The main purpose of this paper is to analyze observations of the 630.0-nm (red line) and 557.7-nm (green line) zenith airglow intensities measured during the month of October 2002 over Arecibo, Puerto Rico. We begin by describing an improved model for calculating the intensity of the red and green airglow lines that takes into account the role of molecular ions. We show, however, that at least for the data used in this study, it is not necessary to include the effects of molecular ions in our calculations. From observations of the airglow emissions on quiet days, we infer the general characteristics of the red-line intensities, which show a minimum before midnight and a peak after midnight. These results are consistent with a decrease in NmF2 and an increase in hmF2 before midnight, followed by the so-called midnight collapse. We then focus on the storm day of October 1--2, 2002, during which large-amplitude variations in both the red- and green-line intensities were observed and reproduced by the airglow model. An additional peak of NmF2, together with an hmF2 decrease, was observed before midnight, associated with the passage of a large-scale atmospheric gravity wave. The NmF2 nighttime increase requires plasma flux from the plasmasphere, which we find can be as large as 109 cm-2 s-1. The plasmasphere needs an additional source of plasma in order to provide such a large flux, and we explain this by considering the role of the meridional wind in the plasma exchange process. Strong changes in the shape of the F2 region were observed during the downward and upward motion of the F2 layer during the storm period. We have found simple analytical solutions for the height distributions of the electron density in the F2 region by including the effects of recombination and diffusion. These height distributions are in excellent agreement with the measured profiles. Finally, we discuss mesospheric green-line fluctuations and show that good agreement can be obtained for the storm conditions if small adjustments are made to the eddy diffusion coefficients in the mesosphere, which we associate with the passage of the gravity wave. However, we find that some green-line behavior during quiet-time conditions is difficult to explain.