Rainfall and lack thereof has been a center of interest (and controversy?) of late in Seattle, with some rain going for hours and only counting as a trace, while a short 14 minute rain can get the elusive 0.01" needed to end our dry streak.
How does the rain gauge determine if the rain was enough to measure or not? It depends on the weight of the accumulated water.
The Automated Surface Observation Stations (ASOS) used at Sea-Tac Airport and many of the larger NOAA/FAA sites all use rather new equipment called "AWPAG" -- Automated Weather Precipitation Accumulation Gauge.
They just in the past few years replaced Heated Tipping Buckets (HTB -- this is the government we're talking about and they never met an acronym they didn't like.) Seattle was upgraded on May 1, 2006.
AWPAG is newer technology that collects rain, then uses a precise electrical sensor that can exactly determine the amount of rain based on the weight of the water it has collected. It's coated in an anti-freeze to prevent precipitation from freezing in cold weather and in general, has shown to be much more accurate in snow and freezing rain than the older HTBs.
How does it measure snow? The rain gauge is also heated so that snow melts into water and is then measured as rainfall equivalent. That's how the climate book might note we had 0.12" of "rain" on a date where it snowed all day. (Incidentally, a rough rule of thumb is 10 inches of snow = 1 inch of rain, but that ratio can vary depending on the water content of the snow.)
If rain is observed or sensed by the equipment, but not enough collects to measure 0.01", it counts as a "Trace" and does not count as measurable rain. That is how it can drizzle for 2 hours and change in late July but not count as a rainy day, yet 14 minutes of rain Sunday night was enough water to get the elusive 0.01 inches. (There have been days when the morning mist or drizzle has been heavy enough to register 0.01". And yes, those situations count officially as a rainy day too.)
Most other rain gauges, including the vast majority of home weather stations, use the old tipping bucket method.
The way those work is there is a tiny little "sea-saw" rain collector inside an electronic rain gauge. The top of the gauge is a funnel that has a certain radius at the top and bottom. The water then gets funneled down to a lever that is equally divided with walls on three sides, and ends left open.
The lever then sits on a fulcrum point that tips up and down like a see-saw. Enough rain has to collect on side of the lever to be enough weight to tip the scale down that way -- that dumps the water out the end and is calibrated as such to register 0.01" inch of rain. The rain then collects anew on the "high side" of the lever until it has enough water to tip the scale down the other way, registers another 0.01", then collects the other side. (This image will give you a general idea of how they work.)
Now, imagine thunderstorms and microbursts in the Midwest - that little rain level would be flipping back and forth on overtime! Indeed, it was found these rain gauges tended to under-report heavy rainfall events and also wasn't perfect at processing snow and freezing rain events, thus the NOAA switch to the new AWPAG technology.
If you're a glutton for punishment, here are the specs of the ASOS in all their glory government-acornymed detail.