In 1989, the TIME general circulation model was used in order to estimate the effects of doubled concentrations of mesospheric carbon dioxide and methane on the mesosphere and lower thermosphere. The model predicted the mesosphere to cool by 10 K, and the themosphere - more dramatically -- by 50 K. In the following year, the question of whether there would be a "Greenhouse Effect in the Ionosphere" was answered by predicting, that the F2-layer peak would lower by 15 to 20 km as a result of a cooled thermosphere. A number of papers have been published thereafter trying to reveal long-term trends in the data of ionosondes world-wide. These studies are based on estimates of the F2-peak height from the MUF factor and the critical frequencies of the E and F2 layer by means of various empirical formulae, which are applied to the data of large sets of ionosondes. However, the applicability of the formulae to a given station has to be verified in each case, esp. for high-latitude stations. In the present paper I re-analyse long-term trends in F2-peak altitude by means of several different algorithms in order to study the dependence of the trend magnitude on the data processing.