The 1991 Survey

Compiled by

Chris A. Rutkowski


Roy Bauer

Grant Cameron

Lorne Goldfader

Gordon Kijek

Clive Nadin

Christian Page

Michael Strainic

Guy Westcott

Bonnie Wheeler

Published by

Ufology Research of Manitoba

Box 1918

Winnipeg, Manitoba

Canada R3C 3R2

The 1991 Canadian UFO Survey

After a very positive response from the publication of both the 1989 and 1990 Canadian UFO Survey, it was decided to continue the systematic collection of raw UFO report data in Canada and prepare yearly reports for general circulation. It has always been felt that the dissemination of such data would be of great advantage to researchers, so it is presented here once again as data with some analysis.

As in 1989 and 1990, a letter requesting data for analysis was sent to all Canadian ufologists believed to be actively investigating reports. A notice was also placed in the Swamp Gas Journal. Names and addresses were obtained from UFOROM files, the mailing list of the Swamp Gas Journal and from ufologists in contact with UFOROM. In 1989, twenty-five such letters were sent out. In 1990, with additional names and addresses added to the list, over fifty letters were mailed. About the same number were mailed for the 1991 Survey. As in previous years, the response was poor; it is now suspected that only a small fraction of "active" ufologists actually investigate cases and maintain useable records.

In 1989, 141 UFO reports were obtained for analysis. In 1990, 194 reports were recorded. In 1991, 165 reports were received. These reports came from contributing investigators' files, press clippings and the files of the National Research Council of Canada. The NRC routinely receives UFO reports from private citizens and from RCMP, civic police and military personnel. The number of cases in 1991 represents a 15% decline from the previous year, but if we assume an average of 167 cases per year, the variation is uniform in either direction, and we can suggest that numbers of UFO reports in Canada remain at a constant level.

In 1991, there were apparent significant increases in the number of reports in Alberta, while there was an apparent decrease in reports in Manitoba and Saskatchewan. The rest of the Provinces appear to have had average numbers of reports in 1991.


1989  15 16 18 22 34 28 1 - 3 3 - 1
1990  76 9 10 20 21 36 7 3 5 4 1 2
1991 59 22 7 6 30 16 9 1 7 4 1 -

The monthly breakdowns of reports during each year show slightly different patterns from those of previous years. In 1989, there was a significant increase in UFO reports in the late fall, with other months maintaining what appeared to be a fairly constant "normal" level of reports. But 1990 saw two major increases in report numbers in two months: April and August. The "normal" level of monthly report numbers appeared to be constant in other months, with minor fluctuations. In 1991, reports peaked in August, but there was no single obvious trough, and there were an abnormally large number of reports in the winter months.


  J F M A M J J A S O N D
1989 13 6 6 9 5 9 5 5 12 32 27 9
1990 17 7 6 47 10 10 9 47 15 16 10 -
1991 13 7 17 12 7 12 16 25 16 12 11 17

An analysis by report type shows a similar breakdown to that found in previous years. The numbers of cases of a particular type remained roughly constant except for the category of Nocturnal Lights, which exhibited nearly a twofold increase in 1990. However, numbers of NLs were closer to a 1989 level in 1991.


1989  84 20 16 10 7 - 2 2    
1990  141 24 15 2 1 - 4 3    
1991 110 26 13 7 4 1 2   1 1

For those unfamiliar with the categories, a summary follows:

NL (Nocturnal Light) - light source in night sky

ND (Nocturnal Disc) - light source in night sky that appears to have a definite shape

DD (Daylight Disc) - unknown object observed during daytime hours

CE1 (Close Encounter of the First Kind) - ND or DD occurring within 200 metres of a witness

CE2 (Close Encounter of the Second Kind) - CE1 where physical effects left or noted

CE3 (Close Encounter of the Third Kind) - CE1 where figures/entities are encountered

CE4 (Close Encounter of the Fourth Kind) - an alleged "abduction" or "contact" experience

EV (Evidence) - a case where physical traces left by an event are the primary claim

RD (Radar) - UFOs observed on radar

PH (Photograph) - photographs of a UFO, but no actual sighting

The category of Nocturnal Disc was created by UFOROM for differentiation within its own report files. Similarly, Evidence is also an ad hoc creation, and may not be applicable by other researchers. Normally, Evidence would include such physical traces as "crop circles", "landing rings" and "saucer nests". However, in 1990 there was a great increase in the numbers of such traces discovered in North America, and it was decided to treat these as separate from UFO reports in these Surveys. [For the record, there were 27 "crop circles" and related traces discovered in Canada in 1990, and 39 in 1991. Many of these were investigated by UFO researchers, and a few were reported to the NRC. UFOROM is associated with the North American Institute for Crop Circle Research, which investigates such cases and publishes reports on its findings.]

The breakdown by evaluative conclusions for 1991 cases can be shown to be similar to results from previous years. There were three operative categories: Insufficient Information, Possible or Probable Explanation, and Unknown. Readers are warned that a classification of Unknown does not imply that an alien spacecraft was observed; no such interpretation can be made with certainty, based on the given data (though the probability of this scenario is admittedly never zero). In most cases, the evaluations are made subjectively by both the contributing investigators and the compiler of this report. The category of Unknown is adopted only if the contributed data or case report contains enough information that a conventional explanation cannot be satisfactorily proposed. This does not mean that the case will never be explained, but only that a viable explanation is not immediately obvious. In 1991, two cases were positively explained by investigators, and not included in the breakdown below.


  1989 1990 1991
  # % # % # %
Insufficient Information  74 52.5  90  46.4  80 49.1
Possible/Probable Explanation  47  33.3  78  40.2 69 42.3
Unknown  20  14.2  26  13.4 14 8.6

The hourly distribution of cases tended to follow the same pattern for 1991 as in previous years. There appears to be a continuous curve, with a peak near 2200 hours local and a trough around 1000 hours local.

The average number of witnesses per case went down from a value of 2.12/case in 1989 to 1.4/case in 1990, then up again to 1.91/case in 1991. It is not known what this may indicate. It is possible that there might have been a tendency for only one of a pair or group of witnesses to report an incident, and hence this value would wax or wane depending on the social factor. This may have been true in the NRC files, which may not reflect the total number of witnesses sharing a UFO experience. However, these figures show that a typical UFO experience has more than one witness, supporting the contention that UFO sightings represent observations of physical phenomena.

The category of Duration is interesting in that it represents the subjective length of time the UFO experience lasted. Naturally, these times are greatly suspect because it is known that people tend to misjudge the flow of time. However, some people can be good at estimating time, so this value has some meaning. Although an estimate of "one hour" may be in error by several minutes, it is unlikely that the correct value would be, for example, one minute (disregarding the claims of "missing time" during the abduction category of experiences). Furthermore, there have been cases when a UFO was observed and clocked accurately, so that we can be reasonably certain that UFO events can last considerable periods of time. The average duration of a sighting can be calculated as a summation of all given durations then divided by the number of cases with a stated duration. The resulting value for 1991 is about 12 minutes, down from 19 minutes in 1990. This surprisingly long duration is likely due to the large number of sightings lasting only a few seconds contrasted with a few that lasted several hours.

In cases where a colour of an object was reported in 1991, the most common colour was white (61 cases), followed distantly by green (18 cases). Other colours were also represented, although there is a noticeable change from previous years, when red or orange were dominant colours. Since most UFOs are nocturnal starlike objects, the abundance of white objects is not surprising. The green coloured objects were, in general, bolides, which were seen in significant numbers in 1991.

Summary of Results

As with previous annual Surveys, the 1991 Survey does not offer any positive proof of the physical reality of UFOs. However, it does show that some phenomenon which is called a UFO is continually being observed by witnesses. The typical UFO sighting is that of two people observing a moving, distant white or red light for a period of over 15 minutes. In most cases, the UFO is likely to be eventually identified as a conventional object such as an astronomical object. However, in a small percentage of cases, some UFOs do not appear to have an easy explanation and they may be given the label of "unknown".

What are these "unknowns"? An additional classification is useful to try and better understand this kind of report. In the gathering of data for the study, contributors were asked to give a value for their personal Evaluation of the reliability of the report. This value is noted as "E" in the case listing. This value gives the likelihood that the UFO experience "really" occurred as described by the witness. Granted, it is impossible for any investigator to judge this absolute value; often, a subjective value for two categories of "strangeness" and "probability" is assigned. The Evaluation value is another subjective value imposed by the investigator or compiler (or both) with a scale such that the low values represent cases with little information content and observers of limited observing abilities and the higher values represent those cases with excellent witnesses (pilots, police, etc.) and also are well-investigated. Naturally, cases with higher values are preferred.

In addition, as an experiment, 1991 cases were also coded according to the methods proposed by Jacques Vallee. In several of his books, Vallee adapted the Hynek UFO classification system to a wider range of experiences. Vallee created a 5 x 4 array of UFO experiences: ANomalies 1-5; FlyBys 1-5; MAneuvers 1-5; and Close Encounters 1-5. His classification is that distant sightings are better classified according to the behaviour of the UFO(s) rather than conditions such as daytime, nighttime, radar, etc. Each Canadian case was thus given a coding such as FB2, MA1, etc., and this is noted in the case listing.

Furthermore, Vallee suggests another method of weighting or assigning credibility (or, as described above, reliability) to each case. His classification, called the SVP Credibility Rating, or SVP (which some have chosen to call the "S'Il Vous Plait"), uses three categories: Source reliability (where the report comes from; original investigation, newspaper, rumour, etc.); site Visit (on-site investigation by researcher, casual observer, no one, etc.); and Possible explanations (data consistent with natural causes, generally consistent, not at all, etc.).

The inclusion of such parameters was an test to see how they could be adapted for use in the statistical treatment of UFO data. In general, it was observed that the Vallee system is more complex than needed for simple UFO data which are largely NLs. In these cases, it is easier to rely upon the original Hynek classification. However, in cases beyond mere NLs, the more specific and descriptive Vallee system can serve to define a UFO experience according to its extraordinary characteristics, and is more useful than the simpler Hynek system.

Cases were coded and entered into a WATFILE database on a common PC clone environment. The coding key is as follows:

Example: 9910115 1636 BC Fort Nelson FB1 1 0 1 Yellow 0.03 1 NL 8 P Daylight etc.

Field: 1 2 3 4 5 6 7 8 9 10 11 121314 15

Field 1 is the date, in YYYMMDD format (UFOROM is now coding to allow for the next millennium).

Field 2 is the time, according to the 24-hour clock, local time.

Field 3 is the Province.

Field 4 is the Geographical Location, or common map name.

Field 5 is the Vallee classification.

Fields 6, 7 and 8 are the Vallee SVP, as described in the text.

Field 9 is primary colour of the object(s).

Field 10 is the duration of the sighting, in minutes and seconds

(i.e. 2 minutes, 15 seconds is 002.15)

Field 11 is the number of witnesses.

Field 12 is the Hynek case type: NL, ND, DD, C1, C2, C3, C4, RD, EV, PH.

Field 13 is the subjective evaluation (reliability) of the case, as described in the text.

Field 14 is the Conclusion given to the case: I = Insufficient information for an assessment; P = Possible or probable explanation, given the facts; E = Explained; or U = Unknown or unexplained at the present time.

Field 15 contains any short comments that distinguish individual cases.

In the 1991 study, only 12 cases (7.3%) were high-reliability unknowns. This agrees well with the 1989 results (4.9%) and with the 1990 results (4.6%), but is slightly higher for reasons that could include sampling techniques and inconsistent subjective evaluations. The 1991 high-reliability cases were the following:

FB1 ND Case 9910319 Powell River, BC

MA1 NL Case 9910329 Cranbrook, BC

MA1 NL Case 9910329 Edmonton, Alberta

FB1 NL Case 9910330 Surrey, BC

MA1 ND Case 9910502 Iron Springs, Alberta

FB1 NL Case 9910506 Lawn, Newfoundland

MA4 C1 Case 9910700 White Rock, BC

MA1 ND Case 9910725 Carleton, Quebec

CE2 C2 Case 9910831 Granum, Alberta

MA1 ND Case 9910903 Beaverlodge, Alberta

MA1 NL Case 9911030 South Walshingham, ON

AN1 RD Case 9911228 Toronto, ON

The interpretation of this list is that these cases were among the most challenging of all the reports received in 1991. It should be noted that many UFO cases go unreported, and that there may be ten times as many UFO sightings that go unreported as those which get reported to public, private or military agencies. Furthermore, it should be noted that some cases with lower reliability ratings suffer only from incomplete investigations, and that they may well be more mysterious than those on the above list.

We have also learned that UFOs are constantly being reported at a rate of about ten per month across all of Canada, and one or two per month in most provinces. Witnesses range from farmhands to airline pilots and from teachers to police. Witnesses represent all age groups and racial origin. What is being observed? In most cases, only ordinary objects. However, this begs a question. If people are reporting things that can be explained, then the objects they observed were "really" there. Were the objects we can't identify "really" there as well? If so, what were they?

These are questions only continued and rational research can answer, and only if researchers have the support and encouragement of both scientists and the public.

Further Comments

It is most instructive to compare the UFOROM analyses with those of the National Sighting Research Center of New Jersey, headed by Paul Ferrughelli. The NSRC results have been reported in a series of publications, a recent one being the National Sighting Yearbook 1990. The NSRC collected UFO reports from newspaper clippings and UFO publications, and analyzed the raw UFO data. Because of the difference in data sources, a comparison with the UFOROM results will not be true. However, it is still interesting to compare the two studies.

The NSRC found a total of 195 UFO reports in 1990. This number is nearly identical with that of Canada for the same year. However, because of the larger population, it is likely that the USA had many, many more sightings that were never accessed through the NSRC's sampling technique.

The NSRC study revealed essentially a reversed monthly distribution for UFO reports compared with Canada. Whereas US sightings peaked in the winter and had a noticeable trough in the summer, the Canadian peaks are traditionally in the summer. However, the 1991 Canadian distribution was much more even, with peaks in both winter and summer. Grouping the two studies together yields a monthly distribution with troughs in mid-summer and mid-winter, with slight variations month-to-month. It is possible to speculate that with adequate report sampling, there would be no monthly variation in the number of sightings, except for major flaps which would be more noticeable in an international survey.

Like the Canadian study, the American data was unevenly distributed throughout the country, with most reports coming from just two states, Florida and Indiana. The Florida flap is likely due to the Gulf Breeze reports which receive a great deal of media attention. The distribution of sighting duration was nearly identical to the Canadian study. The average duration of a typical UFO sighting is about 15 minutes.

For the hourly distribution of UFO cases, the American study found a symmetrical distribution with a pronounced peak at 9 PM local time and a trough at around 9 AM local time. Canadian distributions are normally about one hour later in each peak, but are otherwise identical in distribution. Breakdown by Hynek classification yields identical distributions within both American and Canadian studies, with NLs being overwhelmingly predominant.

In summary, Ferrughelli's analyses of American UFO data yield results remarkably similar to the UFOROM Canadian studies, despite the differences in collection procedures. The most marked discrepancy between the two studies was in the monthly distribution of UFO reports. This was probably an artefact of the NSRC sampling technique, which did not involve solicitation of UFO reports from investigators but relied upon newspaper accounts for many of its cases. The two studies are complementary, and will aid further research into the UFO phenomenon.

[Reference: Ferrughelli, P. National Sighting Yearbook 1990. National Sighting Research Center, 60 Allen Drive, Wayne, NJ 07470.]