Surface weather observations

Why can a METAR and a tower wind disagree?

Use the direction reference, sampling location and averaging period before comparing the numbers.

Mean wind versus a gust

Surface wind is measured with an anemometer and direction sensor, or an ultrasonic wind sensor, exposed to represent the relevant aerodrome or runway area rather than a building wake. The standard surface-wind reference height is about 10 m above the ground. Wind direction is the direction from which the air arrives, not the direction towards which it travels.

An ICAO/EASA METAR normally reports the preceding 10-minute mean direction and speed; a marked discontinuity can shorten the representative averaging interval. The reported gust is the maximum short-period speed, assessed with a 3-second average, during the preceding 10 minutes or relevant interval since that discontinuity. It is included when it exceeds the mean by at least 10 kt. Thus 24015G25KT gives a 15 kt mean and a 25 kt maximum gust, a 10 kt excess—not an extra 25 kt to add to 15.

Do not use one instantaneous spike as a 3-second average, or the gust as the mean. The local routine/special-report threshold can be 5 kt where noise-abatement procedures apply; the normal METAR rule remains 10 kt. Instrument smoothing, obstructed exposure and rapidly changing conditions all affect representativeness.

Report reference versus the wind used by the pilot

The meteorological wind in METAR/SPECI and local MET REPORT/SPECIAL is expressed relative to true north. The runway wind passed by ATS for take-off and landing normally uses a 2-minute mean, is representative of the relevant runway section and is normally passed to the pilot in magnetic degrees under the applicable ATS/AIP convention. True references are used in specified areas or procedures; check the published convention rather than converting twice.

The shorter interval responds more quickly to recent change. Different runway sensors may also see different flows. A METAR and the latest tower report therefore need not match even when both are correct. First check observation time, true/magnetic reference, sensor location and whether you are comparing a mean with a gust. A general statement that every wind spoken on the radio is magnetic is unsafe: route weather forecasts and information relayed verbatim retain their stated reference.

What is the observer or instrument actually assessing?

Separate the physical ability to detect a target from the coverage of the observing system.

Meteorological visibility and ground visibility

Visibility for aeronautical meteorological purposes is the greater of two distances: the greatest distance at which a suitably sized black object near the ground can be seen and recognised against a bright background, and the greatest distance at which lights of approximately 1,000 candelas can be seen and identified against an unlit background. The two components account for different day/night/background conditions; they do not mean that all terrain or every dim light will be visible that far.

Ground visibility is visibility at an aerodrome as reported by an accredited observer or automatic systems. It is distinguished from flight visibility assessed forward from the cockpit in flight. A ground report can be limited by its location and does not guarantee the same slant range from an approach or visibility everywhere along a route.

A human observer compares known-distance landmarks or suitable lights, considers illumination and all relevant directions, and uses visibility sensors where provided. A transmissometer measures how much a beam is attenuated along a known path. A forward-scatter meter samples scattered light from a small volume and derives optical extinction through its calibration. These are local measurements: an isolated fog patch beyond a sensor can be missed, and a patch at a sensor can be unrepresentative of the rest of the field. Automated reporting needs the declared sensor coverage and limitations to be understood.

Units and reporting resolution

Visibility is expressed in metres or kilometres in ICAO/EASA reports; US-style METARs use statute miles, not nautical miles. One statute mile is 1,609.344 m. For example 3 SM is about 4.83 km, not 3 NM. European METAR visibility uses four digits in metres below 10 km; 9999 means 10 km or more, not an exact 9,999 m.

Reported resolution is not instrument precision. Values are rounded down in specified steps: 50 m below 800 m; 100 m from 800 m to below 5 km; 1 km from 5 km to below 10 km. Below 50 m is coded 0000. A physical estimate of 4,850 m is therefore reported 4800. These conventions differ from RVR reporting steps.

Can prevailing visibility be good while one approach is poor?

Combine sectors by angular coverage, not by averaging their kilometre values.

At least half the horizon

Prevailing visibility is the greatest visibility value reached over at least half the horizon circle or half the surface of the aerodrome. The qualifying sectors may be adjacent or separated. It is not the lowest value everywhere, the best single direction, nor an arithmetic mean of sector values. For equal-width sectors, count how many meet or exceed a candidate distance, then find the greatest distance for which at least half qualify.

Suppose four of eight equal 45° sectors have 8 km, two have 4 km and two have 1 km. At least half have 8 km, so the prevailing value is 8 km even though the worst sector is 1 km. A suitably reported lower directional value can alert a pilot to that discrepancy. Under the EASA convention, a differing minimum is also reported when it is less than 1,500 m, or less than half the prevailing value and less than 5,000 m; give its direction when possible. Rapid fluctuations preventing a prevailing assessment are handled by reporting the lowest value without a direction.

A regional reporting example must be read under its own rules: UK CAP 746 is useful for observing practice but is not the EU legal authority. Use Part-MET for the EASA convention. A good prevailing value cannot override a limiting reported RVR or substitute for the conditions required at a particular runway and time.

Why is runway visual range not simply the visibility in metres?

The pilot’s runway view depends on light and marking detection as well as atmospheric extinction.

Definition and instrument chain

Runway visual range is the range over which a pilot on the runway centre line can see the runway surface markings or the lights delineating the runway or identifying its centre line. It relates to a specific runway and viewing task, while ground and prevailing visibility describe the general near-surface atmosphere. The same fog can produce different visibility and RVR because the targets, light intensity and background differ.

An instrumental RVR system measures atmospheric extinction using a transmissometer or a forward-scatter meter, measures background luminance, and uses the appropriate runway-light intensity to calculate the range. The sensor does not put a miniature pilot in the cockpit or directly observe every runway light. A transmissometer has a source and receiver separated by a known baseline; a forward-scatter meter samples a small volume with a prescribed scattering geometry. Both need calibration, clean optics and representative siting. Patchy fog, snow contamination and unrepresentative sampling can limit the result.

Marking contrast and light detection have different physical relationships. Increasing lights in a conceptual example can improve light range without improving optical clarity; it cannot be assumed to remove fog. Do not multiply visibility by one universal number to obtain RVR. Approved conversion rules, where applicable for a particular operation, have restrictions and do not replace available reported RVR.

Where the instruments belong

Installations sample the runway touchdown zone, and where required the midpoint and stop end. The touchdown representative site is about 300 m beyond the threshold; representative stop-end sampling is about 300 m before the other end. Exact siting and any additional sites depend on runway length and local meteorology. The EASA guidance describes instrumental assessment at about 2.5 m above the runway and not more than 120 m laterally from the centre line. These are siting references, not the length of the optical beam and not permission to place obstacles in a runway protected area.

Local runway reports can give touchdown, midpoint and stop-end values separately, so a pilot can see variation along the runway. A METAR normally carries the touchdown-zone value for each reported runway, up to four runways; it cannot be assumed to contain every operationally relevant measurement. The required availability and use of each value depend on the operation, aircraft/operator approval and current aerodrome information.

Where should you look for the latest runway conditions?

Treat the report source, representative zone, unit and averaging time as part of the value.

Units, channels and updates

RVR is reported in metres in European/ICAO reports; feet are used in some other systems, notably US runway reports. The unit must be explicit where needed. A value of 1,200 ft is only about 366 m; it must never be read as 1,200 m. In standard ICAO METAR groups the unit is metres unless the national format explicitly uses FT.

RVR reaches pilots through METAR/SPECI, current ATIS or D-ATIS, and ATS voice reports; suitable authorised data-link services can also relay it where available. METAR is a periodic aerodrome report and can be older or represent a different averaging interval from the latest tower display. Compare UTC time, runway designator, touchdown/midpoint/end label, unit, tendency and any beyond-range or unavailable indication. Ask ATS for missing operational information rather than treating absence as good conditions.

A standard example R27L/0650U identifies runway 27 left, 650 m and an upward tendency; it does not establish that midpoint and stop-end values are also 650 m or that the next few minutes will improve. M and P indicate below or above the system’s reportable limit. Ten-minute mean RVR is used for METAR, with special handling after a marked discontinuity; local reports use shorter averaging, normally one minute. RVR changes can require updates or special reports between routine observations.

Can a visibility sensor alone tell fog from heavy rain?

Combine visual, optical and other observations to classify the weather.

Human and automated observations

An observer identifies precipitation type and intensity, visibility-reducing phenomena, nearby weather and evidence such as thunder or lightning using the view, sound, suitable lighting, instruments and local procedures. An automated station combines relevant sensors: optical present-weather detectors analyse particle scattering or interruption; precipitation sensors and temperature assist phase identification; lightning networks can support thunderstorm detection. A single visibility measurement does not uniquely determine the phenomenon.

Different mixtures and droplet/particle sizes can challenge a sensor. Freezing drizzle, ice pellets, snow, blowing particles and shallow or nearby fog may be difficult to distinguish without appropriate sensor coverage or human assessment. AUTO indicates an automated report; missing or unavailable elements must not be read as no significant weather. Human reports also have limits, particularly at night, in precipitation and when a hazard is hidden by cloud.

Compare an observation with other evidence rather than forcing one label from one variable. A low visibility with wet suspended droplets supports fog; the same range with dry dust supports a different code and hazard mechanism. A detector’s inability to see distant convection is not evidence of a clear approach or route.

Does a low scattered layer define the ceiling?

Connect visual cover, measured base and the datum used in the report.

Observe type, amount, base and top

A human observer assesses cloud form and the fraction of sky covered, using all visible layers. One okta is one eighth of the sky. FEW represents 1–2 oktas, SCT 3–4, BKN 5–7 and OVC 8. These are cover categories, not cloud thickness, precipitation intensity or the fraction of the runway covered by fog. Layer amounts are assessed separately according to reporting rules; simply adding coded layers is misleading.

A ceilometer sends light pulses upward and detects backscatter. The return time locates a cloud base: distance equals light speed multiplied by round-trip time divided by two. It samples a limited column; repeated measurements and algorithms may estimate cover, but overhead sampling can miss off-axis cloud and does not directly measure the entire sky or cloud tops. Cloud-base estimates can also use known terrain, visual references or pilot reports. Tops are assessed using satellite information, appropriate radar, soundings and aircraft reports, with method-specific limits; an infrared brightness temperature is not automatically a unique top height.

Cloud base is the lower boundary of a cloud layer. In a routine aerodrome report its height is normally given in hundreds of feet above aerodrome elevation; local reports may use the specified runway-threshold datum under applicable rules. Cloud altitude above mean sea level requires the datum to be added. A BKN020 layer at a 1,000 ft-elevation field is about 3,000 ft AMSL, not 2,000 ft AMSL.

The ceiling is a subset of the cloud report

Ceiling is the height above ground or water of the base of the lowest layer below 20,000 ft covering more than half the sky: BKN or OVC. In an obscured sky, reported vertical visibility is used in the relevant ceiling assessment. FEW or SCT below a higher BKN layer does not by itself establish a lower ceiling. With SCT008 BKN020 OVC070, the cloud ceiling is 2,000 ft above the aerodrome datum, while the lower scattered cloud can still affect an approach or terrain clearance.

METAR and TAF cloud groups identify amount and base, with CB or TCU added when appropriate; ordinary genera such as Stratus or Altocumulus are not named in the standard cloud group. CB and TCU matter even at heights which would otherwise be omitted. SIGMET does not catalogue ordinary cloud layers: it warns of specified significant phenomena, including qualifying thunderstorms described by their obscuration, embeddedness, frequency or squall-line character and their extent/levels. A forecast CB group and an observed CB group have different times and certainty.

Why is VV003 not a cloud base at 300 ft?

An obscured sky can prevent a discrete base or cover assessment.

The upward range

Vertical visibility is the vertical distance into an obscuration that can be seen from the ground. It is assessed when the sky is obscured and a cloud-base/amount observation cannot be made in the normal way, for example in dense fog, heavy precipitation or blowing snow. A ground observer may use an upward reference or suitably configured cloud/visibility equipment; a ceilometer can estimate penetration into an obscuring layer rather than finding a distinct cloud boundary.

VV003 means a reported vertical visibility of 300 ft in the ICAO METAR convention, which codes hundreds of feet. It does not prove a physical cloud base at that height or promise that an aircraft at 400 ft will be clear of the obscuration. VV/// indicates that vertical visibility cannot be assessed, not unlimited visibility. The sky may remain obscured even if a horizontal sensor sees some distant lights.

Vertical visibility can be expressed in feet or metres according to the reporting system; read the stated units and code convention. Local plain-language reports can include explicit FT or M. Compare VV with the applicable approach/ceiling criteria, and use current RVR or visibility as required; no universal conversion links the upward range to the horizontal runway view.

Why is a hot tarmac surface not the reported air temperature?

A representative air sample requires exposure, shielding and calibration.

Temperature, dew point and relative humidity

Air temperature is measured with a calibrated thermometer, commonly an electrical resistance sensor, in a ventilated radiation shield or screen at a representative near-surface exposure. Solar heating, reflected radiation, engine exhaust, nearby buildings and wet or poorly ventilated sensors can bias it. The objective is the temperature of the surrounding air, not the temperature of a sunlit runway surface. Temperature and dew point in European METARs are degrees Celsius; M denotes a negative value.

Dew point is the temperature to which the sampled air must be cooled at essentially constant pressure and water-vapour content to reach saturation with respect to liquid water. A hygrometer can measure humidity from which dew point is derived; a chilled-mirror instrument detects condensation at a controlled surface. A psychrometer uses dry- and wet-bulb measurements with pressure-dependent psychrometric relationships. Relative humidity is a percentage ratio relative to saturation at the actual air temperature; it is not an absolute water mass per cubic metre.

Dew-point temperature can be stated in Celsius or Fahrenheit: °F = 1.8 × °C + 32. A 10 °C dew point is 50 °F; it is neither 10% RH nor 50% RH. RH can change when temperature changes even if the air’s vapour content is unchanged. A small temperature/dew-point spread suggests near saturation but does not alone prove fog at the runway or ice-free conditions aloft.

Can you apply the ideas?

Four original questions. Decide from the facts, then review the explanation.