FLIGHT CATEGORIES 14 OCT 2000Z
VFRMVFRIFRLIFR
Illustration
METAR remarks (RMK) decoded: US and Canadian examples
Decode the RMK section: AO2, SLP, the T group and peak wind in US reports, and cloud types such as SC6AC2 and density altitude in Canadian ones.
12 min read
Everything after RMK in a METAR is a remark. The body of the report has a fixed order. The remarks add what the body cannot say: the cloud types in Canada, the temperature to a tenth of a degree and the peak wind in the US, and the sea-level pressure in both countries. This page decodes the remarks. For the groups before RMK, start with how to read a METAR.
Here is a real report from Montréal / St-Hubert (CYHU), 14 October 2022 at 20:00 UTC, from the Iowa Environmental Mesonet (IEM) archive:
The report has two remarks: the sea-level pressure (SLP127) and the density altitude (DENSITY ALT 200FT). It has no cloud types. AUTO marks a report from an automated station, and the TC AIM says in MET 8.2 that the AWOS cannot report cloud types.
At the same hour, Ottawa (CYOW) sent a report without AUTO (real report, IEM):
CU3 is a Canadian cloud remark: cumulus covers 3 eighths of the sky. It is the scattered layer at 5,500 ft.
What RMK is, and who uses it
RMK starts a section of “information included by national decision” (MANOBS 10.2). So each country’s manual sets its own remarks. This page covers the Canadian and the US rules.
| Country | Rules | What the remarks hold |
|---|---|---|
| Canada | TC AIM MET 8.2; MANOBS chapters 4, 7, 9, 10 and 12 | Cloud types and their eighths, general weather remarks, the sea-level pressure, the density altitude |
| United States | FMH-1 12.7 | Automated, manual and plain-language remarks (12.7.1), then additive and maintenance data (12.7.2) |
Three general rules help you read them:
- Directions go clockwise, in both countries.
PRFG SE-Nmeans that fog partly covers the aerodrome from the southeast to the north (TC AIM MET 8.2). FMH-1 12.7(f) uses the eight points of the compass, also clockwise. - US times can be short. A time in the same hour as the report can be in minutes past the hour only. A time in another hour takes hours and minutes (FMH-1 12.7(b)). The hour is not required when you can infer it from the report time (FMH-1 12.7.1(d)), but it can be there: the Burlington report below gives
1108for 11:08 in its 11:54 report. - US distances are in statute miles. One exception: automated lightning remarks are in nautical miles (FMH-1 12.7(d)).
Canadian remarks: cloud types and eighths
Canadian remarks include, where observed, “layer type and cloud or obscuring phenomena (in eighths of sky covered or oktas)” (TC AIM MET 8.2). The observer judges each layer’s amount and opacity (TC AIM MET Table 8.3). The cloud types come from TC AIM MET 8.2 and MANOBS Table 7-3:
| Code | Cloud | Code | Cloud |
|---|---|---|---|
CI |
Cirrus | NS |
Nimbostratus |
CS |
Cirrostratus | ST |
Stratus |
CC |
Cirrocumulus | SF |
Stratus fractus |
AS |
Altostratus | SC |
Stratocumulus |
AC |
Altocumulus | ACC |
Altocumulus castellanus |
CU |
Cumulus | CF |
Cumulus fractus |
TCU |
Towering cumulus | CB |
Cumulonimbus |
Obscuring phenomena take weather codes, such as FG fog, SN snow, HZ haze and FU smoke (MANOBS Table 7-4). When a station sees CB or TCU, the remarks always report it, even when it is only embedded or distant (TC AIM MET 8.2).
How the digits match the cloud groups
The body gives each layer the summation amount: the sum of the eighths of all the layers at and below that level (MANOBS 7.2.6 and 7.3). The remark gives each type on its own. So you can add the digits from left to right.
Here is the sample report of TC AIM MET 8.2, from Saskatoon (CYXE). It is a real-format example, not a real report:
| Remark | Running total | Group in the body |
|---|---|---|
SF5: stratus fractus, 5 eighths |
5 eighths: broken | BKN008 |
NS3: nimbostratus, 3 eighths |
5 + 3 = 8 eighths: overcast | OVC040 |
The TC AIM decode says the same thing: broken at 800 ft, then, “combined with the lower layer, overcast clouds at 4 000 ft”. The cover words come from TC AIM MET Table 8.2: few is less than 1/8 to 2/8, scattered 3/8 to 4/8, broken 5/8 to less than 8/8, overcast 8/8.
The example in the METAR guide works the same way. SC6AC2 gives 6 eighths for BKN028, then 6 + 2 = 8 for OVC060. All the cloud remarks on this page add up this way. The ceiling always comes from the body: the lowest layer reported as BKN or OVC, or the vertical visibility (MANOBS 7.2.7). More in what is a ceiling.
VIS NW 3/8 in the CYXE sample is a sector visibility: 3/8 SM to the northwest, against 3/4 SM prevailing. A sector goes in the remarks when it is half or less, or double or more, of the prevailing visibility (MANOBS 4.5.1.3).
Fog in the remarks
Fog takes eighths too, and in these real reports its eighths add up with the cloud eighths. The reports come from Québec City (CYQB) on 15 October 2022 (IEM) and show a fog layer that shrinks:
| Time (UTC) | Type | Body | Remark |
|---|---|---|---|
| 10:48 | SPECI | 1/2SM FG VV001 |
FG8: fog hides the whole sky. |
| 11:00 | METAR | 1/8SM FG BKN001 BKN150 |
FG4ST1AC1: fog 4 and stratus 1 make 5 (broken at 100 ft); altocumulus 1 makes 6 (broken at 15,000 ft). |
| 12:15 | SPECI | 1 1/4SM BR VCFG FEW002 SCT150 |
FG1ST1AC1: fog 1 and stratus 1 make 2 (few at 200 ft); altocumulus 1 makes 3 (scattered at 15,000 ft). |
The 12:15 report in full:
VIS W-N 5 is another sector visibility: 5 SM from the west to the north. The prevailing visibility is 1 1/4 SM, so the sector is double or more (MANOBS 4.5.1.3). For mist, fog and how fog forms and clears, see fog types for pilots.
Other Canadian remarks
| Remark | Meaning | Source |
|---|---|---|
SLP134 |
Sea-level pressure, 1013.4 hPa. It is the last mandatory field of a METAR. | TC AIM MET 8.2; MANOBS 9.2.2 |
DENSITY ALT 200FT |
Density altitude, 200 ft. TC AIM MET 8.2 writes DENSITY ALTITUDE. |
MANOBS 12.2.6.3 |
SLP MISG, PRES MISG, DENSITY ALT MISG |
The sea-level pressure, the pressure (altimeter) data or the density altitude is missing. | MANOBS 12.2.6.3 |
PRESRR, PRESFR |
Pressure rising or falling rapidly: 2.0 hPa per hour or more. Get the latest altimeter setting. | MANOBS 9.2.1; TC AIM MET 8.2 |
FROIN |
Frost on indicator: frost since the previous report. | TC AIM MET 8.2 |
WND EST |
The wind sensors at a staffed site do not work, so the observer estimated the wind. | TC AIM MET 8.2 |
VIS VRB 1/4-3/4 |
Variable visibility, from the lowest value to the highest. | MANOBS 4.5.2, 4.6 |
LTNG DIST NE |
An AWOS detected lightning more than 10 SM and up to 30 SM away, to the northeast. | TC AIM MET 8.4.3(a) |
ICE, ICE INTMT, ICE PAST HR |
An automated station detected icing. | TC AIM MET Table 8.3 |
DENSITY ALT gives the density altitude at the station. The density altitude guide explains when the remark appears and how to use it.
Automated stations give fewer remarks. TC AIM MET Table 8.3 says that AUTO reports describe no clouds or obscuring phenomena. They can give variable visibility, icing, precipitation amounts, rapid pressure changes and the location of lightning. The sea-level pressure and the density altitude are the same as in a staffed report.
SLP is not the altimeter setting
TC AIM MET 8.2 is direct: “SLP does not directly relate to altimeter setting, as the SLP is based upon actual temperatures, while the altimeter setting is based upon the ICAO standard atmosphere.” In the US, the station reduces its pressure to sea level with the 12-hour mean temperature (FMH-1 11.4.4).
To read SLP, put back the missing first digits. The group holds the last three digits of the pressure in hectopascals, with the tenths (MANOBS 9.2.2):
SLP127is 1012.7 hPa (TC AIM MET Table 8.3).SLP982is 998.2 hPa (FMH-1 12.7.1(w)).SLPNOmeans the sea-level pressure is not available (FMH-1 12.7.1(w)).
In the CYHU report above, A2990 is 29.90 inHg, about 1012.5 hPa (computed with 1 inHg = 33.8639 hPa). SLP127 is 1012.7 hPa. The two values are close at this station, but they come from different methods. Set your altimeter to the A group, not to SLP. More on altimeter settings: QNH, QFE and the altimeter setting.
US remarks: automated, manual and plain language
FMH-1 12.7.1 lists these remarks. They usually come in this order (FMH-1 12.7(g)).
| Remark | Meaning | FMH-1 |
|---|---|---|
AO1, AO2 |
Automated station without (AO1) or with (AO2) a precipitation discriminator | 12.7.1(c) |
PK WND 28045/15 |
Peak wind since the last METAR: 280° at 45 kt, at 15 minutes past the hour | 12.7.1(d) |
WSHFT 30 FROPA |
Wind shift that began at 30 minutes past the hour, with a frontal passage | 12.7.1(e) |
TWR VIS 1 1/2 |
Tower visibility, 1 1/2 SM (SFC VIS for surface visibility) |
12.7.1(f) |
VIS 1/2V2 |
Prevailing visibility varying between 1/2 and 2 SM | 12.7.1(g) |
VIS NE 2 1/2 |
Sector visibility: 2 1/2 SM in the northeast | 12.7.1(h) |
OCNL LTGICCG OHD |
Occasional lightning, in-cloud and cloud-to-ground, overhead | 12.7.1(j), Table 12-5 |
RAB05E30SNB20E55 |
Rain began at :05 and ended at :30; snow began at :20 and ended at :55 | 12.7.1(k) |
TSB0159E30 |
Thunderstorm began at 01:59 and ended at 02:30 | 12.7.1(l) |
CIG 005V010 |
Ceiling varying between 500 and 1,000 ft | 12.7.1(q) |
PRESRR, PRESFR |
Pressure rising or falling rapidly: at least 0.06 inHg per hour, and 0.02 inHg or more in total | 12.7.1(v), 11.4.6 |
SLP982, SLPNO |
Sea-level pressure 998.2 hPa; not available | 12.7.1(w) |
NOSPECI |
A manual station that takes no SPECIs: no change is reported until the next METAR | 12.7.1(y) |
SNINCR 2/10 |
Snow depth up 2 in during the past hour, 10 in on the ground | 12.7.1(z) |
AO2 is not AUTO. AO1 or AO2 is in every report from an automated station (FMH-1 12.7.1(c)). A precipitation discriminator is a sensor that tells liquid, freezing and frozen precipitation apart (FMH-1 glossary). AUTO in the body means something else: a fully automated report “with no human intervention or oversight” (FMH-1 12.6.4). So a report can carry AO2 without AUTO, as the Burlington example below does. If NOSPECI is new to you, read what triggers a SPECI.
Lightning. The frequency words are OCNL (less than 1 flash a minute), FRQ (about 1 to 6) and CONS (more than 6). The types are CG (cloud-ground), IC (in-cloud), CC (cloud-cloud) and CA (cloud-air) (FMH-1 Table 12-5). Automated stations in both countries code lightning by distance, but with different units:
| Lightning detected | US automated station (FMH-1 12.7.1(j)) | NAV CANADA AWOS (TC AIM MET 8.4.3(a)) |
|---|---|---|
| Close | TS: within 5 NM |
TS: 6 SM or less |
| Near | VCTS: 5 to 10 NM |
VCTS: more than 6 to 10 SM |
| Far, in the remarks | LTG DSNT and the direction: beyond 10 and less than 30 NM |
LTNG DIST and the direction: more than 10 to 30 SM |
US additive data and sensor status
Additive groups come only from designated stations. Maintenance groups come only from automated stations (FMH-1 12.7.2).
| Group | Meaning | FMH-1 |
|---|---|---|
P0009 |
Precipitation since the last METAR: 0.09 in. P0000 is less than 0.01 in. |
12.7.2(a) |
60217 |
Precipitation in the past 3 or 6 hours: 2.17 in. 60000 is a trace. |
12.7.2(a) |
70125 |
Precipitation in the past 24 hours: 1.25 in, in the 12:00 UTC report | 12.7.2(a) |
4/021 |
Snow depth on the ground: 21 in | 12.7.2(a) |
T00261015 |
Temperature 2.6 °C, dew point −1.5 °C | 12.7.2(d) |
10142 |
Highest temperature in the past 6 hours: 14.2 °C | 12.7.2(e) |
21001 |
Lowest temperature in the past 6 hours: −0.1 °C | 12.7.2(f) |
401001015 |
Highest and lowest in the past 24 hours: 10.0 °C and −1.5 °C | 12.7.2(g) |
52032 |
Pressure tendency over 3 hours: code 2 (increasing steadily or unsteadily), 3.2 hPa | 12.7.2(h), Tables 12-7 and 12-8 |
RVRNO, PWINO, PNO, FZRANO, TSNO |
Sensor missing or not working: RVR, present weather, rain gauge, freezing rain, lightning | 12.7.2(j) |
VISNO LOC, CHINO LOC |
Second visibility or ceiling sensor not working; LOC is where the sensor is, such as RWY11 |
12.7.2(j) |
$ |
The station needs maintenance | 12.7.2(k) |
The T group
T starts the group. Then each value takes four digits: a sign digit and three digits in tenths of a degree. The sign digit is 1 below 0 °C and 0 at 0 °C or above (FMH-1 12.7.2(d)). So T00261015 reads as 0 026 (+2.6 °C) and 1 015 (−1.5 °C).
The body rounds the same values to the nearest whole degree, and −0.5 °C becomes M00 (FMH-1 12.6.10). A real Burlington report at 13:54 UTC on 1 November 2025 (IEM) had T00671006 in the remarks and 07/M01 in the body: 6.7 °C and −0.6 °C.
The pressure tendency
The 5 group gives the character of the pressure change over 3 hours (FMH-1 Table 12-7), then the amount in tenths of hectopascals (Table 12-8):
| Code | Pressure now, compared with 3 hours ago |
|---|---|
1, 2, 3 |
Higher |
0 |
Higher or the same (increasing, then decreasing) |
4 |
The same (steady) |
5 |
The same or lower (decreasing, then increasing) |
6, 7, 8 |
Lower |
Worked example: a full US report
Here is a real report from Burlington, Vermont (KBTV), 1 November 2025 at 11:54 UTC (IEM):
| Group | Meaning |
|---|---|
AO2 |
Automated station with a precipitation discriminator |
PK WND 30030/1108 |
Peak wind since the last METAR: 300° at 30 kt, at 11:08 UTC |
SLP048 |
Sea-level pressure 1004.8 hPa |
60001 |
0.01 in of precipitation in the past 6 hours |
70118 |
1.18 in of precipitation in the past 24 hours |
T00610006 |
Temperature 6.1 °C, dew point 0.6 °C (the body shows 06/01) |
10067 |
Highest temperature in the past 6 hours: 6.7 °C |
20061 |
Lowest temperature in the past 6 hours: 6.1 °C |
53040 |
Pressure 4.0 hPa higher than 3 hours ago. Code 3: decreasing or steady, then increasing; or increasing, then increasing more rapidly. |
$ |
The station needs maintenance |
This is the 12:00 UTC report. FMH-1 12.6.3 names a report by its scheduled time: its own example, KDCA at 08:55 UTC, is the 09:00 report. The FMH-1 glossary calls the 00:00, 06:00, 12:00 and 18:00 UTC reports “6-hourly”. So the 6 group gives the precipitation of the past 6 hours (FMH-1 12.7.2(a)), and the report has the 6-hour highest and lowest temperatures (1 and 2, FMH-1 10.5.2). It also has the 24-hour group (7), which FMH-1 12.7.2(a) puts in the 12:00 UTC report.
Say it in plain words:
Automated station that can tell precipitation types apart. Peak wind 300° at 30 knots at 11:08. Sea-level pressure 1004.8. A hundredth of an inch of precipitation in 6 hours, 1.18 inches in 24. Temperature 6.1, dew point 0.6. High 6.7 and low 6.1 over 6 hours. Pressure up 4.0 hectopascals in 3 hours. Maintenance needed.
In the ClearMETAR app, you can tap each group of a report, remarks included, to open its explanation. The app decodes the Canadian cloud types and every remark group of this Burlington report.
For the codes in the body of a report, such as BR, VCFG or BKN, see METAR codes and abbreviations.
Sources
- Federal Meteorological Handbook No. 1 (FCM-H1-2019): 10.5.2, 11.4, 12.6.3, 12.6.4, 12.6.10, 12.7, Tables 12-5, 12-7 and 12-8, glossary, Office of the Federal Coordinator for Meteorological Services and Supporting Research
- TC AIM (TP 14371) 2026-2, MET 8.2, 8.4.3, Tables 8.2 and 8.3, Transport Canada
- MANOBS, 8th edition, Amendment 3 (September 2026): 4.5, 4.6, 7.2, 7.3, 7.4, 9.2, 10.2, 12.2.6, Environment and Climate Change Canada
- ASOS/METAR archive: CYHU, CYOW and KBTV routine reports; CYQB routine and special reports, Iowa Environmental Mesonet, Iowa State University
Rules checked on
Common questions
What does AO2 mean?
AO2 marks a US automated station with a precipitation discriminator, a sensor that tells liquid, freezing and frozen precipitation apart; AO1 marks one without it (FMH-1 12.7.1(c) and glossary).
What is SLP in a METAR?
SLP is the sea-level pressure in hectopascals, written as its last three digits with the tenths, so SLP127 is 1012.7 hPa (MANOBS 9.2.2, FMH-1 12.7.1(w)). It is not the altimeter setting: SLP uses the actual temperature, and the altimeter setting uses the ICAO standard atmosphere (TC AIM MET 8.2).
What does the T group in the remarks mean?
It gives the temperature and the dew point to a tenth of a degree Celsius, and a sign digit of 1 means below 0 °C, so T00261015 is 2.6 °C and −1.5 °C (FMH-1 12.7.2(d)).
What does PK WND mean?
PK WND is the peak wind since the last METAR: PK WND 28045/15 is 45 knots from 280° at 15 minutes past the hour (FMH-1 12.7.1(d)).
What do remarks like SC6AC2 mean in a Canadian METAR?
They give each cloud type with the eighths of sky it covers: SC6AC2 is stratocumulus 6/8 and altocumulus 2/8 (TC AIM MET 8.2). The body gives the running total for each layer, so 6 makes the broken layer and 6 + 2 = 8 makes the overcast layer (MANOBS 7.3).