Speculations

An alternative would be to propose an entirely "new" phenomenon tailored to preserve all significant features of the sightings, possibly having no physical connection to atmospheric optics or EQL. Although we think this is much less likely, and is arguably less economical n1 Occam's Razor advises against invoking "new entities" where existing entities suffice. Defining "new" is fraught with ambiguity in an area where "existing" entities (for example, EQL phenomena) are themselves speculative, poorly understood and only tentatively (not universally) accepted on the basis of observational data often less well established than those here being assessed. EQL remains largely a phenomenological category rather than a physical category (see Section 6.k) and is perhaps best thought of as a small core of cases surrounded by a less well-defined periphery of related accounts of "earthlights" merging into folk tales (see, e.g: Devereux, Paul, Earthlights, Turnstone Press, 1982). The same can be said of "ball lightning" (BL), and, to a much greater extent of course, of"unidentified flying objects". The phenomenological differences between sub-groups of these categories may sometimes be small compared to their internal variation, suggesting that there are sociological origins for these ambiguous boundaries, which are overlaid on - and possibly act to obscure - physical ones.., we cannot rule it out.

It is interesting that both of the theories we consider to have at least some potential to lead to an explanation are theories that place light-emitting or light-scattering phenomena of some type close to the triangulated locations in the Channel Islands area. To some extent this reflects the internal consistency of the prima facie sighting geometry (Section 3) and the sighting on a nearreciprocal bearing from the Jetstream. But the result does not depend on it. A major obstacle to any mainly atmospheric-optical theory - such as mirage or mock-mirage of sun reflections beyond the horizon - is the lateral displacement of the two similar n2 In the mathematical sense, i.e., identical except in terms of angular scale. images relative to one another, and even more importantly their steady lateral motion relative to one another, over an arc amounting to perhaps 3°. This behaviour is very explicitly reported and we can find no justification for discounting it. But we are unable to find any reference for an atmospheric-optical mechanism, even a speculative one, that could explain it. On the other hand, this behaviour would be quite naturally explained by parallax in terms of the prima facie sighting geometry permitted by placing phenomena (such as the two Band 3 phenomena listed) in the triangulated locations.

At the same time, there are historical accounts of observations that do appear to invite atmospheric-optical explanations but are beyond the abilities of any known mechanism to explain. A possibly relevant example is the remarkable "double sun" discussed by Minnaert n3 M.G.J. Minnaert, Journal of the Optical Society of America, 58 (1969): 297.. A photograph allegedly taken by a passenger from the deck of a ship in the Indian Ocean shows the sun near the horizon in clear air with, beside it, a second sun, a perfect duplicate image at precisely the same elevation over the horizon. The story is that this was witnessed by 20 or 30 other passengers. We are not aware of any proper explanation of this phenomenon in terms of conventional atmospheric physics. Physicist Philip Morrison n4 Morrison, P., The Nature of Scientific Evidence' (1969 AAAS Symposium), in: Sagan, C. & T.Page (eds) UFOs-A Scientific Debate, Cornell U.P. 1972, p.287. dramatised the problem by pointing out that the effect might be simulated by suspending a gigantic sheet of flat plastic at the proper angle over the sea about 1000yds from the vessel. If the report is credible n5 It's reasonable to point out that the chain of evidence is rather weak in terms of the standards generally applied by scientists to reports of unidentified flying objects', for example. It is not clear that a hoax or opportunistic prank can be ruled out. A reflection in window glass might produce the same novel effect, and the "20 or 30 other witnesses" have probably not been traced and deposed under oath. there seems to be no possible explanation other than an extraordinary ice-halo reflection.

A dissimilar phenomenon but one inviting the same sort of speculation is the group of UAPs observed by crew and passengers of a BOAC Stratocruiser at FL19 near Goose Bay, Labrador, on a June evening in 1954. This event has some interesting similarities to the Channel Island UAPs.

In this case a linear array of dark shapes was seen that appeared to climb into the clear from below a broken stratocumulus deck and remained visible silhouetted against the bright "silver" sunset sky off the left wing over a distance of 85nmi. One large object changed shape in a "jellyfish-like" manner whilst six other smaller ones, disposed either side of it, moved relative to it and to one another n6 An account by the pilot, Capt Howard, showing drawings of the phenomena taken from his logbook was published in Everybody's Weekly, 11Dec, 1954. https://www.ufocasebook.com/1954ufomothership.html. The UAPs appeared to remain close to 0° elevation relative to the aircraft and at the same bearing from the aircraft for 22 minutes (by lining them up against the cockpit window post the navigator noticed a small deviation at one point, but this could possibly have been due to yaw in the aircraft axis). They appeared to be about 5nmi away. Towards the end of the sighting the smaller shapes appeared to "enter" the larger, which then dwindled and vanished. The UAPs were apparently not seen from an approaching F-94 closing head-on with the Stratocruiser at 20nmi, and there was apparently no ground or airborne radar contact with the UAPs (although the F-94 did acquire the Stratocruiser on its airborne radar at that range).

Thayer n7 Thayer, G.D, Optical & Radar Analyses of Field Cases, in: Gilmor, D. (ed.) Scientific Study of Unidentified Flying Objects (Condon Report), Sect. III, Ch..5, p.207. comments that "certain facts in the case are strongly suggestive of an optical mirage phenomenon" but adds that the persistence of the display at the same bearing over 85nmi would be "quite unusual" for any mirage. This seems a little understated! Capt Howard was certain that the UAPs were solid objects, not mirage images, and they were seen at first moving "in and out of a broken layer of stratocumulus cloud. As we watched, these objects climbed above the cloud . . . ." If accurately described this behaviour seems most unlike optical mirage. Nevertheless, we are inclined to agree with Thayer who finally places the report in "the category of some almost certainly natural phenomenon, which is so rare that apparently it has never been reported before or since."

Similarities with the Channel Islands UAPs include the extended visibility from a moving aircraft at an approximately constant bearing, horizontal motions of the UAPs relative to one another, confinement of the UAPs to within a few degrees of a horizontal plane containing the aircraft, and an apparent climb to the horizontal from a small depression angle.

But unlike the Labrador phenomena our UAPs were: Brilliant, not dark silhouettes; had stable, sharply-defined shapes instead of "jellyfish-like" fluid outlines; remained visible through 2000ft of altitude change; and were apparently also observed on a near-reciprocal line of sight from Capt Patterson's Jetstream far to the south.

None of these cases is very easy to explain by optical refractive index anomalies, but in the Channel Islands case it seems especially difficult (Section 6.d). Partly this is because of the difficulty of finding plausible light sources at distances that would not place severe constraints on the optical geometry of mirage or mock-mirage (6.d.i, ii, iii & iv), partly because of the duplication of identical very peculiar images, their rather well-defined steady lateral relative motion past each other, and the very wide-baseline triangulation of sightlines to UAP#1.

In the Labrador case lateral motions of the UAPs were observed, but of a rather chaotic kind, with the "about six" smaller objects seeming to switch about from left to right of the larger one in different patterns, whilst remaining approximately in a horizontal line. Finally they "suddenly vanished". The First Officer said it "looked as though" they shrank into the big "jellyfish" just before it began itself to dwindle in place, vanishing in a few seconds. It seems possible that such a display might be caused by some type of superior mirage, with looming and towering causing small silhouetted pieces of irregular terrain beyond the normal horizon to be lifted into view in a randomly changing sequence. Apparent relative horizontal motions in such a case could be an illusion as different parts of the skyline are selected in sequence by the optical duct.

On other hand, the Indian Ocean double-sun described by Minnaert involves a clear horizontal separation of two very well-defined, stable images. This is itself not explainable by any known mirage mechanism n8 Atmospheric scientist Andy Young suggests some very rare small-radius ice halo caused perhaps by an unusual crystal configuration, but agrees (emails to Martin Shough, 06.02.08) that a similar mechanism is ruled out in the case of the Channel Islands UAPs, because the freezing level was above 10,000ft and no line of sight from any observer to any possible light source can have passed through an ice crystal layer (see Section 5 and Section 6a & b)., but at least the perfect image(s) do(es) not display any relative motion.

In the Channel Islands case, however, we have as it were the worst of both worlds: Well-defined, stable images with identical internal detail, separated both vertically and horizontally, and displaying steady relative horizontal motion. If it seems scarcely feasible to explain the doublesun as a kind of lateral "looming" due to sharp horizontal refractive index gradients, it is surely difficult even to imagine the atmospheric structure that would cause two such stable images to move steadily past each other and swap places laterally, never mind explain observations on near-reciprocal sightlines.

Nevertheless for completeness we did consider the possibility that such a very extraordinary structure - a sharp vertical layer of RI discontinuity approximately representing Philip Morrison's imaginary "plastic sheet" - located in the north Channel Islands area between the Trislander and the Jetstream, might in principle explain both the laterally-displacing UAPS seen from the former and the UAP sighted on a near-reciprocal sightline from the latter. The idea would be that the "yellow/beige" object seen by Capt Patterson was actually a mirage image of the bright yellow Trislander, at that time many miles to the N. If this intermediate structure could be imagined to act as an atmospheric one-way mirror, might it even be possible that the "brilliant yellow" UAPs seen by Capt Bowyer and passengers were reflections of their own sunlit aircraft? Capt Patterson indicated (Appendix B) that the colour of his UAP was not dissimilar to that of an Aurigny Trislander (see Fig 36) seen at the range of Alderney in conditions of haze n9 It's worth noting that the only large ship possibly in the area, the Mv Bretagne (Section 4), was painted not yellow but white, with some red trim. See https://www.brittany-ferries.co.uk/index.cfm?articleid=149.

It was not hard to find problems with this speculation:

1. The bearing of the object seen by Capt Patterson ~2nmi to the W of Alderney is about 10° to the left of island. Angular and linear distance estimates may be unreliable, but Alderney is a nonnegotiable back-stop: The UAP was to the left of the island. However the Trislander position proven by the radar plot at the time of Capt Patterson's observation (Section 3, Fig.7) was well to the right of the island. The angle between these bearings is about 25°. There is no known or conjectural mirage mechanism that can refract ray paths laterally by 25°, even fleetingly, never mind for a long duration. Horizontal deviations of only seconds of arc or possibly in extreme cases minutes of arc might occur due to horizontal temperature fluctuations, and this is just transitory image wander, like stellar scintillation n10 In fact a couple of minutes earlier, when the hypothetical angle of lateral refraction would have been much closer to a favourable grazing angle, Capt Patterson was unable to see anything in the area (see Section 2).
   . The horizontal temperature gradients required are huge and completely unphysical. Even the strongest vertical gradients only cause bending (vertically) of <1.0°.
2. Towering of superior mirage can cause image enlargement, but of course only in the plane of refraction. Horizontal analogues of looming and towering are totally unknown and there is no mechanism for such. Refraction in two directions orthogonal to one another cannot occur in any case. A lensing effect like this would require the index of refraction to vary symmetrically around the line of sight. This is meteorologically unheard of, and such symmetry would anyway be inconsistent with the requirement to asymmetrically displace the entire image sideways by many degrees.
3. But let us suppose an unknown mirage mechanism capable of bending raypaths horizontally through 25 degrees: The distance to the Trislander from the Jetstream is over 30nmi. At that range, even supposing the most favourable orientation (a side-view of the 50ft fuselage) the Trislander subtends about 0.01°, very tiny. But in reality it is almost nose-on, and at about the same altitude, so the aspect presented to Jetstream is the most unfavourable, i.e. basically just the nose. (The wing section would be in the region of 1/10,000° thick at 30nmi, or in the order of 0.1 arcsec, which is smaller than the smallest visual angle detectable by the human eye in optimum laboratory conditions using a black line against a uniform bright background. The optical refraction cannot be invoked to fatten this horizontal wing section because our theory requires the ray bending to occur at 90° to it, right to left, not up and down.) So the available yellow area is (say) 10 feet of the Trislander's nose, which is less than 10arcsec at 30nmi, or about 1/260th of the diameter of the moon. (For comparison the disc of Jupiter is 30-45 arcsec, Mars and Venus around 15 arcsec. )
4. 10arcsec is about a factor 5 smaller than the threshold of detectability for a young non-myopic adult eye in ideal conditions. But Capt Patterson estimated that the object he saw (with the naked eye) was an oval or oblong shape which, by comparison with the island of Alderney, would have had a maximum horizontal dimension of about 0.5nmi at the same range. Alderney, in this perspective, would have subtended about 7° in width from 20nmni range, indicating an angular width of about 1.3° for the object, or more than twice the apparent diameter of the moon. He says this is a maximum possible size. When he estimates a comparison with a Trislander fuselage at the same range (~20nmi) he gives a smaller size, maybe 4 or 5 times the size of a Trislander's 50ft length, i.e. about 200-250 ft, which is about 0.1°, or 1/3 the diameter of the moon. The direct angular scaling against the adjacent island is in our opinion likely to be the more reliable. In any case, the average of these rather wide brackets is something comparable to the diameter of the moon.
5. An atmospheric-optical explanation of the sightings by Capt Bowyer and passengers as a"reflection" of their own Trislander would require at least 100% efficient backscatter at nearnormal incidence - "at least" because of the extreme near-specular brilliance of the yellow UAPs, much brighter than incoherent scattering from the Trislander's yellow paint could possibly be (the Trislander's windshield angle is ~50° away from the angle for specular sun reflection near 0° elevation). We imagine that this would add many orders of magnitude to the already outlandish refractive index gradients required by 1) above.
6. Even if a temperature domain wall with an amazing power reflection coefficient existed between the Trislander and Guernsey, we question whether it could simultaneously refract by 25° in one direction, whilst being perfectly transparent to rays passing unrefracted through it from the opposite direction. (The island of Guernsey was seen both behind and adjacent to the UAPs and appeared normal to Capt Bowyer.) It is also very unclear how such a phenomenon could explain two reflected images of the Trislander or their relative motions.
7. Our meteorological model shows a French coastal inversion decaying south of the Channel Islands and a small inversion (vertical gradient of course) of maybe 2 - 3°C/kft along the line of sight between the Jetstream and the Trislander, and capped by the 2000ft haze layer therefore below the altitudes of both aircraft. It seems to fit the general synoptic weather situation. Nothing in this model or the observations supporting it hints at even the remote possibility of anomalous vertical temperature domain walls in the atmosphere. There are no doubt other geometrical and physical objections but in our opinion the above points suffice to render the notion completely unrealistic.

In summary, we have tried our hardest to explain the observations but none of the theories we have explored sits comfortably with all significant features reported.

An unusual mock-mirage of brilliant sun-glitter reflections from the sea near the French coast was considered, and might be worth the cost of discounting Capt Patterson's sighting were it not for Capt Bowyer's explicit description of lateral image motions. This feature is effectively impossible for mirage; even so, we put the theory in the category of "barely plausible" to acknowledge its other attractions.

We score two other theories as "somewhat plausible" because they seem to have potential to explain the lateral apparent motion as well as at least some, perhaps a majority, of the other significant features. These are

• Secondary scattering, by a haze layer, of specular sunray reflections from greenhouse glass on Guernsey
• Earthquake lights

But a potential to explain is not an explanation. It may prove possible for other investigators to adapt these theories and so improve the fit with observation, or further work might thoroughly rule out one or both of them.

Finally we note that either of these theories could be consistent with the apparent absence of unambiguous ATC or weather radar detection. But although we have found no evidence of such detection, as mentioned in Section 4 the raw ATC radar data has not yet been thoroughly investigated to the point of ruling out all possibility of significant echoes n11 We anticipate that this issue will be clarified in a forthcoming report by the French government agency GEIPAN. It is our informal understanding, as of the date of writing, that GEIPAN's analysis has not found any significant echoes in the area.. Moreover, the complexity of the radar and software environment does mean that, in this case, absence of evidence would not necessarily be sufficient evidence of absence, so it may not be straightforward to exclude other theories solely on this basis.