The long-term seasonal trend in NO2 density near 10 mbar, as measured by the Solar Mesosphere Explorer (SME), is compared to photochemical model predictions of the trend throughout the first 3 months of 1982. The general increase in the observed NO2 is found to be caused by a shift in the partitioning of odd nitrogen in favor of NO2. The model is also used to infer the odd nitrogen (NOx:NO+NO2+NO3+2¿N2O5) O5) mixing ratio, which is seen to decrea rapidly throughout the period. This rapid decrease is found to be caused in part by improper modeling of the photodissociation rate of NO2. Indirect measurements of the photodissociation rate of NO2 from SME show the rate to change with solar zenith angle at angles greater than 70¿. Only when this dependence is taken into account do the variations in the odd nitrogen mixing ratio become consistent with theory.