An analysis of the response of lower thermospheric nitric oxide (NO) at auroral latitudes to the auroral storm of September 19, 1984, is presented. A comparison of data from the Solar Mesosphere Explorer (SME) taken 1 day after the storm (September 20) with data obtained 1 day before the storm (September 18) revealed a factor of 3 increase in NO. In order to model this response, particle data from the NOAA 6 and 7 satellites are used to assess the time history of the auroral energy input along each SME orbital track. The deduced fluxes and characteristic energies are used as input to a time dependent one-dimensional photochemical model. In addition, the NCAR thermospheric general circulation model (TGCM) is used to calculate the response of the background neutral atmosphere to auroral forcings such as Joule and particle heating. It was found that particle precipitation accounted for 90% of the increase in the peak NO density, although Joule heating was more important at the higher altitudes (>140 km). The results of the model calculations predict an NO enhancement; however, the amplitude of the response as well as the absolute magnitude of the calculated NO density greatly exceed the observations. Two possibilities are proposed to explain this discrepancy. The first is that the yield of N(2D) from electron impact on N2 may only be 50%, rather than the 60--75% previously assumed. The second is that vertical winds of the order of 1--5 m s-1 may be generated in an E region auroral arc. It is shown that such winds could be important in damping out the NO response in increased particle precipitation. ¿ American Geophysical Union 1989