Case 35

South Pacific

Fall 1967

Investigators: Levine, Low, and others

BACK to Chapter 2


The events began with a visual sighting about 8:00 p.m. of a stationary object with colored lights over the ocean. Missile-tracking radars were asked to look for the object; they immediately picked up many unidentified targets, most of them moving, and tracked them. Most moving targets permitted radar lock-on. They moved at speeds up to 80 knots, and sometimes returned very strong echoes. Several additional visual sightings were reported. Most sightings were made over the ocean, but some targets appeared to the east and north, over land. The radar targets were still being observed when the equipment was closed down about 2:30 a.m. Yet no aircraft were known to be in the area, and three flights of fighters sent in to investigate found nothing unusual.

An unusually strong temperature inversion provided favorable conditions for both visual and radar mirage effects. Mirages of ships below the normal horizon appear to account adequately for the stationary or slow objects. The higher, faster radar targets were consistent with birds, which tracking-radar operators had not had occasion to look for before. Similar radar observations were reported on two subsequent days.


Project Blue Book had notified the Colorado project of this interesting visual and radar sighting at AFB A. It was also reported that, in a test three nights after the sighting, it had been estab- lished that radars at the base could once again observe "bogies"


similar to those sighted on the night of the original sighting. Project investigators and others visited the site on two different dates. On the latter day, the following were present: R. T. H. Collis, Roy Blackmer, and Carl Herold of Stanford Research Institute; Marx Brook of New Mexico Institute of Mining and Technology; Roger Lhermitte of the Environmental Science Services Administration; and Low and Levine of the Colorado project. On the first date Low and Dr. Robert Nathan of the Jet Propulsion Laboratory had visited AFB A.

Observers. The AFB A sightings were exceptional because of the high professional qualifications of the observers. Two were officials of the Western Test Range, each having had 17 yr. of exper ience as a naval aviator. One of them had 10,000 hr. as an air intercept and final approach controller; the other also had been an air intercept controller. A third, who was Range Air Control Officer on the night of the first sighting had had 11 yr. experience with ground and airborne electronics systems. Six others were radar operators employed by private contractors on the base, all of whom had had extensive experience in radar operation. They displayed impressive understanding of the sophisticated radar systems they were operating and good comprehension of radar engineering principles. Another witness was of the security force, without extensive technical training.

Radars. The following radars were involved in the sightings:


Details of the sightings. 2000 to 2045 For one-half hour a missile range official observed from his home an object at azimuth 290°. He called another official, also at home three miles to the south, who confirmed the sighting at azimuth approximately 280° and altitude 10° to 15°. The second observer reported that the object seen through 7 X 50 binoculars, appeared the size of a large thumbtack, elliptical in shape having a red and green light separated by a distance about the wing span of an aircraft. But the object was stationary, and fuzzy like a spinning top.

2045: Observer two called Range Control Operations (located at an altitude of 900-1,100 ft.). The range control officer confirmed the visual observation. To him it appeared to have white, red, and green or blue colors that did not vary. They "looked like the running lights on a stationary object." He gave its bearing as 290°, range, several miles, altitude approximately 10,000 ft., and suggested that the object looked like a helicopter.

2045: FPS-16 radar in search mode locked on two strong targets, one moving around and one stationary. The stationary target appeared in the general direction of the visual sighting, but the optical position was not determined with sufficient accuracy to establish that this was a simultaneous optical-visual sighting. The original interpretation was a helicopter, with another assisting.

2100: The range control officer checked for possible air traffic in the AFB A area with several other air bases. All reported negatively.

2100: Using its FPS-16 in lock-on automatic mode, base D reported strong targets headed toward AFB A. Because of the narrow beam of the radar the targets were presumed to be in line.

2100: TPQ-18 radar at AFB A was brought into operation, and saw many targets. One, at 8 n.m. range, 4,000 ft. altitude, 290° azimuth, and 4°.6 elevation proceeded south at low speed. One strong target approached and went directly overhead. At one time, the TPQ-18 saw


four targets. Base D saw as many as eight. AFB A and base D did not establish that they were looking at the same targets.


a. Dozens of targets were seen. Speed ranged from 0 to 80 k. with rapid changes in altitudes. The radars would lose their tracking "locks" on the objects, and then re-engage.

b. The target that went directly overhead produced an extremely strong 80 dB signal. Three persons went outside the radar shack, but were unable to see any object. On the TPQ-l8 radar one of the strongest targets appeared to separate into eight objects after which it was necessary to switch to manual to gain control to separate the signal.

c. NORAD surveillance radar at AFB A operates at a frequency quite different from the tracking radars. It saw no targets, but its operator reported clutter or possible jamming.

d. Base D reported a target "bigger than any flat-top at three miles."

e. As the radar activity increased, the number of visual obser- vations decreased.

(only the most interesting are described)

a. Many objects were sighted, but they declined in frequency as the radar activity increased.

b. One visual appeared to move toward the observers so alarmingly that one of them finally yelled, "Duck."

c. One object, dull in color but showing red, white, and green, moved generally south and finally out of visual range.

d. Another, the color of a bright fireball, moved on a zig-zag course from north to south. Two radar operators reported, "The radar didn't get locked onto what we saw. By the time the radar slaved to us, the object was gone visually, and the radar didn't see anything... It looked like a fireball coming down through there. Like a helicopter coming down the coast, at low elevation. We got the 13-power telescope on it." Then it grew smaller and smaller until it disappeared. Duration 1.5-2 min. Moved only in azimuth. Brighter than a bright


star. Like aircraft landing lights except yellower. This sighting occurred between 0100 and 0200 on the second night. A balloon was released about this time, and the winds were right to accord with the sighting; but the weather officer thought it could not have been a balloon, because the report did not indicate that the object rose, and a balloon would have risen at approximately I,000 fpm.

f. Two other radar operators reported having seen an object that traversed 45° in a few seconds, "making four zigs and four zags," and then, after reappearing for one second, disappeared to the north.

2310: Air Defense Command scrambled the first of three flights of fighters to investigate the situation. The tape of the conversations with the radar sites and other bases gave evidence of considerable confusion at this time.

The fighters were handed off to AFB A Range Control by the FAA at a nearby city and controlled locally. Range Control tried to vector the fighters in on the bogies, but found it impossible to do so very systematically. By the time the second flight came in, the controllers were so busy with the aircraft that they no longer observed any unidentified targets. They did observe a moderate amount of clutter in the west and southwest quadrant. None of the fighter pilots saw anything. One pilot observed something repeatedly on his infrared detector, but only at distance. As soon as he would close in, the object would disappear. Another aircraft did "lock-on" to a target which was found to be a ship.

Weather. The weather officer reported that there was an inversion layer at 1,800-2,200 ft. (The unidentified targets generally were reported to be above the inversion). All observers indicated that the night was exceedingly clear. The project's consulting meteorologist reports:

The following is a summary of weather conditions surrounding UFO visual and radar sightings near .... [AFB A] between 7:30 P.M. and midnight on .... [the date of the first sighting].


Fig 4

Figure 4: Vandenberg Weather

Click on thumbnail to see full-size image.



Radiosonde and wind data from--

.... [AFB A, island A, city A]

Surface weather observations surrounding the times of sightings from--

.... [city B, C, D, E; AFB A, B, C; base D]


In a weather sequence which moved a trough line and a low pressure center southeastward from northwestern Utah to northwest Texas.... [the day prior to the first sighting], a dome of high pressure formed over the Great Basin and a surge of warm air moved from northeast to southwest.... Most of the surge of warm air moved southwestward from the southern part of the .....Valley between midnight.... [the day before the sighting] and 3:00 P.M. ....[the day of the sighting]. Weather stations near the coast from ....[city B] to ....[city D] all showed abnormally warm temperatures at a time of day when ordinarily a sea breeze would have created a cooling influence.


Using surface wind data from various coastal stations it is possible to reconstruct an approximate pattern of the forward edge of the warm, dry air which moved out over the ocean from a general northeasterly direction. For most stations, fairly strong northeasterly winds were maintained through 11:00 A.M. (see Fig. 4) with northeast winds continuing until 3:00 P.M. at the surface at ....[AFB B].


The upper wind flow from 1000' to 7000' was still from an easterly component at ....[island A] shortly after 3:00 P.M. By 4:00 P.M. air was still moving from an easterly component between 3000' and 10,000' over....[AFB A]. Near the surface westerly winds were beginning to move the warm air back toward the east and southeast. This air had been cooled and some moisture had been added during its stay over the ocean.

During most of the afternoon hours the modified air moved from the ocean back over the coastal area. Some of the strongest evidence of the bulge of warm air over the ocean is indicated by the warm, dry air that moved over ....[city D] between the hours of noon and 5:00 P.M. With surface wind directions from 240° through 300°, temperatures held above 80° with maximum of 90°. A portion of the heating of this air would have been caused by dynamic heating as it it moved downslope from the .... mountains.

The abnormality of the warm air is indicated in Figures 5 and 6 by the approximate difference in air temperatures between 6:00 A.M. and 8:00 P.M. The blue profile of normal.... temperature [the date of the first sighting] was made up from long term average maximum and minimum temperatures and an assumed sea breeze influence. The red shaded area indicates the approximate abnormality of warm temperatures on this day as warm, dry air moved from land toward the ocean as compared with typical weather for.... [the date of the first sighting]. The hatched area shows the abnormality remaining after the air had been modified by its path over water.


Fig 5

Figure 5: Time/Temp Charts

Click on thumbnail to see full-size image.


Fig 6

Figure 6: Time/Temp Charts

Click on thumbnail to see full-size image.



When warm, dry air is forced to move from a land mass out over cooler water it creates a narrow boundary of mixing as moisture is picked up from the ocean developing small turbulent eddies of cooler, more moist air near the ocean surface. This is accompanied by very rapid fluctuations of refractive index. At the upper edge of the bulge of warm, dry air there would be another more difuse boundary where some- what less sharp differences in both temperature and moisture would be present. However, there would be corresponding fluctuations in refractive index.

The Glossary of Meteorology defines a mirage as "a refraction phenomenon wherein an image of some object is made to appear displaced from its true position...The abnormal refraction response for mirages is invariably associated with abnormal temperature distribution that yield abnormal spatial variations in the refractive index. Complex temperature distributions produce correspondingly complex mirages."

The layer of warm, dry air above cooler water from the ocean would have been particularly conducive to anomalous propagation of any radar unit scanning the atmosphere at low angles. A somewhat less important segment of the air mass capable of producing anomalous propagation on the radar would have been the upper boundary of the bulge of warm dry air. The following is quoted from Battan's book on RADAR METEOROLOGY under the heading of Meteorological Conditions Associated with Non-standard Refraction. "There are various ways that the index of refraction can be modified to give rise to anomalous


propagation... When warm, dry air moves over cooler bodies of water, the air is cooled in the lowest layers, while at the same time mois- ture is added. In this way strong ducts are produced. These conditions are frequently found over the Mediterranean Sea as air blows off the African continent. Extreme anomalous propagation has been experienced in this region. For example, there have been days when centimeter radar sets have 'seen' ground targets at ranges of 400-500 miles, even though the horizon was at perhaps 20 miles. In conformance with meteorological terminology, superrefraction brought about by the movement of warm, dry air over a cool, moist surface may be called 'advective superrefraction.' By the nature of the processes involved, it can be seen that such conditions can occur during either the day or the night and last for long periods of time. The duration would depend on the persistency of the glow patterns producing the advection."

Figure 7 contains the wind and temperature profiles for ....[island A] and ....[AFB A] beginning with release times of 3:15 P.M. and 4:08 P.M. PST respectively on ....[the date of the first sighting]. At ....[AFB A] (shown by the solid lines of temperature, dew point, wind direction and velocity) dry air prevailed for all levels above the surface at: 4:00 P.M. (For the lowest point on the profile, surface temperatures reported at 7:30 P.M. have been substituted). The vertical sounding of temperature, dew point, wind velocity and direction for ....[island A] are indicated by the dashed lines in Figure 7. Temperatures even warmer


Fig 7

Figure 7: Wind/Temp Profiles

Click on thumbnail to see full-size image.


than over ....[AFB A] were reported in the ascent above ....[island A]. For emphasis, the area shaded in red indicates how much warmer the temperatures were over ....[island A] than at ....[AFB A] during the mid-afternoon hours. Ocean water temperatures between 58° and 59° were being reported, which is considerably cooler than the warm, dry air having temperature in the 80's as it moved from land to over the water.


It is the author's opinion that the surge of very warm, dry air may have caused a mirage and visual observations could have been correspondingly distorted in the vicinity of ....[AFB A] between 7:30 P.M. and 8:30 P.M. It is more certain that the air mass conditions prevailing over the water continuing through at least midnight in an arc from south of ....[AFB A] swinging eastward to the coastline could have produced anomalous propagation echoes on radar. Visibility observations were generally 12 miles or greater at all stations and no clouds were reported by the observer at ....[AFB A] between 7:00 P.M. and midnight. ....[base D] reported a few stratus clouds offshore in the Remarks Column beginning at 7:00 P.M. continuing through 11:00 P.M.

Evaluation and Conclusions:

Further radar tests. Three days after the first sighting, under weather conditions similar to the first day but with more wind, more clouds, and lower temperatures, the FPS-16 radar at....[AFB A] was operated to determine if similar targets could be seen again. Targets having the same general characteristics were acquired, but they were


not as strong as the earlier sightings. Two other operators, working unofficially with a different radar, indicated that they observed "some of the same sort of stuff."

On the night of the investigators' second visit, similar targets were acquired on the FPS-16 and TPQ-18 radars. The radar experts among those present (Blackmer, Brook, Collis, Herold, Lhermitte) immediately requested that printouts be obtained giving information on signal strength. This information could not be compared with earlier sightings because the operators had not taken steps to print out the data from the other observations.

General conclusions. The AFB A series of sightings is remarkable for two reasons; first, because of the extraordinarily high qualifications of the observers, and second, because of the availability of hard instrument data. No other UFO case in the records of the Colorado project contains so many numbers, representing such quantIties as range, azimuth, elevation, and velocity. Information from which signal strengths could have been computed also would have been available had the operators thought to print it out, but they did not. To relate signal strengths and ranges for these events, it was necessary to go back to the tape of the conversations and find the reports of signal strengths, which, when assigned precise times (fortunately, the tape contained good timing references), could be compared with the printouts of range, which also included timing references. Information on the visual sightings was, except for the high credibility of the observers, comparable to that in other reports of UFO sightings in the Colorado files: i.e., no reliably measured quantitative values were available from such sightings.

Mirage conditions. The detailed weather study by Loren Crow was not available at the time of the second trip to AFB A, so that it was not known at that time that the atmospheric conditions were in fact quite unusual. Fig. 7 of the Crow report indicates that at AFB A, although return air flow at the surface was well established by the late afternoon of the original sighting, the flow at 2,000 ft. was still from the northeast, so that a thin sheet of warm, dry air lay over the


cool, moist air. This sheet of air extended southward almost to the island, where there was return flow from the surface to 3,000 ft., but easterly flow persisted from 3,000-10,000 ft. There were strong gradients of moisture and temperature at both stations. Crow has pointed out that the temperature and moisture contrasts probably were even greater than those shown, because the surface measurements were not made at the surface, but at some distance above it. Altogether the weather report indicates that conditions were very favorable indeed for optical mirage and scintillation and for anomalous radar propagation.

It should be noted that the incident that set off the entire sequence of events was an optical sighting at 8:00 p.m. It appears highly probable that the observer saw the running lights of a ship below the normal horizon, but made visible as a result of mirage. The conditions for such a mirage were present, but it must be pointed out that both the first two witnesses insisted emphatically that the object appeared at an elevation of about 10°. That is too high for a mirage of a ship's lights below the horizon. Hence, either their reports of the elevation angle were incorrect, or some other explanation must be found. However, even experienced observers tend to overestimate elevation angles.

A further fact is of interest, and that is that, in the Operations Control Center on the date of the second visit to AFB A, one of the operators of a search radar declared that he never saw any ships, that the shipping lanes were too far off the coast for ships to be seen by radar from that location, although the antenna was at an altitude of approximately 1,000 ft. He thereupon switched to his most distant range (80 mi.) and immediately a sprinkling of blips appeared at extreme range. They turned out to be ships, their identity confirmed by their slow speed. Since there is no reason to suppose, from a quick study of weather conditions that night, that anomalous propagation had anything to do with the observation of ships, it must be concluded that they could be seen any time. The only reasonable explanation of the operator's statement that he never saw ships on the scope is that


he had never looked for them. Both the original witnesses indicated that large ships never were seen visually from the coast, and that is undoubtedly correct, because they would be below the horizon. Computations show, however, that, under mirage conditions, the running lights of ships would be visible at the 80 mi. range the radars had indicated.

Some of the visual sightings obviously were not of ships. However, they were impossible to evaluate on the basis of the limited and subjective descriptions given. In this connection, it is significant to note the importance of quantitative instrument observations or records in such investigations. The visual objects could not be evaluated with much confidence, for lack of definitive evidence; but abundant quantitative radar records made it possible to identify most of the radar targets beyond serious doubt.

Birds. The behavior and characteristics of the unidentified radar targets appeared to be consistent with the hypothesis that most of them were birds. Individual birds would produce signal strengths consistent with those observed. (The targets observed the night of the second visit to AFB A, according to calculations made by Dr. Lhermitte, yielded a radar cross section of approximately 10 cm.2). The velocities and coherent tracks of the targets also suggested consistency with the bird hypothesis.

In view of the remarkable inversion conditions on the date of the original sighting, it is highly probable that some of the radar targets were effects of anomalous propagation (radar mirages). Temperature and moisture gradients were quite sufficient to produce echoes from atmospheric discontinuities.

At first, even the radar experts were puzzled by the radar data, because the remarkably strong echo signals returned by some of the targets suggested much larger objects than birds. Their confusion was resolved when it became apparent from comparisons of range data and concurrent signal strengths that the very strong signals were always associated with targets at close range. A radar echo


declines in strength proportionally to the fourth power of the distance of the target from the antenna, so that even a small target at unusually short range can produce a very strong signal. Also, the pulse power of the tracking radars was much greater than that of the more familiar search radars, and they were normally used to track relatively distant rockets. Consequently, their use in the unaccustomed search mode drew attention to the deceptively strong signals from very near targets.

No attempt had been made during the sightings to associate ranges and signal strengths. Had someone asked, "When you get an 80-dB signal, what range do you read?" the evening probably would have ended differently. Future radar operating procedures might very well provide that, when unidentified targets are causing concern, ranges and signal strengths be correlated. Apparently no formal procedure existed at the time of the sightings for use in identifying unusual radar targets such as insects, sidelobe echoes, anomalous echoes from object on the ground, etc. In the absence of such a procedure, the operators involved in this case handled the situation reasonably.


Some comments in a letter from Mr. Collis are particularly pertinent:

I think that the .... incident could be a landmark case in the whole area of UFO studies. It combines so many factors. Firstly, the incident involved a whole complex of associated events which were reported by the most respectable observers. It combined multiple radar and multiple optical sightings. It occurred very recently and a substantial amount of recorded, data is available-- i.e., the TPQ 18 radar records and the meteorological data. At least in part, the radar echo phenomena were repeatable and were observed by


design on subsequent occasions. It was sufficiently strange to cause interceptor aircraft to be sent off to investigate it in the heat of the moment, and also to cause the local and visiting experts considerable perplexity even in the cool light of day. We thus have a wonderful opportunity not only to study the physical nature of the incident but also to study the psychological implications of such incidents.

It would seem that most of the inexplicability of the events in this case (and possibly in many others) arises not from the facts themselves, (i.e., the specific sightings, etc., at any given instant) but in the interpretation made and significance attached to them when they were considered in inappropriate juxtapositions. The way in which this was done at the time under operational pressures and even subsequently provided, in my opinion, a most important object lesson.

It does indeed! The lesson is that the "flap" could have been avoided if the radar operators had been acquainted with the kinds of targets they might pick up in search mode, especially during anomalous atmospheric conditions. It is unlikely that such a "flap" will occur again at AFB A in such circumstances; but it can happen elsewhere unless this experience is communicated through appropriate operating procedures or in some other manner, to other operators of powerful tracking radars.