Satellite data show large declines in global (4.5%) and midlatitude (10%) ozone in the mid-1980s and during 1992 and 1993. Analyses of ozone, temperature, and aerosol records and two-dimensional chemical transport simulations have been carried out to develop an understanding of the causes of these changes. Simulations include contemporary homogeneous and heterogeneous chemistry. Also included are the effects of trace gas increases, dilution and denitrification associated with the Antarctic ozone destruction, solar cycle effects including relativistic electron precipitation (R. EP), variable diabatic transport fields and temperature, and variable sulfate aerosol surface area density and acidity. Simulated global and midlatitude ozone agree very well with observations for the entire period. Mid-1980s near-global ozone declines calculated by the model were found to be due to solar cycle (including R. EP) effects,-1.9%; volcanic effects,-1.5%; dilution effects,-1.1%; transport and temperature effects,-1%; and trace gas effects,-0.2%. The maximum effects of these different processes occur at different times. The observed 10% reductions in midlatitude ozone are reproduced in the simulations and are primarily due to 1 to 2-year transport and temperature variations.

Measurement of short‐lived photochemically‐produced species in the stratosphere by solar occultation is difficult because the rapid variation of such species near the terminator introduces ambiguities in interpreting the measured absorption in terms of meaningful atmospheric abundances. These variations produce tangent path concentrations that are asymmetric relative to the tangent point, as opposed to the symmetrical distribution usually assumed in most inversion algorithms. Neglect of this asymmetry may yield an inverted profile that deviates significantly from the true sunset/sunrise profile. In the present paper, the influence of this effect on solar occultation measurements of CℓO and NO is examined. Time varying altitude profiles of CℓO and NO were calculated with a time‐dependent photochemical model to simulate the distribution of these species during a solar occultation measurement. These distributions were subsequently used to calculate simulated radiances for various tangent paths from which mixing ratios were inferred with a conventional technique that assumes spherical symmetry. These results show that neglecting the variation of CℓO in the retrieval process produces less than a 10 percent error between the true and inverted profile for both sunrise and sunset above 18 km. For NO, errors are less than 10 percent for tangent altitudes above about 35 km for sunrise and sunset; at lower altitudes, the error increases, approaching 100 percent at altitudes near 25 km. The results also show that average inhomogeneity factors, which measure the concentration variation along the tangent path and which can be calculated from a photochemical model, can indicate which species require more careful data analysis.

Of all tropospheric species, ozone (O3) comes closest to being naturally present at toxic levels. In addition, O3 controls the ultraviolet flux reaching the Earth's surface and affects the temperature of the surface and atmosphere. For these reasons, O3 was an important species of the paleoatmosphere. Surface and atmospheric levels of paleoatmospheric O3 were calculated using a detailed photochemical model, including the chemistry of the oxygen, nitrogen, and hydrogen species and the effects of vertical transport. Surface and tropospheric O3, as well as the total O3 column, were found to maximize for an atmospheric oxygen level of 10−1 present atmospheric level (PAL). Coupled photochemical/radiative-convective calculations indicate that the radiative effects of O3 corresponding to an oxygen level of 10−1 PAL resulted in a globally-averaged surface temperature increase of 4.5 K.