What Do Precipitation Percentages in Weather Forecasts Actually Mean?

Last updated 7 days ago. Our resources are updated regularly but please keep in mind that links, programs, policies, and contact information do change.

A high rain chance can make an outdoor plan look like a washout. The National Weather Service’s percentage describes the chance of measurable precipitation at a location during a stated period. It does not say how long the rain will last. Even a 100 percent chance can mean rain for just a minute. For a walk or an outdoor event, the percentage is the first thing to check, not the only one: the forecast’s timing and expected amount show what the rain would disrupt.

The percentage is a chance at a location

The National Weather Service (NWS) calls this number the probability of precipitation, or PoP, and defines measurable precipitation as at least 0.01 inch. Its definition includes liquid precipitation and the water equivalent of frozen precipitation. The threshold measures water, so snow depth is not the amount being counted.

Precipitation includes rain, sleet, snow, hail and drizzle. The percentage alone does not say which one to expect.

A 40 percent chance of rain, the NWS explains, means a 40 percent probability of at least 0.01 inch of rain at the specific forecast point. A 30 percent chance of rain means the same thing with a lower probability: a 30 percent chance of measurable precipitation at the location during the forecast period. Neither number promises a dry outing, and neither predicts that the entire outing will be wet.

The NWS illustrates a 20 percent forecast as a one-in-five chance of measurable precipitation at the location during the specified period. A low chance does not rule rain out; weigh how much getting caught in it would matter. A very high chance still needs the forecast’s amount and timing before it is useful for planning.

Why the same chance can describe different storms

One NWS explanation multiplies two factors: how certain the forecaster is that precipitation will form or move into an area, and how much of the area it is expected to cover. In one NWS example, 80 percent certainty and 50 percent expected coverage produce a 40 percent chance at any given location. In another, 40 percent certainty that widespread precipitation reaches the area, with 100 percent expected coverage, produces the same 40 percent point probability. In the first, the forecaster is more sure that rain will develop but expects it over less of the area; in the second, the rain would cover the whole area but is less likely to arrive.

Either way, the result is a point probability: the chance of rain at one spot, not the chance that rain falls somewhere in a larger forecast region. The percentage cannot say which neighborhood will stay dry.

The forecast period changes the question

The NWS definition uses a specified time period, normally 12 hours unless otherwise stated. The NWS Hourly Weather Graph, reached from a location’s forecast page as described below, shows precipitation probabilities on a graph for a selected forecast point. A chance for a whole daytime period and a chance for one hour are different measures: the longer window gives rain more opportunities to happen. Compare the place and time window before comparing the percentages.

PoP also does not describe the percentage of the forecast period during which it will rain. A high chance over a broad window can still leave plenty of dry time. For a walk, ceremony or outdoor shift at a particular time, the hourly forecast is more closely matched to the decision than the broad daytime number.

Do not add the hourly percentages to manufacture a daily chance. Each one refers to its own time window, and adding them ignores that the same weather event can span several hours. Use the forecast’s own probability for the broader period and its hourly details to examine the part of the day that matters.

A higher chance does not mean heavier rain

The NWS warns that a higher PoP is not an indicator of precipitation intensity. Its hourly weather graphs can show rainfall amounts and snowfall amounts alongside precipitation chances. A forecast can be confident that some rain will fall without expecting much of it. For a rain-sensitive task, look for the expected amount as well as the chance.

On its liquid precipitation amount graphics, the NWS expresses the forecast in inches for the period shown, using the amount forecasters consider the most likely outcome. An amount in inches answers how much water is expected, while PoP answers the chance of meeting the measurable threshold.

A rain percentage is also no substitute for NWS hazard alerts. An NWS watch signals increased hazardous-weather risk whose occurrence, location or timing remains uncertain. An NWS warning signals a hazardous event that is occurring, imminent or likely and poses a threat to life or property. Read any watch or warning for the location before deciding that a low rain chance makes an outdoor activity acceptable.

How forecasters build the estimate

NWS forecasters assess current conditions, then use numerical, statistical and conceptual models together with local experience to judge how those conditions will change. When models disagree, forecasters determine which perform best for the situation or seek a blended solution.

The National Blend of Models combines model forecasts from the NWS and other forecasting systems. The blend is designed to create a consistent starting point for the forecast, and it gives forecasters a suite of information to use for their own forecasts.

An ensemble is a set of model forecasts made with slightly different starting conditions or model versions. Forecasters use differences among ensemble members to obtain information about forecast uncertainty.

The United Kingdom’s Met Office gives an example of how an ensemble yields a percentage: if six of 24 members predict more than 5 millimeters of rain at a location during a defined period, the chance of that event is estimated at 25 percent. The event in that example is a specified rainfall amount, so its probability answers a different question from the NWS chance of any measurable precipitation.

New observations and revised model guidance can change the estimate. The National Oceanic and Atmospheric Administration (NOAA), which includes the NWS, describes a process of combining real-world observations with computer models to improve the models’ accuracy. A zone forecast describes expected weather for an area with similar climate and weather conditions. The NWS explains that its zone forecasts are issued again when the forecast database is updated to reflect changing conditions. When comparing an earlier rain chance with a later one, check the forecast issue or update time as well as the location and period.

When forecasts differ, first check whether they describe the same location, period and precipitation threshold. The Met Office’s hourly chance uses at least 0.1 millimeter in one hour, while its three-hour forecast uses 0.3 millimeter in three hours. Environment Canada defines its chance using 0.2 millimeter of rain or 0.2 centimeter of snow at a point during the forecast period. A percentage from another service should be read with that service’s definition, rather than with an assumed universal threshold.

Read the forecast around the number

The NWS uses these words to express the chance of precipitation:

Those words express degrees of uncertainty, rather than the strength of the rain. At 80, 90 and 100 percent, the same NWS list uses none.

Isolated and scattered describe something else, how much of an area the precipitation covers: isolated means up to 20 percent of the area and scattered 30 percent to 50 percent of it. The NWS uses those two terms for convective precipitation, which includes rain showers and thunderstorms.

Clicking a location on an NWS forecast map produces a forecast for that spot, and clicking another location may give a different forecast. From the point forecast page, the NWS directs readers to scroll down and select Hourly Weather Graph. The graph can show wind, gusts, precipitation types and amounts, temperature and other conditions alongside the rain chance. Select the actual destination and the hours of the planned activity, then compare the chance, amount and any hazard alerts.

Carrying an umbrella and moving an entire event indoors involve different tradeoffs, so they need not use the same probability as their trigger. Check the latest forecast before acting, and choose a backup that fits what rain would cost rather than one percentage rule for every activity.