A 7 day barometric pressure graph shows when the pressure will fall and how fast, and a fall is what warns of wind and rain.1 But much of what moves on it isn’t weather at all: pressure rises and falls twice a day on its own, and only some of the dips matter.
The pressure is lowest around 4 a.m. and 4 p.m. and highest around 10 a.m. and 10 p.m.1 On a week-long graph, that rhythm draws a row of small, even ripples.
They mean nothing for your plans. A dip into the afternoon followed by a climb into the evening is the daily cycle, the same tomorrow as today. The weather shows in the line underneath the ripples, so compare the same hour from one day to the next and see whether the whole line sinks or climbs.
The closer you live to the equator, the larger the ripples.1 In the tropics, the day-to-day changes are so small that, outside a tropical storm, a pressure graph says little about the coming weather.1
Once you look past the ripples, a real fall is easy to spot. What it warns of depends on how steep it is.
A sudden fall of the barometer means high wind, says a rule from 1848 that NOAA still quotes.1 A big fall on a fair day that stays low means much rain within a few days, and probably wind.1 In very hot weather, a fall means thunder.1
To put a number on “sudden”, forecasters measure the change over three hours. The Met Office, the UK’s weather service, grades it in four steps.2
| Change over 3 hours | Met Office calls it |
|---|---|
| Under 0.05 inHg (0.1–1.5 hPa) | Falling slowly |
| About 0.05–0.10 inHg (1.6–3.5 hPa) | Falling |
| About 0.11–0.18 inHg (3.6–6.0 hPa) | Falling quickly |
| More than 0.18 inHg (more than 6.0 hPa) | Falling very rapidly |
A line that drops more than about 0.1 inHg (3.5 hPa) within a three-hour stretch is falling quickly, and that’s the day to expect wind. The same grades work the other way for a rise.
If your body reacts to the weather, the fall is also what to watch. The guide to barometric pressure headaches shows how to grade the drops beside your own attacks.
How steep does a fall have to be before it brings a storm you shouldn’t be out in at all?
Meteorologists call a storm a bomb cyclone when its central pressure drops very fast, and the bar depends on how far north you are. At New York’s latitude, it is a fall of about 0.53 inHg (17.8 hPa) in 24 hours. At 60° north, it is 0.71 inHg (24 hPa).3
Your graph shows the pressure where you are, not at the storm’s center. Even under such a storm, it may never fall that far. But when your own line falls anywhere near that far within a day, a strong storm is passing close by. Check your weather service’s warnings, and move what you can’t do in a gale: a sailing trip, a drive over a pass, a ridge walk.
A fall that steep is worth seeing days ahead, while plans can still move. Sonuby’s Health Report draws the pressure for your location, or any place you search for, as one hourly graph for each of the next 14 days, with that day’s high and low. When each day’s low sits below the last, the line is sinking, and the day with the steepest drop is the one to plan around. Tap any hour to read its value, and pick hPa or inHg in the settings.
Each day’s graph runs hour by hour, with the day’s lowest and highest pressure marked on it, so a steep fall stands out days ahead.
For the next few hours, the Live Overview graph covers three hours back and 21 ahead, which is where you see how quickly a fall is happening on the day itself.
The forecast converts every reading to what it would be at sea level, so that places can be compared.1 A barometer at home in the hills shows a lower number. Read the change on both, and leave the exact figure to the forecast.
The Live Overview shows how fast the pressure is moving today.
How much of the graph’s second week can you rely on?
NOAA puts a five-day forecast right about 90% of the time and a seven-day forecast about 80%. A forecast for ten days or more is right only about half the time.4 Weather models have to estimate how the air will change, and those estimates get less reliable the further ahead they look.4
So treat the graph in three parts. Plan on the first five days, including the hour a fall starts. Days six and seven tell you a fall is likely, though not exactly when. Days eight to fourteen are a heads-up, since a drop there may move by a day or vanish. Check it again as it comes closer, and don’t cancel anything on it yet.
Once a week, scan the whole two weeks for the falls that cut through the daily ripples, and note the days they land on. Then each evening, check tomorrow and the day after on the hourly graph. Where a steep fall still sits on a day you planned to be on the water or a ridge, move the plan now rather than in the morning.
If you feel the weather in your head or joints, keep the plan you agreed with your doctor for such days. See them if your symptoms change.
Only slightly. Standard pressure at sea level is 29.92 inHg (1013.25 hPa), so 29.88 inHg is just below it.1
For the weather, high pressure usually comes with fair weather and low pressure with unsettled, stormy weather.5 For your body, no level is better or worse for everyone.
It differs from person to person. Changes in pressure and in temperature have each been linked to how bad knee osteoarthritis pain gets.6 Comparing your symptoms with a pressure graph for a few weeks shows your own pattern. The guides to migraine weather forecasts and arthritis weather forecasts go further.
Yes. Sonuby Weather is free on iOS and Android, and shows barometric pressure as an hourly graph for each of the next 14 days. The barometric pressure guide walks through each screen where it appears.
Resources:
[1] NOAA National Weather Service JetStream: Air pressure
[2] Met Office: Coast and sea glossary
[3] NOAA National Ocean Service: What is bombogenesis?
[4] NOAA NESDIS: How reliable are weather forecasts?
[5] National Weather Service: Pressure
[6] McAlindon, T., Formica, M., Schmid, C. H., & Fletcher, J. (2007). Changes in Barometric Pressure and Ambient Temperature Influence Osteoarthritis Pain. The American Journal of Medicine, 120(5), 429-434. https://doi.org/10.1016/j.amjmed.2006.07.036