Captain Alejandro Franz  (third from left) was the first to publicly conclude the source of the UFOs (From Franz website)

Flying Oil wells?

by Tim Printy August 2004

Updated October 2004

Interesting discussions elsewhere

Several skeptical discussion groups however did begin to peck at the data available and noticed several glaring items being publicly ignored by the UFOlogists. Unlike the secretive UFOlogists on UFO Updates, these groups made it possible for everyone to see the full tape (thanks to Thomas Strunch on the James Randi Forum)! Most important is that an individual in the same forum by the name of Richard Gemmell (AKA "Wipeout") began to notice that the Infrared targets bearings were in the direction of the Gulf of Mexico near Campeche. Further efforts by other members of this forum began to pursue this line with one member by the name of "Mummymonkey" hitting on the possibility that they may be oil well platforms out at sea. Indeed, located in the gulf were numerous oil wells, which vent gases and burn them off. Gemmel started this line of reasoning as early as May 14th and appears to be the first person who suggested this connection. One could also suggest that Julio Herrera of Mexico's Autonomous University was hitting about this idea when he suggested they were electrical flashes that might have been created in conjunction with gases being released by oil well platforms at sea. He made his suggestion on May 13th. Either way, it can be accurately stated that several people were zeroing in on the potential explanation as early as mid-May. How long would it be before UFOlogists made the same connection?

UFOlogists scoff at a likely theory

In late May, Captain Alejandro Franz broke the news on the UFO updates list that these oil wells could be the source of the images. Captain Franz had previously exposed another Maussan event, a collision of an airplane with a "UFO", as a hoax. While Franz had a few errors in his evaluation, the concept was there for other UFOlogists to pursue. Unfortunately, such a prosaic explanation was met with the usual closed-minded derision that UFOlogists often suggest skeptics employ. Dr. Maccabee, who was apparently involved in an in-depth analysis of the video, suggested that the clouds and angle of elevation (2-3 degrees) demonstrated the lights were in front of the clouds and above the horizon. Ray Stanford stated the flames would be evident in the video images as "flickering". He also erroneously pointed out that the radar would not track oil well fires. He did not seem to yet fathom that the radar contacts were independent of the FLIR images despite this being clear in several discussions. Still other UFOlogists referred to Captain Franz suggestion that this was "pelicanism"/"debunking". Richard Hall, who had expressed some skepticism on the case, now began to take up the "pelicanism/debunking" chant:

I have some problems with Maussan's activities and reputation, but in this case the objective evidence trumps all those pre- emptive debunking comments. At an absolute minimum, we have a very interesting, well- documented event that any scientist worth his/her salt should be intensely curious about and willing to investigate objectively, not reflexively debunk it. (Hall)

Hall seemed unwilling to accept any prosaic explanation unless it comes from some form of respected UFOlogists like Dr. Bruce Maccabee. Any explanation offered by skeptics or amateurs appear to be rejected immediately.

The tape was analyzed by Dr. Maccabee but he noticed difficulties because of several problems with the way the data was being recorded. Noticable in the video is that the camera's azimuth readings tended not to rapidly respond to the camera's motion introducing some minor errors when trying to make precise measurements. This could have been due to the azimuth recording mechanism or minor changes in the aircraft's heading introducing minor errors. Dr. Maccabee also noted that the camera's elevation angle was not accurate either and recorded several objects at angles lower than they should have been.

Franz did not retreat from his position. He would acquire the flight data and azimuth readings to produce a very convincing map of the event that demonstrated that the FLIR was pointed towards the oil well fields. He published this on a second website:


It is indeed interesting that all sorts of possible explanations were mentioned but the oil well idea was not even suggested on the UFO updates forum before Captain Franz.

Arguments against oil well fires

UFOlogists most significant arguments against the oil well fire theory are:

With regard to the flickering/shape comment, this seems obvious. Captain Franz's maps indicate that the oil wells were roughly 100 miles distant. At this distance, a flame 25 feet across would only subtend an angle of about 10 seconds of arc across (0.003 degrees). Add to this that the camera had to look through 100 miles of atmosphere and it seems unlikely that the FLIR could resolve such a target. Close examination of the video shows that there does appear to be a change in intensity on several of the targets indicating a "flickering" one might expect from a fire. This argument appears to be without merit.

Bruce Maccabee was the one who suggested the objects were visible in front of clouds:

However, the most convincing visual evidence that the objects are not at ground level appears in the video around 17:06:43-45 at which time the lights/object emerge from a cloud and, behind them is a dark area (cloud shaded from the sun). After they, one after another, pass the dark cloud they also pass in front of a bright cloud (lit by the sun). (Maccabee Re: UFOs)

Looking at the video, one can see the objects appear to pass in front of clouds but they also dim as they pass through this area.

Frames showing the lights interaction with clouds. Frame grabs using "Snatch-it" and video taken from video located at Thomas Strunch's webpage

The clouds described are not very thick and a likely explanation is that lights were visible through thin, wispy clouds and not in front of thick clouds. The lights tend to brighten as they exit the thin bright cloud described. Bruce Maccabee would eventually drop this argument:

You can bet I looked and looked for a provable situation in which the lights were in front of a cloud. Unfortuntely, I could find none. I would agree that there were a couple ambiguous portions of the video which I initially thought were evidence of the lghts being closer than a cloud. But then upon further study I realized that I couldn't prove it because clouds are not opaque when thin. A cloud can reflect sunlight and appear "thick" while actually being quite "thin". On the other hand, the dimming of the lights by intervening clouds occurred numerous times and triangulation indicated distances to the lights that were much greater than distances to the clouds. (Maccabee Re: Mexican)

While this argument has some merit, there is a likely explanation.

The range of the FLIR was brought up in several discussions in various forums. It was questioned if it was possible that the FLIR could detect the oil well fires from a distance of over 100 miles. There is no doubt that the visual horizon from 10,000 feet is over 140 nautical miles meaning the oil wells were not beyond the optical horizon. Additionally, if one examines the video, one can see that the FLIR was seeing very distant objects throughout the video. Clouds, rivers, shorelines, and even the moon, which is about 250,000 miles away, are visible. I had originally thought that it was possible the CCD chip was recording visible wavelengths as well as the infrared. Modern digital cameras employ an infrared filter to screen out this wavelength of light. However, Thomas Strunch stated this was not the case based on his discussions with a FLIR expert. In either case, these "cold" targets are visible from great distances in the video. Therefore, it is highly likely that high temperature oil well flames should have been easily recorded from this distance.

Brad Sparks is the only person I know that openly suggested that the camera is pointing in the wrong direction. He stated in a 3 August posting on UFO updates, that the oil well fires were at 290-300 degrees true azimuth between times 1652 and 1709 and the camera was pointed at 310 azimuth. Checks of the aircraft's heading at times 17:07 showed the aircraft moving at bearing of about 82 degrees. The bearing for the camera was centered around -139 degrees. If one does the basic math, we determine the camera was pointed towards -57 degrees from true north (303 azimuth). Looking at geographic coordinates of the aircraft and the oil wells using this webpage, I determined the angle to the key oil wells (AKAL-C/AKAL-J/NOHOCH-A) was about 304 degrees azimuth at time 17:07 (because of the motion of the aircraft, the bearings towards these oil wells would have been at an azimuth of greater than 304 prior to time 17:07). This is a pretty good match and seems to indicate the camera WAS pointed in the right direction. Sparks did not explain how he arrived at his bearings of 290-300 but they appear to be incorrect.

The angle of elevation argument seemed to be the best one to downplay the idea that the images were of oil well flames. Brad Sparks suggests that it is impossible for the images to be oil well flames because of this:

The moon was at +1.3 degrees elevation at 5:19:32 PM (CST) when the FLIR showed it too low at -1 degs, a calibration error of about 2.3 degrees that in fact only makes the problems _worse_ for the oil well theory. The FLIR obviously was set to register elevations slightly _too low_ so that means that instead of rescuing the flying oil wells it makes the _actual_ elevation angles of the IR "lights" _higher_ by 2.3 degs. Subsequently Bruce Maccabee calculated the sun position for a solar reflection at 4:46:59.5 PM that also verifies that the FLIR elevations were reading about 2 degrees too low in elevation angle (-20 degs instead of the correct and higher -18 degs elevation).... (Sparks Re:Oil)

When I examined the solar reflection image, I computed a value of -2.5 degrees, which is close to Sparks -2.3 degrees for the moon. This gives the impression that the -2.3 degree correction needs to be equally applied across the entire range of azimuth. However, this is not how a calibration error would be applied. The instrument's calibration would be based on setting the camera's horizon. The instrument would be made level at a specific azimuth (probably 0 degrees) and the elevation calibrated to zero. This would establish the instrument's horizon. If the instrument's horizon were calibrated incorrectly such that it was off by -2 degrees, this would change as the azimuth changed towards the zero point where the calibrated horizon crosses the actual horizon. This probably occurred at the +/-90 degree mark (depending on where the initial calibration had been done). As one neared this azimuth the value of correction would become a less negative and approach zero. As one moved past this mark (towards the point 180 degrees away from the calibration point), it would become a positive value and make the FLIR indicate HIGHER than actual. To apply the -2.3 degree uniformly across the entire azimuth means the FLIR would never attain a value of -90 degrees in elevation, which does not conform with a standard instrument calibration procedure. If the -2 degree error is due to incorrect calibration, one can state that a negative error towards the front of the plane would be positive towards the rear of the aircraft. Contrary to what Sparks states, this positive error would support the oil well flames hypothesis and not make it worse.

There may be another reason for the angle error instead of faulty calibration. It may be the case that the camera is calibrated in relation to the aircraft's horizon and not the actual horizon. If the plane's angle of attack (AOA) were slightly positive, then the camera's horizon (0 deg elevation) would be higher than the actual horizon. As the camera rotated around in azimuth towards the +/- 90 mark, the correction would approach zero. As the camera rotated past +/- 90 degrees, the camera's horizon would now dip below the actual horizon. This means the horizon would now appear as a positive elevation vice negative.

How the aircraft's angle of attack could affect the elevation angle in relation to the actual horizon.

Fortunately, the sun reflected at an azimuth around -45 degrees. The objects in the FLIR video are centered near -135 degrees or just about 90 degrees away. This means that this was equidistant from the zero mark at 90 degrees and the correction can be applied with an opposite sign to get an approximate correction for the FLIR images of the objects. That is, the objects being shown 1-2 degrees in elevation are actually below or near the horizon depending on the magnitude of the correction.

When Dr. Claude Poher also suggested this in his paper about the images, Sparks fumed how this was impossible:

Poher desperately and ridiculously tries to argue that the slight 2-deg error in Elevation is a mechanical error that will be highest in the direction of the Moon and lowest in the opposite direction from the Moon, without ever considering a digital electronic error that does not change with direction of pointing the FLIR camera. And he deliberately disregards the Sun position data in Maccabee's report as that would undercut his baseless theory by showing the correction factor for Elevation does not change depending on what direction the camera was pointing at. (Sparks Re: Dr. Poher)

Sparks suggests the error is digital in nature based on two readings with the camera pointing towards the front of the plane. It is not based on any measurements to the rear of the plane. Without the actual calibration procedure for the FLIR, it is difficult to make any blanket statements concerning calibration errors. However, based on my understanding of how most encoders work, my evaluation seems more correct.

Dr. Bruce Maccabee also has a problem with angles of elevation and amplified on this in his work, "A study of the March 5, 2004 Radar and FLIR sightings during a surveillance flight by the Mexican Department of Defense". He points to the images of the "twins" visible in the video showing a three degree angle of elevation. Dr. Maccabee correctly points out that the angle shown is 1-2 degrees higher than the actual elevation but states this is not enough to account for the three degree indication. As a result, he concludes that the "twins" were above the optical horizon. However, he is not really being honest about this value. At time 17:03:44-53, the angle of elevation is three degrees but before and after this, the elevation indication is two degrees. Additionally, the elevation angle changes to one degree at time 17:03:59! Later observations of these lights at time 17:05:19 put the "twins" again at one degree elevation. The "twins" do not change their relative distance from the cloud tops in these images, so it appears that the plane is causing the change in elevation. Examining the entire tape indicates that the high value Dr. Maccabee cherry-picked had more to do with the plane banking or changing it's angle of attack than the actual elevation being this high. The true angle of elevation for the "twins" appears to be more like 1-2 degrees, which is consistent with most of the elevation angles observed for the lights during this time period. This places the lights at the horizon or just below it, which is consistent with the angle one might expect from oil well flames at a distance.

Santiago Yturria takes another route concerning the angle of elevation and suggests that the error associated with the elevation angles does not explain the values displayed:

The oil flames speculation is based on the alleged FLIR disadjustments and a wrong calibration as well as the Radar malfunctioning implying the FLIR camera was pointing to the ground level all the time without moving up and down by the operator during the incident even that the elevation levels were reading in the FLIR screen counters... (Yturria Re: 29 7 Oct)

Yturria expounded that the elevation angle is off but is indicating lower than it should when looking to the side of the aircraft:

Assuming you studied the FLIR footage as you claim, you may have noticed the C26A left propeller at 17:16:33. The FLIR camera is located underneath the aiplane. Do you know what would have the camera's elevation angle been in order to film the plane's left propeller along with the luminous object? According to logic it must have been pointing upwards is'nt? So at 10,500 fts. and the FLIR camera pointing upwards what do we got? I hope you can comment on this. (Yturria Re: 29 7 Oct)

In this clip, the left propeller is visible and the angle of elevation is -1 to -2 degrees at an azimuth of about -97 degrees. However, Yturria does not explain what he concludes from this. I believe he is trying to conclude that the elevation angle was indicating much lower than it actually was. However, on the Merlin, the propellors extend well below the plane of the FLIR as shown by this image:

Frame grabs using "Snatch-it" and video taken from video located at Thomas Strunch's webpage and located at

We can not determine what part of the blade is being imaged. Although it looks like that the upper half might be the culprit in the frame above, this is probably not the case. The motion of the blades and depth of field focus can produce some misleading images. Additonally, one would expect to see various parts of the wing or engine nacelle in various parts of the video where the camera pans backwards. Instead, all that appears to be visible is the blur of the propeller blades, which tends to indicate that one is looking at the lower half. We also do not know if the plane is flying straight and level. Any changes in the aircraft's pitch, yaw or roll would make for more problems with the elevation angle. Considering the lower half of the blades extend below the FLIR's horizon, it seems possible that the blur is from this section of the propeller and, as a result, a value of near zero seems perfectly in line with the pitch of the aircraft introducing these errors.

All these arguments being presented are not adequate to falsify the oil well fire hypothesis. Do the tapes support the hypothesis?

Works Cited

Franz, Alejandro. Mexican Air Force FLIR images are oil well flames! Available WWW:

Maccabee, Bruce. "Re: UFOs or simply oil well flames?" 27 May 2004.UFO Updates Mailing list. On line posting. Available WWW:

-. "Re: Mexican FLIR video and story for sale" 20 September 2004. UFO Updates Mailing list. On line posting. Available WWW:

Hall, Richard. "Re: UFOs or simply oil well flames?" 29 May 2004.UFO Updates Mailing list. On line posting. Available WWW:

Sparks, Brad."Re: Oil well flames objectively investigated" 17 June 2004.UFO Updates Mailing list. On line posting. Available WWW:

-."Re: Dr. Poher's Mexican Airforce UFO video" 3 August 2004.UFO Updates Mailing list. On line posting. Available WWW:

Yturria, Santiago. "Re: 29 photos confirming oil flame identification" 5 October 2004.UFO Updates Mailing list. On line posting. Available WWW:

-. "Re: 29 photos confirming oil flame identification" 7 October 2004.UFO Updates Mailing list. On line posting. Available WWW:


NEXT: Is it possible?

Back to My skeptical opinion about UFOs