Monsoon Season Clock

Two calendars, a month apart

Almost everything confusing about the Arizona monsoon comes from one fact: there are two seasons, and they overlap but do not line up. The operational season is a fixed calendar window. Monsoon season, in the National Weather Service's own words on its Phoenix page, "officially extends from June 15th through September 30th." That is 108 days, and it is the same 108 days every year.

The other season is the weather, and it is defined by moisture. Before the window was fixed, the season began on the first of three consecutive days whose average daily dew point reached 55 °F at Phoenix, dew point being the measure that actually tells you whether the air has monsoon in it. This site computes that crossing from NOAA's 1991–2020 hourly normals, and it lands on 16 July. The dew point falls back below 55 °F on 8 September.

So the calendar season opens 31 days before the air turns moist, and closes 22 days after it dries out again. That is 53 of the season's 108 days at the ends, and it is not a mistake in either calendar. One is a planning window, chosen so that everyone from the forecast office to the evening news is talking about the same dates. The other is the weather.

Why the driest weeks come first

The consequence of that gap is that the beginning of monsoon season is the driest part of it. June at Phoenix Sky Harbor averages 0.02 inches of rain, and half of that month sits inside the window. July averages 0.91 inches and August 0.93, the two wettest months of the Phoenix year, and between them they carry about a quarter of the annual total.

What the early weeks do have is heat and instability without much moisture: the hottest part of the year, with storms that are high-based and isolated. That combination is why the build-up is the season for dry lightning and dust rather than for rain, and why a wall of dust arriving in June is not a contradiction of the calendar.

How the dew point curve was built

There is no published daily dew point normal, so this site derived one from a dataset that is published. NOAA's 1991–2020 hourly normals for Phoenix Sky Harbor give a normal dew point for each of the twenty-four hours of each calendar day. Averaging those twenty-four values gives a daily mean dew point normal for every day of the year, and the resulting series is what the chart draws.

The onset and end dates then come from applying the old rule to that series: find the first run of three consecutive days at or above 55 °F, and the last such run. The method is written into the site's own data file alongside the numbers, so the derivation can be checked rather than taken on trust. Two things about it are worth stating plainly. First, it is a climatological normal, not a record of any season that happened. Individual years cross the threshold two or three weeks either side of 16 July, and a dry year may barely cross at all. Second, the 55 °F figure itself is the part this site could not verify against a current NWS page, which is why it is labelled everywhere it appears.

What the strip is for

The 108 columns are the season, one per day, with the three acts shaded and the dew point drawn across them. It exists because the offset is much easier to see than to explain: the shaded moisture season sits inside the calendar season with a visible gap at each end, and no paragraph makes that land the way the picture does.

Set the date to a day in the build-up and the answer box says so, the dew point reads in the forties or low fifties, and the marker sits well to the left of the crossing line. Set it to a day in the wet core and the same three things move together. The tool is not predicting anything (every number on it is a thirty-year average), but it does answer the question people actually ask in June, which is whether the season has started yet.

Rain in the Valley does not fall evenly

Every rainfall figure on this page is for Phoenix Sky Harbor, and that is a real limitation rather than a footnote. Monsoon storms here are small and violent: a single cell can drop an inch on one part of the metro while a gauge five miles away records nothing, and the airport is one point in a basin that runs a hundred miles across. The normal is a useful long-run average and a poor guide to what happened on your street.

That is the reason this site keeps a rain log as a separate tool. If you want to know how your own corner of the Valley compares with normal, the answer has to come from a gauge you own, recorded month by month. The normals are the yardstick, not the measurement.

Flood risk outlasts the storms

The last act of the season is the one that catches people. Storm frequency falls away through September, which makes the month feel like the end of the danger, and it is not. After two months of storms the ground is saturated and sheds water rather than absorbing it, the washes and channels are primed, and a single slow-moving cell in September can produce a flash flood that a similar cell in July would not.

NWS Phoenix frames the season around exactly this: the monsoon is when the flood and storm hazards are most likely, and the season's end date is a date on a calendar rather than the day the risk stops. Treat the wind-down as the part of the season when the storms are fewer and the consequences are not.