Most eruptions have occurred within ±3.5 days of the time of the full moon (or new moon) and the moon at perigee, which is when the stronger fortnightly tides occur. We conclude that there is a correlation between the times of eruptions at Fuego and the fortnightly lunar tidal maxima.
Also, the hours of the beginning and duration of many recent eruptions of Fuego have been compiled. The results of the comparison of the beginning of many eruptions since 1957 to the nearest calculated semi-diurnal minimum (of vertical tidal gravity acceleration) show that 9 out of 16 events occurred within ±2 hours of the minimum. This correlation also showed up well during the several stages of the 1974 eruption. The probability of 9 successes out of 16 events by chance is 2.0%. The eruptions are thus significantly clustered around the semi-diurnal minimum.
A further aspect of the influence of the Earth tides involves changes in the length of the day, as measured by an atomic clock (Melchior, 1966). Challinor (1971) reported semi-annual variations in the length of the day which are though to be caused by variation in the angular momentum of the atmosphere and by Earth tides. The maximum amplitude of the variation occurs in mid-January, mid-April, and mid-October. The mid-April and mid-October variation maxima occur at the times of maximum braking effect due to the bodily tide torque. Hamilton (1973) noted a correlation of increased volcanic activity near these dates. The influence of the rainy season, which occurs from May to October (plus a likely time lag) does not seem to be an adequate alternative explanation. No correlations with longer tidal cycles are evident, because components of the longer cycles cannot be easily separated from the fortnightly tidal maxima.
(Martin and Rose, 1981)