DIURNAL VARIABILITY OF MESOSPHERIC OZONE AS MEASURED BY THE UARS MICROWAVE LIMB SOUNDER INSTRUMENT - THEORETICAL AND GROUND-BASED VALIDATIONS

Citation
P. Ricaud et al., DIURNAL VARIABILITY OF MESOSPHERIC OZONE AS MEASURED BY THE UARS MICROWAVE LIMB SOUNDER INSTRUMENT - THEORETICAL AND GROUND-BASED VALIDATIONS, JOURNAL OF GEOPHYSICAL RESEARCH-ATMOSPHERES, 101(D6), 1996, pp. 10077-10089
Citations number
20
Categorie Soggetti
Metereology & Atmospheric Sciences
Volume
101
Issue
D6
Year of publication
1996
Pages
10077 - 10089
Database
ISI
SICI code
Abstract
Diurnal variability of mesospheric ozone as measured by the 183-GHz ra diometer of the UARS microwave limb sounder (MLS) instrument for the n orthern midlatitudes in October 1991 and 1992 is compared with theoret ical calculations of diurnal amplitudes produced by two photochemical models and with ground-based microwave measurements made from Bordeaux (France, 45 degrees N) in October 1988, 1989, and 1990 and the Table Mountain Facility (California, 35 degrees N) in October 1990, Great ca re has been taken in comparing all the data sets within the same frame , i.e., interpolating onto the same vertical grid (pressure or altitud e), using the same units (concentration or mixing ratio) and degrading the vertical resolution of some data or models (convolution of the ve rtical profiles with appropriate averaging kernels). MLS diurnal varia bility generally agrees to within 10% with ground-based and model resu lts at 0.5, 0.2, 0.1, and 0.05 hPa (approximately 55, 60, 65, and 70 k m, respectively). Although modeled diurnal changes at 55 +/- 8 km are closer to the ground-based Bordeaux measurements than to the MLS data at 45 degrees N, MLS results are closer to ground-based Table Mountain Facility data at 35 degrees N at 0.42 and 0.22 hPa (similar to 55 +/- 8 and similar to 60 +/- 8 km, respectively)than to models. At 0.1 and 0.042 hPa, MLS diurnal changes are weaker than ground-based and model variations, but daytime O-3 mixing ratios are found to be in very goo d agreement for all data sets.