Date: Thu, 19 Nov 2009 13:59:33 +0000
From: Martin Shough <>
Subject: USCGC Sebago Nov 1957

DATE: November 5 1957   TIME: 0510 local       CLASS: R/V
shipboard radar/deck visual

LOCATION:           SOURCES: Thayer, Condon 165
Gulf of Mexico         Mebane (app. Michel FSSLM '58 242)
                                Lorenzen SEIOS 1966 101

RADAR DURATION: 27 mins. (intermittent)

EVALUATION: Blue Book - Venus/aircraft
                           Thayer - aircraft/AP/meteor

PRECIS: At 0510 the U.S. Coast Guard Cutter Sebago was in the
Gulf of Mexico some 200 miles S of the Mississippi delta at 25
degrees 47' N, 89 degrees 24' W on a heading of 23 degrees true,
when a radar target was acquired on a 290 degree true azimuth at
a range of 14 miles, heading S. The target turned in towards the
ship, closing to 2 miles, then returned N along the ship's port
side. The target was lost at 0514, and average speed was
calculated as 250 mph.

  Two minutes later at 0516 a second target was picked up,
bearing 188 degrees, range 22 miles, and was plotted to a
position 190 degrees, 55 miles, where it was lost. Departure
speed was measured at 650 mph.

  At 0520 a stationary target was displayed at 350 degrees, 7
miles range. At 0521 a visual object "like a brilliant planet"
was observed from the deck for about 5 seconds travelling S-N at
31 degrees elevation between 270 & 310 degrees azimuth. At about
the same time the radar target moved slowly NE, and finally
accelerated rapidly, moving off the scope at 0537, 15 degrees,
175 miles range.

NOTES: The most reliable published source is Thayer, drawing
directly on the Blue Book case file, and the above precis
reflects this. The 1958 account by Mebane appears to contain some
errors of timing. Certain circumstantial details in Mebane are of
interest, however, drawn from a press report and a radio
interview broadcast by WBZ, Boston, on the following day.

  The initial explanations offered for the Sebago report in USAF
press releases on November 15 & 17 were the planet Venus, two
unspecified aircraft - a piston-engine light plane and a jet -
plus some possible "false" targets. According to Thayer, the
first radar target "behaved generally like an aircraft", and he
inclines to accept the Air Force view that it probably was an
aircraft, possibly from Eglin AFB to the N. No specific flight
could be identified as the culprit, however. The second target,
which Blue Book apparently concluded was probably a jet, is
attributed by Thayer to anomalous propagation. Thayer also
explains the third target as AP, and the concurrent visual as
"undoubtedly a meteor".

  Whilst these coincidences may on the face of it seem
improbable, and the whole melange a little desperate, Thayer
points out that radio refractivity data for Key West, Florida
(the nearest applicable soundings) show the possibility of
unusual propagation conditions, with marked temperature/humidity
stratification conducive to partial reflection echoes. Further,
he argues that since the moving visual object appears to have
been seen at a time when radar showed a  stationary target, and
apparently at a different azimuth, the two events cannot be
related, and the brevity of the visual sighting is suggestive of
a meteor.

   It is only prudent to point to some reservations about this
scenario, however.

  An alternative and quite plausible explanation of the visual
sighting (in fact proposed by Tulane University astronomer Dr. J.
F. Thompson in an interview for the New Orleans Times-Picayune as
early as November 6 1957) is Sputnick 2. The second Soviet
satellite had been launched on November 3 into an inclined
65-degree elliptical orbit with a perigee of 140 miles, and
according to Thompson would have been visible west of the
Sebago's position travelling roughly NNW at about that time.

  Thayer relates from the report that "the third radar target
remained stationary for about 1 min." before moving off to the
NE. It is implied that this minute elapsed after the visual
object had been observed at 0521, and thus the target movement
would have begun not earlier than 0522. But the statement is
ambiguous. The stationary target was acquired at 0520; if it then
"remained stationary for about 1 minute" its movement would be
consistent both in time and approximate heading with the moving
object observed visually. This is a moot point, but it might also
be noted that times are only given to the minute, and the
difference between approximately 0521 and approximately 0522
could only be seconds. Therefore, given that the visual
observation was made by members of the crew other than the radar
operator (four hands who had gone up to the bridge to look for
the object); given the possibility of small errors in independent
timing; given that times are only cited to the minute; given that
the precision of a phrase like "about one minute" is a debateable
estimate of elapsed time; and given that the material discrepancy
is only about a minute - then to conclude that the visual and
radar movements were definitely inconsistent as to time may be to
expect too much of the information available.

  The visually observed heading S-N at 31 degrees above the
horizon might, as we have said, be consistent with the radar
heading SW-NE. However, the visual report has the object moving
from 270 degrees to 310 degrees, whereas radar reports the target
moving from 350 degrees to 015 degrees. How accurate are these
values, and how significant is the discrepancy?

  The radar bearings are given "true" - that is, in relation to
true longitudinal north. These true values can be read off
directly from the bearing ring if the heading marker - a bright
scope trace - is aligned to the ship's true course (so-called
"north-up" presentation). With the heading marker aligned to 0
degrees ("heading-up"), indicated bearings will be relative to
the ship and would have to be corrected by the operator.
Furthermore it is essential that the heading marker is
continually stabilised to the true course if true readings are to
remain true. This can either be done manually by using a
picture-rotate control, or automatically by a linkage with the
ship's gyro compass; but even if automatically stabilised it is
advisable to check picture orientation from time to time, since
the compass will only correct it relative to its initial setting.
With the heading marker switched off an operator might
conceivably become confused, momentarily, as to whether displayed
bearings were true or relative. And on a manually aligned scope
an excited operator might neglect to correct bearings for yaw.
Therefore there are possible sources of error. However, given
that this was a Coast Guard vessel and that the operator would
presumably be well-trained, there is no reason not to assume that
the marker was correctly set on an automatically stabilised
display. The bearings cited are probably accurate.

   But the possibility exists that the visual bearings are
translated from positions off the bow, which is quite common
practice at sea. The figure of 270 degrees might be taken as
supporting this guess, since this would correspond to exactly 90
degrees to port and is the kind of "cardinal point" approximation
that might well be given by a visual observer recalling a
fleeting observation off the port beam and offering a rough guess
as to the start-point of a trajectory which was initially seen
out of the corner of his eye. If this were the case, then the
quoted values would have to be increased by the 23-degree heading
of the ship to give true azimuths, yielding bearings of 293
degrees and 333 degrees true. Some allowance might also be made
here for what is, ex hypothesi, an approximation of the bearing
angles of a transient light seen in the pre-dawn dark, and it is
well known that even experienced observers can be quite
inaccurate in estimating visual angles, even in relaxed
conditions. Perhaps the most that ought to be said is that a
light was noticed heading approximately N somewhere off the port
(W) bow, and it is far from certain that this is inconsistent
with the initial position of the radar target some 33 degrees off
the port bow.

  There is a fairly important inconsistency in reported speeds,
however. According to the radar report, the target moved off
relatively slowly from a range of 7 miles, only accelerating
rapidly towards the end of its track some minutes later. But the
visual observers estimated that the light travelled 40 degrees in
five seconds: at a constant range of 7 miles, an angular rate of
8 degrees a second translates to a speed of 3600 mph. And this
figure does not allow for the significant radial vector of the
target, which - if light and target were one and the same - would
drive the true speed very much higher still. We will show later
that the mean speed of the radar target was about 650 mph, which,
given that it accelerated from zero to its maximum speed shortly
before going off-scope, demands that its initial "slow" speed was
significantly lower than this figure. Thus, radar-visual
consistency would demand that the visual witnesses made at least
a factor ten error in their estimate of angular rate.

  As regards the hypothesis of partial radar reflection from
inversion strata, which Thayer suggests to explain the second and
third targets, it should be noted that the refractivity data
quoted were indeed, as he concedes, "taken at some distance from
the ship's position" - in fact, some 400 miles from the ship's
position, and even given that subtropical atmospheric patterns of
this sort "tend to extend in rather homogeneous form over large
horizontal distances", one has to admit that there is a good deal
of speculation here. Furthermore, the refractivity profile on
which Thayer concentrates as being especially likely to generate
strong partial reflections was taken at 1800 CST on the following
evening; the stratification of the more relevant 0600 CST Key
West profile, whilst still significant, is not nearly so marked.

  Thayer adduces support for the AP hypothesis from the fact
that the two latter targets appeared "suddenly" on the radar,
well inside its maximum range (at 22 miles and 7 miles
respectively), suggesting thereby that these were probably
phantom echoes. However, all three targets appeared well within
range, including the first (at 14 miles) which Thayer
nevertheless believes was probably an aircraft. This behaviour -
which can be explained in terms of targets entering the top edge
(this being a marine radar) of the radiation pattern - cannot
therefore be held to be uniquely diagnostic of AP. His assertion
that the targets "were, with the possible exception of the first
one, erratic and unpredictable in their movements" finds no clear
basis in the report; the latter two targets  moved on roughly
constant headings, the third being tracked NNE to the maximum
range of the display (175 miles), whereas the first - the
"aircraft" - meandered south, then east, and then roughly north.
The latter two targets displayed high speeds during departure,
and one appeared to accelerate from a standing start, but it is
mischievous to describe such movements as "erratic and
unpredictable", and the first target was at least no less

  The behaviour of the two later targets is interesting in the
context of partial reflection echoes, which tend to move at twice
the wind-speed at the layer, generally with the wind or at an
acute (<90 degree) angle to the wind. At least, the direction of
movement will have some component vector related to the heading
of the wind. The headings of the radar targets can be
reconstructed from the range and bearing data given: In the case
of the second target, its heading was 192 degrees, or about SSW;
the heading of the third target detected a couple of minutes
later was diametrically opposite, 17 degrees or about NNE.
Further, the measured speed of the second target was 650 mph,
which would equate to winds of over 300 mph, and whilst this
might be dismissed as a misreading of sporadic echoes on a
relatively fleeting track the third target was painted in
movement for some 15 or 16 minutes, which yields an average speed
over the 170 mile track of, coincidentally, about 650 mph.
Finally, the third target maintained station for one or two
minutes before moving off and accelerating, which is not easy to
explain as an effect of strong winds driving waves on the surface
of an inversion layer.

  To summarise so far, some elements of the AP explanation are
questionable, and it is not proven that the visual sighting was
unrelated to the approximately concurrent radar target although
they are markedly inconsistent in terms of estimated speeds.

  The visual sighting could have been a meteor or Sputnick 2,
and the first radar track apparently did not display any
characteristics which could not equate with an aircraft, even
though no responsible aircraft could be identified. The radar
targets could possibly be explained as noise tracks, although
more information about scope presentation and movement would be
desirable and the duration of track 3 is possibly excessive.
Track 2 could possibly have been a jet flying at high altitude,
entering and leaving the top edge of the radiation pattern (a low
altitude jet would probably be displayed for longer by a radar
designed pricipally to detect shipping and coastal features).
Track 3, with a stationary episode, was almost certainly not a
jet as this would require a steep climb or dive on a radial
heading which preserved constant slant range and azimuth for as
long as 2 minutes, a highly improbable circumstance.
(Multiple-trip echoes from a jet beyond the unambiguous range
could give a spuriously high ratio of displayed minimum:maximum
speeds if, for example, it turned from a tangential onto a radial
heading. Any non-radial motion would be slower than true, since
the angular rate is preserved but at spuriously short displayed
range. However, given an angular rate which was imperceptible on
the scope for one or two minutes at typical mid-'fifties US
marine radar scan-rates of 15-20 rpm, yielding at least 15
consecutive paints, a jet-speed target would have to be at such
an immense range that one doubts if it could possibly return a
detectable signal on such a navigation radar, of low power
typically on the order of 30-40 kW or less. If one supposes a
target with a proportionately immense radar crosssection - say a
flight of several large, well-aspected military transports or
bombers, integrated below the resolution of the display - then an
inconsistency emerges with the displayed speed of 650 mph. The
actual true speed of such a  multiple-trip target with a
non-radial vector would be significantly in excess of this

  The speed of track 3 rules out a helicopter, and no VTOL jets
were flying in 1957. Birds, insects, meteor-wake ionisation, CAT,
balloons or other windborne objects are inappropriate.
Multiple-trip echoes of Sputnick 2 at a true slant range of (at
least) several hundred miles are highly unlikely on a low-power
marine radar, and anyway could not explain the extended
stationarity and subsequent extreme acceleration of the target.
Multiple reflection "ghosts" offer no useful explanation of an
echo which was stationary for up to 2 minutes and then
accelerated through Mach 1 with a 25-degree change in azimuth:
the possible reflection geometries of "ghosts" are complex, but
suffice it to say that this behaviour requires, amongst other
conditions, at least one moving aerial reflector (i.e., aircraft)
in the radiation pattern, which as the primary reflector would
present a stronger echo than the "UFO" blip itself and would
appear to be "shadowing" the UFO at slower speed closer to the
ship. No such target was reported. Whatever secondary reflector
we might hypothesise (another aircraft, ship etc.), it is highly
improbable that this kind of reflection geometry could be
maintained for upwards of 15 minutes. And finally, the stationary
episode cannot be explained by the same "ghost" echo, requiring
either a quite different primary reflector or another explanation

  As a postscript, it is worth adding that another incident
involving the ship SS Hampton Roads took place that evening not
far (about 180 miles) from the location of the Sebago incident.
The ship was at 27 degrees 50' N, 91 degrees 12' W when a "round
glowing object" was sighted at apparently high altitude at 1740
LST. It was observed for 10 minutes, and was lost to sight as
dusk drew on at 1750. This object was explained by Blue Book as a
probable balloon drifting with the upper winds, which is
certainly plausible. In the absence of more detail, it is also
possible that this was another sighting of Sputnick 2,
disappearing as it moved into the earth's shadow.

  In conclusion, although a conventional explanation of the
Sebago sightings might require a series of coincidences, and
although some elements of that explanation remain open to
question, nevertheless the visual sighting in particular is of
low strangeness, and its correlation with the radar track is
somewhat doubtful. Considered alone the radar data available,
whilst interesting, cannot be said to strongly support an
unconventional interpretation.

STATUS: Insufficient information