Les Nuages: Guide dIdentification

Clouds are not merely decorative — they are the atmosphere's most visible expression of the physical processes that create weather. Every cloud tells a story about temperature, humidity, wind, and stability at the altitude where it forms, and learning to read that story transforms the sky from a static backdrop into a dynamic narrative of atmospheric change. Cloud identification is one of the oldest meteorological skills, practised by sailors, farmers, and shepherds for millennia before instruments existed, and it remains one of the most useful: a person who can identify the ten main cloud types can predict weather changes, estimate wind conditions at altitude, and assess the risk of severe weather with a reliability that surprises those who assume forecasting requires technology.

The Classification System: Luke Howard's Legacy

The modern classification of clouds dates to 1802, when Luke Howard — a London pharmacist and amateur meteorologist — proposed a Latin-based naming system that, like Linnaeus's biological taxonomy, provided a universal language for describing natural forms. Howard identified three fundamental cloud forms: cirrus (from the Latin for "curl" — wispy, high, delicate), stratus (from "layer" — flat, horizontal, sheet-like), and cumulus (from "heap" — puffy, vertical, three-dimensional). These three forms, combined with the prefix "nimbus" (rain-bearing) and modifiers for altitude, produced the ten-type system that the World Meteorological Organisation (WMO) adopted and that remains the international standard today.

Howard's genius was not merely taxonomic but observational: he recognised that clouds are not random shapes but expressions of specific atmospheric processes. A cirrus cloud exists because conditions at high altitude (cold temperature, low moisture) allow only thin, wispy ice crystals to form. A cumulus cloud exists because convective uplift — heated air rising in columns — creates the vertical development that produces a puffy, tower-like shape. A stratus cloud exists because stable atmospheric conditions prevent vertical development, forcing moisture to spread horizontally in layers. By linking cloud form to atmospheric process, Howard created a classification that is simultaneously descriptive (what does the cloud look like?) and diagnostic (what is the atmosphere doing?).

The ten genera are further divided into species (based on shape and structure), varieties (based on transparency and arrangement), and supplementary features (based on associated phenomena like virga, precipitation trails, or mammatus formations). The full WMO International Cloud Atlas catalogues over 100 distinct cloud types, but for practical weather observation and prediction, the ten genera and an understanding of their basic meteorological significance are sufficient for most purposes.

High Clouds: Cirrus, Cirrostratus, Cirrocumulus

High clouds form above approximately 6,000 metres (in mid-latitudes; higher in the tropics, lower at the poles) and are composed entirely of ice crystals because temperatures at these altitudes are well below freezing year-round. Cirrus — the most recognisable high cloud — appears as thin, white filaments, streaks, or hooks against the blue sky. The delicate, fibrous appearance results from ice crystals falling through layers of different wind speed and direction, producing the wind-sheared trails that characterise the type. Cirrus often appears 24–48 hours before an approaching warm front, making it one of the most reliable visual indicators of weather change: the old saying "mares' tails and mackerel scales make tall ships carry low sails" reflects centuries of empirical recognition that cirrus (mares' tails) often precedes deteriorating conditions.

Cirrostratus is a thin, translucent sheet of ice crystals that covers part or all of the sky, often producing a halo around the sun or moon (the 22-degree halo). Cirrostratus is the cloud that makes the sky appear whitish or milky without obvious structure — easy to miss if you are not looking for it but diagnostically significant because it typically indicates the leading edge of an approaching warm front. When cirrus thickens progressively into cirrostratus, and cirrostratus thickens into altostratus, the classic warm-front cloud sequence is in progress and rain or snow can be expected within 12–24 hours.

Cirrocumulus — small, white patches or ripples arranged in regular patterns — is the least common of the high clouds and the most aesthetically striking. The "mackerel sky" pattern of cirrocumulus results from small-scale instability at high altitude, producing tiny convective cells that create the regular, fish-scale pattern. Cirrocumulus is short-lived, often transforming into cirrus or cirrostratus within minutes, and its presence indicates atmospheric conditions that are changing — making it a signal, like cirrus, that current weather is unlikely to persist.

Middle and Low Clouds: The Working Atmosphere

Middle clouds — altostratus and altocumulus — occupy the atmospheric layer between approximately 2,000 and 6,000 metres and are composed of water droplets, ice crystals, or a mixture of both. Altostratus is a grey or blue-grey sheet that covers the sky uniformly, thin enough to show the sun as a bright spot (a "watery sun") but thick enough to eliminate shadows. It is the cloud of approaching rain: when altostratus thickens and lowers, becoming nimbostratus, continuous precipitation begins. The altostratus-to-nimbostratus transition is one of the most reliable cloud-based weather predictions available — if altostratus is thickening, rain is coming, almost without exception.

Altocumulus appears as white or grey patches, sheets, or layers of rounded masses, often arranged in lines or waves. It is one of the most varied and most common cloud types, and its diagnostic significance depends on context. Altocumulus castellanus — with turret-like vertical extensions rising from a common base — indicates instability at mid-levels and, on summer mornings, is one of the best predictors of afternoon thunderstorms. A morning sky showing altocumulus castellanus over Greece in summer is a reliable signal that convective development will occur later in the day, potentially producing the afternoon thunderstorms that are common over the mountains and occasionally affect coastal areas.

Low clouds — stratus, stratocumulus, and nimbostratus — form below 2,000 metres and are composed almost entirely of water droplets. Stratus is the grey, featureless cloud that produces overcast skies and drizzle — the cloud of grey days, low ceilings, and the depressing flatness that northern Europeans associate with winter. Stratocumulus, its more structured cousin, appears as grey or white patches with darker areas, arranged in rounded masses or rolls. It is the world's most common cloud type, covering approximately 20 percent of the ocean's surface at any time, and its role in Earth's radiation budget — reflecting solar energy back to space while allowing terrestrial heat to escape — makes it one of the most climatologically important cloud types despite its visually unremarkable appearance.

Cumulus and Cumulonimbus: The Vertical Giants

Cumulus and cumulonimbus clouds are the products of convection — the vertical transport of heat and moisture from the surface into the atmosphere — and they span multiple altitude levels, making them the most visually dramatic and meteorologically significant cloud types. Fair-weather cumulus (cumulus humilis) — the small, white, cotton-wool clouds of sunny afternoons — forms when surface heating creates columns of rising air that reach their condensation level and produce individual cloud puffs. These clouds indicate fair weather, limited vertical development, and an atmosphere that is conditionally stable: the rising air reaches a level where it is no longer warmer than its surroundings and stops ascending.

When atmospheric instability is greater — when the rising air remains warmer than its surroundings through a deep layer — cumulus grows vertically into cumulus mediocris and cumulus congestus (towering cumulus), producing clouds that extend from their bases at 1,000–2,000 metres to tops at 5,000–8,000 metres. These clouds produce significant updrafts, turbulence, and localised showers, and they are the precursors to the most powerful cloud type: cumulonimbus.

Cumulonimbus — the thunderstorm cloud — is the atmosphere's most powerful single-cell feature, extending from near the surface to the tropopause (12–15 kilometres in mid-latitudes, up to 18 kilometres in the tropics). Its defining feature is the anvil — the flat, spreading top that forms when the updraft reaches the tropopause (the stable boundary between troposphere and stratosphere) and spreads horizontally. A mature cumulonimbus produces thunder, lightning, heavy rain, hail, strong downdrafts, and occasionally tornadoes. The energy contained in a single large cumulonimbus cloud is equivalent to approximately 10 Hiroshima-sized atomic bombs — a comparison that conveys the extraordinary concentration of atmospheric energy in these structures.

Clouds as Weather Predictors: Reading the Sky

The practical value of cloud identification lies in its predictive power. Specific cloud sequences indicate specific weather changes, and a person who can read these sequences can anticipate weather with useful accuracy 12–24 hours in advance. The classic warm-front sequence — cirrus thickening to cirrostratus, lowering to altostratus, thickening to nimbostratus — predicts rain with near certainty. Morning altocumulus castellanus predicts afternoon thunderstorms. Lenticular clouds (lens-shaped clouds forming downwind of mountains) indicate strong upper-level winds that may bring weather changes. Rapid cumulus development on a summer morning indicates instability that will likely produce afternoon showers or storms.

In Greece, cloud reading is particularly useful because the country's complex terrain produces local cloud patterns that generic forecasts may not capture. Cumulus development over the mountains on summer mornings indicates where afternoon thunderstorms will develop — typically over the highest terrain first, then potentially spreading to lower elevations and coasts by evening. The appearance of lenticular clouds over the Pindus or Taygetus mountains indicates strong winds aloft that may bring changes within 24 hours. And the sudden clearing of clouds from mountain summits that have been capped all morning often indicates that the atmosphere is drying and conditions will improve.

The limitations of cloud-based prediction are important to acknowledge. Clouds indicate current atmospheric conditions, not future ones: they show what the atmosphere is doing now, from which future conditions can be inferred but not precisely predicted. A cirrus-to-nimbostratus sequence almost always produces rain, but the timing — whether rain arrives in 12 hours or 24 — cannot be determined from clouds alone. Cloud observation complements but does not replace forecast products, and the wisest approach is to use both: check the forecast for the general outlook, then use cloud observation to refine the timing and assess whether the forecast is tracking as expected.

Clouds and Climate: The Largest Uncertainty

Clouds are the single largest source of uncertainty in climate projections — a fact that surprises many people who assume that the science of climate change is more settled than it is. The uncertainty arises from the dual role of clouds in Earth's energy budget: they reflect incoming solar radiation back to space (a cooling effect) and they trap outgoing terrestrial radiation (a warming effect). The net effect depends on the type, altitude, thickness, and coverage of clouds — and predicting how these properties will change as the climate warms is one of the most challenging problems in atmospheric science.

Low clouds — particularly marine stratocumulus — are net coolers: their high reflectivity bounces more solar energy back to space than the modest amount of terrestrial radiation they trap. If warming reduces low cloud coverage (as some models predict), the cooling effect diminishes and warming accelerates — a positive feedback that would increase climate sensitivity. High clouds — particularly thin cirrus — are net warmers: they are too thin to reflect much solar energy but effective at trapping terrestrial radiation. If warming increases high cloud coverage, the warming effect amplifies. The balance between these opposing effects — and how it will shift as the climate changes — is the question that most influences the range of climate projections, from the manageable (~2°C by 2100) to the severe (~4.5°C or higher).

The challenge is one of scale: individual clouds operate at scales of metres to kilometres, while climate models operate at scales of 50–100 kilometres. A model grid cell contains thousands of clouds, none of which can be explicitly simulated — they must be parameterised, represented by statistical approximations that capture their average behaviour but may miss the fine-scale processes that determine their response to warming. Improving cloud parameterisations is the most active frontier in climate modelling, and the clouds that Luke Howard classified in 1802 remain, over two centuries later, the atmospheric features that most challenge our ability to predict the future of Earth's climate.