Skip navigation.
Home resources

Beaufort Letters

When recording weather in a diary, log or similar medium, it might be useful to know that there exists a system of 'shorthand' which can be used for the purpose. This page lists the letters, the method of using them, combinations etc., and some hints/tips.

The system of abbreviations was first used by Francis Beaufort (later Admiral Sir Francis) early in the 1800's, and intended for use at sea. The scheme has been considerably revised since those days.

It is accepted practice to record phenomena at a particular time in the following order:

  1. State of Sky.
  2. Thunderstorm (if present).
  3. Precipitation (if present).
  4. Atmospheric obscurity (if present).
  5. Any other phenomena.

The following is the list of abbreviations used, ordered in sections as above: NB... you can click on the subject lines above to go directly to that section of the document.



b Total cloud amount: 0 to 2 oktas (eighths of sky covered)
bc Total cloud amount: 3 to 5 oktas
c Total cloud amount: 6 to 8 oktas*
o Uniform thick layer of cloud, completely covering the sky (i.e. 8 oktas - no gaps: mainly used with St, Ns and thick As)


* There are often occasions when large amounts of cirrus cloud cover the sky, but the day is bright or even quite strongly sunny. The system does not distinguish between c = cloudy/grey and c = cloudy/bright + fine, so it is often useful to note in plain language what the 'character' of the sky is, e.g. c(bright, fine with diffuse sunshine).

As an optional extra, plus / minus signs can be used to indicate whether the cloud cover has been generally increasing / developing over the past hour (+), or decreasing / dissolving (-), thus: bc+, b-etc.

State of sky letters are always recorded, unless there is deep fog or thick falling snow, such that the sky state is not possible to determine. In this case, the state of sky letters are left off.

Observing cloud amounts: Estimating the amount of sky covered with cloud is not an easy task - even overcast / 8 oktas and sky clear are not straightforward!

  • Try to observe from a point where the entire horizon can be seen: not easy especially in an urban / suburban environment.
  • Before putting pen to paper, scan the sky to get a 'feel' for the cloud-scape: note particularly where clouds are at different levels.
  • Beware of under, or over estimating cloud amounts. Very fine / thin cirrus (Ci) is often under-estimated as the eye is drawn only to the obviously visible elements, missing the very fine filaments elsewhere. Watch out in cirrostratus (Cs) situations: if the sun can be seen through the cloud, but you are sure that the cloud covers the entire sky, then this is 8/8 of cover, not 7/8. The same applies for thin altostratus (As)- the cloud often covers the sky from horizon to horizon, but the sun can be seen weakly through the cloud. Conversely, cirrus (Ci) and cirrocumulus (Cc) can never be 8 oktas cover - these types as defined will always have some gaps.
  • When cirriform clouds alone are present, ( or the dominant cloud type ), it is useful to use some form of filter to view the sky - sunglasses for example. However, do not look directly at the sun: a useful technique is to hold a hand at arms length against the solar position to avoid both accidental viewing of the sun, and to obscure from the edge of sight the brightest light coming from the sun.
  • Cumuliform cloud amounts are often over-estimated because the towering vertical extent of large cumulus (Cu), or cumulonimbus (Cb) give an impression of more 'horizontal' cloud cover than there really is. It is the base of the cloud that is assessed.
  • Divide the sky mentally into half, and decide whether a cloud fills a full half, or something under, or something over. In this way, the amounts 3, 4 and 5 oktas can be adjudged. If the amounts are much under, or much over half cover, further mentally divide that area to assess the difference between 1 and 2 oktas, and 3 oktas, or between 5 oktas, and 6 or 7 oktas.
  • Night-time observing is of course something else! Only great practice will allow for useful observation of cloud amounts, and you must allow your eyes to become accustomed to the darkness. The length of time varies on who you talk to, but at least 10 minutes is required, but for operational observers, this length of time is often impractical. On starry nights, watch for under / overestimating cirrus (Ci), and indeed stratocumulus (Sc) and altocumulus (Ac) can be mis-observed because stars are seen through thin / well broken layers, and amounts less than reality are noted. Don't be afraid to make a note to the effect that you are not sure!



l Distant lightning (storm too far away for sound to reach observer)
t Thunder heard; no lightning seen
t l Thunderstorm


l, lightning seen (thunder not heard), is obviously more likely to be observed at night. During the day, if lightning is so bright as to be visible, then the thunder will also be heard, and 'tl' should be recorded.

A Thunderstorm is regarded as being as being 'at' a particular point from the first hearing of thunder, to 10 minutes after the last hearing of thunder, in cases where there is doubt about whether a storm has passed by or not. However, when it is obvious that a storm has passed, then a note can be made to that affect, even if distant thunder can still be heard.

Lightning and / or precipitation may or may not be observed - the important thing for reporting thunder is the sound of same. See note below regarding the coding of the intensity of thunderstorm.

The intensity of a thunderstorm is a matter of great subjectivity. It is based upon the lightning / thunder frequency only, the precipitation is assessed separately. However, in some situations, the 'vividness' of the lightning display and 'loudness' of the accompanying thunder [ notwithstanding the actual frequency of same ], may be relevant to assessment, though some plain language comment should be made. In particular, it is useful, if possible (and safe), to note things like whether the lightning is cloud-to-cloud (CC), cloud-to-ground (CG) etc., and also the actual number of discharges detected per minute.

PRECIPITATION (i.e. rain, snow, hail etc.)



d Drizzle, freezing drizzle(*) [ " fairly uniform precipitation composed exclusively of fine drops of water, less than 0.5 mm diameter, very close to one another; the effect of their individual impact on water surfaces is imperceptible " ]
dr Drizzle and rain mixed
h Hail, small hail, snow pellets, diamond dust, ice pellets($)
hr Hail($) and rain mixed
hs Hail and snow($) mixed
r Rain, freezing rain (*) [ " precipitation of liquid water particles, either in the form of drops of more than 0.5 mm diameter or of smaller, widely scattered drops. " ]
rs Rain and snow mixed, or partially melted snow ('sleet')
s Snowflakes($)
sh Snow grains($) ( known in some texts as 'granular snow' )


(*) Freezing rain/drizzle, is liquid precipitation that freezes on contact with surfaces chilled below zero deg.C.

Precipitation is recorded by type, intensity and continuity.


The type of precipitation is indicated by the appropriate letter, or combination of letters if there is a mixture of precipitation. Example: d = drizzle, r = rain ; dr = drizzle and rain : If the precipitation is in the form of SHOWERS, i.e. falling from cumulus, cumulonimbus or altocumulus, then the prefix 'p' is used [originally from 'passing'], thus a shower of snow is ps; a shower of rain and snow mixed is prs etc. Showery precipitation is usually marked by abrupt beginning and end, and by rapid fluctuations in intensity. The particles are often larger than those falling in non-showery situations.

$ Notes relating to the observing/recording of 'solid' precipitation, and cloud types associated:
( notes are appended to the bottom of this page ---- click on the link above to go straight there )


There are four categories of intensity: slight, moderate, heavy and violent. To indicate slight precipitation, the letter denoting the type of precipitation is followed by a subscript 'o', thus; slight rain = ro, slight shower of rain and snow = proso etc. Note that in mixed precipitation, the subscript is applied to both elements (see also the note at the end of this paragraph). For moderate precipitation, the letter alone is used, thus moderate shower of rain = pr. For heavy precipitation, use capital letters. Thus for heavy snow, use S. For violent phenomena, usually applied to showery/convective precipitation, use a subscript 2, thus a violent shower of rain = pR2. On a general note relating to mixed precipitation, when two types of precipitation co-exist, the intensity of the 'heaviest' type governs the coding: thus for a shower which contain a moderate fall of rain, with a slight accompanying fall of snow, this would be noted as: prs. However, some plain language remark should be made in such a situation to amplify.

When some method of measuring the rate of rain / snow falling is available, then the following should be used to decide upon the intensity:

RAIN: (i.e. from layer clouds)

  1. Slight Rain: rate of accumulation is slow, not more than about 0.5 mm per hour.
  2. Moderate Rain: enough rain to form puddles in a short time, and giving a rainfall rate of between 0.5 and 4 mm per hour.
  3. Heavy Rain: Makes a noise of roofs, a splashing off hard surfaces (roads etc.) and gives an accumulation of more than 4 mm per hour.

SHOWERS: (i.e. from cumuliform clouds)

  1. Slight shower: less than about 2 mm per hour.
  2. Moderate shower: from about 2 mm/hr to 10 mm/hr.
  3. Heavy shower: about 10 to 50 mm/hr.
  4. Violent shower: above about 50 mm per hour.

SNOW: (applicable to both layer and cumuliform type clouds - assumes no drifting.)

  1. Slight snow: flakes are sparse. Rate of accumulation not exceeding 0.5 cm per hour.
  2. Moderate snow: visibility impaired. Rate of accumulation 0.5 to 4 cm/hr.
  3. Heavy snow: visibility much reduced - often to or below fog limits - with accumulations more than 4 cm per hour.

As regards noting the the intensity of thunderstorms with rain / snow, the intensity of the thunderstorm is considered separately from that of the accompanying precipitation. The intensity of a thunderstorm is judged by the frequency of the thunder and lightning, so if there is a thunderstorm in progress with little thunder / lightning activity, but heavy rain, this would be noted as: tloR. Note that any intensity qualification for a thunderstorm is applied to the combined letters, tl, not to each letter. So a 'slight' thunderstorm is noted as: tlo.

Continuity (i.e. how long does it go on for!):

When precipitation falls from layer cloud, then the continuity of such precipitation is noted in the following manner:

  1. When considering the past hour, if the precipitation (i.e. rain, snow etc.) has been of a 'start-stop' variety, then the letter for type ( e.g. r ) is prefixed by the lower case ' i '. Thus for intermittent slight rain, this would be noted as..... iro, and for intermittent moderate rain and snow mixed as..... irs etc.
  2. When, over the past hour, the precipitation has continued at that intensity, without a break, then the letter denoting the precipitation type is repeated, thus for slight rain that has continued for an hour without a break this is noted as..... roro, and continuous heavy snow by..... SS etc.
  3. If noting precipitation not specified as continuous or intermittent, the Beaufort letter is used alone. For example R = heavy rain; doro=slight rain and drizzle mixed.

When there is a change of type and / or intensity, then this is indicated by successive use of letters as appropriate. This is best demonstrated by an example:

Consider light snow that has continued for over an hour at the same intensity without a break, this then turns to slight rain / snow mixed, but for a brief (less than one hour) period turns to heavy snow (sky obscured), before becoming light snow for less than one hour, then dies out. The sequence of Beaufort letters would be:

csoso, croso , S , cso , c

Note that each group of letters is separated by a comma, though will see in some texts that the comma should only be used for 'clarification'. In my view, it is best to always use it to avoid any possible confusion.

ATMOSPHERIC OBSCURITY (i.e. factors reducing visibility, other than precipitation.)


f Fog, ice fog ^(visibility < 1000 m)
fe Wet fog(visibility < 1000 m)
fg/fs Shallow land/sea fog(visibility above the fog >=1 km)
F Fog, ice fog ^(visibility < 200 m)
ks Drifting or blowing snow(%)
kz Dust or sandstorm (visibility < 1000 m)
m Mist (visibility >=1 km) [ Relative Humidity >~95% ]**
z Haze


(%) The distinction between the two is that blowing snow is raised well above the normal (adult) 'eyeline', and thus reduces the overall visibility markedly, and drifting snow remains below the 'eyeline', and does not materially affect the prevailing visibility.

^Ice Fog: fog consisting of minute ice crystals (as opposed to water droplets for water fog).
When the visibility varies with direction around an observer, such that in one area the visibility is below 1000 m and elsewhere it is at or above, then fog patches exist, and the letter for the type of fog is prefixed by 'i', thus ' if '

As with precipitation, changes in occurrence, thickness etc., of fog is noted by successive use of appropriate letters: thus bcif, cf, ff, F ..... indicates that fog patches extend to the whole area, which last for at least an hour without appreciable change, then thickens further to reduce the visibility below 200 m.

** on the subject of the relative humidity with mist, this can become a bone of contention where several observers are noting the same/similar events. The figure quoted is meant to be a guide, not a mandatory limit.

and although not strictly belonging to this group:

v Abnormally good visibility


What the definition of 'abnormal' is when talking about visibility can be the subject of many an argument between observers. In a polar airflow over many parts of Scotland, 60 km or more would not be abnormal, but in the hazy Thames Valley basin, it might well be.



e Wet air, without rain, snow etc. falling
g Gale (mean speed 34-47 knots over a 10 minute period)
G Storm (mean speed 48 knots or more over a 10 minute period)
i Intermittent (used with precipitation and fog)
j Phenomenon within site but not at the location of the observer (**)
kq Line Squall(***)
p Shower -- see notes against precipitation for usage.
u Ugly, threatening sky (in addition to bc/c/o etc.)
w Dew deposit (make sure that 'dew' is really being deposited,
and not just 'guttation' water extruded from plant leaf surfaces on some cold nights.
x Hoar frost deposit*
y Dry air -- less than 60 % relative humidity

* The letter ' x ' (hoar frost) is used when a white, crystalline deposit of ice is observed on solid objects, after a cold, clear night. Hoar frost occurs when water vapour condenses (sublimates) directly to the ice phase without an intermediate liquid phase (though the process is usually initiated by a vapour-to-liquid phase). Hoar frost should not be used for the freezing of water already present, or for the glaze produced by freezing rain / drizzle, or for the rime produced during freezing fog episodes - all these phenomena should be noted in plain language separately.
** The letter ' j ' is used in combination with various other letters to record phenomena occurring within sight of, but not at the station; thus jp indicates a shower within site but not at the observing point;

for example:

jp .... precipitation within sight (often used with showers); however note that if the 'distant' shower was at the point of observation earlier, and has moved away, jp is not used in this context.

jf .... fog within sight (visibility at the station 1000 m or more )

jks .... drifting snow within sight.

*** A squall is differentiated from a gust by its greater duration: generally lasting for several minutes before decaying again. Squalls are often associated with the passage of fronts, particularly cold fronts, or well defined troughs, or with the 'gust front' from a well defined/mature supercell Cb. To qualify as a line squall, other marked changes are often observed, e.g. change of wind direction, fall of temperature etc. The following definition is used when estimating wind speeds using the Beaufort scale of wind speed: " .... a sudden increase of wind speed by at least three levels of the Beaufort scale, the speed rising to F6 or more and lasting for at least one minute."

$ Notes relating to the observing/recording of 'solid' precipitation, and cloud types associated:

[ for more on each type, including how the precipitation is formed, click on the name. ]
  1. SNOW / SNOW-FLAKES: Precipitation composed of white, ice crystals - often aggregated into clusters/clumps depending upon the temperature in the lowest layers of the atmosphere. Roughly, at temperatures above about - 5 degC, then clustering is likely, but at lower temperatures, individual crystals may be expected. CLOUD OF ORIGIN: As, Ns and occasionally Sc (usually with a high convective element - e.g. downwind of the North Sea in winter); Cu and Cb in showers.
  2. SNOW GRAINS: (also called granular snow) Very small (dia: < 1 mm ) white and opaque grains of ice. The grains are flattish or elongated. When the precipitation reaches the ground, they do not bounce or break up and usually fall in small amounts - not sufficient to cover the ground and not from shower cloud. CLOUD OF ORIGIN: Usually St, Sc (shallow, but sub-zero top) or perhaps Fog.
  3. SNOW PELLETS: (also called soft hail) White and opaque grains of ice. These grains are spherical or conical. The diameter is larger than that of Snow Grains (q.v.), being in the range 2 to 5 mm. The grains are easily crushed and rather brittle and tend to bounce when falling on hard surfaces, sometimes breaking up in the process. Snow pellets generally occur mixed with other shower-type precipitation: e.g. rain or snowflakes, when surface temperatures are around zero deg.C. CLOUD OF ORIGIN: Cu, Cb and sometimes Sc.
  4. ICE GRAINS: (also called Ice Pellets [ a ] or just Ice Pellets) These transparent or semi-transparent pellets of ice usually bounce when hitting a hard surface and make a sound on impact. They are usually spherical and have a diameter less than about 5 mm. Frozen raindrops, or largely melted and refrozen snowflakes. The freezing process is thought to take place close to the ground. CLOUD OF ORIGIN: As, Ns.
  5. SMALL HAIL: (also called Ice Pellets [ b ]) The appearance and behaviour is similar to Ice Grains above. Pellets of snow, encased in a thin layer of ice, which may be the result of interception of super-cooled water droplets, or the pellet itself partially melting, then re-freezing. CLOUD OF ORIGIN: Cb or vigorous Cu
  6. HAIL: Small balls/pieces of ice falling either separately or massed together in irregular lumps. The diameter is very variable, from 5 mm up to 50 mm, or (in really extreme cases more.) Hailstones are composed almost exclusively of transparent ice, or a series of transparent layers, alternating with semi-transparent/cloudy ice, indicative of the formation process whereby the hailstone has spent some time inside the parent Cb, at varying altitudes gathering coats of ice from different regions of the cloud, and having markedly different temperatures from region to region. Where possible, the diameter of the hailstones should be noted, and photographs taken against a scale/ruler etc.
  7. ICE NEEDLES: (or ice prisms, or diamond dust) unbranched ice crystals, in the form of needles, columns or plates: so tiny that they often appear to be suspended in the air, and can give rise to spectacular visual effects when the sun shines through them - indeed it is not necessary for cloud to be present as they are often observed (in polar latitudes) under a clear sky. Need very low temperatures, and conditions are often not right in maritime NW Europe - better inland.