Cloud Cover, Oktas and Ceilings: How Sky Condition Is Actually Measured

Cloud is reported in a manner which appears straightforward and is, in practice, the element of the weather report I most often found to be misunderstood.

The confusion is rarely about the codes themselves. It concerns what the codes are measuring, and in particular the difference between the amount of cloud present and the existence of a ceiling.

The okta

Cloud amount is measured in oktas, meaning eighths of the visible sky dome. An observer — or, increasingly, an automated ceilometer system — assesses what proportion of the sky is obscured by cloud at each level.

Two further codes appear in automated reports. NSC indicates no significant cloud — no cloud below 5,000 feet or below the highest minimum sector altitude, whichever is greater, and no cumulonimbus. NCD indicates that an automated system has detected no cloud at all.

VV followed by three digits, such as VV002, indicates vertical visibility where the sky is obscured and no cloud base can be determined. This is common in dense fog, and it should be read as a more restrictive condition than an overcast layer at the same height.

Height, and the datum which matters

The three digits following the amount code give the height of the cloud base in hundreds of feet. SCT018 is scattered cloud at 1,800 feet.

The critical point is the datum. Cloud height in a METAR or TAF is given above aerodrome level, not above mean sea level.

Ceiling: the distinction that matters most

A ceiling is defined as the height above ground of the lowest layer reported as broken or overcast.

Layers reported as FEW or SCT do not constitute a ceiling. This is not a technicality; it reflects the fact that an aircraft may reasonably expect to see the ground through a sky which is less than half covered.

Consider two reports:

SCT004 SCT010 SCT025
BKN018

The first describes three separate layers, the lowest at 400 feet, and has no ceiling at all. The second describes a single layer at 1,800 feet and has a ceiling of 1,800 feet. By flight category, the first is VFR and the second is MVFR — notwithstanding that the first sky looks considerably more cluttered from the ground.

I would encourage anyone using automated category indications to verify this against the underlying report, particularly when the indication seems inconsistent with the observed conditions.

Multiple layers

Layers are reported in ascending order of height, and a maximum of three or four layers is generally given. The reporting convention requires that each successive layer reported must represent a greater cumulative amount than the one below, since an observer cannot see through a lower layer to assess a higher one accurately.

This produces a useful reading rule: once a layer is reported as OVC, no further layers will appear above it in the report. The observer cannot see them.

The significant cloud types

Only two cloud types are identified by name in a routine report, and both are appended as a suffix to the amount and height group.

CB denotes cumulonimbus. TCU denotes towering cumulus.

These are the only two types considered operationally significant enough to name, which itself indicates their importance. As discussed in our guide to the cumulonimbus, the reported quantity is largely irrelevant where these suffixes appear. FEW020CB describes one or two oktas of cumulonimbus, which is to say a thunderstorm, with everything that entails.

A practical reading sequence

When assessing the cloud groups in a report, I would suggest the following order.

1. Scan for CB or TCU first. If either is present, that becomes the governing consideration and the remainder of the cloud assessment is secondary.
2. Identify the lowest BKN or OVC layer. That is the ceiling, and it is the figure relevant to approach minima.
3. Note the aerodrome elevation and convert to a sea-level figure if the height is to be used alongside chart altitudes.
4. Consider the lower FEW and SCT layers in the context of visual manoeuvring, circling approaches and visual reference on final.
5. Compare against the previous observation to establish whether the situation is improving or deteriorating.

That final step is the one most frequently omitted and, in my experience, the one which most often provides the earliest indication that a plan requires revision. A single report describes a moment. Two reports describe a trend, and it is the trend upon which sound decisions are built.