The current classification of clouds is based on the pioneering system developed by Luke Howard (1804). In this work, seven basic cloud genera were typified and classified according to their appearance (shape, colour, texture) and evolution. Afterwards, the use of balloon flights made it necessary to add the height of the clouds as a qualifying criterion, an idea which had been previously proposed by the French scientist Jean Baptiste Lamarck. In 1855, the French meteorologist Emilien Renou proposed altostratus and altocumulus as new types of clouds. The last type of cloud proposed was the cumulonimbus, put forward by the Danish amateur meteorologist Philip Weilbach in 1880.
The current International Classification of Clouds published by the World Meteoro-logical Organization (WMO, 1975; 1987) includes the ten basic cloud genera with later nuances. The cause of cloud formation remained a secondary issue or was not even considered at all: the assumption is that they are produced by natural mechanisms. In the 1956 edition of this Classi-fication, the concept of ‘mother cloud’ was introduced as additional information for meteorological observations. For example, one form of cirrostratus may be called cirrostratus cumulonimbogenitus when it originates from the spreading out of the upper part of a cumulonimbus cloud. However, whether the ‘mother cloud’ had a natural or anthropogenic origin was not indicated.
Since the end of the nineteenth century, human activity has injected large amounts of water vapour, aerosols and hot air into the atmosphere, mainly as a result of combustion processes. Under certain spatial and temporal conditions, these activities enhance cloud formation. For instance, aerosols modify optical atmospheric properties, and consequently cloud properties. Aerosols tend to cool the lower troposphere by reflecting sunlight; but they warm the upper levels by absorbing the sunlight. It is possible that anthropogenic aerosols may act as condensation nuclei, enhancing cloud formation and even rainfall. An increase in aerosol concentration or changes in their composition could affect the climate of Earth and its water resources (Kaufman et al., 2002).
Human activities can enhance any of the triggering mechanisms of cloud formation: forcing air upwards (consequently lowering its temperature), humidification (adding moisture to air) and aerosol activation (increasing condensation nuclei which starts the condensation process). Depending on the situation, these factors may be found working simultaneously.
Forcing air upwards from chimneys at factories (cooling towers) and air humidification from combustion could be the main causes for the formation of some low clouds. In large urban and industrial areas, humidification and aerosol activation due to pollutant emissions could produce some low anthropic (that is, related to human activity) clouds. In coastal areas, where humidity can be high, pollutants may act as condensation nuclei to form a diffuse stratiform cloud. Other good examples of low-level anthropic clouds created by these mechanisms are the ‘pea soup’ fogs that occasionally occurred in some UK cities, especially London, until the 1950s. Wildfires also enhance vertical upward movements and inject into the atmosphere large amounts of aerosol that produce convective low clouds. Aircraft flying at high altitudes influence air humidification and the engines emit aerosol particles, both of which form high anthropic clouds. These clouds are usually called aircraft condensation trails (contrails) and sometimes occupy a large part of the sky: their evolution often shows an appearance of cirrus or cirrostratus. The persistence and extension of contrails could also be an indication that a weather disturbance is approaching. Because of the widespread extent of human activity, it may be useful to differentiate between natural and anthropic cirrus and cirrostratus, bearing in mind the significant impact that contrails have on the planet's energy balance (Hartmann et al., 1992; Hu and Stamnes, 1993; Gamier et al., 1997; Stuben, 2006).
It would indeed be interesting, whenever possible, to identify whether clouds have a natural or anthropic origin. The use of a prefix for designating the origin of a cloud has already been internationally accepted, as in the case of pyrocumulus and pyrocumulonimbus: clouds caused by fire. We propose that all clouds originating from human activities should be designated with the prefix anthropo- (a-), to differentiate them from those whose origin is natural. The knowledge of local weather observers would be the key to determining whether or not it is an anthropocloud, a cloud in which the triggering mechanism is anthropogenic, although it will not always be possible to determine this once the cloud has moved away from its source.
Weather observers already indicate some cloud details in their observations, such as family and species, as well as special cloud characteristics, such as virga, mamma, velum, tuba, etc. A natural further step would be to use the above-mentioned prefix a- to differentiate the anthropic origin in cloud formation, as we will show in the rest of this article which will discuss and provide examples of different clouds of anthropic origin (low, medium and high clouds).
The condensation trails of aircraft and high-level -anthropoclouds: aCi, aCc, aCs
Contrails are probably one of the most frequent examples of anthropic clouds. They represent a change in atmospheric composition and may produce an impact on Earth's climate because they affect the Earth's radiation budget by reflecting and absorbing solar radiation and delivering and absorbing infrared radiation (Minnis et al., 1999, 2003; Ponater et al., 2002, 2005). They may also redistribute moisture up to the top of the troposphere, mainly at heights between 6000 and 12 000 metres. Penner et al. (1999) analysed contrails and their effects as one of the largest outstanding uncertainties in air traffic's impact on the atmosphere. Contrail coverage distribution and optical depth are some of the variables that must be accurately quantified. With commercial jet air traffic expected to continue increasing at a rate between 2% and 5% per year through to 2050 (Penner et al., 1999), it is important to reduce the uncertainties in the parameters that determine the impact of contrails on climate.
Marquart et al. (2003) have investigated the trend of aircraft condensation trails and its influence on global warming, using mathematical models. They suggest that the contribution of contrails to global sky cloud coverage was 0.06% in 1992, and will be 0.14% in 2015 and 0.22% in 2050. That will increase radiative forcing: in 1995 this was about 3.5Wm–2, and the estimations for 2015 and 2050 are around 9.4Wm–2 and 14.8Wm–2 respectively. Minnis et al. (1999) estimated that the global radiative forcing of line-shaped contrails is about 17.5mWm– 2; Stuber and Forster (2007) calculated 2.0mWm–2. The difference is because the contrail optical depth may take a wide range of values over various seasons, geographical regions and altitudes (Marquart et al., 2002; Palikonda et al., 2005), which renders the determination of something like a typical or mean figure quite difficult.
The anthropic origin of some high clouds, despite their distance from the surface (above 6km), is due to the injection of water vapour and cloud condensation nuclei by aircraft. It can form all three genera of high clouds: cirrus (Ci), cirrocumulus (Cc) and cirrostratus (Cs). The most frequent of the high-level clouds caused by contrails is Cc: we would call these clouds anthropoCirrocumulus, and designate them as aCc. Cirrus and cirrostratus are formed from aCc (aligning and spreading out of contrails, respectively), but they could be considered as -independent types and be named anthropoCirrostratus (aCs), and anthropoCirrus (aCi).
Figure 1 shows a fibrous structure similar to cirrus formed from a contrail. This aCi can be formed when small ice crystals from contrails disperse laterally, creating fibrous structures due to strong winds in the high layers of the troposphere. Often several contrails adopt a structure similar to Cc, as thermal differences between engine exhaust and the surrounding air generate local convection – leading to aCc (Figure 2). In some atmospheric conditions, mainly when a front is approaching, this aCc may expand, forming a tiny layer of aCs. Figure 3 shows an -example of this kind of anthropic cloud: in this instance, ice crystals in the central part are dispersed by the west wind, causing fibrous forms like cirrus aCs, surrounded by prevailing forms of anthropoCirrocumulus (aCc).
When there is enough humidity at high levels in the troposphere, contrails may persist for several hours and evolve into several high anthropic cloud types. In these situations, aCc is the most common cloud genera formed directly behind the engine exhaust. If the ice crystals persist and are dispersed by the wind horizontally into fibrous structures then this would result in aCi; persistent aCi may spread and grow further into a layer of aCs. Figure 4 shows anthropic clouds caused by several contrails: all of the cloud here was due entirely to the evolution of contrails.
Medium-level anthropoclouds: aAc, aAs, aNs
Human activity is very unlikely to be involved in the formation of, or have an influence on the evolution of, a cloud of great vertical and horizontal extent such as nimbostratus (which is classified as a medium-level cloud although its base may be at low levels). This probably also applies to altostratus. Therefore, anthropo- varieties of these medium level clouds probably do not exist. However, -low-level contrails (around −5000–6000 metres altitude) may be identified by an observer as anthropoAltocumulus (aAc), rather than either aCc or aCi. aAc is therefore probably the only human induced medium-level cloud that is possible, and that an observer would be able to positively identify.
Low-level anthropoclouds: aCu, aSc, aSt
Low clouds are the most likely genera to be formed due to human activities - especially cumulus and stratocumulus. These clouds could be named anthropoCumulus and anthropoStratocumulus, and designated by aCu and aSc, respectively. Figures 5(a) and (b) show an aCu formed in cold, unstable air by the thermal energy and moisture from a geothermal plant. Figures 6, 7(a) and (b) show aSc. The emissions from an industrial area have formed a convective column that spreads out into aSc under an inversion layer a few hundred metres above the surface. Notice in Figure 6 the clearly convective structure as compared to the other stratus clouds in the background.
Stratus clouds of pure anthropogenic origin are unlikely, because anthropic clouds tend to be associated with a degree of instability in the atmosphere, and stratus is typical of stable conditions. Stratus formation linked to human action (anthropoStratus, aSt) does, however, occur in areas with high density industry and population: the emission of large amounts of condensation nuclei from combustion (traffic, factories) leads to the formation of stratus clouds if there is enough moisture. It happens, for instance, during hot and humid summer days in some metropolitan areas near water sources, where a thin stratus layer is formed. Figure 8 shows an example of aSt formed in a stable atmosphere: the emission of moist air from a milk treatment factory (at the lower right corner of the picture) was sufficient to initiate a small convective cloud which could be observed just above it. A weak breeze extended this cloud horizontally, due to the stable, stratified conditions in the lower troposphere. The combination of all these factors was responsible in the formation of this aSt, which had a clear anthropic origin as no other cloud was observed in the surroundings. This aSt is a good example of the importance of a local weather observer in differentiating some anthropic clouds from those originating from natural causes. This type of cloud could extend several kilometres away from the source, to areas where it would be difficult, or even impossible, to differentiate it from a natural cloud; only the knowledge and experience of local weather observers would enable the origin of the cloud to be identified.
The influence of clouds on the Earth's energy budget has an important role and has been studied by several authors. Only in some special cases, such as pyrocumulus and contrails, has the origin of the clouds been taken into consideration. However, the combustion of a large amount of fossil fuel injects condensation nuclei, water vapour and heat into the troposphere and enhances cloud formation, whilst the increase in air traffic over recent decades has brought about an increase in the formation of condensation trails, which can be persistent.
The identification and classification of clouds formed mainly by human activity (anthropoclouds) can contribute to the study of future trends regarding these clouds, as well as the contribution our activities make toward cloud formation, the role of cloudiness in the Earth's energy budget and its contribution to climate change.
We have shown several examples of anthropic clouds, proposing their differentiation using the prefix a- to differentiate the anthropic origin for some of the ten basic clouds defined in the International Cloud Atlas: anthropoCumulus (aCu), -anthropoStratocumulus (aSc), anthropoStratus (aSt), anthropoAltocumulus (aAc), anthropoCirrus (aCi), anthropoCirrocumulus (aCc), and anthropoCirrostratus (aCs). However, anthropoNimbostratus (aNs) and anthropoCumulonimbus (aCb) are unlikely.
We are aware that the increasing automation of meteorological observation networks around the world is leading to a decrease in the number of human observers, which will make it difficult to differentiate between natural clouds and anthro--poclouds. In our opinion, the weather observer can still be a fundamental tool for research in meteorology and climate studies and we believe that the classification of anthropoclouds should be promoted.
This work was carried out under the Spanish MICINN project CGL2009-08609. We would like to thank the reviewers whose comments helped us to improve the manu-script.