Abstract. The turbopause is the demarkation between atmospheric mixing by turbulence (below) and molecular diffusion (above). When studying concentrations of trace species in the atmosphere, and particularly long-term change, it may be important to understand processes present, together with their temporal evolution, that may be responsible for redistribution of atmospheric constituents. The general region of transition between turbulent and molecular mixing coincides with the base of the ionosphere, the lower region in which molecular oxygen is dissociated, and, at high latitude in summer, the coldest part of the whole atmosphere. This study updates previous reports of turbopause altitude, extending the time series by half a decade, and thus shedding new light on the nature of change over solar-cycle timescales. Assuming there is no trend in temperature, at 70° N there is evidence for a summer trend of ~ 1.2 km decade−1, but for winter and at 52° N there is no significant evidence for change at all. If the temperature at 90 km is estimated using meteor trail data, it is possible to estimate a cooling rate, which, if applied to the turbopause altitude estimation, fails to alter the trend significantly irrespective of season. While studies of atomic oxygen density, [O], using mid-latitude timeseries dating from 1975, show positive trends which can be explained by a lowering of the turbopause, [O] exhibits negative trends since 2002 that, although at a different latitude, are compatible with the observed increase in turbopause height reported here.