Photographic evidence can contribute to a better understanding of the UFO phenomenon if the evidence has sufficiently strong credentials that the possibility of a hoax can be ruled out. It is also highly desirable that the photographic evidence be accompanied by strong witness testimony, but it is very difficult to meet these requirements (as in the case of remotely operated scientific monitoring stations) because of the unpredictable nature of UFO events (events that give rise to UFO reports). In order to be confident of the authenticity and flawless operation of the equipment and acquisition, it is necessary to plan an observational program very carefully. This approach has been adopted by Strand and is discussed further in Section 6. However, such equipment must normally be run in an automatic mode so it is unlikely that there will be witness testimony to accompany the data acquisition.

On the other hand, photographic and similar evidence are sometimes acquired in connection with unexpected and incomprehensible UFO events. In these cases, there will normally (but not invariably) be witness testimony but, since the data acquisition was not planned, the equipment, operation and analysis will probably not be optimal and there may indeed be some question concerning the authenticity of the claimed data.

Haines presented in some detail one case in which an intriguing photograph was obtained, but the intriguing aspect of the scene was unknown to the photographer at the time the photograph was taken. This event occurred on October 8, 1981 at about 11:00 am Pacific Daylight Time on Vancouver Island, British Columbia, Canada. It has been described in detail elsewhere (Haines 1987), and a copy of that article is to be found on the Web Site (see Section 15).

In 1984 Haines received on loan, directly from its owners, two connected frames of 35 mm color negative film. The lower number frame shows a child standing in front of a fireplace, and the higher number frame shows a daytime view of a mountain with evergreen trees on the bottom and a white cloud near the top of the mountain. The intriguing aspect of the latter frame was that it showed a silvery oval-shaped object set against the blue sky. The photographer and her family were making a rest stop in a Canadian provincial park and the exposure was made on impulse because of the beauty of the scene.

Haines and his father, Donald Haines, spent four days with the principals of the case visiting their home and the site where the photograph was taken (north of Campbell River, British Columbia) exactly two years later. Fortunately, the weather conditions were comparable with those of October 8, 1981. Donald Haines, a registered civil engineer and land surveyor, carried out a land survey of the relevant area.

The object appeared to be a disk with the near edge tipped downward, possibly with a rounded "dome" or protuberance on its upper surface. Richard Haines provided detailed information concerning the camera, the lens and the film. Haines had analyzed the negative using a microdensitometer; the blue sky and cloud were quite bright and the brightest spot on the disk was even brighter. The luminance gradient of the brightness of the disk was measured and found to be consistent with what would be expected for a diffusely reflecting metal object, with a shape similar to that indicated by the photograph and the known position of the Sun. The color photograph was also analyzed by making black and white enlargements on different wavelength-sensitive papers. The negative was also digitally scanned using a Perkin-Elmer scanning densitometer, using three separate color filters which matched the film's three dye layers.

Haines was especially diligent in looking for evidence of a double exposure, but found no such evidence. He also looked for a possible significant linear alignment of pixels or grains which might result from the presence of a thin supporting line or thread, assuming that the object was a small model hanging beneath a balloon, but no such evidence was found. Haines tested for differential edge blur, such as might be produced by linear motion during the exposure, but found no such blur.

Haines also attempted to identify the object in the photograph as something mundane. He considered, in particular, the possibility that a Frisbee had been thrown into the air and photographed. The principals did own a Frisbee, but it was dull black, not shiny, and the principals steadfastly denied having produced the photograph in this way. Haines experimented with several other Frisbees. He attached a dome to the top of one Frisbee and tried to fly it, but it would fly no more than about ten feet before losing lift. Haines also calculated that a Frisbee would have displayed noticeable edge blurring in the photograph.

This case is instructive in showing what detailed analyses of a photograph can be made using modern analytical equipment, but it suffers from the severe drawback that there is no witness testimony to accompany the photograph. While the panel was impressed with Haines' thorough analysis of the evidence he had available, there was some concern that a film defect or blemish may have been introduced during processings, and there was considerable discussion concerning the crucial point that an object that had appeared on the photograph was apparently not seen by the photographer or by her companions. The picture was taken with a single-lens reflex camera, which means that the object must have been in the field of view of the viewing screen as the photograph was being taken. Haines explained that there is published research which shows how perceptual "blindness" can occur even when physical objects are clearly present in the environment. Louange also pointed out that an object that is angularly small, stationary, and not expected to be present, is not as likely to be noticed as a similar object that is moving.

The panel expressed the opinion that detailed analysis of photographic evidence was unlikely by itself to yield evidence sufficient to convince a neutral scientist of the reality of a new strange phenomenon unless a number of additional detailed conditions are met (see Appendix 2). They also expressed concern that, now that modern digital techniques are easily available in photo laboratories, it may never be possible to rule out possible hoaxes without convincing, corroborative eye-witness accounts.

For further information about photographic cases, see Section 15 and Appendix 2.

When witnesses of unidentified aerial objects are debriefed by investigators, some of the most striking statements concern the luminosity of the phenomenon, according to Vallee. It is not unusual to hear expressions such as "it lit up the whole landscape," or "every object in the area stood out, thrown into intense relief," but it is normally difficult to go beyond these subjective statements to obtain reliable quantitative estimates of the luminosity of the phenomenon. Vallee summarized data for six cases of unexplained aerial phenomena that have been reported by qualified observers over a 20 year period, with a view to making estimates of the optical power output. Vallee's estimates range from a few kilowatts to many megawatts.

Case #1 occurred on August 27, 1956, near McCleod, Alberta, Canada. The witnesses were two Royal Canadian Air Force pilots who were flying in a formation of four F86 Sabre jet aircraft. The planes were flying due west over the Canadian Rockies at 36,000 feet about one hour before sunset. One of the pilots saw a "bright light which was sharply defined and disk-shaped," that resembled "a shiny silver dollar sitting horizontal," situated below the planes but above a thick layer of clouds. It appeared to be considerably brighter than sunlight reflecting off the clouds. The duration of the sighting was estimated to be between 45 seconds and 3 minutes. The first pilot to notice the object reported the observation to the flight leader and then took a photograph on a Kodachrome color slide. This case and this photograph were subsequently analyzed by Dr. Bruce Maccabee (Maccabee, 1996). Maccabee has presented an argument against the propositions that the phenomenon is due either to reflection of sunlight by the clouds or to lightning. From the available data, Maccabee estimates the luminosity of the object (the power output within the spectral range of the film) to be many megawatts.

Case #2 occurred in late September 1965 at Fort-de-France (Martinique). Two French submarines accompanied by a supply vessel were returning to France from Norfolk, Virginia, stopping at Martinique. One evening, according to the report, when there was a dark sky and clear weather, a large luminous object arrived slowly and silently from the west, flew to the south, made two complete loops in the sky above the vessels, and vanished like a rapidly extinguished light bulb. The object was observed by a highly qualified helmsman from the deck of one of the submarines. He took six pairs of binoculars to the conning tower and distributed them to his companions. There were in all 300 witnesses, including four officers on the submarine Junon, three officers on the submarine Daphne, a dozen French sailors, and personnel of the Martinique weather observatory. The appearance of the object was such as to suggest either a large ball of light or a disk on edge. Its color was that of a fluorescent tube, and its apparent luminosity was that of the full Moon. It moved slowly and horizontally, at a distance estimated to be about 10 kilometers, and left a whitish trace in the sky similar to the glow of a television screen. After the object first vanished, the halo remained visible for a full minute. Some time later the halo reappeared and the object then emerged as if "switched on." After further maneuvers, the object flew away.

Based on the descriptions of the witnesses, Vallee estimates that the luminosity of the object was of the order of 2 megawatts.

Case #3, that occurred at Voreppe, France on November 5, 1976 at 8:10 p.m., was investigated by GEPAN/SEPRA (GEPAN 1976; see also Appendix 1). The director of a physics laboratory at the Nuclear Research Centre in Grenoble saw a luminous disk in the sky as he was driving. Several other witnesses reported a similar observation on the same day. The main witness, considered to be a reliable scientist, gave a precise description of the position (in front of mountains), size and speed, as well as of the luminosity of the disk compared to the luminosity of the moon. The illumination of the landscape was reported to have been brighter than the illumination produced by the full moon when it is at the zenith. From this fact, and from the relevant geometrical considerations, the GEPAN/SEPRA investigator estimated the minimum transmitted luminous energy to be 6 kW if the estimated altitude of 500 m was correct, or 24 kW if the altitude was 1,000 m.

Case #4, that was also investigated by GEPAN/SEPRA, occurred at Gujan-Mestras, France, on June 19, 1978 at about 1 a.m. GEPAN/SEPRA was advised by the Gendarmerie that three witnesses had reported seeing a large luminous object that had emitted a loud noise. They also reported that the public lighting in the town had been extinguished for a few minutes as if triggered by morning light. The GEPAN/SEPRA investigators carried out a site investigation and made measurements of the triggering threshold of the photo-cells that controlled the public lighting system. This information led the investigators to an estimated radiated energy in the range 40 kW to 5 MW.

Vallee reviewed briefly two other cases: Case #5 occurred on December 30, 1966 in Haynesville, Louisiana, and Case #6 occurred on August 24, 1990 at Greifswald, Germany. Vallee cautioned the panel that the estimates of luminosity presented at the workshop are raw approximations derived from a comparison of the estimated intensity in the visible band with the intensity of known sources, such as the full moon and automobile headlights, and from assumptions concerning the distance and perhaps size of the source.

The panel noted that the human eye is a very poor device for measuring absolute luminosities: the state of dark adaptation of the eye affects the amount of light reaching the retina, and different parts of the retina respond differently to light. Furthermore, the above luminosity estimates were apparently based on the assumption of isotropic emission. This may be a reasonable assumption for a natural phenomenon, but could be inappropriate if a case involves a technological device. For instance, aircraft landing lights are highly anisotropic. A 1 kW source that is beamed with a half-angle of 3.6 degrees has the same intensity as a 1 Mw isotropic emitter. Furthermore, the distance estimates may be quite dubious. Hence the power estimates derived for the above cases must be considered quite uncertain. The most promising cases will be those for which some form of physical interference took place (such as an effect on a public lighting system), but these call for detailed investigation by specialists familiar with such systems.

For further information about luminosity estimates, see Section 15.

Velasco presented information on radar cases drawn in part from the files of GEPAN/SEPRA (see Appendix 1). He pointed out that one catalog (the "Weinstein catalog" now under development at GEPAN/SEPRA), with 489 cases in all, contains 101 (21%) radar/visual cases (cases that involve both radar detection and visual observation), and the files of the US Air Force Blue Book project contain 363 cases of which 76 (21%) are radar/visual cases. Since 1945, reports of aeronautical cases have been collected by order of the French Air Force Chief of Staff. From 1977 on, information from civil and military observations made in French air-space have been sent to GEPAN/SEPRA (see Appendix 1). It should be noted that civil radar information usually refers only to objects containing a transponder, whereas military radar equipment can detect any object greater than two square meters in radar equivalent surface area. From 1982 on, twelve French aeronautical cases were reported to GEPAN/SEPRA. Of these, only three or four cases may be considered to be radar-visual cases of the UFO type.

One of these cases is particularly interesting. This case occurred on January 28, 1994, about 70 kilometers southeast of Paris, at a height of 11,700 meters, under excellent meteorological conditions. An object was first noticed by a steward who happened to be in the cockpit, and his observation was then confirmed by the copilot. The captain then saw the object. It was above the thick layer of altocumulus clouds at 10,500 meters. The captain described the object as resembling a gigantic disk (diameter about 1000 meters, thickness about 100 meters) with slightly fuzzy edges. The witnesses suddenly lost sight of the object when the edges appeared to go out of focus and the object disappeared.

Corresponding radar information was obtained from the military air traffic control (ATC). The object was positively detected by radar for a period of 50 seconds. The apparent speed of the object was measured first as 110 knots, then as 84 knots, and subsequently as zero. The altitude of the object was not recorded by radar. The radar was also tracking a nearby commercial aircraft and appeared to be in good working order. There appears to be good correspondence between the radar measurements and the visual observations.

Von Ludwiger also presented information concerning radar evidence, drawing in part on the results of studies that he had carried out in association with other members of the Mutual UFO Network (MUFON) Central European Society (MUFON-CES). For a certain period of time, they were able to obtain records from both civil and military ATC radar systems. The Swiss Military ATC was particularly cooperative, and provided several hundred hours of radar data over the time period 1993 to 1996. Radar data were also obtained from Belgian sources through the good offices of Professor A. Messens (SOBEPS, 1991). Military ATC radar systems provide three-dimensional information, whereas civilian ATC radar systems provide only two-dimensional information. Furthermore, the functioning of civilian ATC radar systems normally depend upon the cooperation of a transponder in the object being tracked. For this reason, civilian ATC radar records are usually not helpful for the study of unidentified objects. Moreover, there is the general problem that ATC systems are designed to register only targets for which the flight characteristics fall within certain parameter ranges. For instance, any object that moves faster than Mach 4, or does not follow a smooth trajectory, will be rejected by the system whether civil or military, and so will not be tracked. A further limitation, relevant to the study in hand, is that conditions for a good radar record and conditions for a good visual sighting are quite different. An object can best be seen if it is at low altitude, but radar systems normally do not detect objects at low altitude.

In the United States, the Federal Aviation Administration (FAA) radar routinely records on tape all targets, not just aircraft with transponders. Of course, radar systems record only objects that are sufficiently close and have high enough altitude. Although it is unlikely that private investigators will be able to obtain regular access to these records, such access has been granted on occasion. Such data can be very helpful in providing physical evidence for cases that have reliable witness testimony, in which case the records can be compared to witness testimony to determine whether an object seen visually was also recorded on radar, and — if so — to obtain accurate velocity estimates.

According to von Ludwiger, there are many events, involving both visual observations and radar responses, in Swiss airspace but the radar records are not publicly available. However, one case for which radar records were released occurred on June 5, 1996 at about 2:30 p.m. Six employees, including radar operators, of the military ATC at Dubendorf, Switzerland observed from their building in Klothen a large silvery disk apparently at a distance of 1700 meters. It appeared to be rotating and wobbling at an altitude of 1300 to 2000 meters. There was a corresponding recording of a target by three radar devices.

Von Ludwiger also mentioned a number of other cases of radar targets, some of which followed curious trajectories unlike those of conventional aircraft. Recognition of these anomalous trajectories typically came some time after the events when the radar data were analyzed. Von Ludwiger considers this to be one reason that (except for two cases) it was normally not possible to find corresponding visual observations. Von Ludwiger considers that for many of these cases the most likely explanation involves anomalous atmospheric refraction of the radar signals, but that some cases for which the radar records showed very long connected trajectories may have been produced by real objects. (See Appendix 4.)

The panel concludes from these presentations that the analysis of radar records is a very specialized activity that requires the services of radar experts. (See, in this connection, Appendix 4.) The panel also notes that information from military radar can be obtained only with the cooperation of military authorities, and that most military authorities do not offer this cooperation. Although intriguing cases have been presented by both Velasco and von Ludwiger, further study of this phenomenon by means of radar-visual cases may not be feasible unless the relevant authorities recognize the mission of an official UFO research organization (as has been done in France) and give the investigators clearance for access to some of the unexploited raw data. It would be necessary for the research organization to help implement adequate software modules that can read and store available data in a mode of operation that does not interfere with the primary mission of the system.

Strand summarized the design and operation of the Hessdalen Project. Hessdalen is a valley in central Norway, 120 kilometers south of Trondheim. The valley is 12 kilometers long and a maximum of 5 kilometers wide. The hills to the west and to the east rise to about 1,000 meters above sea level. Most people in the valley live at a height of about 800 meters.

In December 1981 the inhabitants of the Hessdalen valley began to report seeing strange lights. They were sometimes visible three or four times a day. There were hundreds of reports during the period 1981 to 1985, but the phenomenon began to decrease during 1984, and since 1985 there have been comparatively few sightings. Most observations were on winter nights: there were comparatively few during the summer or during the day.

Witnesses reported observations that seemed to fit into three different categories:

Type 1: A yellow "bullet," with the sharp end pointing down.
Type 2: A strong blue-white light, sometimes flashing, always moving.
Type 3: A pattern comprising many light sources with different colors that moved as if they were physically connected.

In 1983, a small group with five participants set up "Project Hessdalen." They received assistance from the Norwegian Defense Research Establishment, the University of Oslo, and the University of Bergen. They carried out field work in the Hessdalen valley from January 21, 1984 to February 26, 1984, when up to 19 investigators were in the field at the same time. The project then involved three stations with observers and their cameras, some cameras fitted with gratings to obtain spectroscopic information. At the principal station, observers used the following equipment: cameras, some fitted with gratings; an infrared viewer; a spectrum analyzer; a seismograph; a magnetometer; radar equipment; a laser; and a Geiger counter.

Lights that were recorded to be below the contours of the mountains must have originated in the Hessdalen region, but lights that were recorded to be above the crest line may have originated at a great distance. Without triangulation or other information, it is impossible to determine the distances of the lights. However, some of the events that were seen as lights were tracked also by radar. If taken at face value, the radar measurements would imply speeds up to 30,000 kilometers per hour. (However, see Appendix 4.)

During a period of four days, unknown lights were seen on 10 occasions, and the flux-gate magnetometer registered 21 pulsations, of which 4 appear to correspond with the observations of lights, suggesting an association between some of the unknown lights and magnetic disturbances. The gratings on the cameras were intended to obtain spectroscopic data: the spectra appear to be continuous, with no indication of either emission lines or absorption lines.

Observations continue to be reported from the Hessdalen valley; the rate is now about 20 reports per year. An automatic measurement station, for installation in Hessdalen, is now being developed and prepared at Ostfold College (Norway), which is the present base of Project Hessdalen. This station will include a CCD-type camera in the visible region. The output from the CCD-camera will be fed automatically to a computer which will trigger a video recorder. This automatic station will hopefully prove to be but a first step in the development of a network of stations.

As a result of this presentation, the panel concluded that there would be merit to designing and deploying a not-too-complicated set of instruments. These should be operated according to a strict protocol in regions where the probability of significant sightings appears to be reasonably high. It is recommended that, as a first step, a set of two separate video recorders be equipped with identical wide-angle objectives and installed on two distant fixed tripods to help eliminate the possibility that some of the apparent motions detected by video recorders are due to the operators' hand movements or ground vibrations. It would also be useful to set up two identical cameras, one of which is fitted with a grating. However, experience so far at Hessdalen indicates that a grating may not be adequate for obtaining spectroscopic information. In view of the great importance of spectroscopic data, it would be highly desirable that special equipment be developed and deployed for obtaining high-resolution spectroscopic data from transient moving sources. This may be a nontrivial problem.

If it proves possible to obtain useful results from a small system, such as suggested above, one may be able to make the case for the design and implementation of a permanent surveillance network. This should be designed as a multi-purpose system so that costs and data can both be shared. This could resemble the Eurociel project that was studied in Europe in the 1980s at the request of GEPAN/SEPRA. (See Appendix 1.)

The panel notes that in cases that involve repeated, semi-regular sightings of lights (such as are said to occur at Hessdalen in Norway and at Marfa in Texas), it is difficult to understand why no rational explanation has been discovered, and it would seem that a small investment in equipment and time should produce useful results.