Weather Forecasting ... On-Line

Public Forecast Terminology

Introductory Comments

If you routinely issue public forecasts, you need to have a set of forecast terms that convey to your user forecast information in a clear and concise manner. This means that you need a set of terminology that are explicitly defined. This allows similar forecasts to have similar meanings and also permits straight-forward verification of your forecast product.

The purpose of this web page is to define the typical terminology used in public forecasts along with some editorial comment and options. These terms are similar to National Weather Service usage but have been modified based on this author's experience and opinions. Please realize that there may be regional variations in the use of these terms.

Forecast Content

The typical public forecast contains a forecast for the following weather elements:

  • sky cover/cloud cover
  • present weather (precipitation and obstructions to vision), as needed
  • temperature, usually maximum or minimum, as appropriate
  • wind direction and wind speed
  • human comfort issues, as needed
  • non-convective watches and warning, as needed
  • Although many forecasts begin with cloud cover and then address present weather, temperature, and wind, in that order, there is no specific order to the parameters included in a forecast. If some unusal or dangerous weather is expected, this element should be mentioned at the beginning of the forecast statement. Each of the above forecast elements is discussed separately below.

    Sky Cover

    Sky Condition Term Cloud Coverage METAR Sky Cover
    Cloudy 8/8 OVC
    Mostly Cloudy
    Considerable Cloudiness
    5/8 - 7/8 BKN
    Partly Cloudy
    Partly Sunny
    3/8 - 4/8 SCT
    Mostly Clear
    Mostly Sunny
    1/8 - 2/8 FEW
    Clear or Sunny less than 1/8 CLR or SKC

    The amount of cloud in the sky is an important part of a forecast. The table above lists the typical terminology used and the amount of sky cover associated with each term. The above definitions are consistent with the defintions for the sky cover in METAR observations. This allows for easy verification of the sky condition forecast. Some variations on these terms can be used if you want to move away from the usual verbage:

  • In some cases where precipitation is the dominant weather parameter, sky cover can be omitted and cloudy skies implied. For example: "Periods of rain this morning ... giving way to partly cloudy skies this afternoon" implies cloudy skies during the morning period.
  • For situations where sunny or mostly sunny skies are expected, terms such as "an abundance of sunshine will dominate this afternoon" or "a mixture of clouds and sun" can be used.
  • If the amount of cloud cover varies during a forecast period, particularly during the first period of the forecast, a combination of the above terms may be appropriate. For example, you might use the phrase "cloudy skies this morning giving way to sunshine this afternoon" in your forecast.
  • One could argue about where the line between "partly cloudy" and "mostly cloudy" should be drawn. As defined above, "partly cloudy" is more sky than cloud while "mostly cloudy" is more cloud than sky.

    The term "fair skies" often used to describe the state of the sky. It implies that there is less than 4/8 cloud cover, most of which is middle or high cloud. It is a rather vague term and probably should be avoided.

    Present Weather (Precipitation and Obstructions of Vision)

    Present weather includes both precipitation and obstructions to vision that may impact the public. If these elements are not expected to occur, they are not mentioned in the forecast.

    Precipitation: Precipitation forecasts can take several forms in a forecast statement:

  • probability of precipitation (POP)
  • an areal coverage description, or
  • a categorical forecast.
  • POP is defined as the likelihood of occurrence (usually expressed as a percent) of a measurable amount of precipitation (liquid or water equivalent of frozen) during a specified period. Measurable precipitation is 0.01 inch or more. POP values are usually specified for a period of 12 hours.

    POP forecasts can be explicitly stated as a numerical value or described using an expression of uncertainty. Numerical values should be rounded to increments of 10 percent. The table below shows the expression of uncertainty that corresponds to specific POP values.

    POP Percentage Expression of Uncertainty
    0, 10 percent no descriptor
    don't mention precip
    20 percent slight chance
    30, 40, 50 percent chance
    60, 70 percent likely
    80, 90, 100 percent no descriptor
    1st period: categorical

    Although there is a small chance for precipitation to occur when the POP is 10 percent or less, it is common practice not to mention precipitation in a forecast when the POP is 10 percent. Depending upon your perspective, some forecaster also apply this prohibition at the 20 percent level and do not mention precipitation unless the POP is 30 percent or higher.

    When the POP is 80 percent or above and the forecast is for the first 12 hours from issue time, it is common practice to use categorical terms instead of a POP. Terms like "occasional light snow" or "periods of rain" indicate a high probability of precipitation without the need to explicitly state a POP value. For longer period forecasts (12 hours and beyond), POP values should be used in lieu of categorical descriptors. Do not use POP values and categorical terminology in the same forecast period.

    For convective precipitation, areal qualifiers (shown in the table below) should be used during the first 12 hours of a forecast instead of POP values. The areal qualifiers define how much of an area is expected to be affected by convective precipitation.

    Percent Coverage Areal Qualifier
    10 percent isolated or few
    20 percent widely scattered
    30, 40, 50 percent scattered
    60, 70 percent numerous
    80, 90, 100 percent categorical

    It does not make sense to combine POP values with area qualifiers. For example, what are you really saying if the forecast states: Isolated thunderstorms; chance for precipitation 30 percent? Does this mean there is a 30 percent chance of isolated thunderstorms, which is a slim chance of anything, or a 30 percent chance for thunderstorms with an isolated coverage? To avoid confusion, do not use areal qualifiers and POP values in the same forecast period. Limit areal qualifiers to high probability convective events during the first 12 hours of your forecast to indicate how much of your forecast area will be affected.

    Obstructions to Vision: If there is some type of adverse weather anticipated, such as dense fog, include these elements in your forecast. Dense fog occurs when the visibility is less than 1/2 mile. Sometimes dense fog is localized while at other times is may be widespread. In either case, it can be a hazard to driving.


    Maximum and Minimum Temperature: Maximum and minimum temperatures are a routine part of public forecasts. These temperature forecasts should be as specific as possible. A broad range in a temperature forecast can verify but convey little information. The forecast range for an area will be larger than the forecast range for one location. It is recommended that area forecasts use a range not larger than 5 degree Fahrenheit while the forecast range for a point should not be larger than 3 degrees Fahrenheit. Some users may want a specific high or low temperature for a point.

    The table below shows the meaning of typical phrases used to describe temperature ranges.

    Temperature Phrase Implied Range
    lower 50s 51, 52, 53
    middle 60s 64, 65, 66
    upper 70s 77, 78, 79
    30s 30 to 39
    near 90 89, 90, 91
    around 25 24, 25, 26

    Have your even heard a forecast that says "the temperatures tomorrow will be in the 100s?" Does this mean that the temperature can reach as high as 199 degrees? When making a forecast over 100 degrees Fahrenheit, it is best to use specific a range like "103 to 106 degrees." A phrase like "100 degrees plus this afternoon" is a little vague but does not imply values well above typical summer values for the area.

    Details and Non-Diurnal Trends: It is common to forecast specific temperatures for specific times during the day. These data are usually based on the Model Output Statistics (MOS) guidance that is readily available to a forecaster. How you include this information in your forecast will depend upon who your user is.

    Normally the daily maximum temperature occurs during the afternoon while the daily minimum temperature occurs around sunrise. Temperature events such as rising nighttime temperatures or a morning high with cooling during the afternoon due to a strong cold front passage should be reflected in your forecast verbage.

    Temperature Descriptors: It is common practice to use descriptive terms to highlight or emphasize temperature departures from normal. The table below lists some of these descriptive terms and a suggested usage.

    Descriptive Phrase Summer Season Winter Season
    hot high temperatures above 100oF not applicable
    very warm more than 8 deg F above normal more than 11 deg F above normal
    warm 4-8 deg F above normal 6-11 deg F above normal
    near normal within 4 deg F of normal within 6 deg F of normal
    cool 4-8 deg F below normal 6-11 deg F below normal
    cold more than 8 deg F below normal more than 11 deg F below normal
    very cold not applicable low temperatures below 0oF

    The intent of the above table is to provide you with a starting place for setting temperature ranges for each descriptive term. The original intent was to express these ranges in terms of the monthly standard deviation (s.d.) for the daily high and low temperatures. Specifically, near normal would be within one s.d. of normal; warm and cool would be 1 to 2 s.d. from the normal; while very warm and cold would be greater than 2 s.d. from the mean. This approach needs to be investigated further and refined.

    In applying the above table, keep in mind that temperatures vary more in winter than in summer, i.e., higher standard deviations are observed during the winter months. Similarly, high and low temperatures should be evaluated separately.

    The recommendations in the above table are focused on the Central Plain. If used in other locations, modification may be needed. Similarly, during the Spring and Fall seasons, the ranges given should be somewhere between the Summer and Winter values. Consider the above table as a starting place for your definition of terms.


    Winds are usually forecast in 10 mph ranges such as 5 to 15 mph or 10 to 20 mph. The use of a range supports the variability found in wind speed. If wind gusts greater than 25 mph are expected, the term "gusty" should be included in the forecast.

    There are times when a descriptive term is used along with the range. The table below list some common expresssions for specific wind speed ranges.

    Sustained Wind Speed Wind Terminology
    < 5 mph
    no specific direction
    light and variable
    < 10 mph,
    with a direction
    5 - 15 mph
    10 - 20 mph
    no descriptor
    15 - 25 mph breezy (mild weather)
    brisk (cold weather)
    gusts > 25 mph blustery
    (cold weather only)
    20 - 30 mph windy
    30 - 40 mph very windy
    35 - 65 mph gale-force
    > 75 mph hurricane-force
    convective gusts > 58 mph severe

    If a front passes through your forecast area and a significant wind shift is expected, mention it in your forecast.

    Human Comfort

    During the winter and summer season, human comfort terminology is often included in the forecast in the form of wind chill or heat index values. It is important to include these terms to warn your users of the danger associated with cold winter winds or high heat stress conditions.

    The table below provides you with the impact of heat index values on people.

    Heat Index Possible Heat Disorders for
    People in Higher Risk Groups
    130oF or higher heatstroke/sunstroke highly likely
    with continued exposure
    105o - 130oF sunstroke, heat cramps or heat exhaustion likely,
    and heatstroke possible with prolonged exposure
    and/or physical activity
    90o - 105oF sunstroke, heat cramps, and heat exhaustion possible
    with prolonged exposure and/or physical activity
    80o - 90oF fatigue possible with prolonged
    exposure and/or physical activity

    Watches and Warnings

    It is common practice, particularly within the National Weather Service, to include watches, warnings, and other advisories in a public forecast. How you treat these products will depend upon who the user of your forecast is.

    Time Periods

    Forecasts are issued for specific time periods such as "today" ... "tonight" ... "tomorrow" ... etc. What time frames do these terms apply to? Generally, daytime runs from 6 am to 6 pm while nighttime is from 6 pm to 6 am. One could argue that day time is from sunrise to sunset, and nighttime from sunset to sunrise. How you define these periods will likely depend upon your user.

    What time period does "evening" refer to? During the warmer months, evening likely runs from around 6 pm to about one hour after sunset, perhaps as late as 10 pm. During the colder months, the sun often sets before 6 pm and this defintion does not work. During this time of year consider evening as the period from sunset to around 8 pm. How do you define evening?

    In any case, however you define your forecast periods, these periods need to be defined explicitly so that the verification process can accurately evaluates your forecast.

    Concluding Remarks

    This web page provides you with a basic set of forecast terminology that can be used in a public forecast. For some elements there are several ways to apply the ideas presented. Use this page as a starting point for developing you own set of forecast terminology.

    In every situation, be clear and concise in expressing your forecast so that there is no question about what your forecast really says.

    Return to the Operational Weather Topics page

    last updated on 3/23/08