The nightly and seasonal variability of gravity wave activity and spectra in the mesopause region are studied with 10 years of sodium lidar observations. From the linear layer density response to gravity wave forcing, the lidar data were analyzed to get the atmospheric density perturbations and their spectra. The atmospheric density perturbation, density variance for fluctuations with vertical scales between 2 and 10 km, and amplitudes of density perturbation spectra at m = 2π/8 km and 2π/4 km all exhibit large nightly variability as well as large seasonal variations, with the semiannual maxima occurring near the equinoxes. The mean RMS atmospheric density perturbation and the mean RMS horizontal wind perturbations over our site are 5.1% and 25 m/s, respectively. The growth lengths of the density perturbations in spring and autumn are lower than those in summer and winter, and the annual mean value is 38 km. The annual mean density shear variance is about 15 (%/km)2, and the maxima occur near the equinoxes. The mean Richardson number is about 1.0. The mean value of the RMS vertical wind perturbation is 0.85 m/s with a maximum occurring at the end of the year. The m spectra show power law shapes, and their range of variation is between -2.06 and -3.81 with an annual mean value of -2.93. The ω spectra also show power law shapes, and their range of variation is between -1.06 and -2.32, with an annual mean of -1.64. The mean amplitudes of density perturbation spectrum, Fa(m) (m = 2π/4 km), and of the horizontal wind fluctuation, Fu(m) (m = 2π/4 km), are 1.35(m/cycles) and 3 ¿ 105(m2 s-2/(cycles/m)), respectively. The value of λz* averaged around autumn equinox is 14.8 km, which is lower than the value of 16.8 km, averaged around spring equinox. The annual mean of T* is 23.5 hours. The fact that the joint (m,ω) spectra are not separable, together with the large variability found in the m spectra slopes, is not compatible with linear instability theory but is compatible with Doppler spreading theories and diffusive filtering theory.