field disturbances, make it difficult to accurately measure O concentration profiles.; This research has quantified the effect of the three major disturbances, seen as errors, and attempts to correct for each. Computations show that the largest errors are introduced by flow field disturbances. With a strong dependence upon measurement attitude and mission geometry, errors in the ram direction are often in excess of 300%. Errors as large as 100% are introduced by Doppler shift of the lamp output, and errors upwards of 40% are introduced into the absorption measurements due to contamination. The approach discussed herein can potentially reduce errors in both the fluorescence and absorption measurements to within 15--20% of undisturbed values. Furthermore, since the corrections are applied across the entire measurement region, direct comparison of upleg and downleg measurements is now possible. |
