At right you see a chart of the past week's barometric pressure at our home. There is a regular variation in the pressure that occurs twice a day (a diurnal cycle). The peak-to-peak variation is roughly 0.7 in/hg, or 2.2 millibars.
When I first saw this variation I thought it was some kind of heating effect on my instrument, or some other data artifact, and not a real pressure variation. However, the more I studied it, the stranger it looked. What, for instance, could account for the cycle being twice a day? So I did what I always do when faced with such a mystery: I googled.
It turns out that this diurnal pressure cycle is very well-known, and it's real. I've never seen it before because (a) the variation is quite small, and (b) until this past year I've never owned a sensitive enough recording barometer.
Although the diurnal cycle turns out to be very complex in nature, I can summarize what I discovered as follows:
--The large peaks (daily at around 10 AM PST) are mainly caused by solar heating of the atmosphere.
-- The lesser peak (daily at around 10 PM PST) are mainly caused by a solar tidal effect.
-- The solar tidal cycles (unexpectedly) dominate the lunar tidal cycles because of a resonance with a normal mode of atmospheric oscillation of approximately 12 hour period that closely matches the solar cycle, greatly amplifying it.
Because solar heating is the dominant underlying mechanism, there are some aspects of the atmospheric tides that are quite different than the oceanic tides:
-- The 10 AM peak is far larger at the equator than at higher latitudes.
-- Inland locations have greater 10 AM peaks than ocean locations (because there are greater heating effects on land masses).
There are also pressure variations with periods longer than a day. There is a pronounced annual cycle that relates to the seasonal change of temperature -- but its polarity varies by altitude. For locations below about 1300 feet in altitude have lower pressures in the summertime; locations above 1300 feet have higher pressures in the summertime (and, I suppose, right at 1300 feet there's no difference at all!). There are also several cycles --not well studied--that seem to be related to lunar cycles.
This turned out to be far more interesting, and complicated, than the instrumentation problem I suspected! You can read more about these phenomena here, here, here, and here.
