A note on the diurnal averaging of aeronomical models