Reading Aviation Weather: METARs & TAFs Explained

What is a METAR, and why it matters

A METAR is an observed aviation routine weather report — a snapshot of the actual weather at an airport at a specific moment, issued roughly once an hour (in practice, observed and transmitted in the 10 or so minutes before the hour). It is an observation, not a forecast: it tells you what a sensor or a human observer measured. When conditions change significantly between the scheduled hourly reports, a station files a SPECI — a special, off-schedule report — so you learn about a fast-moving change immediately instead of waiting for the next observation.

METARs matter because they are the ground truth behind every preflight decision. A forecast can be optimistic; the METAR tells you what is genuinely happening at your departure, destination, and alternate right now. It is written in a terse, standardized international code so that a report from Denver, Frankfurt, or Tokyo is read the same way. Once you can read it fluently, you stop translating abbreviations and start seeing the weather — the wind, the ceiling, the trend toward fog — directly in the string.

The good news: a METAR is not random. It is always read left to right in a fixed order, and every element has a predictable shape. Learn the order once and any report becomes a checklist you run top to bottom.

The full anatomy of a METAR, in order

Here is a real, fair-weather report from Denver. Below it, the same string is broken into its twelve numbered elements. Every METAR follows this sequence; some elements (like RVR or a COR modifier) simply don’t appear when they aren’t needed. We color-code the fields here and reuse the exact same colors in the diagram, the code chips, and both worked decodes later — one visual grammar for the whole lesson.

Wind: direction, speed, gusts, and variability

Wind is coded as dddssKT — three digits of direction, two (or three) digits of speed, then KT for knots. The direction is degrees true (the tower converts to magnetic when it reads you the wind on frequency, but the METAR text itself is true). So 09008KT is wind from 090° true at 8 knots. Wind always names the direction it blows from.

Gusts matter for takeoff and landing performance and for your personal crosswind limits — a steady 15 is a very different runway than 15 gusting 25. When the direction is genuinely swinging you’ll see a variable range appended, such as 18010KT 150V210; if the whole wind is light and shifting, it collapses to VRB. But the runway you use only “feels” the part of the wind along and across it — so resolve the reported wind onto your runway heading before you commit:

Visibility & RVR

In the United States, prevailing visibility is reported in statute miles followed by SM: 10SM is ten statute miles, 3SM is three. Fractions appear as you’d write them — 1/2SM, 1 1/2SM. A leading M here means “less than”: M1/4SM is less than a quarter mile. (Watch that M — in the temperature field the same letter means “minus.” Context tells you which.)

RVR — Runway Visual Range — is a separate, instrument-measured horizontal visibility down a specific runway, reported in feet when visibility is low. R06/2400FT reads: on runway 06, the RVR is 2,400 feet. RVR only appears in poor conditions and governs many instrument-approach landing minimums, so its presence alone is a signal the weather is marginal.

Present-weather codes (with a full decode key)

Present weather is where new pilots feel lost — but the codes are built from a simple grammar. Read each group as intensity → descriptor → phenomenon and chain the pieces together. +SHRA is heavy (+) showers (SH) of rain (RA). TSRA is a thunderstorm (TS) with rain (RA). FZRA is freezing (FZ) rain (RA). -DZ is light drizzle. Learn the grammar once and you can read combinations you’ve never seen before.

Step 1 — Intensity & proximity (the prefix)

Step 2 — Descriptor

Step 3 — Phenomenon

Sky cover & how to find the ceiling

Sky condition describes cloud layers from the ground up, each as a coverage code plus a height in hundreds of feet above ground level (AGL). Coverage is measured in eighths of the sky (oktas): picture the sky dome divided into eight parts and count how many hold cloud. BKN012 means a broken layer (5–7 eighths) at 1,200 feet AGL.

Convective clouds get a suffix: CB for cumulonimbus (thunderstorm cloud) and TCU for towering cumulus. So BKN040CB is a broken layer at 4,000 feet made of thunderstorm cloud — a very different day from plain BKN040.

Finding the ceiling: cross out FEW and SCT, ignore anything above

The ceiling is the height of the lowest BROKEN or OVERCAST layer (or a VV value). FEW and SCT are never ceilings — there’s enough gap in the sky to be considered “below” them. Scan upward from the ground, skip past every FEW and SCT, and stop at the first BKN or OVC. That layer is your ceiling — and every layer above it is irrelevant.

The same three-step scan handles any report. In our Denver example FEW070 SCT120 BKN250, you cross out FEW and SCT and the ceiling is 25,000 ft — the first (and only) broken-or-overcast layer. In a low-ceiling case like SCT006 BKN012 OVC030, you skip the SCT layer, stop at BKN012 for a 1,200 ft ceiling, and the OVC030 above it doesn’t matter. Read up, skip FEW/SCT, stop at the first BKN/OVC.

Temperature, dewpoint & the spread

Temperature and dewpoint are reported in degrees Celsius, separated by a slash, with M meaning minus: 24/09 is 24 °C over a 9 °C dewpoint, and M04/M07 is −4 °C over −7 °C. The single most useful thing this field gives you isn’t either number alone — it’s the spread between them.

The dewpoint is the temperature to which the air must cool to become saturated. A small temperature–dewpoint spread means the air is close to saturation, so fog, mist, and low cloud become likely — especially overnight and near dawn as the temperature falls toward the dewpoint. A wide spread means dry air and generally better visibility.

This is the field that lets you anticipate. If an evening METAR shows a 2 °C spread and calm winds, the morning may well bring fog before your planned departure — a forecast the numbers hand you for free.

Altimeter setting

In the U.S. the altimeter setting is coded as A followed by four digits of inches of mercury with the decimal implied: A3012 = 30.12 inHg, and A2992 = 29.92 inHg. You dial this into the Kollsman window so your altimeter reads true field elevation. Internationally you’ll see Q plus hectopascals instead — Q1013 = 1013 hPa.

Altimeter setting is not an altitude — it’s the pressure reference. But it feeds directly into performance: pair it with weight and balance and runway numbers using the weight & balance calculator and the performance calculator before a high-density-altitude departure.

Remarks (RMK)

Everything after RMK is supplementary detail. Two you’ll see constantly identify the station type: AO2 is an automated station with a precipitation-discriminating sensor (it can tell rain from snow), and AO1 is automated without one. Another common entry is sea-level pressure, coded SLPppp: SLP184 decodes to 1018.4 hPa (prepend a 9 or 10 to land nearest 1000).

Remarks are where a station adds nuance a machine or observer thought you should know — pressure trends, when a thunderstorm began, peak wind gusts. Don’t skip them; they often carry the detail that tips a marginal decision.

The four flight categories

Ceiling and visibility together sort every report into one of four flight categories — a fast, standardized read on how much of a challenge the weather poses. You apply both the ceiling and the visibility criteria and take the more restrictive (worse) of the two.

METAR vs TAF, and TAF change groups

A METAR tells you what’s happening now. A TAF — Terminal Aerodrome Forecast — tells you what’s expected. A TAF is a forecast for the area within roughly a 5-statute-mile radius of the airport, valid for 24 to 30 hours, and it reuses almost all the same codes as a METAR. The big new idea is change groups: the TAF states a base forecast, then layers on how and when conditions are expected to shift.

The change groups, on one timeline

Four change groups describe four different shapes of change. This is exactly where students conflate them — so here they are on a single timeline: FM is a clean step, BECMG is a ramp across a window, TEMPO is brief spikes, and PROB is a dashed “maybe” block.

Here’s a TAF that uses all four:

Read top to bottom: valid from the 14th at 1800Z to the 15th at 2400Z; base conditions light east wind and more than 6 SM; from 2000Z the wind veers and picks up; temporarily between 2000–2300Z mist may drop visibility to 5 SM; the wind gradually becomes southerly early on the 15th; and there’s a 30% chance of thunderstorms with rain from 0600–0900Z. Note P6SM means “greater than 6 SM.”

Where to get aviation weather

METARs and TAFs are free and everywhere. Get them before every flight from a source you trust, and cross-check the observation against the forecast.

ATIS, ASOS, and AWOS give you the current observation over the radio as you approach a field — the same data as the METAR, spoken aloud. On the ground, decode and sanity-check any raw report instantly with the METAR decoder.

Two worked decodes, token by token

Now we run the whole checklist on two contrasting reports — a clear VFR day and a low-IFR morning — using the same field colors from the anatomy diagram, so the structure of each report reads with one visual grammar.

Example 1 — a clear, VFR day at Denver

Category (worse of two): ceiling 25,000 ft is VFR (>3,000) and visibility 10 SM is VFR (>5). Both rows are VFR, so the report is VFR. A textbook good day: light wind, high thin cloud, dry air.

Example 2 — low IFR at San Francisco

Category (worse of two): the ceiling is 600 ft, which falls in the 500-to-<1,000 ft band — that’s IFR. Visibility 3 SM sits at the bottom of the MVFR band (3–5 SM). You take the worse of the two rows, so the IFR ceiling outranks the MVFR-boundary visibility and the report is IFR. Notice the 2 °C spread had already warned of the low, moist conditions before we computed anything — and it warns that if the temperature nudges toward the dewpoint, this could sink toward LIFR fog.

Common mistakes & misconceptions

These are the errors that trip up almost every student pilot at least once. Get ahead of them.

How this drives a go/no-go decision

Reading a METAR is only useful if it changes what you do. Here’s the reasoning chain from raw text to decision, the way a careful pilot runs it:

1. Category first. Derive VFR/MVFR/IFR/LIFR at departure, destination, and alternate. Does it match your ratings, currency, and the aircraft’s equipment?
2. Wind & gusts. Resolve the wind onto the runway heading and compare the crosswind component (and gust factor) to your personal limit. Run the numbers through the performance calculator.
3. Trend from the spread. A small temp–dewpoint spread plus a falling temperature warns of fog forming before you arrive — even if it’s clear now.
4. Density altitude. Hot, high, or low-pressure? Convert temp and altimeter to density altitude with the density altitude tool and confirm you’ll actually climb.
5. Weight & balance. Verify loading against the day’s performance with the weight & balance calculator.
6. Forecast the window. Read the TAF across your whole flight window — a TEMPO or PROB30 for thunderstorms at your ETA is a no-go even if the current METAR is calm.

Frequently asked questions