Noctilucent Clouds (NLC)
Noctilucent clouds are immediately recognizable, even when being seen for the first time. The name suggests it all: they are night-shining clouds. From mid-latitudes (Φ > 50°), they can be seen during the summer in the twilight arch which moves around the north (or south, in the southern hemisphere) horizon as the night progresses. In form much like cirrostratus clouds, they are usually silvery-white or pale blue in colour and they stand out clearly behind the darker twilight sky. Ordinary (i.e. tropospheric) clouds are dark silhouettes under these conditions; noctilucent clouds shine. The reason for this is that noctilucent clouds are very high in the atmosphere and remain in sunlight long after the Sun has set at ground level.
Noctiluccnt clouds were first noted during the summer of 1884, some months after the volcanic explosion which destroyed Krakatoa Island in 1883. From the first, they were recognized as being quite different in character from clouds lying low in the atmosphere. It is slightly puzz1ing why the clouds were not written about in the years before 1885 (Gadsden 1983, 1984; Schröder 1975). Current views on the cause and development of noctilucent clouds suggest that the Krakatoa explosion had no direct effect in producing the noctilucent clouds; the event is significant only in having led observers to take careful note of the twilight sky and what was to be seen in it (Schröder 1975; Austin 1983).
Noctilucent clouds have an appearance that impresses an observer immediately and there are published descriptions that express almost astonishment at the way they shine in the sky. Certainly, astronomers and aurora-watchers are eager to contribute reports and simple measurements to the corpus of noctilucent cloud research. As it happens, one of the most extensive and brightest displays of noctilucent clouds in recent years occurred 100 years ago almost to the night after Smyth’s enthusiastic report (Smyth 1886). Gavine (1987) received over 30 individual reports of the display over NW Europe on July 23/24 1986. Clearly, noctilucent clouds have lost none of their attraction in 100 years.
How, when and where noctilucent clouds are seen
In anticipation of proper justification of the statements later on, reference has been made above to “summer... twilight”, to “mid-latitudes (Φ > 50°)” and to “very high in the atmosphere”. We shall see that noctilucent clouds are essentially a polar phenomenon and that the clouds are blown away from polar regions to disappear at mid-latitudes.
At noctilucent cloud level, the air pressure is a few millionths of that at sea level. The upper atmosphere is also very dry, with water molecules present only as a few in the million of the molecules making up the surrounding air. The partial pressure of water vapour at these heights is perhaps 10 picobars. The actual amount of water available for cloud formation in the upper atmosphere is therefore minute. Clearly, for clouds not only to exist there, but to be dense enough to be seen, there must be unusually cold conditions. This is confirmed by direct (rocket-borne, in situ) measurements of temperature.
At the height of noctilucent clouds, the atmosphere is relatively transparent to solar radiation and the local air temperature is set principally by mixing, by radiation from the air itself and from the Earth’s surface and the lower atmosphere, and by bulk movement in the vertical direction. The global circulation of the upper atmosphere, largely the result of solar heating in the stratospheric ozone layer well below the noctilucent cloud region, imposes upward (cooling) movement of the air over the summer polar regions, with downward (warming) movement over the winter polar regions. Hemisphere to hemisphere flow closes the circulation; for details, the reader should consult a textbook such as that of Houghton (1977).
Consequently, the upper atmosphere shows the paradoxical behaviour of being colder in the summer than in the winter. This is the result of solar heating at these high levels being less important than the temperature changes caused by expansion or compression when air rises or falls.
Thus, there are already two inherently conflicting influences on the ability to see noctilucent clouds. If the observer goes to very high latitudes, noctilucent clouds may be difficult to observe because they are just forming and have not had the time to permit cloud particles to grow to observable size. If, on the other hand, the observer is at too low a latitude, the noctilucent clouds may have moved equatorwards, out of the regions of low temperature, and have evapo rated. A compromise is called for, and observable noctilucent clouds are to be found at latitudes of approximately 60°–75° (Gadsden 1982; Schröder 1975).
The amount of water vapour in the upper atmosphere is difflcult to estimate with precision. The Earth’s atmosphere contains water vapour which is present at all heights through the action of diffusion upwards from sea level. The tropopause acts as a “cold trap” for water vapour: much of the upward flux of water vapour will be frozen out at or below the tropopause.
If we assume that the atmospheric pressure at the tropopause is typically 100mb, and that the tropopause is at a temperature of 200K (at which the saturation vapour pressure of water over ice is 1.7µb), the relative amount of water vapour just above the tropopause will be 17 parts per million (ppm). This, the so-called mixing ratio, is the number of water molecules per given number of atmospheric molecules.
Water vapour in the upper atmosphere comes, therefore, from a dry, cold region low down. An upward flux (and the negative gradient of mixing ratio, necessary to keep the flux going) is maintained through “evaporation” of hydrogen atoms from the topmost part of the atmosphere, the exosphere. As the water molecules diffuse up in the atmosphere, they may be dissociated (at stratospheric heights) by solar ultraviolet radiation to give hydrogen atoms and hydroxyl radicals. In the mesosphere, therefore, there is a complicated balance in the system of chemical reactions which ozone and atomic oxygen, among others, play a prominent role.
What measurements are available to allow estimates of the mixing ratio at 80km and above suggest that 3ppm is probably as good a number as any to keep in mind.
Temperatures as low as 111K (-162°C) have been measured at a few kilometres above a noctilucent cloud. At such low temperatures, water molecules will cluster together quite effectively to form ice particles although the flux of molecules to a surface is small. The number of water molecules in the atmosphere above, e.g. 80km is approximately 7 × 1015m-2, that is a mass of 0.2µg m-2. The entire water content above 80km could freeze out to give just one cloud particle of radius 0.042mm for each square metre of cloud layer (or more particles, of course, inversely as the cube of their assumed radius). One does not expect, therefore, very dense noctilucent clouds and measurements show that their optical thickness is usually in the vicinity of 10-4; noctilucent clouds are transparent and scatter less than one part in a thousand of the sunlight incident on them.
They cannot be seen from sea level during the day. Light that is scattered low in the atmosphere completely hides the tiny proportion of sunlight that is scattered from a noctilucent cloud. After sunset, the sky darkens and the sky brightness is down by a factor of several hundred at the end of civil twilight (which is defined as being the time when the centre of the solar disc, with no allowance for refraction, is at a zenith distance of 96°, that is, 6° below the horizon). At this time, noctilucent clouds can be distinguished, albeit with difficulty, right across the observer’s sky, from the bright twilight arch in the direction of the set Sun to the opposite part of the sky where the so-called Earth shadow is rising (Minnaert 1954).
The invisibility of noctilucent clouds during daytime and early twilight imposes another constraint, and a strict one, on successful observation. Noctilucent clouds occur only in summer (but this statement is not quite true) and only at high to middle latitudes, At latitudes higher than 61°, civil twilight lasts all night in the middle of summer and the sky does not become dark enough to permit noctilucent clouds to be seen. At high latitudes, therefore, noctilucent clouds are not to be seen from sea level except at the very beginning and at the very end of the “observing season”, which is usually taken to he the months of May, June, July and August in the northern hemisphere, and the months of November, December, January and February in the southern hemisphere. Statistics of visual observations assembled by Fogle (1965; Schröder 1966c) show that 57°N is essentially the best latitude for seeing noctilucent clouds, whlch occur typically some 3°–5° further north. This statement is true for both hemispheres, because there is poor summer weather at latitudes of 55°–60° in the southern hemisphere.