2013 Aguadilla Puerto Rico UAP Final Draft v5s

2013 Aguadilla Puerto Rico
The detailed analysis of an Unidentified Anomalous Phenomenon

captured by the Department of Homeland Security.
This report is a detailed analysis of a Homeland Security thermal video taken from an aircraft as
it tracked an unidentified object. What you will see in the infra-red is an object that seems
capable of traveling at night without lights, at times below tree-top altitude, at speeds
approaching 100 mph, and apparently without risk of impacting objects as it passes by.
The report was written under the auspices of the Scientific Coalition of UFOlogy (SCU). The
SCU is a think tank of scientists and researchers stretching across organizations, governments
and industries to scientifically and publicly explore unknown anomalous phenomenon known
around the world as Unidentified Flying Objects (UFOs), Unidentified Aerial Phenomenon
(UAPs), the French, Spanish, and Italian equivalent to UFOs (OVNIs), and Unidentified
Submersed Objects (USOs).
Scientific Coalition for UFOlogy is open to all scientific based analyses of this report and are
willing to provide all the information that we have on this phenomenon to any other serious
researchers.
TABLE OF CONTENTS
SECTION
PAGE
EXECUTIVE SUMMARY.......................................................................................................... 1-4

PURPOSE........................................................................................................................................ 4
BACKGROUND.......................................................................................................................... 4-7
ANALYSIS AND CHARACTERIZATION.............................................................................. 7-42
DISCUSSION...........................................................................................................................42-47
ACKNOWLEDGMENTS............................................................................................................. 47
APPENDICES
Appendix A Author Acknowledgements................................................................... 49-51
Appendix B WESCAM Model MX-15D.................................................................. 52-54
Appendix C FOIA Requests and Replies.................................................................. 55-64
Appendix D Delay of Fed Ex Flight 58.....................................................................65-67
Appendix E Auxiliary Witness Communications...................................................... 68-73
Appendix F Analysis of Radar Information.............................................................. 74-87
Appendix G Object Location, Seed, Size.................................................................. 88-98
Appendix H Modeling of Object............................................................................. 99-118
Appendix I Alternate Speed Calculation Using Background Objects...................119-123
Appendix J Water Transit...................................................................................... 124-138
Appendix K Estimated UAP Temperature.............................................................139-152
Appendix L Line-of-Sight Evaluation................................................................... 153-159
2013 Aguadila Puerto Rico UAP
April 25 2013
Executive Summary
On April 25, 2013, at about 9:20 pm local time, an unknown object at low altitude flew
directly across the Rafael Hernandez airport runway at Aguadilla, Puerto Rico, causing a
delayed departure of a commercial aircraft. There was no squawking transponder signal to
alert the aircraft tower, nor was there any communication with the tower to prevent a
dangerous situation with departing and arriving aircraft. Fortunately an airborne U.S.
Customs and Border Protection aircraft captured the object on infrared video. This report is
an analytical evaluation of that video as well as witness statements and radar data of the area.
An original copy of a thermal video was obtained from an official source on October 20 th of
2013. The source of this video evidence was vetted and identified. The source wishes to
remain completely anonymous to ensure no issues arise with the source's employers. The
individual's occupation, address, and background history were verified by the authors of this
report as legitimate. Extensive efforts were made to ensure that this video did not contain any
classified information and none was found. The three minute video detailed the flight of an
unknown object that crossed into northwestern Puerto Rico from the Atlantic Ocean,
traversed the Rafael Hernandez airport airspace two times, and returned into the Atlantic
Ocean where it appeared to repetitively submerge.
The thermal video imaging system is a standard reconnaissance video system typically used
in military, law enforcement and civilian applications. The thermal video was taken from a
De Havilland Canada (DHC)-8 DHC-8 Turboprop aircraft that was controlled by the U.S.
Customs and Border Protection (CBP). The authenticity of the video used in this report was
corroborated using radar data obtained from the U.S. Air Force (USAF) 84th RADar
Evaluation Squadron (RADES) group. The radar data displayed the tracking aircraft that took
the thermal video. All times and
locations of the tracking aircraft
were consistent with the thermal
imaging video on screen data and
the USAF 84th RADES group
radar information. Details are
available on pages 10-12 of this
report.
An in-person interview with the
source indicated that the pilots of
the DHC-8 Turboprop took off on DHC-8 Turboprop.Courtesy of Homeland Security
a routine mission and as they
veered to the northwest saw a pinkish to reddish light over the ocean that was in their vicinity
and approaching toward the south. Concerned that the control tower had not alerted them to
incoming traffic they contacted the tower. The tower confirmed that they had a visual
sighting of the light but did not know its identity. According to the source, once the object
came close to shore, the light on the object went out. At about that same time the thermal
imaging system was engaged to follow the object.
April 25 2013
Thermal Imaging Video

Analysis of the thermal imaging video revealed irregular characteristics of an unknown
object not similar to any known natural or man-made objects. The video was split into 7027
individual frames so that careful analysis could be made. Evaluation of the video frame by
frame enabled detailed characterizations of the object. The objects size, speed, location,
infrared (IR) emissions, directional movement, and other properties were compared against
all known possible explanations including state-of-the-art drone capabilities as well as the
possibility of a hoax.
Radar Data
In order to support the validity of the thermal video and to look for unknown targets in the
area, a Freedom of Information Act (FOIA) request was made to the USAF 84 th RADES
group to obtain all FAA originated radar in the area during the time frame in question. This
request was granted. A second FOIA request for radar data from military radar sources in the
area was denied. Using the radar data that was provided, it was possible to validate that the
times and locations displayed on the thermal imaging equipment of the CBP aircraft matched
a government aircraft that was detected by radar. Additionally, unknown target(s) over the
ocean and two to three miles to the north and northwest of the Rafael Hernandez airport were
detected. None of these targets had transponders. This information supports the witnesss
claim that the Border Protection aircraft and the control tower sighted an unknown aircraft
moving from the ocean to the south.
Size, Speed, and Location
The size and speed of the object were determined at points in the video when the locations of
the object could be accurately determined. This allowed for an exact calculation of distance
and angular size of the object. With that information, basic trigonometry was used to
calculate the objects size. The object was between three to five feet in length and its speed
varied between approximately 40 mph to 120 mph. Its median speed was roughly 80 mph.
One of the object's flight characteristics the authors found to be significant was the objects
speed through the water which did not vary as it impacted the water. Its speed through the
water reached a high of 95 mph and averaged 82.8 mph. Details as to how these parameters
were calculated are on pages 16-24.
Interaction with Water
There was very limited interaction with the water, visible within the infrared video, when the
object impacted the ocean. Its speed immediately prior to impact was 109.7 mph. Frame by
frame analysis indicated that there might be a slight wave or movement of the water as the
object entered the ocean. It is unknown at the time of this report if the U.S. or another nation
has developed the ability to diminish water displacement caused on impact. It is more
difficult to explain the lack of significant1 deceleration as it entered the water despite the
absence of an identifiable power supply. See pages 25-30 for additional information.
Significant deceleration as an unpowered object, such as bullet, striking

the water.
1
April 25 2013
Splitting Into Two Parts

As can be seen in the video, the object splits into two parts shortly after entering the ocean
and then briefly re-emerging. Frame by frame analysis ruled out the possibility of a reflection
or of a second object emerging from the water. The objects thermal image actually grew in
size momentarily before it split into two parts. Both parts moved through the air and water at
the same speed as the original object. There exists no aircraft, projectiles, or other technology
known to the authors of this report to have these characteristics or capabilities. The authors
discuss this unusual characteristic in detail on pages 31 to 39 of this paper.
Power Source
The unknown in the video displayed qualities that require some type of power source. Over
the course of more than four miles the object reached speeds of almost 120 mph, made
multiple changes in direction, reduced and increased its speed, entered and exited the ocean
at speeds of over 100 mph, and finally split into two parts. In this thermal video, black
represents the hotter objects in a given frame and white the cooler objects. The unknown in
the video emits more heat than the ambient air and even after submerging in the ocean it
continues to emit heat after it exits. However the heat generated is generally less than what is
seen from jet engines and automobiles in the video. There is no exhaust plume or any other
indication of an aircraft. This is not characteristic of objects with ordinary power sources.
The objects speed, maintenance of momentum, directional changes, and its ability to sustain
high velocities in water eliminates all aircraft, blimps, balloons, wind-blown objects, any
species of bird, mammal, or other natural/man-made phenomenon. See pages 40 to 41.
Maneuverability
The object's ability to maneuver at speeds of 80-100 miles per hour (mph) though residential
and commercial at low altitude is of interest. A notable characteristic of the object is its
apparent tumbling2 as it moves through the air, which gives it a very non-aerodynamic
appearance. This tumbling action ends prior to the object's entry into the water and as it
moves through the water. The object also apparently accelerated while underwater. At
01:23:37 hours in the video the object can be seen to disappear behind a tree momentarily,
which places its altitude at below 40 feet. The ability to fly at that altitude at night and
between trees requires precise control of movement and a highly responsive propulsion
system particularly given the lack of control by aerodynamic devices (like wings). In terms
of our technology, advanced sensors or GPS satellites in communications with an on board
microprocessor might partially explain such maneuvers. Even more difficult to explain would
be the willingness of any government or organization to advertise this capability through a
residential area where malfunctions during flight could result in harm to the civilian
population and expose an advanced military technology.
The Authors
The six authors of this report all have scientific backgrounds including degrees in chemistry,
physics, mathematics, and environmental science. Their work backgrounds are also scientific
with experience in the air defense industries, semiconductors as well as various patents.
The exact nature of the IR emissions from the unknown object is unknown.
Tumbling could be an appearance due to the variable nature of the IR
emission from that object.
2
April 25 2013
Together they have 86 years combined experience studying the UFO phenomenon. A copy of
their backgrounds is listed in Appendix A.
A minimum of 1000 man hours were spent in the analysis of this video over the course of one
and a half years. Every effort was made to objectively evaluate the data obtained and ensure
the protection of the sources identity, according to his wishes. Non-Disclosure Agreements
were signed by all parties which stipulated details would remain secure.
Conclusion
The object witnessed by CBP and tower personnel and recorded on the CBP DHC-8 aircraft's
thermal imaging system is of unknown origin. There is no explanation for an object capable
of traveling under water at over 90 mph with minimal impact as it enters the water, through
the air at 120 mph at low altitude through a residential area without navigational lights, and
finally to be capable of splitting into two separate objects. No bird, no balloon, no aircraft,
and no known drones have that capability.
The authors are open to any reasonable explanation that addresses the various characteristics
displayed by this object. The full analysis and associated appendices can be read for the
detailed analysis that contributed to the above summary conclusions.
I. PURPOSE
The purpose of this report is to analyze the characteristics of an unknown flying object,
recorded by a thermal camera system, in an effort to determine if the object in the video can
be explained.
II. BACKGROUND
Thermal Imaging Video A special investigator known to the authors of this report received
information from a personal contact about a sighting of an unknown aerial object by a pilot
who was employed with the U.S. Customs and Border Protection (CBP), a branch of the
Department of Homeland Security (DHS), in Aguadilla, Puerto Rico. The special
investigators contact was not a direct witness but rather an acquaintance of the direct
witness(s). The source is considered the secondary witness. The pilot and crew of the DHC-8
Turboprop aircraft are considered the primary witness(s). On October 21, 2013, the
secondary witness provided the special investigator an original copy of the AVI 3 video file
depicting the unknown object over Aguadilla, Puerto Rico. According to the secondary
witness, the encounter occurred when the pilot was beginning a routine mission and saw a
pinkish to reddish light approach from the ocean towards the south. The local time of this
event was 9:20 pm on April 25, 2013. The object was visually detected and then tracked
using the planes on board thermal imaging video system. The secondary witness indicated
that the pilot could not discern a defined shape of the object but the object did possess a
reddish/pink colored light source. The light source turned off as the object entered the Rafael
AVI is Audio Video Interleave; a multimedia format introduced by
Microsoft in 1992.
3
April 25 2013
Hernandez airport airspace. From this point the object was exclusively observed through the
thermal imaging system of the DHC-8 Turboprop aircraft until the object entered the water
and was observed to split into two equal parts and gradually disappear under the water. There
were a total of four witnesses to the event on the aircraft and an unknown number of airfield
and Federal Aviation Authority (FAA) tower personnel.
Forward Looking InfraRed Systems, Inc. (FLIR) was initially contacted to determine if the
thermal imaging system was manufactured by their company. A photo obtained of the
thermal camera system was shown to a FLIR representative. The FLIR employee indicated
that it was not their thermal system but it belonged to L-3 Wescam, a Canadian company.
Wescam Inc. is a subsidiary of L-3 Communications Holdings, Inc. A Wescam representative
confirmed that it was their state of the art Wescam MX-15D thermal imaging system. This
system uses a InSb sensor with sensitivity in the 3-5 micron range. 4 A specification manual of
that system is included in the Appendix B. The capabilities of the system and its output
parameters can be seen on a Wescam video located on YouTube5.
The camera's video output parameters include the latitude/longitude coordinates of the
aircraft, date, time, azimuth heading of the aircraft, azimuth bearing to the target, and the
altitude above sea level of the tracking aircraft. The imaging system also provides the
latitude/longitude of any object within the cross-hair reticle of the camera, the altitude above
sea level, and the distance in nautical miles.
The video of the unknown consists of 3 minutes and 54 seconds of video imagery of which 2
minutes and 56 seconds displays the object arriving from over the ocean, traversing land, and
then disappearing back into the ocean. The entire video was broken into individual frames for
analysis of the unknown object. There were a total of 7027 frames with each frame equating
to approximately 1/30 of a second exposure. Breaking the video into individual frames
allowed for detailed evaluation of the object's characteristics. Each individual frame is
comprised of a set of 345,600 (720 x 480) picture elements (pixels) whose individual values
can range from 0 to 255. A given pixel value corresponds to some relative intensity of
infrared radiation which formed the image of the object. Low pixel values reflect warmer
temperatures (shown in black) while high pixel values correspond to cooler temperatures
(shown in white). It is important to understand that the image formed via these wavelengths
of infrared is not visible to the human eye but this does not mean the object could not have
been seen within visible wavelengths. It does mean, however, all the video image provides as
evidence can only be found within the infrared wavelengths given.
Radar Data Radar data was requested through the Freedom of Information Act (FOIA) in
November of 2013 from the U.S. Air Forces 84 RADES group as a means to both verify the
validity of the thermal imaging video and to look for any unknown targets in the area of
operation of the CBP aircraft. Radar data was obtained from three FAA sites in the area of
Puerto Rico. The primary radar site was QJQ, which is a long range radar located at 3417 feet
elevation. Its coordinates are 181607N and 654531W. Radar data was also obtained
from SJU located near San Juan, Puerto Rico. This radar site only receives secondary radar,
NATIBO Collaborative Point Paper on Border Surveillance Technology,
December 2007, p.14
5
https://www.youtube.com/watch?v=eZcFUYMAWBY
4
April 25 2013
also known as transponder signals. It was only useful in verifying the location of the CBP
aircraft. The third radar site information was obtained from radar on St. Thomas Island. It is
120 miles to the east of Puerto Rico and was too distant to provide meaningful information. A
copy of the FOIA request and the Air Force reply is included in Appendix C along with an
example of the data received from the Air Force. The entire file consists of 20 columns by
33,113 rows and is in a Microsoft Excel format.
The radar data from the QJQ site provided both primary and secondary radar (transponder
code of the aircraft and its altitude), date, time, latitude/longitude coordinates of the target,
distance to the target and the azimuth bearing of the target from the radar site. The radar
revolution rate was one sweep every ten seconds.
There is also a military radar installation located on the premises of Rafael Hernandez
airport. It is on the west end of the runway and is known as the Punta Borinquen radar site. A
FOIA request was made in May of 2014, again to the USAF 84 RADES group. This radar
site is the closest to the unknown object in the thermal video and would have been very
useful information. Unfortunately, the Air Force denied the request for that information. A
copy of the Air Force response is in Appendix C.
Control Tower Logs A FOIA request was made in March of 2014 for the airports control
tower logs, known as the Daily Record of Facility Operation, on the night of April 25, 2013.
The tower at the Rafael Hernandez airport is known as the Aguadilla Tower. The FAA
responded that their logs are maintained by a private company, Robinson Aviation, which is
not required to respond to FOIA requests. A copy of the FOIA request and the denial from the
FAA is in Appendix C. Robinson Aviation was contacted for information regarding the
control tower logs and did not reply to requests. According to the conversation between an
investigator and the Aguadilla Tower manager, the records (logs and recordings) from the
tower were destroyed 90 days from the date of any event. The tower manager also indicated
they were aware of the events of April 25, 2013, and were not willing to participate further in
the investigation.
Weather and Astronomical Conditions At 9:50 pm the surface temperature was 79F, the
humidity was 74%, barometric pressure at 30.05, scattered clouds, visibility of 10 miles, and
the wind was out of the east at 8-13 mph. 6 Upper wind speeds were measured out of San
Juan, which is 50 miles to the east of Aguadilla. At 8 pm local time the upper wind speeds
from 400 feet to 3200 feet were similar and were out of the east northeast at 12 to 18 mph.7
Sunset was at 6:48 pm and astronomical twilight was at 8:04 pm. There was a full moon that
night that rose at 6:53 pm. By 9:20 pm the moon was ESE at an elevation of 30 degrees.
Geography and Geology The area geology near the event location is that of Tertiary
limestone making up the majority of the nearby coast lines and visible topography. The
object traversed areas of steep and gradual inclines from the Atlantic Ocean. Much of the
http://www.wunderground.com/about/data.asp
University of Wyoming, Department of Atmospheric Science.
http://weather.uwyo.edu/upperair/sounding.html
6
7
April 25 2013
coastline in this specific area consists of a combination of erosional beaches and steep cliff
faces. The regional geologic structure is considered karst as well as the topography
characteristics. Puerto Rico sits near the strike-slip fault of the North American Plate a
Caribbean plate which is located sub parallel to the Puerto Rico trench. This area off the
coast where the unknown object entered the ocean consists of many geologic transition areas,
fault system and deep trench passageways.8
Limitations There were various factors that limited either the amount of information for
analysis or the quality of that information. Those limitations are listed as follows:
1. A FOIA request was sent to the Air Force but they would not provide radar data from the
military radar station that is located at the west end of the runway.
2. The FAA replied negative to a FOIA request for the airport control tower logs because they
had turned over maintenance to a private company, Robinson Aviation. The private company,
exempt from FOIA regulations, would not respond to requests for information.
3. Although a basic overview of the Wescam MX-15D video system is available, a detailed
manual that describes specifics of the system was not available to the public.
4. The readout of the Wescam system's latitude and longitude coordinates is rounded to the
nearest second. This rounding can produce a maximum potential error of 60 feet in the
location of the aircraft or its sited ground coordinates. This potential error was taken into
consideration in all calculations.
5. There is a consistent one second delay between latitude/longitude values displayed on the
Wescam video and the latitude/longitude values as reflected by the true coordinates of the
objects shown in the video. Due to its consistent variation, this shift is believed to be due to
an inherent system delay that does not affect the system's operational capabilities and was
taken into account in all calculations.
III. ANALYSIS AND CHARACTERIZATION

Witness Testimony
The witness testimony was from a secondary witness whose direct
testimony cannot be discussed in this report due to a request of anonymity. Additionally, it
was the secondary witness who contacted this team's investigator labeled as special
investigator and provided the original video. The primary witness is referred to in the
remainder of the report as Witness A. Similarly, the secondary witness is Witness B.
Witnesses A is an officer and pilot employed by the CBP division of Homeland Security.
Witness A was the pilot of the aircraft and one of four crew members that witnessed the event
from the aircraft. Witness B, who asked not to be identified, was the only means of contact
with Witness A who was not willing to talk to us at the time of the initial investigation of this
report. Questionnaires were given to both Witnesses A and B. Witness B, in turn, provided one
8
http://earthquake.usgs.gov/hazards/products/prvi/2003/documentation/
April 25 2013
of the questionnaires to Witness A. The completed questionnaires were then returned to this
investigative team by Witness B.
Multiple phone calls were made with Witness B as well as a personal meeting with two of this
team's investigators on February 15, 2014, at an undisclosed location. The following
testimony is from Witness B and it describes what was seen prior to the thermal video.
Witness A and his crew took off on a routine aircraft patrol of the Puerto Rican coast on the
night of April 25, 2013. The DHC-8 Turboprop aircraft took off from the runway heading
east at 9:16 pm. The aircraft contained four crewmen including the primary witness, a
copilot, and two instrument operators (one manning the on-board radar system and the other
manning the thermal image mounted camera system). Witness A looked out his left window
and saw a pinkish to reddish light over the ocean northwest of the airport. The light was
moving towards the airport. He believed the light to be at a higher elevation than his aircraft,
which was at 1600 to 2100 feet, based on the radar data and the thermal video system
engaged a moment before. The pilot confirmed visual contact with the tower personnel. The
tower personnel also confirmed visual contact. As the target approached shore, its light went
out. The pilot then requested monitoring of the craft with the on-board surveillance
equipment. According to the reporting witness the on-board radar did not pick the object up,
but the thermal imaging camera did detect the object. (The CBP's DHC-8 aircraft are
equipped with SeaVue Marine search radar primarily for detecting seacraft. 9,10)At this time,
Witness A no longer had visual contact with the object but did see the object in his thermal
imaging display in the cockpit along with the thermal imaging display in the rear of the
aircraft under control of the instrument operator. He continued tracking the object while on
routine patrol in the aircraft. The pilot made no attempt to intercept the unknown target nor
did the target seem to react in any way to the tracking aircraft.
Witness B stated the close presence of this unknown object caused the delay of a commercial
aircraft's departure from the airport. This statement from the witness could not be verified
since the authors of this report were denied access to the airport tower logs. However, the
statement's claim was supported when it was found that Fed Ex flight 58 was scheduled to
depart the airport at 9:10 pm but did not actually depart until 9:26 pm. It would also be
logical to believe the tower would delay departures if there was an unknown aircraft in the
airport's immediate airspace. Nonetheless, it cannot be known for certain that this departure
delay was due to the unknown object. There was only one arriving flight during this time
period. It was MartinAir flight 5713 that landed at 9:00 pm prior to the onset of this event.
No arriving flights were affected. This information was obtained from the FlightStats, Inc
database and is available in Appendix D.
Witness B indicated the video provided in this report was the entire unedited video and that
knowledge of this video was widespread within the CBP office located at an undisclosed
location. According to Witness B, Air Force Intelligence was contacted and subsequently was
provided a copy of the video. Air Force Intelligence offered no explanation to CBP and
recommended other agencies to contact. The identity of those agencies is not known. It is not
http://www.cbp.gov/sites/default/files/documents/FS_2014_DHC-8%20Bombardier.pdf
10
http://www.raytheon.com/capabilities/products/seavue/
April 25 2013
clear, based on discussions with Witness B, whether the evidence for this event had an official
security level. Since the event, Witness A indicated there have been no follow up
investigations by any other government agencies nor has any debriefing ensued with any of
the CBP officers.
Auxiliary Witness Testimony During research and field investigations pertaining to the
subject video evidence, additional indirect witness testimony was obtained regarding the
event and other similar events near and above Rafael Hernandez airport in Aguadilla, Puerto
Rico. Witness A indicated another independent fellow CBP pilot was east of the base and on
his way back to the airport about 15 to 30 minutes before the primary witness's sighting. This
officer witnessed a formation of pinkish/red lights flying extremely low over the airfield in
an unusual flight pattern. According to Witness A, the fellow pilot made a call to the base to
notify personnel of his observations. Additionally, according to Witness A, the primary
witness's son witnessed a light similar to the observed unknown object exit and enter the
ocean just off the coast north of the airport one to two evenings after the main event of April
25, 2013.
An anonymous email was sent to investigator and researcher Morgan Beall from an
individual with the alias John and from a secure email address 11. The IP address within the
email sent to the email server was received showing only the service provider server
locations. The writer stated he or she had worked for the CBP and could vouch that the video
was real. The writer's subsequent statements supported their claims. The writer mentioned the
specific model of the thermal imaging IR system used, an L-3 MX 15D and the writer
named the CBP DCH-8 maritime patrol aircraft specifically, which the research for this
report had already confirmed through Witness A. The writer goes on to describe the events of
April 25, 2013 with information we have only been able to glean directly from cooperative
witnesses to this event. The information provided by this anonymous writer is considered
credible and corroborates information from Witness A and B's testimonies. Uniquely, the
writer mentions the unknown object first appeared as a forward flying horseshoe shaped
craft and gradually changed its configuration to a spherical shape before entering the water. It
is not known if this is his or her interpretation of the video or if it is information witnessed by
airport personnel or other privy information to which the witness had access. Conversely to
this report's observations and conclusions, the writer makes a statement that the object did
not split into two parts but rather the original object was met by a second craft and both
proceeded to enter the water together. It is suspected this witness is either an active duty CBP
person on site or is acquainted with personnel actively serving on the base.
Another anonymous communication was posted under a YouTube commentary section next
to a lesser quality copy of this video at https://www.youtube.com/watch?v=Hee70AwwUJ8.
This statement was posted in Spanish in June of 2014 by an individual with the alias Red
Bill. This individual is suspected to have some inside knowledge of the event because
contrary to the original user's post of the video, Red Bill correctly states the source of the
video as the U.S. Customs and Border Protection and correctly states that the video was
taken from an airplane and not a helicopter. He also indicates (unverified by the authors of
11
john@truth.com is a fake email used under a secure or untraceable email
address.
April 25 2013
this report) that the video was analyzed in Quantico, Virginia and claims that two other
videos were made in the same area on different dates. All attempts to initiate communication
with Red Bill were unsuccessful.
A third anonymous communication was sent to John Greenewald Jr., creator of the web site
Black Vault (http://theblackvault.com/), in October 2014. The statements made in that
communication were nearly identical to that of the anonymous email sent to investigator
Morgan Beall. It is suspected that all three of these anonymous communications may be from
the same person. It was confirmed these communications were not made by Witness B when
questioned directly about the content and the source. The factual information concerning the
agency involved, the aircraft used by CPB and the object described is considered credible.
It should be noted that all primary research and witness interviews by the authors of this
report were completed prior to any of the afore mentioned leaks of a video and commentary
on YouTube and Facebook.
Radar AnalysisVerification of Thermal Imaging Video Information As noted earlier
in this report, data used in this analysis was from the primary radar site known as QJQ and is
a long range radar located 90 miles ESE of the area of interest and is at 3,417 feet elevation.
The radar is a FPS-20E and has a range of 200 nautical miles. 12 The radar was manufactured
by Bendix and is an L-Band radar that operates at 1280-1350 MHz and has a transmission
power of 2.0-2.5 megawatts.13 Based on the lowest altitudes detected of identified aircraft in
the area of interest, this radar is capable of detecting objects, if near the airport, at 400 feet
altitude. A graph and discussion of how this information was derived is in Appendix F.
The radar data was used to verify that there was a government aircraft on the day, time and
location as noted in the thermal imaging video. Hoaxes were eliminated once the aircraft that
took the video had been verified on radar from QJQ site.
It is a straightforward exercise to determine whether the aircraft on radar is an exact match to
the aircraft that filmed the thermal video. The video provides the exact time and location of
the aircraft as it was taking video of the unknown object. The radar data can verify if an
aircraft was present at the same time and location.
The CBP aircraft's location at specific times using the thermal video's time and
latitude/longitude stamps of the CBP aircraft was compared against radar data to verify the
existence and location of the aircraft. Radar data confirmed an aircraft tracking the same path
and time as shown on the thermal video. Figure 1 shows the path taken by the CBP aircraft.
The aircraft traveled north over the ocean once it departed the airport (north is at the top of
the map), then gradually to the southwest before traveling back over land and to the south.
Radar data indicated the transponder number of this aircraft as 4406. This transponder
number itself indicates that the aircraft is a military or law enforcement aircraft. FAA Order
Lincoln Laboratory, MIT, Correlated Encounter Model for Cooperative
Aircraft in the National Airspace. October 24, 2008.
13
OCEANA NAS, Harvey Clute, Jr., Bendix Engineer.
12
10
April 25 2013
7110.66 stipulates that all transponder codes between 4401 and 4433 are controlled by FAA
Order 7110.67, which is named Special Aircraft Operations by Federal, State Law
Enforcement, Military Organizations and Special Activities.
Based on the radar data, there is no doubt that the thermal video is a real video taken by a law
enforcement or military controlled aircraft. Because we have time and distance information,
the speed of the aircraft can be calculated. The aircraft's speed always varied from 180 mph
to 240 mph, which indicates it is not a helicopter but is a fixed wing aircraft. All of this
information supports the information in the thermal video as well as the story told by the
CBP witness. (A detailed analysis of this work and the aircraft's calculated speeds is included
in the Appendix F.)
FIGURE 1: Tracking aircraft's location based on thermal and radar data. The name Tracker represents
the CBP aircraft and the value to the right is Zulu time at that location. Radar data supports the time and
location coordinates provided by the thermal imaging video.
11
April 25 2013
Radar AnalysisVerification of Witness Testimony of Visual Sighting The witness

indicated that the unknown aerial object was sighted just after takeoff and at the beginning of
his standard nightly patrol. Visual confirmation of an approaching red light was made both by
the pilot and the control tower.
Radar data shows that the aircraft made an extra search pattern around the airport before
commencing what appeared to be its standard patrol and operational activities down the
Puerto Rican coast. Although the thermal video shows the aircraft's path for only four
minutes, the radar data shows the aircraft's path prior to and after the thermal video being
engaged. In the Google Map in Figure 2 the aircraft images in blue represent both the
matching thermal and radar data while the aircraft images in red are only radar data. It is
clear that before the thermal imaging video was engaged (red colored plane), the aircraft
circled the airport and then engaged the thermal video on its second pass (blue colored
plane). This supports the witness testimony that the pilot was aware of an unknown target in
the area, searched for the unknown target, and after finding it, engaged the thermal video
tracking system prior to resuming the aircraft's normal course near the coast. Data that is
discussed in the next section indicates that there is evidence to support the pilot's and control
tower's claim of an unknown object in the area.
FIGURE 2: Radar only data of a law enforcement or military aircraft shown in red with
Thermal Imaging & Radar data of the aircraft's location show in blue. The aircraft locations in
red show the aircraft's flight pattern before engaging its thermal imaging video.
12
April 25 2013
Radar AnalysisVerification of Unknown Targets

Radar data was reviewed for any primary data without a transponder code that would signify
an unknown radar track in the area of interest. Primary radar tracks are those created by the
actual reflection of the radar beam from a target. Known aircraft such as the law enforcement
or military aircraft transmit a transponder code, which appears in the radar data, is also
known as secondary radar. The radar picked up 50 primary radar strikes (no transponder) to
the north and northwest of the airport of what appears to be a single object from Zulu time
00:58hrs to 01:14hrs, a 16 minute period of time. The CBP aircraft, which transmitted a
transponder code, departed the airport runway at 01:16hrs. These 50 radar tracks (the radar
sweeps every twelve seconds) of this unknown object are visually displayed in Figure 3. The
amount of information in Figure 3 requires considerable commentary.
The first four radar strikes of the unknown target, seen at the far left area of Figure 3,
occurred after each twelve second sweep of the radar and are designated as a, b, c, and d. The
unknown target was not picked up for the next four sweeps, which equates to 48 seconds of
no radar contact. The fifth radar strike designated as 1a++++ 14 indicates the unknown was at
the same location as it was one minute earlier. That does not necessarily mean that the object
was stationary because the accuracy of the radar is only within 1/8 mile. If those first four
target strikes are a single object then the movement indicates 7800 +/- 660 feet moved in 36
seconds or a speed of 135-160 mph. The altitude of the object is not known but based on the
radar's minimum altitude detection limit at that distance, the object must have been about 800
feet altitude or higher. See Appendix F for detailed information.
The sixth radar strike occurs immediately after the fifth radar strike, i.e. the next twelve
second sweep of the radar. Beginning with this sweep of the radar, the object shows up on
almost every sweep of the radar for the next ten minutes. It could be that the first six radar
strikes were not related to the next 42 radar strikes identified as 1b thru 1aq. Symbols 1b
through 1aq are in consecutive order15 and represent consecutive radar strikes 12 seconds
apart. In total there were 42 radar strikes out of 50 possible in a ten minute period of time to
the northwest of the airport. This level of activity would be sufficient to cause concern for
anyone monitoring the radar system.
Within 18 minutes of this flurry of radar activity, the law enforcement aircraft with the
thermal video imaging capability took off from the airport just before 01:16hrs, which is
when the aircraft was first detected on radar at an altitude of 800 feet at the eastern end of the
runway. Figure 3 reveals that the CBP aircraft made an extra sweep into the area where the
unknowns were picked up on radar. Likely, the pinkish-reddish light seen to the northwest of
the airport by the pilot was the unknown target seen on radar and the object later recorded by
the CBP aircraft's thermal imaging video; it would be too coincidental to think otherwise.
The unknown target that appeared on radar for 16 minutes does not display characteristics
expected of ordinary aircraft in flight. The speed variation and sudden changes in direction
do not support mundane aircraft. Nonetheless, there are characteristics that can be attributed
to the unknown target.
14
15
The four plus signs indicate that four radar sweeps were missed.
1b 1c 1d 1y 1z 1aa 1ab 1ac 1ap 1aq.
13
April 25 2013
First, this target's appearance on radar occurred at the right time and location to likely have
been the object visually confirmed by the control tower and the CBP aircraft. Second,
although the target jumped around, its overall directional movement was from the northeast
to the southwest. Third, the target strength was strong as it was detected on almost every
sweep of the radar for eight of the ten minutes it was on radar. Lastly, the target was no
longer detected on radar during the time that the unknown was detected on the thermal
imaging video. At that point in time the object was believed to be below the Pico Del Este
radar's detectable altitude of 800 feet.
The authors of this report have looked for other explanations for the unknown radar strikes to
the northwest of the airport. A temperature inversion is a possible cause of false radar returns.
These occur when the upper air temperature is higher than lower air temperature. This
possibility is discussed in Appendix F and discounted due to the lack of any temperature
inversion layer in the area. One of the strongest arguments against some type of anomalous
propagation is the consecutive radar returns every 12 second radar sweep within a small
geographic area for a solid eight minutes coupled with the lack of these returns prior to this
incident and the lack of these returns after the unknown is picked up on the thermal video at a
lower altitude over land. It seems reasonable to consider the possibility that the visual
confirmation of the object by the pilot and the control tower, the detection of these unknown
radar returns on FAA radar data, and the detection of the unknown object on the thermal
video are all related to the same event and the same object. No other reasonable explanation
has yet been found.
14
April 25 2013
FIGURE 3: Radar plot of unknown that showed up off shore prior to the departure of the aircraft
with thermal imaging capabilities. Tracks are designated in order of time beginning with a-d
(segregated because of distance from the other radar tracks), followed by 1a-1aq, and followed by
2a, 3a, and 4a (segregated because of significant time delays of greater than one minute between
radar tracks.
The radar sweeps every twelve seconds. Each + after a radar hit indicates that the target was not
detected in the previous radar sweep. A designation such as 1ac,1af indicates that two different
radar sweeps occupied approximately the same physical location to within 1/8 of a mile of each
other. (There is no difference between green & yellow circles and is due to a Google Earth issue.)
15
April 25 2013
Object Size The object in this video was tracked using a state of the art Wescam MX-15D
multi-sensor multi-spectral targeting system. The MX-15D was mounted on the underbelly of
a DHC8 turbo prop aircraft operated by U.S. Customs and Border Protection. This system
has high definition thermal imaging, short range IR for enhanced haze penetration, a laser
rangefinder and illuminator, and stabilization features. The video lasted more than three
minutes and due to familiar objects in the background, the approximate size, speed, and path
of travel of the object were identified. The camera's video output included the latitude and
longitude coordinates, azimuth heading, and the altitude above sea level of the tracking
aircraft. It also provided a target latitude/longitude, an altitude above sea level, and the
distance in nautical miles as well as meters. Due to the capabilities of this particular camera
its sale outside of the United States requires approval from the U.S. Government.
The video consists of 3 minutes and 54 seconds of video imagery of which 2 minutes and 56
seconds displays an unknown object arriving from over the ocean, traversing land, and then
disappearing back into the ocean. The entire video was broken into individual frames for
analysis of the unknown object. There were a total of 7027 frames with each frame
approximately 1/30 of a second exposure. Breaking the video into individual frames allowed
for detailed evaluations of the object's characteristics.
Specific information is provided as to how the size, speed, and location of this object were
determined. The basic determinations were based upon trigonometry related to the actual
object size, angular size, and distance of the object. If two of those variables are known then
the third variable can be calculated.
The angular size of the object was calculated from the angular size of each pixel in the video
at a given magnification. The angular size of a pixel was determined from several different
objects of known size, known distance, and the number of pixels that made up the object's
length in a video frame. Using the known distance and size, the angular size of the known
object was calculated in degrees. Dividing this by the object's length in pixels provided the
angular size in pixels at that particular magnification. A value of .001483 +/- .000045 per
pixel was obtained. An example of one of the known objects and distances used is shown in
Figure 4. The angular size of a pixel is proportional to the magnification used in the video. A
detailed discussion of the technique and calculations used is available in Appendix G.
The angular size of the object can be used to calculate the object's true size if we know the
distance to the object. Although distance and altitude of the object is shown on the thermal
video display, these values are actually the distance to the terrain behind the object. An
example is shown in Figure 5 where the altitude of zero feet is clearly that of the ocean as is
the distance of 3.5 nautical miles. The unknown object's true altitude in Figure 5 is some
value greater than zero and its true distance is some value less than 3.5 nautical miles.
The distance can be accurately calculated whenever the object is at a known or a very low
altitude. This occurs towards the end of the video when the object passes behind a telephone
pole, behind trees, and then finally enters the water. During these periods of time there are
means to measure the distance.
16
April 25 2013
FIGURE 4: Frame 0892. Known tank size and distance along with unknown.
FIGURE 5: Frame 0141. Unknown object near center of cross-hairs.
17
April 25 2013
For example, in the right triangle shown in Figure 6 the camera is at point 'C'. The cross-hairs
of the camera are pointing towards 'A'. Any object in the cross-hairs (represented by point
'D') of the camera could be at any location along line 'AC'. However, when foreground
objects such as trees or a telephone pole or the water surface itself interact with point 'D' then
one knows that point 'D' is close enough to point 'A' (point 'A' is on the ground) to allow for a
reasonably accurate determination of the distance from the object to the camera.
FIGURE 6: Right triangle

Calculations of the object's size were done on multiple frames whenever the object was a
known distance from the ground which allowed accurate values of the object's distance and
its size. Two examples are shown in Figure 7, where the object is seen four seconds after it
had exited the water, and in Figures 8-10 that show the object moving behind a tree. In both
examples the distance of the object is known because its approximate altitude is known; with
a known distance then the object's size can be determined. The size values obtained for the
object varied significantly from a minimum size of 3.0 feet to a maximum size of 5.2 feet.
The variation in size could be due to either varied angular sides of the object as it is appears
to be tumbling or temperature variations as seen by the IR camera that could distort the
object's apparent shape. Regardless, the variations in apparent size seen by the IR camera are
due to properties intrinsic to the object.
Object Path The path taken by this object during the video cannot be ascertained simply
by plotting the latitude/longitude coordinates that are displayed by the thermal imaging
system based on the cross-hairs. Those coordinates are driven by a laser range finder, which
is not striking the object itself but the ground and other large objects in the background. This
was done empirically using known objects in the background and verifying the
longitude/latitude coordinates belonged to the background object. Witness B also supported
this when he indicated to the authors that their laser range finder is used for ground targeting
and only rarely has it been able to capture aerial targets. As a result, when the object is at
altitudes above about 40 feet there can be significant differences in the actual distance
between the object and the camera. This is the same issue discussed in the determination of
the object's size. A powerful tool used in this analysis was Image J software that allows one
to zoom in on groups of frames and run/reverse/stop the video at higher magnifications. This
software was developed by the U.S. government.16
ImageJ 1.47v. National Institute of Health, July 8, 2013.

http://imagej.nih.gov/ij
16
18
April 25 2013
FIGURE 7: Frame 5085. Unknown object in center of cross-hairs.
FIGURE 8: Frame 2697. Unknown object visible.
19
April 25 2013
FIGURE 9: Frame 2705. Unknown object disappears behind trees.
FIGURE 10: Frame 2713. Unknown object reappears from behind trees.
20
April 25 2013
The path was determined from various frames in the video where the approximate altitude of
the object was known. This occurs during the last half of the video when the object's altitude
was less than 40 feet and was descending into the ocean. Details of the calculations that
identified the object's path are displayed in Appendix G.
The best determination of the object's possible paths is shown by the brightest of the three
blue lines in Figure 11, a Google Earth image of the northwest coast of Puerto Rico. The
airport that is seen in the image is the Raphael Hernandez Airport and is a joint civil-military
airport located in Aguadilla, Puerto Rico. The top of the page faces west and the right hand
side of the page faces north. The dark blue aircraft icons indicate the actual locations of the
aircraft with the thermal camera. The locations were verified by both the thermal camera
system's latitude/longitude values and by radar from the Pico Del Este radar site. In Figure
11, the numbers next to the aircraft represent the time in Zulu (aka Greenwich Mean Time)
hours that the plane was at that specific location. A corresponding UAP (Unidentified
Anomalous Phenomenon) location is on the map for the same time period. The UAP
locations marked in red are exact locations of the object at those times due to accurate
altitude values being available. The first exact location of the object is marked in red on the
map at time 01:23:37 as the object passed behind a tree as shown in Figures 8-10. The UAP
locations marked in orange represent approximate locations of the object within 500 feet. The
UAP locations marked in yellow with a time value next to them and the darker blue line
connecting them represent a best guess of the object's location based on the previous path
of the object and its known direction from the aircraft. The brighter blue line begins at a
question mark that represents the uncertainty of the object's location at the beginning of the
video. The object's route does raise the possibility that its origin could have been its final
destination or its origin could have been up to one mile farther to the west as shown by the
other two light blue lines. The light blue lines connect possible routes taken by the unknown
object that are farther to the west. The yellow colored UAP locations represent a higher level
uncertainty of the object's position than those colored in red.
The object approached the island of Puerto Rico out of the north from the ocean. Its exact
origin is unknown. Heading south it crossed the airport runway, turned east, then north again
to recross the airport runway on its way back out to sea. The object's path reflects a complete
180 degree change in direction over land and a continual drop in altitude during the last half
of the video. The object left land at 01:24:04 and changed direction again heading northwest.
At 01:24:13 the object impacts the water and travels just below the surface. The object's
movement through the water can be seen if you look carefully at the video. It exited
momentarily at 01:24:18 before moving just below the surface again. Once more, the object
can be followed until it exited the water at 01:24:31. The object changed its direction while
underwater from northwest to west. (It is believed the reason the camera is able to follow the
object while underwater is due to the Bernoulli 17 hump created by any object moving
underwater. This is discussed further in Appendix H.) Eleven seconds after exiting the water
at 01:24:42 the object split into two equal parts, both the same size as the original, as it
continued to the west. One of the two parts entered the water at 01:24:52 while the other part
17
Stefanick, T.; "Strategic Antisubmarine Warfare and Naval Strategy"; Institute for Defense and
Disarmament Studies; (1987): Appendix 3
21
April 25 2013
FIGURE 11: Google Map of three possible paths of the unknown object.
22
April 25 2013
changed direction to the southwest before also disappearing into the water at 01:25:04.
Object Speed What is the energy source that propels the object in this video? The
movements made by this object require some type of power source. The object traversed over
four miles during the video and during that process changed direction; south to north to west
then finally towards the southwest. No type of propulsion is evident from the thermal
imaging video yet some form of propulsion is required for the object to maintain and vary its
speed, change directions multiple times, and move in and out of the water. Again, the source
that propels this object is not evident.
The speed of the object is most accurately calculated during the latter half of the video when
the object's location can be more accurately determined. The calculation of the object's speed
is straightforward and established from given distances and times. The thermal video
system's clock and latitude/longitude locations allowed for the calculation of time and
distance values. The clock accuracy is to the nearest 1/30 second due to the frame rate of the
video. The main error is that the latitude/longitude values are in degrees, minutes, and
seconds so that the location is a digitally displayed to the nearest second. The accuracy is
within 0.5 seconds of a degree, which is roughly 51 ft. The rounding error is taken into
consideration for the speed and distance calculations.
Table 1 shows the time of the latitude/longitude measurement, the distance traveled since the
last measurement, and the calculated speed of the object. Although the speed of the object is
fairly constant and normally varies from 70 mph to 110 mph, it is clear that the object
accelerates and decelerates during this portion of the video, which again indicates some type
of power source must be present. Some of the speeds shown in the table are noted as being
through water. These speeds are average speeds and do not examine the possibility of higher
speeds within short time bursts, which sometimes appear during the video.
ZULU
TIME
ALT.
LAT./LONG.
DISTANCE
TRAVELED
TIME
DIRECTION
OF TRAVEL
SPEED
01:23:37
25 ft 18:30:08N 67:07:13W
------
------
------
------
01:23:49
16 ft 18:30:19N 67:07:18W
1213 ft
12.0
north northwest
69 +/- 3 mph
01:23:54
16 ft 18:30:24N 67:07:20W
542 ft
5.0
north northwest
74 +/- 7 mph
01:24:07
< 1 ft 18:30:46N 67:07:19W
2230 ft
13.0
north
117 +/- 3 mph
01:24:20
**
18:30:56N 67:07:34W
1761 ft
13.0
west northwest
92 +/- 3 mph
01:24:42
**
18:30:52N 67:07:57W
2246 ft
21.7
west
70 +/- 2 mph
01:24:46
18:30:52N 67:08:01W
384 ft
3.95
west
66 +/- 9 mph
01:24:51
18:30:53N 67:08:04W
305 ft
5.28
west
39 +/- 7 mph
01:24:57
**
18:30:51N 67:08:08W
434 ft
6.03
west southwest
49 +/- 6 mph
TABLE 1: Speed of object at known positions.

* Speed underwater.
**Speed through water and air.
23
April 25 2013
Table 1 uses 4 to 13 seconds in each calculation of the object's speed based on the latitude
and longitude coordinates provided by the thermal imaging video.
Graph 1 shows speed averages over a total of about 800 frames. Average speeds were
calculated for collections of about 50 frames each (covering approximately 1.5 seconds) from
the 800 frames. Then, from all of the average speeds calculated, 15 sets of three each were
used to generate a moving average. The purpose of the moving average was to emphasize
longer term trends which may be seen in the curve of Graph 1. Even given the longer term
speed trends of Graph 1, the object's speed is still seen to vary within a short period of time.
The speed of the object was also measured using a completely different method based on the
object's ground speed by comparing its relative movement against background ocean waves
whose speed was negligible. This method showed speed variations between frames 3769 to
3843 of 70 mph to 130 mph, which is comparable to the speeds found using latitude and
longitude coordinates. Details on this second method are available in Appendix I. These two
different methods clearly establish that this object moved at speeds above, through, and under
the water that cannot be explained by simple conjectures such as a balloon, bird, or windblown object.
Graph 1: Speed measured every 50 frames
Object's Interaction with Water Although most of the video concerns the object moving
through the air, there are portions in the latter part of the video when the object interacts with
24
April 25 2013
the ocean. These include the object's actions immediately prior to impact in the ocean;
entering the water; motion underwater; exiting the water; and object division.
Object's Interaction with Water: two seconds prior In the two seconds prior to ocean
impact, there is no indication of the object slowing down. The object's speed prior to impact
is near the highest speed measured during the latter part of the video. Frames 3700 through
3750 were used as the points of reference for measuring the object's speed prior to frame
3769 when the object began to impact the water. In order to minimize errors due to rounding,
time to the nearest second and latitude/longitude to the nearest degree second, extrapolations
were made using individual frames to provide higher accuracy. The speed calculated during
the two seconds prior to impact was 109.7 mph with an error of +/- 11 mph. The details
behind this calculation are shown in Appendix J.
Object's Interaction with Water: moment of impact In today's understanding of science,
it is impossible to enter, leave and move through a fluid and not affect it. There is no visual
indication within the video that the object immediately slows down on impact, (Graph 1
shows this minimal impact as the object slows by only 10% as it first impacts the water until
total submersion.), creates an expected significant splash, or reacts with the water in any
obvious fashion. The mystery around this lack of a splash is further compounded since the
object doesn't appear streamlined but is more of an oval shape. The lack of a visual effect
could be due to our difficulty in translating a heat signature into the more normal visual
picture. A very detailed discussion of the object's shape is covered in Appendix H, Modeling
of the Object.
At 01:24:13 Zulu hours, as shown in Frame 3769 and those following (see Figures 12 and
13) an object larger than three feet, traveling over 100 miles per hour, hit and entered the
ocean seemingly with little or no splash. Although present science knows ways to minimize
the splash, eliminating it is not possible. Effectively a splash is taking a volume of water and
drastically increasing its surface area. Since both evaporative and radiative heat transfer are
proportional to surface area, a splash enables that volume of water to become cooler. The
change in temperature discussed here is very small and is nearly invisible in infrared as can
be seen in Figure 14. In this figure, the red circle outlines the unknown object that has just hit
the surface of the ocean. The red arrow indicates the object's direction of travel. As was
stated above, even with a 300X zoom, no significant cooling (lighter shades) of the water can
be seen.
A three-dimensional view of pixel intensities is helpful in looking for a splash. An Image J
tool18, "Surface Plot", was used to create a 3-D view. The software converts the pixel
intensities (heat variations) in the IR frame to height variations with the lighter (cooler)
pixels being represented as hills and the darker (hotter) pixels as valleys. The red outline in
Figure 14 was provided to allow a direct comparison of that picture with the surface plot
shown in Figure 15.
ImageJ 1.47v. National Institute of Health, July 8, 2013.

http://imagej.nih.gov/ij
18
25
April 25 2013
FIGURE 12: Object's initial impact with water shown in Frame 3769
FIGURE 13: Object's < 1/6 of a second later in Frame 3773
26
April 25 2013
Although still small, the cooler areas representing the splash are seen as raised areas around
the upper-right corner of Figure 15. It is believed these represent a splash rather than simply
cooler areas of the UAP since they only show up in these plots where the UAP is entering the
water. The object in Figure 14 is moving to the left and slightly down. This raises an
interesting observation. Rather than the splash being in front of the UAP, it is trailing. Since a
frontal splash could not be found in this or any later frames, it is believed the splash was
caused by the lower middle or back lower portion of the UAP. The UAP was angled such that
it sliced into the water with little or no splash at a speed close to 109 mph. This feat requires
a technology that would be at the forefront of the U.S. Navy's current capabilities. If it has
been developed then it would most assuredly be highly classified.
FIGURE 14: Frame 3769 - 300X zoom
FIGURE 15: Frame 3769 Surface Plot
27
April 25 2013
Object's Interaction with Water: movement through water Although the video does not
show the object for most of its underwater period, each time the object appears, the camera is
found to be pointed almost directly at the unknown. Since one of the witnesses specifically
rejected the idea that the camera was locked onto the object, this implies that the object has
remained visible to either the camera operator or the pilot or both. It further implies that
while traveling underwater the unknown object has remained relatively close to the surface
throughout. Since these assumptions essentially mean the unknown is at an altitude of sea
level and is placed at the target location printed on each frame, we can utilize the target
location as the unknown's location throughout this period to calculate its speed.
During the first 26.7 seconds that the object is traveling underwater, it covers a distance of
3241.7 feet equating to an average speed of 82.8 mph. This is a slight drop from the object's
aerial speed at impact. Whether that speed drop is due to the resistance in the water or is just
coincidental is not known. The speed calculations made of the object's movement through the
water is detailed in Appendix J. The speed of the object underwater is not beyond our current
technological capabilities. These high speeds are easily exceeded by underwater torpedoes
that reduce water resistance via a process known as super-cavitation.
At this point, it is worth discussing how the thermal imaging video is capable of seeing an
object underwater. Infrared radiation is easily blocked by water and about one millimeter of
water absorbs virtually all of the IR generated by the object; however, that does not
necessarily eliminate detection of underwater objects using infrared.
When a solid object moves underwater, the water must be displaced and some of that water is
displaced toward the surface which then manifests as a moving hump along the surface.
Northrop Grumman is aware of this phenomenon, known as a Bernoulli Hump, and has
mentioned this as one possible method to detect submarines.19 Like the splash discussed
earlier, that surface hump would increase the exposed surface area of the water and therefore
decrease the temperature compared to the surrounding surface water. This thermal effect can
be easily seen in the movie as a moving cool region and is easier to view than in individual
frames. It is caused by the slight bulge in the surface due to water displaced by the motion of
the unknown object. The infrared camera displays a lightly cooler (whitish) area associated
with the object and can be seen in Figure 16. This whitish area, the Bernoulli Hump,
increases with speed and cross sectional size of the object and decreases with depth. Using
this information, the depth of the unknown object can be estimated. The average wave height
off the coast of northern Puerto Rico is one to three feet on a typical day. The height of the
waves seen in the video are then likely to be this height therefore the Bernoulli Hump seen in
the video is also in this range. Consequently, from Graph 2, the maximum depth of the
unknown object is in a range from 9 to 16 feet. See Appendix H for additional information.
Haffa and Patton, Analogs of Stealth, Northrup Grumman Analysis Center

Papers, June 2002, p.14
19
28
April 25 2013
FIGURE 16: Cooler signature of object in frame 3781
Graph 2: Bernoulli hump; height vs depth based on a 3 ft object moving at 83 mph
29
April 25 2013
Object's Interaction with Water: exiting the water Five seconds after the object enters
the water at 01:24:18, it re-emerges for about two seconds and either skims along the surface
or is only partially submerged. During those two seconds, it is clear that the object's
temperature is still hotter than the water around it because it is still a distinct black. It is also
clear that there has been no slowing of the object through the water. The object must have
some type of power source that maintains its movement through the water as well as
maintain its temperature. Figure 17 demonstrates the heat of the object is still present and, as
shown in Table 1, the object maintained a significant underwater speed for a lengthy period
of time; almost a minute. When the object made its next exit at 01:24:31 hours, it continued
to maintain a heat signature on the video.
FIGURE 17: Heat of object indicated by dark color in frame 3937
Object Divides into Two A significant and unusual characteristic exhibited by this object is
the moment when the object splits in half. No indication could be found of a second object
that joins the first does the video evidence suggest the second object is due to reflection. A
careful frame by frame analysis indicates that the object split in half. In less than one second,
the object's thermal image doubled in size; its center of heat then became bimodal; the object
then split into two halves. The process appeared similar to mitosis observed during cell
division with the splitting of the nucleus, the expansion of the cell, and the final separation
into two cells. Due to the significance of this event, a considerable amount of time was spent
illustrating this segment of the video evidence.
30
April 25 2013
The initiation and completion of the splitting process occurs largely within one second. This
second of time was separated into 32 frames (frame numbers 4602 through 4633) allowing
the events to be examined every 1/32 of a second. In order to display the object's appearance
before and after the split, some of the frames displayed are within the immediate one to two
seconds before and after Zulu time 01:24:41 hours.
The object's appearance is illustrated using four approaches. One uses full frame of the video,
the second uses frame magnification that enable individual pixels to be seen, the third
displays pixel values (0-255) with 0 representing the hottest IR signature and 255 the coldest
IR signature, and the last uses a two dimensional contoured surface plot. The frames selected
are representative of the ongoing change in the object and are represented as figures ##A
through ##H. A few frames were not usable because the system's white screen overlay
interfered with the object image.
Table 2 displays the important parameters for each selected frame. The comment section
primarily describes the object's pixel distribution.
Frame #
Time
Comments
4563(A)
01:24:39
Typical display. Hotter in center.
4590(B)
01:24:40
Typical display. Hotter in center.
4611(C)
01:24:41
Size of object begins to increase.
4623(D)
01:24:41
Object's internal heat distribution increases uniformity. Center area

enlarges.
4631(E)
01:24:41
Object's size continues to increase.
4634(F)
01:24:42
Still one object but interior is exhibiting bimodal heat zones.
4640(G)
01:24:42
Bi-modality of center heat zone is now clear.
4652(H)
01:24:42
There are now two separate objects.
TABLE 2: Frames used in evaluation of object splitting in two.
Figures 18A through 18H are the video frames analyzed. The laser range finder reticle, in
white and shaped like four Ts, can be seen at the screen center with the object nearby in all
eight frames. Figures 19A through 19H are enlargements of the previous frames. These
enlargements display the individual pixels making up the object. The object has the darker
(hotter) pixels than its surroundings. Figures 20A through 20H are the pixel values positioned
in their relative screen centered around the unknown object. The pixel values represent the
level of infrared intensity relative to other pixels. The pixel values in the 110-140 range are
from the ocean. Those are much cooler values than the object, which are in the lower and
hotter range from 10-105. Values >160 that occasionally show up are from the laser range
finder reticle, which are the bright white pixels.
31
April 25 2013
Figure 18A: Frame 4563
Figure 18B: Frame 4590
Figure 18C: Frame 4611
Figure 18D: Frame 4623
Figure 18E: Frame 4631
Figure 18F: Frame 4634
Figure 18G: Frame 4640
Figure 18H: Frame 4652
32
April 25 2013
33
April 25 2013
To make the object easier to visualize, the pixel values were color coded from the hottest (red
= values <31) to the next hottest (orange = values 31-60) to the cooler parts of the object
(yellow = values 61-105) all of which are distinct from the cooler areas of the ocean (no
color). When white pixels belonging to the reticle were near the object, they were color
coded as green. The ranges chosen for each color help visualize the changes in the IR
emissions of the object as it split into two halves. The actual edge of the object cannot be
discerned in any absolute way. The coloring of the pixels ends at the edge of the number that
is closest to the arbitrary value ranges of <31, 31-60, etc. The infrared outline of the object
can be seen when there is a significant change in the IR values of the object's periphery
compared to the surrounding ocean.
138
128
129
130
137
132
133
133
136
129
128
142
147
124
82
82
104
136
124
124
128
129
94
52
58
93
136
135
128
108
92
52
24
35
69
136
143
130
97
76
33
11
23
60
136
141
133
103
76
54
43
46
42
136
142
130
104
92
91
93
95
85
134
141
130
107
98
102
115
130
131
134
129
128
117
108
105
112
130
139
129
Figure 20A: Frame 4563; normal image about two seconds

prior to any change in the pixel.
133
116
126
124
106
96
109
123
129
133
116
121
120
99
93
110
122
130
134
116
114
104
73
72
97
114
127
135
117
114
74
21
25
65
102
130
136
119
101
63
27
37
64
97
134
140
123
95
53
24
37
60
95
132
144
129
83
50
35
48
65
98
134
147
132
90
67
63
77
91
113
135
142
136
115
109
102
101
108
118
128
Figure 20B: Frame 4590; normal image about 2/3 of a second

prior to any changes.
34
129
126
126
133
132
127
120
115
117
117
121
127
128
132
144
134
118
119
114
115
116
120
128
121
128
143
120
92
102
104
110
116
119
134
128
123
132
104
69
82
92
108
117
119
April 25 2013
141
132
109
109
82
45
54
66
91
119
119
133
122
95
74
34
24
36
32
77
114
123
126
103
79
52
20
25
42
39
76
114
123
126
97
75
47
25
32
45
48
86
114
124
133
115
89
67
65
71
71
67
89
114
124
134
110
90
85
98
102
102
103
112
115
125
129
113
103
107
115
109
114
121
121
117
125
137
131
110
77
76
74
76
71
101
116
129
139
138
122
93
91
90
94
91
112
122
131
138
142
134
112
110
109
114
115
122
128
131
133
139
137
123
122
122
126
126
127
129
129
Figure 20C: Frame 4611; size begins to increase.

129
127
128
130
122
122
124
127
130
134
134
131
128
128
127
113
110
111
110
131
137
135
133
131
123
120
105
101
99
97
137
145
138
139
122
80
55
34
32
32
37
138
151
141
138
122
73
49
35
35
32
36
120
145
141
140
122
73
51
34
36
33
32
86
128
136
143
122
73
51
31
36
37
34
58
113
132
Figure 20D: Frame 4623; center zone of object expands and the
internal heat distribution increases in uniformity.
132
131
132
100
114
136
150
130
127
127
128
132
130
128
84
98
122
139
125
125
129
131
132
124
117
61
73
98
114
125
123
126
125
132
119
107
44
58
86
97
125
118
125
125
131
89
38
41
42
42
43
50
77
101
111
123
80
32
39
41
39
39
52
75
98
112
116
78
40
43
43
42
41
59
75
98
114
116
90
68
41
38
42
43
63
74
97
111
120
106
97
85
78
83
83
93
96
113
119
Figure 20E: Frame 4631; object's size continues to increase.
35
125
117
115
97
90
95
96
101
103
115
116
130
125
126
119
115
114
113
114
114
118
120
April 25 2013
(green pixels due to white reticule)
186
129
128
126
125
124
124
123
123
127
126
202
114
102
103
102
111
127
130
131
126
125
165
101
79
85
76
93
129
128
127
127
128
136
114
56
60
59
67
95
102
108
113
120
131
90
30
36
42
50
73
80
81
101
121
133
28
31
69
43
18
29
42
53
89
123
133
95
65
89
53
22
30
36
44
82
128
132
123
87
115
88
53
47
53
65
87
124
129
101
114
140
99
71
65
69
75
101
126
131
129
130
127
123
121
124
126
125
133
129
131
129
129
128
126
125
127
129
130
130
132
121
120
116
112
113
115
113
115
121
125
131
123
122
115
118
114
117
115
119
118
122
131
Figure 20F: Frame 4634 at 1/10 of a second after Figure 20E;

the center warmer area has become bimodal.
127
102
88
100
109
138
159
191
203
235
232
123
104
92
79
83
115
132
217
228
255
253
125
100
84
63
73
102
123
94
110
141
141
105
71
36
35
53
81
93
90
119
142
133
91
58
41
70
76
74
62
44
89
110
125
101
75
48
71
64
55
42
31
59
88
126
107
83
61
76
68
57
46
32
62
94
133
116
109
105
98
78
67
69
51
83
108
131
119
116
118
105
92
82
80
100
112
125
135
(green pixels due to white reticule)
Figure 20G: Frame 4640 at 1/5 of a second after Figure 20F;

interior warm zones are now clearly bimodal and separate.
36
169
167
113
107
94
114
134
136
136
127
134
136
135
134
132
77
78
63
107
137
133
130
120
131
137
137
130
126
69
72
46
83
119
111
110
110
122
133
135
107
96
44
49
32
67
125
118
103
96
109
123
130
106
92
52
59
55
80
129
108
63
40
96
111
125
April 25 2013
101
88
77
81
77
89
112
91
50
32
88
102
122
97
83
99
103
101
111
112
93
58
45
85
97
121
120
122
116
108
111
112
102
97
74
52
95
116
123
124
120
118
117
117
119
116
108
84
66
93
113
123
137
128
121
127
122
127
132
124
101
88
81
104
120
142
133
114
123
121
133
141
133
115
102
88
106
118
131
126
116
123
122
127
125
125
128
131
110
113
114
123
123
118
116
116
125
125
128
127
123
118
118
118
Figure 20H: Frame 4652 is 1/3 of a second after Figure 20G with two objects that are
now separated.
The changes seen in these eight frames are representative of all the frames during that same
time period. A ten-frame moving average was used to minimize any intrinsic pixel variation
and/or subjective judgements as to whether a pixel was or was not part of the object. This
change in pixel size is shown in Graph 3. Whenever the reticle cross obscured a significant
part of the object then a note to that effect is shown on the graph. Those frames were not used
in the calculation of the moving average.
Graph 3: Frame by frame moving average of object's change in size
37
April 25 2013
The last of the four approaches used surface plots from ImageJ software to create a 3-D view.
These are displayed in Figures 21A through 21H. Highlighting was done with the LUT
feature that provided six shades used to note the IR heat with the bluer (cooler) pixels being
represented as hills and the redder (hotter) pixels as valleys. The blue area is the water and
the red-orange area is the object with the yellowish-greenish color being a debatable zone of
either the object itself or heated areas around the object. The tall pinkish capped peaks seen
in Figures 21E through 21H are the effect of the laser range finder reticle image. The size of
the area chosen for each frame was kept constant at 13 x 13 pixels so that the change in size,
the bimodal heat zone, and the final splitting of the object would be easier to compare across
the eight surface plots.
Figures 21A through 21E depict the heat signature consistently seen through the unknown's
complete transit. In Figures 21F through 21H a clear representation of the heat signatures can
be seen splitting in to two similar parts as depicted in Figures 20 A through 20H.
Frame by frame analysis provides no evidence any pre-existing and independent second
object arose out of the water; nor is there any indication that the second object is some type
of infrared reflection of the first object. Frame by frame analysis, which was every 1/30 of a
second, did not support either possibility.
38
April 25 2013
39
April 25 2013
Power Source The unknown in the video displayed qualities and behaviors that require
some type of power source. Over the course of more than four miles the object reached
speeds of greater than 110 mph, made multiple changes in direction, accelerated and
decelerated, maintained a temperature significantly greater than the ambient air temperature,
entered and exited the ocean at speeds of over 100 mph, and finally split into two parts. The
unknown object in the video clearly generated heat which statistically remained near its
center with its outer areas at a lower temperature. The heat generated is usually much less
than what is seen in the video from jet engines and automobiles. There is no exhaust plume
or any other indication of an ordinary aircraft power source. It is also evident the unknown
object generates more heat than the ambient air and continues to maintain its heat signature
after submersing in the ocean and re-entering the air. These actions and characteristics cannot
be achieved without some power source.
It was possible to determine approximate temperatures of the object. The thermal video
contains gray scale pixel values between 0 to 255 where 0 is the hottest and is represented by
'black' while 255 is the coldest and is represented by 'white.' Using ImageJ software, the heat
signature of the individual pixels comprising the object could be compared. This type of
display has already been shown in Figures 20A through 20H where the object's lower IR
pixel values of 10-60 in the object's center represent a hotter temperature than the ocean
water's pixel values of 115-140, which represent an ocean temperature of about 79 to 83
degrees during April off the Puerto Rican coast. While we have the temperature of the ocean
and its corresponding pixel values, for these particular frames with a single known
temperature, we can only state that the object's pixel values of 10-60 is warmer than 79 to 83
degrees. Other portions of the video have been analyzed where objects such as cattle, roads
and pastureland are in the same frame and enable the establishment of temperature reference
points that permitted the temperatures of the object to be estimated. In Frame 1251, time
stamped 01:22:49 hours in the video, the unknown object can be seen as well as cattle, an
asphalt road, and a pasture. The temperatures of the latter two could be determined based on
their natural characteristics and known cooling rates while the skin temperature of cattle is a
known value. The details of this analysis are discussed in Appendix K. Graph 4 displays the
temperature values vs. the pixel values using the three known temperatures in Frame 1251.
The eight center pixel values of the object vary from 0 to 8, which on this graph equates to a
temperature of 105 degree Fahrenheit assuming that the thermal video distributes the 255
pixel values linearly. If not then the temperature would be greater than 105 degrees. The
sixteen pixel values surrounding those are slightly cooler with pixel values ranging from 16
to 96. Those temperatures range from 103 degrees to 87 degrees.
40
April 25 2013
Graph 4: Temperature distributions in Frame 1251 at 01:22:49 hours.

A similar but less exact observation can be made by viewing the video from 01:22:33 to
01:22:36 during the time that the object is crossing in front of the airport's tarmac. In those
frames the object's inner temperature is represented by pixel values from 1 to 23 and its outer
temperature pixel values from 6 to 79. The tarmac registers a cooler temperature with pixel
values of 100-115. There is no question that the object's temperature is significantly above
the ambient and its center zone is usually hotter than its outer areas. Whether the object's
warmest temperature is only 105 degrees Fahrenheit or hotter cannot be determined without
knowing the specific algorithm used by the manufacturer of the thermal video which controls
the video's pixel values.
41
April 25 2013
Maneuverability The object's ability to maneuver at speeds of 80-100 miles per hour at low
altitude is of interest. A notable characteristic of the object is its apparent tumbling as it
moves through the air. This tumbling appearance ends prior to the object's entry into the
water and as it moves through the water. At 01:23:37 hours in the video the object can be
seen to disappear behind a tree momentarily, placing its altitude at below 40 feet. The ability
to fly at that altitude at night and between trees requires precise control and a highly
responsive propulsion system particularly given the apparent lack of control by aerodynamic
devices (like wings). In terms of current technology, advanced GPS satellites and
sophisticated vision pattern recognition in communications with an on board microprocessor
might partially explain such maneuvers. More difficult to explain would be the willingness of
any government or organization to expose this capability by traversing a residential area
where malfunctions during flight could result in harm to the civilian population as well as
compromise an advanced military technology.
IV. DISCUSSION AND CONCLUSIONS

Summary The authors of this paper received this video in October of 2013 and have spent
more than a year and a half composing this report. Hundreds of hours of work has been spent
in gathering information and analyzing the video, radar data, witness claims, and other
information related to this event that took place in northwestern Puerto Rico on April 25,
2013. Additionally, our efforts took into consideration protection of the witnesses' identity;
it is unfortunate that stigma often accompanies the reporting of what is commonly termed a
UFO/USO, but what we have referred to as an UAP. It is hoped that the work that went into
this report will inspire future work in the identification of any phenomenon that displays
unusual technological capabilities.
We believe that there is sufficient information in the video to characterize this object as: three
to five feet in size; the shape is circular to oval but changes; air speed varies from 70 to 120
mph; capable of changing direction; internal temperature of about 105 degrees Fahrenheit
usually in the center of the object and exterior temperatures above the ambient air
temperature; capable of traveling at low altitude through a residential area; able to enter the
water with no obvious splash or impact; underwater speed varies from 39 to 95 mph; ability
to exit and re-enter water; and the capability of splitting into two independent parts that
appear to be the same size as the original object based on its infrared signature.
Examination of Possible Explanations Entire classes of animals and man-made objects
may be eliminated by comparison of properties attributed to the unknown object. The most
likely explanations of the unknown object are discussed here.
Examination of Possible Explanations: Hoax One of the first possibilities examined was
that of a hoax. The authors of this report spent hundreds of hours in review of the video,
detailed review of over 5,000 individual frames and that included pixel level enlargements.
During all of this analysis there was never any indication of pixel manipulation. The
background in the video corresponds, in extreme structural and geographic detail, to
42
April 25 2013
hundreds of square miles of the actual location. Every on-screen GPS position corresponded
to real locations verified by satellite images. Additionally, the location, date and time stamps
of independent radar data verified the locations, dates and times of the aircraft taking the
video. As noted earlier in this report in the section labeled Radar Analysis---Verification of
Thermal Imaging Video Information, the radar data supports that there was an aircraft at the
exact time and location indicated on the full three minutes of the video and the radar data
verifies the pilot's claim that he flew a circle around the base before continuing on his
standard surveillance mission along the Puerto Rican coast. The possibility of a hoax is
therefore considered extremely unlikely.
Examination of Possible Explanations: Any Lighter Than Air Devices (Not Powered)
The possibility that the object in the video was a balloon or any other windblown object is
discussed next. The appearance that the object is tumbling is a main attraction of the balloon
theory. Although it is difficult to give this serious consideration, as mentioned previously
there was a poor quality copy of this video released on the internet that caused a lot of
speculation with the leading explanation being a balloon carried by the wind. The balloon
theory posits that a balloon was basically stationary and the movement seen in the video was
actually that of the aircraft as it moved in a semi-circle around the balloon. There are multiple
reasons why the object in the video cannot be explained as a balloon and they are listed as
follows:
1. The object's speed was too great. The actual path of the object was derived in the section
of this report labeled Object Path. Based on portions of the video where the approximate
altitude of the object is known, it was straight forward to calculate the speed of the object
which approached 120 mph at times. The wind speeds were 8-13 mph out of the east at
ground level and 12-18 mph out of the northeast at elevations of 400 to 3200 feet---much too
slow to support a balloon explanation.
2. The object changed directions multiple times, which cannot be explained by a balloon with
winds out of the east or northeast. The pilot saw the object traveling from north to south and
the video also confirmed that. The object then turned east into the wind then headed back
north towards the ocean. Multiple directional changes cannot be explained by a balloon.
3. Temperature information from the thermal video indicates that the object was hotter than
the ambient and the center of it was near 105 degrees Fahrenheit with a cooler exterior. A
balloon would be near ambient temperatures and there would not be the large temperature
gradient as seen in the unknown object. Chinese lanterns would not display the area of heat
seen in the video nor could they continue to burn underwater nor could they split into two
parts while maintaining the same speed well in excess of the wind.
4. The object impacts the water, and this is clearly seen when analyzing the video frame by
frame. The object disappears into the water, travels underwater at an average speed of 82.8
mph then exits back into the air. Prior to impacting the water and after departing the water,
the object's heat signature was still present. A balloon or Chinese lantern cannot enter water
and stay underwater due to its buoyancy, and it certainly cannot maintain a temperature
hotter than the ambient during the process.
43
April 25 2013
5. The splitting of the object into two parts also eliminates a balloon as a possibility.
6. Line of sight movement seen during the video based on the latitude/longitude of the CBP
aircraft eliminates any possibility of the object being a balloon. Some arguments have been
advanced that the motion of the balloon relative to the background is an illusion created by
the motion of the plane circling the balloon. This argument is not valid since the object can
be seen moving past background objects even while the background was stationary in certain
frames. Further, angular analysis reveals that a balloon, traveling at 15 mph, could be no
farther away than 1520 feet from the plane. Using the on-screen GPS data across frames, this
relatively short distance creates major line of sight inconsistencies. The details of this
analysis are found in Appendix L.
The authors of this report do not consider a wind-blown object as a reasonable explanation
for the object in the video due to speed characteristics, directional changes, temperature,
buoyancy issues in water, splitting into two parts, and line of sight issues related the
movement of a windblown object.
Flying Animals One possibility that is supported by the object's temperature is the
possibility of some species of large fast flying bird. This could explain the ability to maintain
a temperature above the ambient, the capability to dive into water, and the ability to change
directions. The key to a bird explanation is the ability of a bird to fly continuously, without
diving, at an average speed of about 80 mph, maximum speeds of up to 120 mph, and the
ability to dive into water at high speed. There are three large birds capable of sustained
horizontal speeds of 80 mph: the golden eagle, the grey-headed albatross, and the peregrine
falcon. The golden eagle is about three feet tall with a wingspan of 6-8 feet and its average
horizontal speed is 28 to 35 mph with a maximum horizontal speed of 80 mph. It does not
live near dense populations of humans and is not native to Puerto Rico. The grey-headed
albatross is almost three feet in size with a 7 foot wingspan and has been recorded flying
horizontally for eight hours at 79 mph with a South Atlantic tail wind. This albatross is native
to the colder areas of the South Atlantic and South Pacific near Antarctica. Lastly, the
peregrine falcon does visit Puerto Rico during the winter. Its body is one to two feet in size
and it has a three foot wingspan. Although it can dive at extreme speeds, its average
horizontal speed is 40-56 mph with a maximum horizontal speed of 65-68 mph. None of
these birds, along with being native to the area, fits all the characteristics of speed and size of
the unknown object. Nor is there ever any indication of flapping wings during this three
minute video, which would be expected over that period of time especially when making
directional changes. Additionally, none of these birds are capable of moving underwater at a
speed of 95 mph. The fastest swimming bird is the Gentoo penguin at a paltry 22 mph. There
is no type of flying animal that can mimic the object seen in the video.
Aircraft Including Drones An explanation worthy of consideration would be some new
type of military drone that is perhaps launched from an ocean platform such as a ship or
submarine. The size of the object at three to five feet fits into the drone category as does its
speed through the air. Currently the Navy is working on a drone capable of traveling in water
and air. The goal is to basically fly as an airplane, splash down and become a submarine,
44
April 25 2013
according to an aerospace engineer at the Naval Research Laboratory. 20 With both flying and
swimming characteristics, it is referred to as a Flimmer. The current model can fly at 68 mph
and a swimming speed that has not yet been tested. It is reasonable to suspect that the Navy's
current capabilities exceed what is released to the public.
It may be within our current technological capability to
build a drone that can match the air and water speeds of
the object in the video. Although the splash seen in the
photo is quite large, there are torpedoes capable of
minimizing their interaction with water and it would not
be unreasonable to suspect that the same capability might
Rendition of the Navy's new
be possible with an advanced drone. This could
Flimmer (fly and swim)
potentially explain the speed, movement capabilities
over land, underwater movement, and the seeming lack
of interaction with the water on impact.
Still to be explained is the thermal heat signature of the
object and its ability to split in half. There is no
indication of the power sources that would be expected
with the familiar types of drones or an air/water drone
such as the Flimmer. Any type of internal combustion
engine, jet engine or rocket would have been
Early Flimmer model
consistently detectable by the thermal imaging system as splashing into the Potomac
seen with the automobiles and jets on the tarmac. This
characteristic eliminates a drone such as the Flimmer. However, the absence of any
comparable heat signature could be addressed by some type of drone more similar to an
artillery shell than an aircraft. It is conceivable such a projectile could change direction
multiple times after launch and perhaps that is related to the apparent tumbling action. It
might even be possible for such a projectile to enter the water with very little impact. As
mentioned before, there are missiles and torpedoes designed with that capability. An
advanced drone without a power source and that is launched could explain the changes in
direction, low altitude maneuverability among trees if equipped with advanced
vision/GPS/navigational technology and the ability to impact water with very little
disturbance. But a non-powered drone cannot explain the increases and decreases in speed
that occur multiple times during the object's flight, the ability to enter and leave water, nor
can it explain the ability to split into two parts. Lastly, a drone powered by a lithium battery
or low temperature fuel cell might be able to fly with a minimal heat signature. A lithium
powered torpedo has traveled at over 50 mph underwater. 21 The military has published aerial
drones such as the Wasp III with speeds of 40 mph so it would expected that classified
battery powered military drones might reach speeds in the air of 100 mph. It might be
possible that a new drone has been developed that can travel in water and air at the speeds of
the object in the video. Still, this leaves two capabilities that such a drone would need to
match the characteristics of the unknown object: the ability to move in and out of water at
high speed; and the ability to split into two parts with both sections capable of independently
Signal Magazine, Fast-Flying Flimmer No Underwater Fluke. December 1,
2014.
21
DCNS Jan. 23, 2013. http://en.dcnsgroup.com/news/a-torpedo-powered-by-alithium-ion-battery-breaks-speed-records-in-complete-safety/
20
45
April 25 2013
traversing through air and water.

Careful consideration has been given to the drone theory but the authors do not believe that
it is a sufficient explanation of what was seen in the video. There are four arguments against
a drone as a possible explanation. The first; the authors' question that an advanced drone
would be tested at night over civilian areas where there is possible exposure of advanced
technology and the risk of loss of the drone when the same testing could be safely achieved
over a military operating area. It is also difficult to believe that our military would accept the
inherent safety risks of flying a drone across airport runways at low altitude with commercial
jets active on the tarmac. Second is the ability of the object to maintain a high speed, even
accelerate, underwater, along with the ability to move back and forth between air and water.
Third is the appearance of the object in the video; there is nothing in the video to indicate the
presence of any type of wings. Lastly, which begs explanation, how one drone splits into two
distinct drones of the same size as the initial drone.
Commercial or military aircraft larger than 8 feet are summarily dismissed as impossible
given the maximum size of the unknown object as well as some of the same arguments listed
in the previous paragraph.
Recommended Actions Further examination and study of this video is warranted as well
as collection of additional information. There are remaining questions. Are there additional
videos from the Puerto Rico area from other dates in the possession of the CBP as claimed by
the anonymous source discussed earlier? Could the FAA provide copies of the Tower logs?
Could the military provide radar from the radar facility located at the airport?
Efforts should continue to enlist professional and academic help for work already done as
well as suggestions for further research into this video. There remains more follow up that
can be done on the software algorithms used in the infrared video system. We should
continue efforts to obtain technical manuals for this equipment. What cannot be gleaned from
technical manuals that we may or may not obtain, we should attempt experimental projects to
determine the visual and measurable limitations of this camera's infrared technology.
Methodologies need to be developed and refined to measure, within this video, object
altitudes, angular and absolute velocities and acceleration both in the air and underwater.
These accelerations need to be compared to the object temperature fluctuations to determine
any correlations. This will help determine any possible relationship between the heat emitted
by the object and its motions. Means of measuring the curvature (sharper turns) in the path of
the object should be explored.
Conclusion The study of unknown anomalous phenomenon often referred to as UFOs
usually carries with it many negative connotations. This negativity has often been brought
about due to claims of aliens and little green or grey men in our midst. These types of claims
are then further dramatized and stigmatized in the media. This stigma prevents the type of
open minded evaluation of aerial phenomena that needs to be undertaken. It is this stigma
along with the fact that this object could be labeled a UFO and a USO that we choose instead
to refer to this as Unidentified Anomalous Phenomenon. This bypasses the arguments and
instead focuses on the fact that it does not fit any logical classification commonly used.
46
April 25 2013
Logically, there should be nothing negative associated with the study of a video that displays
an object that appears to be capable of movements not readily explainable by current
technology. If others can establish a plausible explanation that reasonably accounts for the
characteristics of the object in this video, then so be it. But any explanation must be
supported by a detailed report and not assertions or what-ifs. This video is the best
documentation of an unknown aerial and submerged nautical object exhibiting advanced
technology that the authors of this report have seen.
V. ACKNOWLEDGEMENTS
Robert Powell
B.S. Chemistry
Research Analyst
Morgan A. Beall
B.S. Earth Science
Coordinator
Daina Chaviano
Masters English Literature
Special Field Investigator
Larry Cates
B.S. Mathematics
Computer based analysis & statistics
Carl Paulson
B.S. Physics
Nuclear Physicist Grumman/Lockheed
Richard Hoffman
B.A. Communications
I.T. Strategic Planner Defense Dept
Refer to Appendix A for author biographical information.
47
APPENDIX
48
APPENDIX A
Author Acknowledgements
49
Author Acknowledgments
Appendix A
Robert Powell has a B.S. in Chemistry from Southeastern Oklahoma State University. He has
28 years experience in engineering management in the semiconductor industry. While
working at Advanced Micro Devices he took numerous internal courses related to device
physics, design of experiments, and statistical analysis. He helped the company develop its
first flash memory technology used in todays flash cards. His experience includes managing
a state-of-the-art chemistry laboratory and managing a Research and Development group that
worked on nanotechnology using atomic force microscopes, near-field optical microscopy,
and other techniques. Mr. Powell is also a co-holder of four patents related to
nanotechnology. He has held the position of Director of Research at MUFON (Mutual UFO
Network) since 2007 and is also the head of MUFON's Science Review Board. He is a
member of the National Space Society and the Society for Scientific Exploration.
Morgan Beall has a B.S. in Earth Science from Frostburg State University. He has worked in
the field of environmental consulting and OSHA/EPA regulations for the last 12 years. He
has learned and developed team building skills and discipline both in the classroom and in
the university athletic program. Mr. Beall has been a MUFON field investigator since 2009
and has had vested interest for many years in the investigative field of the UFO phenomenon.
He has conducted many investigations throughout Florida both privately and publicly
through MUFON. He is currently the MUFON State Director for Florida, a MUFON STAR
Team investigator, and has held held many project management positions within the MUFON
organization.
Larry Cates has a B.S. in Mathematics from Jacksonville University. He has 40+ years of
programming experience; systems analysis, application and hardware development. He did
statistical work for Florida Health and Rehabilitative Services which included design of statistical methods and setting up data analysis queries. Mr. Cates has investigated and researched the UFO phenomenon for over eight years including fielding investigations and
analysis of radar data.
Carl Paulson has a B.S. magna cum laude in Physics and spent four years working on a PhD
in Physics at New York University's Dept of Physics Graduate program. He previously
served four years in the United States Air Force. During his career he designed power
systems for the Tokamak fusion reactor at the Princeton Plasma Physics Laboratory and then
worked for Grumman Aerospace in their Advanced Energy department located at Princeton.
This work led to similar positions at Northrop-Grumman and Lockheed Martin.
Simultaneously, Mr. Paulson provided the physics designs for multiple linear accelerators.
Those designs included the entire CWDD linear accelerator at Argonne National Laboratory
and portions of the BEAR accelerator which was placed and operated in space. He is a
MUFON STAR Team investigator and is a member of the MUFON Science Review Board.
Richard Hoffman has a B.A. in Organizational Communications from Wright State
University. He is an Information Technology consultant and strategist. He has been a defense
contractor for over 20 years working primarily for the Army Materiel Command HQ with a
variety of companies. Currently, Mr. Hoffman is a Senior Engineer supporting the U.S. Army
Network Enterprise Technology Command (NETCOM) at Redstone Arsenal in Huntsville.
He has 51 years of experience as an investigator, researcher, writer, and presenter on the
50
Author Acknowledgments
Appendix A
subject of the UFO phenomenon. He was one of the first to join MUFON in 1969 and is
currently the MUFON STAR Team Manager as well as the State Director for Alabama.
Daina Chaviano has a B.A. in English Language Translation from the University of Havana
in Cuba. She began her career as a literary and cultural consultant. While in Cuba she
published several fantasy and science fiction books and became the best-selling author in
both genres in the island's history. She has lived in Miami, Florida, for the last 23 years and
has won several international literary awards. Her works have been translated into 30
languages.
51
APPENDIX B
WESCAM Model MX-15D
52
WESCAM Model MX-15D
Appendix B
53
WESCAM Model MX-15D
Appendix B
54
APPENDIX C
FOIA Requests and Replies
55
1.0
Appendix C
Initial Letter - Investigator Larry Cates Letter to USAF Regarding 84 RADES
56
2.0
Appendix C
Response E-mail - Acknowledgment Letter - FOIA Case # 2014-0638
57
3.0
Appendix C
Phone Call Reference Response - FOIA Case# 2014-4053
58
4.0
Appendix C
FOIA Case# 2014-0638-F Official Response
59
4.1
Appendix C
FOIA Case# 2014-0638-F Example of data provided
60
5.0
Appendix C
Initial Request Letter - Investigator Daina Chaviano
61
6.0
Appendix C
FAA Certified Mail - Return Receipt - FOIA Case# 2014-008277(ES)
62
7.0
Appendix C
April 2014 Request - Regarding 84 RADES
Dear Sir or Madam:

This is a request under the Freedom of Information Act. I am willing to pay up
to $50 for this request for the cost of duplication. If fees will exceed this
amount, please contact me first.
This request is to the 84 Radar Evaluation Squadron (84 RADES) for all primary
and secondary(transponder) surveillance radar information related to the
continuous time period of 23:00hrs Zulu Time on April 25, 2013 through
02:00hrs Zulu Time on April 26, 2013. I request that a radar data extraction
be produced using the following latitude and longitude coordinates: a boundary
box of upper left N19 00 00, W68 00 00 to lower right N18 00 00,
W066 00 00. If possible, please send radar data on a CD in a text or excel
format with data such as date, time, transponder code or lack of, range,
azimuth, altitude, longitude, and latitude.
If you need any additional information regarding this request, please feel
free to contact me via email or telephone.
In order to determine my status to assess fees, you should know that my fee
category is: an individual seeking records for personal use and not for
profit. Thank you for your consideration of this request.
Regards,
Larry Cates
678 773-5449
63
8.0
Appendix C
FOIA Case# 2014-0638-F Official Response - 84th Radar Evaluation Squadron
64
APPENDIX D
Delay of Fed Ex Flight 58
information provided by Flight Stats
65
Appendix D
The Landed On-Time statement refers to FedEx flight 58's arrival at its
destination on time. The delay in departure is shown by the scheduled time
of 9:10pm vs the actual time of 9:26pm.
1
66
Appendix D
67
APPENDIX E
Auxiliary Witness Communications
68
Appendix E
1.0: Anonymous Email to John Greenewald and the Back Vault

This information is location on BlackVault site
(http://www.theblackvault.com/m/events/view/
Anonymous-Letter-Authenticates-Puerto-Rico-UFO-Video-And-Sets-Record-Straight) and
sent sometime in October 2014. Items of interest are highlighted in yellow.
---------------------------------------------------------------------------------------------------------------Recently, I received a letter from an anonymous source, authenticating a UFO video that has
circulated for a couple months. Although he claims that the video is authentic, he does
mention the information circulating about the video's origins are false, and seems to have
quite a bit of knowledge about the videos origin, the technology used, and how the
information about the video that is circulating (like it was shot from a Black Hawk
helicopter) is actually not true. This letter is in regards to a UFO video, shot by an infrared
camera, in Puerto Rico. Special thanks to Jorge Martn, Journalist and UFO researcher in
Puerto Rico, for this higher resolution, and clearer, version of the UFO video: The
anonymous letter is also below:
The anonymous letter, is as follows:
Hello John,
I was reading about your involvement in getting documents from the NSA concerning UFO's.
Sir, if you want undeiable proof of UFO's on earth from a government source it is in the
video below. Try a FOIA request for this video that was leaked onto youtube.
I can vouch that the following video is 100% real. I am remaining anonymous to avoid
government reprisals. https://www.youtube.com/watch?v=Hee70AwwUJ8
If you ever wanted to truly see how aliens are monitoring or studying us, and witness their
technology, this is it. The video is a black and white infrared recording using a L3 MX15
EOIR camera. The IR video uses back hot, meaning the blacker something is the hotter, or
fuller of energy it is. The video was leaked onto youtube because the Federal Agents who
recorded it realized the Federal Government was not interested in disclosing it. The video
was recorded from the screen of a laptop using a iPhone type device. Spanish is heard as
background noise to distort the leakers voices so they are not discovered, so disregard all
audio. Lastly, the poster of the video speculates it was recorded from a CBP Blackhawk. In
reality it was recorded by a CBP DHC8 turboprop maritime patrol aircraft. This is not a
maybe, this is 100% alien technology on earth, in our skies, and under our oceans. The video
was taken in Aguadilla Puerto Rico, and can be verified by calling CBP Caribbean Air and
Marine Branch, and Aguadilla airport control tower (the UFO was over the airport without
permission and tower controllers saw it and tracked it on radar.
69
Appendix E
On April 25, 2013 at 2122 Local/April 26, 2013 0122Z a Customs and Border Protection
Caribbean Air and Marine Branch DHC8 maritime patrol aircraft was on a routine patrol
when it encountered a UFO immediately after takeoff. The object was spotted visually by the
Captain of the aircraft, and the Aguadilla control tower operator. It appeared to have a
strange red light. The Customs and Border Protection crew thought the aircraft might be a
smuggler so they began to follow the UFO, and record a IR video. The UFO circled the
Aguadilla airport and made its way to the ocean. Initially the UFO appears as a forward
flying horseshoe, then as it makes its way to the ocean, it changes its configuration to a more
spherical shape. The UFO skims on top of the ocean, and submerges, unaffected by the
hydrodynamic forces. Watch the video carefully at 01:24:39 (time in upper left corner) when
the object gets really dark (hot) another UFO actually pops out of the ocean and joins the
original UFO in formation. Then both UFOs make controlled entries into the ocean. Alien
technology is no doubt under the ocean near Puerto Rico!
Final note from The Black Vault: Unfortunately, I can not verify the above letter. There was
no contact information whatsoever, and it was sent via my online contact center. Although a
false email address was given, I was able to verify the IP address (unique number given to
every computer on the internet) was from the Miami area - but I will not list the IP address
for obvious privacy reasons. I feel by disclosing in the Miami area - is not a breach of
anything.
70
Appendix E
2.0 Anonymous Email to Florida MUFON State Director from Alias John Truth
to:Xxxxxxxxxxxxxxxxx (Morgan Beall personal email)
date:Sun, Aug 17, 2014 at 3:43 PM
subject:MUFON Florida: UFO recorded by Customs and Border Protection Aircraft
Having worked for CBP I can vouch that the following video is 100% real. I am remaining anonymous to
avoid government reprisals. https://www.youtube.com/watch?v=Hee70AwwUJ8
If you ever wanted to truly see how aliens are monitoring or studying us, and witness their technology,
this is it.
The video is a black and white infrared recording using a L3 MX15 EOIR camera. The IR video uses
back hot, meaning the blacker something is the hotter, or fuller of energy it is. The video was leaked
onto youtube because the Federal Agents who recorded it realized the Federal Government was not
interested in disclosing it. The video was recorded from the screen of a laptop using a iPhone type
device. Spanish is heard as background noise to distort the leakers voices so they are not discovered,
so disregard all audio. Lastly, the poster of the video speculates it was recorded from a CBP Blackhawk.
In reality it was recorded by a CBP DHC8 turboprop maritime patrol aircraft. This is not a maybe, this is
100% alien technology on earth, in our skies, and under our oceans. The video was taken in Aguadilla
Puerto Rico, and can be verified by calling CBP Caribbean Air and Marine Branch, and Aguadilla airport
control tower (the UFO was over the airport without permission and tower controllers saw it and tracked
it on radar.
On April 25, 2013 at 2122 Local/April 26, 2013 0122Z a Customs and Border Protection Caribbean Air
and Marine Branch DHC8 maritime patrol aircraft was on a routine patrol when it encountered a UFO
immediately after takeoff. The object was spotted visually by the Captain of the aircraft, and the
Aguadilla control tower operator. It appeared to have a strange red light. The Customs and Border
Protection crew thought the aircraft might be a smuggler so they began to follow the UFO, and record a
IR video. The UFO circled the Aguadilla airport and made its way to the ocean. Initially the UFO appears
as a forward flying horseshoe, then as it makes its way to the ocean, it changes its configuration to a
more spherical shape. The UFO skims on top of the ocean, and submerges, unaffected by the
hydrodynamic forces. Watch the video carefully at 01:24:39 (time in upper left corner) when the object
gets really dark (hot) another UFO actually pops out of the ocean and joins the original UFO in
formation. Then both UFOs make controlled entries into the ocean. Alien technology is no doubt under
the ocean near Puerto Rico!
71
Appendix E
3.0 Youtube Account Anonymous Informant
The statement below was placed in the YouTube commentary on roughly June of 2014 by
an individual who has some inside knowledge of the event because he indicates that the
video was taken by the U.S. Customs and Border Patrol. The location of the YouTube site
at the time of the postings is https://www.youtube.com/watch?v=Hee70AwwUJ8
Interestingly, this individual claims that there are two other videos made in the same area.
He/she also claims the video was analyzed in Quantico, Virginia.
Two emails were sent to Red Bill through his YouTube homepage. The emails asked if he
could communicate with the researcher of the current investigation. No reply has been
received to the date of this report.
The YouTube page has no information in it and it appears that Red Bill created this name
for the express purpose of making his YouTube comment.
--------------------------------------------------------------------------Red Bill
2 months ago
No se cmo tiene ese video, no fue en un helicptero. Fue en un avin de Us Customs and border
protection. El video original es en blanco y negro, tiene audio, fue examinado en Quantico, Virginia. Hay
2 video ms y son de la misma rea, de diferente fecha y como punto de referencia es la playa surfers
beach.
TRANSLATION:
No way in that video, it was not a helicopter. It was an airplane of the US Customs and
Border protection. The original video is in black and white, has audio, was examined in
Quantico, Virginia. There are 2 more videos and are of the same area, different date and
point of reference is the beach surfers beach.
72
Appendix E
Below is a second comment made by Red Bill on a different YouTube site that has posted
the same video. This comment was made in July of 2014. The site of that YouTube video
was the following at the time of this report:
https://www.youtube.com/watch?v=Pm-Sg_J_hB8
In this comment, Red Bill claims that he/she was there. It is difficult to tell if he/she means
that he was on the airplane or at the airport when this occurred.
Saludos, fue un avin. Lo que gira es la cmara. El avin de color gris de Aduana. Para las dudas yo
estaba ah, cuando paso. Desde el ao pasado esta y hay dos videos de diferentes fechas. Si quiere
preguntar dos semanas antes los vecinos de la Base Ramey llamaron a la polica en relacin a unas
luces que salan del mar. Varios policas llegaron a ver esas luces. Todo se quedo en secreto. Hay una
playa que tiene un portn, hay noches que lo cierran. En ese lugar puedes ver esferas de luces, no
todo el tiempo, pero pasa a menudo.
TRANSLATION:
Cheers, it was a plane. It is the camera that rotates. The plane is a gray colored Customs. For the
doubters, I was there when it happened. From this last year, there are two videos of different dates. If
you want to ask two weeks before residents of the Ramey Base called police regarding some lights
coming out of the sea. Several policemen came to see those lights. All remained in secret. There is a
beach that has a gate, there are nights that close. In this place you can see fields of lights, not all the
time, but it happens often.
73
APPENDIX F
Analysis of Radar Information
74
Appendix F
Radar Information on the Puerto Rico Thermal Video

Introduction
The purpose of this radar analysis is to verify the legitimacy of the video by identifying the
aircraft that took the video through correlation of exact times and radar locations of the
aircraft taking the video. A search will also be made to identify any unknown aerial objects
that were detected on radar.
1.0 Acquisition of Radar Information

The thermal system's latitude/longitude coordinates indicated that the video of this unknown
object occurred over Puerto Rico and the time stamp indicated that the video was taken
between 01:22:07 UTC+1 and 01:26:01 UTC+1 on April 26, 2013. Based on this, a request
was made for all primary and secondary radar data related to the continuous time period of
23:00hrs Zulu Time on April 25, 2013 through 02:00hrs Zulu Time on April 26, 2013, from
FAA radar sites in the vicinity of Puerto Rico. The request included information such as date,
time, transponder code or lack of, range, azimuth, altitude, longitude, and latitude.
Radar data was received for that time period from the following radar sites:
QJQ located 92 miles to the east southeast at Pico Del Este, Puerto Rico, at an elevation of
3417 feet and 1816'07N 6545'31W.
SJU located 75 miles to the east in San Juan, Puerto Rico, at an elevation of 20 feet and
1827'06N 6559'29W.
STT located 144 miles to the east in St Thomas, Virgin Islands.
The data included a time stamp for each radar contact, type of radar beacon, azimuth/range
bearings, latitude, longitude, transponder identification, and altitude. This information is
sufficient to verify if the thermal information matches with an aircraft at the exact time and
location as shown on radar, thus verifying the validity of the thermal video.
2.0 Radar Analysis of Aircraft Matching Time/Location of Thermal Video

It is a straightforward exercise to determine if there is an aircraft on radar that is an exact
match for the aircraft that filmed the thermal video. The thermal video provides the exact
time and location of the aircraft as it was taking video of the unknown object. The radar data
can verify if an aircraft was present at the same time and location. If there was an aircraft
present, then there is no doubt that the video in question was taken by an actual aircraft in
maneuvers over Puerto Rico.
75
Appendix F
The aircraft's location at specific times was obtained from the thermal video and is
represented on the Google Map in Figure 1. The initial thermal video shown in Figure 2
indicates that the aircraft was mile east of the Rafael Hernandez airport at 01:22:07 hours
at an altitude of 1875 feet. The aircraft departed in an easterly direction, turned towards the
north, passed over the coastline in a westerly direction, and finally headed to the south along
the coastline. The last frame of the thermal video shows the aircraft located one mile to the
southeast of the village of Moca at 01:26:01 hours and at an altitude of 4523 feet.
The radar data was examined and an aircraft was detected that matched the signature of the
aircraft that created the thermal video. The transponder number on this aircraft was 4406.
This aircraft showed up on the radar data from all three radar sites. Data from the QJQ radar
site at Pico Del Este was used to correlate against the thermal video. The following table
displays latitude and longitude coordinates from a portion of both the thermal and the radar
data. All of these coordinates are at the same time and are within mile of each other. No
other aircraft was in this area with a similar flight pattern at the time. The date based on Zulu
time is 04/26/2013 for this table.
Time
(Zulu)
thermal Lat/Long
Radar Lat/Long
thermal
Alt
Radar Alt
Speed
01:22:08
1830'11N 6705'48W 1829'52N 6705'51W
1875'
1600'
01:22:25
1831'00N 6706'05W 1830'44N 6706'01W
1912'
1700'
209 mph
01:22:59
1831'09N 6708'04W 1831'09N 6707'55W
1784'
1500'
230 mph
01:23:23
1830'19N 6709'14W 1830'30N 6709'05W
2075'
1700'
239 mph
01:23:49
1828'53N 6709'41W 1829'06N 670'39W
2491'
2300'
238 mph
01:24:07
1827'53N 6709'27W 1828'05N 6709'37W
2561'
2300'
236 mph
01:24:42
1826'20N 6708'15W 1826'27N 6708'25W
3222'
2900'
228 mph
01:26:01
1823'06N 6705'43W 1823'16N 6705'53W
4523'
4200'
211 mph
TABLE 1: Correlation of thermal data to radar data

The transponder code on this aircraft is 4406. This indicates that it is a military or law
enforcement aircraft. FAA Order 7110.66 stipulates that all transponder codes between 4401
and 4433 will be controlled by FAA Order 7110.67, which is named Special Aircraft
Operations by Federal, State Law Enforcement, Military Organizations and Special
Activities.
Based on the radar data, there is no doubt that the thermal video is a real video taken by a law
enforcement or military controlled aircraft. Because we have time and distance information
in Table 1, the speed of the aircraft can be calculated. The aircraft's speed varied from 209
mph to 239 mph, which indicates it is not a helicopter but is a plane.
76
Appendix F
FIGURE 1: Tracking aircraft's location based on video and radar data. The name Tracker
represents the aircraft and the value to the right is Zulu time at that location.
77
Appendix F
Zulu Time and

Video magnification level
Date
Unk Target in
crosshairs
FIGURE 2: Initial thermal video

Aircraft's latitude and
longitude
Aircraft altitude and

bearing
Target distance above

sea level
Target bearing and
Crosshairs sited latitude/
distance
longitude
It is also worth noting that the aircraft made what appears to be an extra search pattern over
the ocean to the north and northwest of the airport before commencing what is probably its
standard patrol and operational activities down the Puerto Rican coast. Although the thermal
video shows the aircraft's path for only four minutes, the radar data shows the aircraft's path
prior to and after the thermal video was engaged. In the Google Map in Figure 3 the aircraft
images in blue represent both the matching thermal and radar data while the aircraft images
in red are only radar data. It is clear that before the thermal video was engaged (red colored
planes), the aircraft circled to the north and northwest of the airport and then engaged the
thermal video on its second pass of the airport (blue colored planes). This may indicate that
the pilot was aware that an unknown target was in the area, searched for the unknown target,
and upon finding it, engaged the thermal video system prior to resuming the aircraft's normal
course down the coast. Data that will be discussed in the next section indicates that there was
a potential reason for the pilot to suspect there was an unknown object in the area.
78
Appendix F
FIGURE 3: Radar only data of a law enforcement or military aircraft shown in red with
Thermal Imaging & Radar data of the aircraft's location show in blue
3.0
Radar's Minimal Detection Elevation near Airport
It has been determined in conversations with the witnesses that the initial indication of an
unknown object came from the Raphael Hernandez Control Tower. That means that at least at
some point, the object had to be high enough such that the radar system used could resolve it
as a target. Since not all agencies are required to answer FOIA (Freedom of Information Act)
requests, this investigation could not determine which radar system made this determination.
The only radar system in the area, other than a military radar system, that could have detected
the target is Pico Del Este located at 3417 feet altitude and approximately 91.5 miles from the
middle of the Raphael Hernandez airport and on the opposite (eastern) end of Puerto Rico. It
is therefore important that the minimum detection height of the radar system be determined.
It is known that in standard conditions the vertical gradient of the index of refraction
decreases with height. This tends to bend the beam down toward the Earth's surface. It is also
79
Appendix F
known that being a oblate spheroid, the Earth's surface tends to fall away from a horizontal
line. Therefore a radar beam will tend to travel farther than normally thought; however, since
the second effect is normally larger than the first, the beam will still slowly move away from
the Earth's surface.
Further complicating this effect is beam divergence. This means the beam spreads as it
moves away from its antenna and widens the envelope in which the beam is able to resolve
targets. Since this spreading is a function of the design of the beam's antenna, this effect
cannot be calculated without knowledge of the complete design plans of the system.
Although the exact beam envelope cannot be determined, given data of targets seen by the
radar system, it is possible to make an approximation of the envelope. Figure 4 is a 2 hour
plot of height verses range of all transponders up to 10,000 feet regardless of azimuth and
time. The maximum height of 10,000 feet was chosen since it highlights the sloped area
under review. This plot will help determine the minimum resolving height of the radar at the
target's location. The red dashed line to the right in Figure 4 is the height based on distance
from the radar. It was found to have a slope of ~114.34 ft/mile and an intercept of ~88.3
miles. The distance of the unknown targets to the northwest of the airport in Figure 4 vary
from 94 to 104 miles from the radar at Pico Del Este. At those distances the radar's minimum
detection altitude would vary from 652 feet to 1795 feet. A target directly over the airport
(91.5 miles from the radar) would need a minimum altitude of 366 feet to be detected.
Figure 4: Minimal altitude detection level of the Pico Del Este radar system QJQ
80
4.0
Appendix F
Radar Analysis of Unknown Target in Area Prior to Aircraft Launch
The data from radar site QJQ was reviewed for any primary data without a transponder code
that identified an unknown radar track in the area of interest. Primary radar tracks are those
created by the actual reflection of the radar beam from a target. Known aircraft such as the
law enforcement or military aircraft that took the thermal video will transmit a transponder
code, also known as secondary radar. Primary radar tracks are identified with the designation
Sch as shown in column 2 (Msg Type) of Table 2. As can be seen in that table, the radar
picked up 50 primary radar tracks of what appears to be a single object from Zulu time
00:58hrs to 01:14hrs, a 16 minute period of time. The CBP aircraft departed the runway with
instructions to look for an unknown to the northwest of the airport at 01:16hrs. These 50
radar tracks (the radar sweeps every twelve seconds) of this unknown object are visually
displayed in Figure 5. The amount of information requires considerable commentary.
The first four radar strikes occurred after each twelve second sweep of the radar and can be
seen at the far left area of Figure 5 and are designated as a, b, c, and d. The object is not
picked up in the next four sweeps, which equates to 48 seconds. The fifth radar strike
designated as 1a++++ (the four plus signs indicate that four previous radar sweeps were
missed) indicates the unknown is at the same location as it was one minute earlier. That
doesn't necessarily mean that the object is stationary because the accuracy of the radar is only
to within 1/8 mile. A list of the radar target's longitude/latitude locations, speed and its
direction of movement is shown in Table 3. Due to the potential 1/8 mile (660 feet) possible
error in the primary radar, the speeds when the object is traveling less than 1200 feet could
vary by almost 100% therefore the speeds shown in Table are not meant to be accurate. A
statistical analysis of those speed numbers gives a mean of 168 mph with a standard
deviation of 97 mph. The altitude of the object is not known but based on the minimum
altitude in which the radar picked up the tracking aircraft, the object must be at 652 feet
altitude or higher.
The sixth radar strike occurs immediately after the fifth radar strike, i.e. the next twelve
second sweep of the radar. Beginning with this sweep of the radar, the object shows up on
almost every sweep of the radar for the next ten minutes; however, if the sixth radar strike
was created by the same object that created the first five then its speed is a minimum of 1700
mph. This calculation is not significantly affected by the radar error due to the large distance
of 30,000 feet that was traveled. This perceived speed is unlikely so it is possible that the first
six radar strikes are not related to the next large set of radar strikes identified as 1b thru 1aq.
Symbols 1b through 1aq are in chronological order and represent the results of each
subsequent 12 second radar sweep. Each plus sign following a symbol indicates that there
was no strike on the previous radar sweep.
The group of radar strikes from 1b through 1aq cover 42 radar hits during 10 minutes. These
radar hits occur with almost every sweep of the radar and they are all in the same general
area.
81
Appendix F
Radar Site: QJQ 26 Apr 2013

Time
MsgType
00:58:16.909
00:58:28.874
00:58:40.902
00:58:52.899
00:59:52.882
01:00:04.941
01:00:16.922
01:00:40.900
01:00:52.866
01:01:04.909
01:01:16.875
01:01:28.871
01:01:40.837
01:01:52.880
01:02:04.970
01:02:16.842
01:02:28.855
01:02:40.883
01:02:52.910
01:03:04.954
01:03:16.997
01:03:29.103
01:03:41.178
01:03:53.159
01:04:05.203
01:04:17.231
01:04:29.243
01:04:53.361
01:05:05.405
01:05:29.507
01:05:41.504
01:05:53.485
01:06:05.529
01:06:17.541
01:06:29.678
01:06:41.737
01:06:53.656
01:07:05.668
01:07:17.696
01:07:29.708
01:07:41.705
01:08:17.835
01:08:41.766
01:09:05.838
01:09:53.762
01:10:29.643
01:10:41.780
01:12:53.743
01:13:53.836
01:14:17.923
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Sch
Rng(nmi)
85.25
85.375
85.875
85.625
85.625
81
81.125
81.25
81
81.25
81.125
81
81.25
81.375
81.375
81.5
81.125
81.125
81.25
81.5
80.875
82
81.625
81.5
82
81.625
81.375
81.625
81.625
82
82
81.875
82.125
82.125
82.125
82.125
82
82.125
82
82.375
81.75
82.5
82.75
82.5
83.125
82.625
83.375
83.5
83.75
84
Az(deg)
280.811
280.635
280.371
280.459
280.459
281.25
281.338
281.777
281.426
280.986
281.602
281.514
281.338
281.162
281.426
281.074
281.162
281.162
281.162
280.898
281.074
281.074
280.898
281.338
280.986
281.074
281.162
280.986
280.635
280.811
280.635
280.635
280.283
281.162
280.635
280.283
280.898
280.459
280.635
280.547
280.811
280.898
280.371
280.283
280.107
278.086
280.02
280.02
279.844
279.844
Lat
Lon
18.31.48.965 N
18.31.34.812 N
18.31.16.718 N
18.31.21.911 N
18.31.21.911 N
18.31.39.674 N
18.31.48.439 N
18.32.26.646 N
18.31.54.354 N
18.31.20.395 N
18.32.10.482 N
18.32.01.690 N
18.31.49.862 N
18.31.36.533 N
18.31.58.658 N
18.31.30.543 N
18.31.33.732 N
18.31.33.732 N
18.31.35.133 N
18.31.15.756 N
18.31.23.597 N
18.31.36.098 N
18.31.17.122 N
18.31.52.708 N
18.31.28.660 N
18.31.31.932 N
18.31.36.533 N
18.31.24.528 N
18.30.54.891 N
18.31.13.778 N
18.30.58.887 N
18.30.57.555 N
18.30.30.367 N
18.31.44.934 N
18.31.00.219 N
18.30.30.367 N
18.31.21.220 N
18.30.45.297 N
18.30.58.887 N
18.30.55.399 N
18.31.11.069 N
18.31.26.683 N
18.30.44.320 N
18.30.34.225 N
18.30.25.533 N
18.27.27.119 N
18.30.20.472 N
18.30.21.723 N
18.30.08.979 N
18.30.11.435 N
067.13.32.938 W
067.13.43.643 W
067.14.19.057 W
067.14.02.078 W
067.14.02.078 W
067.09.01.791 W
067.09.08.066 W
067.09.08.238 W
067.08.58.828 W
067.09.21.664 W
067.09.03.561 W
067.08.57.329 W
067.09.15.816 W
067.09.26.518 W
067.09.22.072 W
067.09.35.733 W
067.09.11.009 W
067.09.11.009 W
067.09.18.764 W
067.09.38.618 W
067.08.56.951 W
067.10.06.759 W
067.09.46.379 W
067.09.31.316 W
067.10.08.216 W
067.09.43.489 W
067.09.26.518 W
067.09.44.940 W
067.09.50.624 W
067.10.11.095 W
067.10.13.925 W
067.10.06.158 W
067.10.27.217 W
067.10.13.044 W
067.10.21.692 W
067.10.27.217 W
067.10.09.661 W
067.10.24.478 W
067.10.13.925 W
067.10.38.629 W
067.09.55.569 W
067.10.40.705 W
067.11.04.719 W
067.10.50.542 W
067.11.32.140 W
067.11.28.651 W
067.11.49.044 W
067.11.56.826 W
067.12.15.055 W
067.12.30.625 W
TABLE 2: Raw radar data that shows targets northwest of the airport
82
Appendix F
FIGURE 5: Radar plot of unknown that showed up off shore prior to the departure of the
aircraft with thermal imaging capabilities. Tracks are designated in order of time beginning
with a-d (segregated because of distance from the other radar tracks), followed by 1a-1aq, and
followed by 2a, 3a, and 4a (segregated because of significant time delays of greater than one
minute between radar tracks.
The radar sweeps every twelve seconds. Each + after a radar hit indicates that the target was
not detected in the previous radar sweep. A designation such as 1ac,1af indicates that two
different radar sweeps occupied approximately the same physical location to within 1/8 of a
mile of each other.
83

Elapsed
Appendix F
--------------------- GPS Points ------------------- Points
Time
---- Start Point ----
----- End Point -----
Heading
Radar
Distance (ft) (Degrees) Speed
11.965 sec
From 18.530268
-67.225816
To 18.526337
-67.228790
1798.27
215.6
102.47
12.028 sec
From 18.526337
-67.228790
To 18.521311
-67.238627
4029.98
241.7
228.44
11.997 sec
From 18.521311
-67.238627
To 18.522753
-67.233911
1799.25
72.1
102.26
59.983 sec
From 18.522753
-67.233911
To 18.522753
-67.233911
0.00
---
0.00
12.059 sec
From 18.522753
-67.233911
To 18.527687
-67.150498
30482.85
86.4
1723.51
11.981 sec
From 18.527687
-67.150498
To 18.530122
-67.152241
1092.37
325.8
62.16
23.978 sec
From 18.530122
-67.152241
To 18.540735
-67.152288
3871.82
359.8
110.10
11.966 sec
From 18.540735
-67.152288
To 18.531765
-67.149674
3408.47
164.6
194.21
12.043 sec
From 18.531765
-67.149674
To 18.522332
-67.156018
4146.99
212.5
234.78
10
11.966 sec
From 18.522332
-67.156018
To 18.536245
-67.150989
5397.00
18.9
307.52
11
11.996 sec
From 18.536245
-67.150989
To 18.533803
-67.149258
1092.07
146.1
62.07
12
11.966 sec
From 18.533803
-67.149258
To 18.530517
-67.154393
2224.02
236.0
126.72
13
12.043 sec
From 18.530517
-67.154393
To 18.526815
-67.157366
1732.21
217.3
98.07
14
12.090 sec
From 18.526815
-67.157366
To 18.532961
-67.156131
2286.90
10.8
128.97
15
11.872 sec
From 18.532961
-67.156131
To 18.525151
-67.159926
3167.60
204.7
181.92
16
12.013 sec
From 18.525151
-67.159926
To 18.526037
-67.153058
2526.17
82.3
143.38
17
12.028 sec
From 18.526037
-67.153058
To 18.526037
-67.153058
0.00
---
0.00
18
12.027 sec
From 18.526037
-67.153058
To 18.526426
-67.155212
798.58
280.8
45.27
19
12.044 sec
From 18.526426
-67.155212
To 18.521043
-67.160727
2811.32
224.2
159.15
20
12.043 sec
From 18.521043
-67.160727
To 18.523221
-67.149153
4296.45
78.8
243.24
21
12.106 sec
From 18.523221
-67.149153
To 18.526694
-67.168544
7186.55
280.7
404.75
22
12.075 sec
From 18.526694
-67.168544
To 18.521423
-67.162883
2821.86
134.5
159.34
23
11.981 sec
From 18.521423
-67.162883
To 18.531308
-67.158699
3915.92
21.9
222.85
24
12.044 sec
From 18.531308
-67.158699
To 18.524628
-67.168949
4463.28
235.5
252.67
25
12.028 sec
From 18.524628
-67.168949
To 18.525537
-67.162080
2527.56
82.1
143.28
26
12.012 sec
From 18.525537
-67.162080
To 18.526815
-67.157366
1781.84
74.0
101.14
27
24.118 sec
From 18.526815
-67.157366
To 18.523480
-67.162483
2228.21
235.5
62.99
28
12.044 sec
From 18.523480
-67.162483
To 18.515248
-67.164062
3058.05
190.3
173.12
29
24.102 sec
From 18.515248
-67.164062
To 18.520494
-67.169749
2822.50
314.2
79.85
30
11.997 sec
From 18.520494
-67.169749
To 18.516358
-67.170535
1536.02
190.2
87.30
31
11.981 sec
From 18.516358
-67.170535
To 18.515988
-67.168377
798.56
100.3
45.44
32
12.044 sec
From 18.515988
-67.168377
To 18.508435
-67.174227
3484.95
216.3
197.29
33
12.012 sec
From 18.508435
-67.174227
To 18.529148
-67.170290
7691.67
10.2
436.59
34
12.137 sec
From 18.529148
-67.170290
To 18.516728
-67.172692
4615.24
190.4
259.27
35
12.059 sec
From 18.516728
-67.172692
To 18.508435
-67.174227
3076.48
190.0
173.94
36
11.919 sec
From 18.508435
-67.174227
To 18.522561
-67.169350
5451.73
18.1
311.86
37
12.012 sec
From 18.522561
-67.169350
To 18.512583
-67.173466
3937.82
201.4
223.52
38
12.028 sec
From 18.512583
-67.173466
To 18.516358
-67.170535
1743.61
36.4
98.84
39
12.012 sec
From 18.516358
-67.170535
To 18.515389
-67.177397
2528.22
261.5
143.51
40
11.997 sec
From 18.515389
-67.177397
To 18.519741
-67.165436
4643.46
69.0
263.90
41
36.130 sec
From 18.519741
-67.165436
To 18.524079
-67.177974
4839.82
290.0
91.33
42
23.931 sec
From 18.524079
-67.177974
To 18.512311
-67.184644
4934.68
208.3
140.59
43
24.072 sec
From 18.512311
-67.184644
To 18.509507
-67.180706
1763.64
126.9
49.95
44
47.924 sec
From 18.509507
-67.180706
To 18.507092
-67.192261
4306.39
257.6
61.27
45
35.881 sec
From 18.507092
-67.192261
To 18.457533
-67.191292
18083.37
178.9
343.62
46
12.137 sec
From 18.457533
-67.191292
To 18.505687
-67.196957
17688.18
353.6
993.67
Skips
+++++
++
++
++
+++
++
++++
+++
TABLE 3: Time, location, and direcitonal movement of targets northwest of the airport
84
Appendix F
The unknown target(s) seen on the QJQ radar were likely the same unknown target(s) that
caused the control tower to request the CBP aircraft to look in the area to the northwest of the
airport. Figure 5 shows that the CBP aircraft flew directly into the area where the unknowns
were picked up on radar. It is reasonable to consider that the pinkish-reddish light seen to the
northwest of the airport by the pilot may have been the cause of the unknown target seen on
radar in the same area and that same object picked up by the CBP aircraft's thermal imaging
video. If those objects were not related then it is a very unusual coincidence that an unknown
object on video was in the same area as an unknown target on radar during a similar period of
time.
The unknown target that was detected on radar for 16 minutes does not display
characteristics that would be expected of an aircraft in flight. The speed variation and sudden
changes in direction are nonsensical. Nonetheless, there are characteristics that can be
attributed to the unknown target. First, this target's appearance on radar occurs at the right
time and location to likely be the object detected by the control tower and the resulting
subsequent alert to the CBP aircraft. Second, although the target jumps around, its overall
directional movement is from the northeast to the southwest. Third, the target strength is
strong as it is detected on almost every sweep of the radar for eight of the ten minutes it is on
radar. Lastly, the target is no longer detected on radar during the time that the unknown is
detected on the thermal imaging video. At that point in time the object is below the Pico Del
Este radar's detectable altitude.
Conclusion
The authors of this report have examined other explanations for the unknown radar strikes to
the northwest of the airport. A temperature inversion is a possible cause of false radar returns.
These occur when the upper air temperature is higher than lower air temperature. This
possibility was examined and discounted due to the lack of any temperature inversion layer
in the area. A copy of the upper atmospheric conditions was obtained and is shown in Table
4. One of the strongest arguments against some type of anomalous propagation is the
continuation of the radar returns within a small geographic area for ten minutes and with
almost every 12 second sweep of the radar, the lack of these returns prior to this incident, and
the lack of these returns after the incident of the unknown object recorded on thermal video
at a lower altitude over land. It seems reasonable to consider the possibility that the control
tower decision to vector an aircraft into the same area as this unknown radar return, the
detection of these unknown radar returns on FAA radar data, the visual by the pilot of an
unknown object with a red light, and the detection of the unknown object on the thermal
video are all related to the same event and the same object. No other reasonable explanation
has yet been found.
85
Appendix F
78526 TJSJ San Juan Observations at 00Z 26 Apr 2013

-----------------------------------------------------------------PRES
hPa
HGHT
m
TEMP
C
DWPT
C
RELH
%
MIXR
g/kg
DRCT
deg
SKNT
knot
THTA
K
THTE
K
THTV
K
55
60
60
65
64
60
60
60
60
59
56
49
45
42
40
38
34
32
31
30
5
330
324
314
310
316
321
328
332
335
334
331
325
327
333
337
339
345
344
343
340
325
310
305
321
325
315
305
303
300
300
304
308
310
310
285
289
300
285
281
280
275
278
280
4
12
12
16
15
13
13
11
9
9
8
7
7
6
6
6
7
7
7
7
7
7
7
7
7
8
9
10
10
11
11
10
9
9
10
11
11
12
13
14
18
17
16
16
22
24
28
32
35
39
39
37
35
34
34
37
36
32
34
33
33
44
48
50
297.7
298.1
298.1
298.1
298.1
298.9
299.3
299.8
300.2
300.2
299.9
301.2
302.3
302.4
302.5
302.6
302.9
302.7
302.7
302.9
304.4
304.9
305.1
306.6
306.6
306.6
307.0
307.2
307.7
307.9
308.1
309.1
309.5
309.6
311.5
311.8
312.8
314.7
315.8
316.5
317.1
317.9
318.2
318.9
321.0
321.2
322.6
323.1
323.4
324.6
325.0
325.3
328.0
328.5
328.8
330.6
331.3
332.2
332.7
333.1
333.3
335.1
335.4
335.9
342.8
343.2
343.1
342.9
342.9
339.3
337.8
335.4
333.6
333.4
334.7
327.6
328.1
332.3
331.8
333.0
327.8
333.3
333.8
332.9
327.3
329.0
329.9
326.4
326.9
327.5
325.1
327.4
323.1
324.3
325.5
318.9
318.9
318.1
316.6
319.6
316.8
316.5
316.8
317.4
318.0
318.9
319.1
319.6
321.4
321.6
323.1
323.7
324.0
325.1
325.5
325.9
328.2
328.8
329.0
331.0
331.7
332.6
333.1
333.4
333.6
335.4
335.7
336.2
300.4
300.9
300.9
300.9
300.9
301.4
301.6
302.0
302.2
302.2
302.0
302.8
303.9
304.2
304.3
304.4
304.4
304.6
304.6
304.8
305.8
306.4
306.6
307.8
307.8
307.9
308.0
308.4
308.6
308.9
309.1
309.7
310.1
310.1
311.8
312.2
313.0
314.8
315.8
316.6
317.1
318.0
318.3
318.9
321.0
321.2
322.6
323.1
323.4
324.6
325.0
325.3
328.0
328.5
328.8
330.7
331.3
332.2
332.7
333.1
333.3
335.1
335.4
335.9
-----------------------------------------------------------------1015.0
1000.0
980.9
947.3
943.0
925.0
914.5
898.0
882.6
881.0
876.0
865.0
858.0
853.0
850.0
845.0
832.0
827.0
823.0
821.4
809.0
792.1
783.0
770.0
763.9
757.0
752.0
744.0
740.0
736.4
733.0
721.0
700.0
696.0
683.0
674.0
670.0
658.7
653.0
641.0
610.8
566.0
550.0
544.7
528.0
523.9
500.0
484.3
475.0
465.4
462.0
435.0
410.0
400.0
395.6
363.9
353.0
320.0
300.0
284.0
280.5
256.1
252.0
250.0
3
137
305
610
649
816
914
1071
1219
1234
1283
1390
1459
1509
1539
1589
1721
1772
1812
1829
1957
2134
2232
2372
2438
2514
2569
2658
2702
2743
2781
2918
3161
3208
3362
3471
3519
3658
3729
3881
4267
4877
5106
5182
5426
5486
5850
6096
6245
6401
6458
6914
7357
7540
7620
8230
8453
9144
9600
9975
10058
10668
10777
10830
25.8
25.0
23.4
20.4
20.0
19.2
18.6
17.6
16.5
16.4
15.6
15.8
16.2
15.8
15.6
15.2
14.2
13.6
13.2
13.2
13.4
12.1
11.4
11.4
10.8
10.0
9.8
9.2
9.2
9.0
8.8
8.4
6.4
6.0
6.2
5.4
5.8
6.2
6.4
5.6
2.3
-2.9
-4.9
-5.1
-5.7
-6.1
-8.5
-10.5
-11.7
-12.3
-12.5
-16.7
-18.9
-20.3
-20.9
-25.4
-27.1
-33.2
-37.3
-40.7
-41.3
-46.1
-46.9
-47.1
20.8
20.5
20.2
19.6
19.5
17.5
16.5
15.0
13.7
13.6
14.3
9.8
9.2
11.4
11.0
11.5
8.2
11.2
11.4
10.8
6.4
6.8
7.1
3.4
3.7
4.0
1.8
3.2
-0.8
0.0
0.8
-7.6
-8.6
-10.0
-16.8
-11.6
-20.2
-30.4
-35.6
-37.4
-37.9
-38.6
-38.9
-41.2
-48.7
-48.2
-45.5
-45.0
-44.7
-46.0
-46.5
-45.7
-54.9
-55.3
-55.0
-52.9
-52.1
-52.2
-52.3
-54.7
-55.0
-57.5
-57.9
-58.1
74
76
82
95
97
90
88
85
84
84
92
68
63
75
74
79
67
85
89
85
63
70
75
58
62
66
57
66
50
53
57
31
33
31
17
28
13
5
3
3
3
4
5
4
2
2
3
4
5
4
4
6
3
3
3
6
7
13
19
21
21
26
27
27
15.49
15.43
15.41
15.38
15.38
13.79
13.11
12.08
11.31
11.23
11.83
8.86
8.58
10.02
9.79
10.18
8.26
10.20
10.39
10.02
7.50
7.91
8.14
6.38
6.57
6.78
5.83
6.52
4.90
5.23
5.56
3.01
2.87
2.58
1.51
2.35
1.15
0.47
0.28
0.24
0.24
0.24
0.24
0.19
0.09
0.09
0.13
0.14
0.15
0.13
0.13
0.15
0.05
0.05
0.06
0.08
0.09
0.10
0.10
0.08
0.08
0.06
0.06
0.06
86

245.0
239.0
233.6
200.0
197.0
187.0
186.0
184.8
169.0
157.0
150.0
124.8
118.7
115.0
114.0
107.0
100.0
86.8
82.3
80.6
79.0
75.0
74.1
71.5
70.0
65.4
63.2
10964
11128
11278
12290
12388
12725
12759
12802
13376
13844
14130
15240
15545
15733
15785
16155
16550
17374
17678
17801
17916
18216
18288
18489
18610
19002
19202
-46.9
-47.3
-48.0
-52.9
-53.3
-52.7
-52.9
-53.0
-55.5
-58.3
-60.7
-69.2
-71.5
-72.9
-72.5
-74.1
-72.9
-75.7
-76.7
-77.1
-75.9
-76.5
-77.1
-78.9
-77.9
-73.9
-73.5
-65.9
-71.3
-71.6
-73.9
-74.3
-78.7
-78.9
-79.2
-82.5
-83.3
-84.7
-87.6
-88.4
-88.9
-88.5
-90.1
-88.9
-91.7
-92.7
-93.1
-91.9
-92.5
-92.9
-93.9
-92.9
-89.9
-90.1
Appendix F
10
5
5
6
6
3
3
3
2
3
3
5
7
7
7
7
7
7
6
6
7
7
7
7
8
7
6
0.02
0.01
0.01
0.01
0.01
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.00
281
283
285
290
291
295
295
295
297
299
300
280
255
263
265
280
300
340
335
7
37
116
135
185
245
165
125
53
56
59
64
65
68
68
68
59
51
46
28
35
37
38
43
30
9
14
13
12
9
8
7
6
3
2
338.1
339.9
341.1
348.8
349.7
355.9
356.2
356.6
361.7
364.6
365.3
369.7
370.8
371.5
373.2
376.9
386.6
397.1
401.0
402.6
407.4
412.2
412.4
412.8
417.4
434.3
439.6
338.3
340.0
341.1
348.9
349.8
355.9
356.2
356.6
361.7
364.7
365.3
369.7
370.8
371.5
373.2
376.9
386.6
397.1
401.0
402.6
407.4
412.2
412.4
412.8
417.4
434.3
439.6
338.2
340.0
341.1
348.8
349.7
355.9
356.2
356.6
361.7
364.6
365.3
369.7
370.8
371.5
373.2
376.9
386.6
397.1
401.0
402.6
407.4
412.2
412.4
412.8
417.4
434.3
439.6
TABLE 4: Upper Air Wind Conditions; San Juan, PR.

Univ of Wyoming, Dept of Atmospheric Sciences.
87
APPENDIX G
Object Location, Speed, Size
88
Appendix G
Introduction
The object in this video was tracked using a state of the art Wescam MX-15D multi-sensor
multi-spectral targeting system. The MX-15D is mounted on the underbelly of a DHC8 turbo
prop aircraft operated by U.S. Customs and Border Patrol. This system has high definition
thermal imaging, short range IR for enhanced haze penetration, a laser rangefinder and
illuminator, and stabilization features. The video lasted for about three minutes and due to
familiar objects in the background, we were able to identify the approximate size, speed, and
path of travel of the object. The camera's video output included the latitude/longitude
coordinates, azimuth heading, and the altitude above sea level of the tracking aircraft. It also
provided the latitude/longitude of any object within the crosshairs of the camera, the altitude
above sea level, and the distance in nautical miles of any object in the crosshairs of the
camera. Due to the capabilities of this particular camera its sale outside of the United States
requires approval from the U.S. Government.
The video consists of 3 minutes and 54 seconds of video imagery of which 2 minutes and 56
seconds displays an unknown object arriving from over the ocean, transversing land, and
then disappearing back into the ocean. The entire video was broken into individual frames for
analysis of the unknown object. There were a total of 7027 frames with each frame equating
to 1/30 of a second exposure. Breaking the video into individual frames allowed for detailed
evaluations of the object's characteristics.
Specific information will be provided as to how the size, speed, and location of this object
were determined. The basic determinations hinge on the trigonometry related to the actual
object size, angular size, and distance of the object. If two of those variables are known then
the third variable can be calculated using trigonometry.
1.0 Angular size of pixels in the video frame

The angular size of an object represents the angle subtended by an object in the sky. This is
measured in degrees, arcminutes, and/or arcseconds. As an example, the angular size of the
moon is approximately 0.5 degrees or 30 arcminutes or 1800 arcseconds.
The angular size of any object in the camera frame can be calculated using the number of
pixels that comprise the object's apparent length. The pixel size is constant for the camera as
long as the magnification is constant. Changes in magnification in the camera result in
linearly proportional changes to the size of the pixels.
In order to determine the angular size of a pixel, frame 892 was used. (See Figure 1.) This
frame displays a tank of known size. The tank is 108 feet in diameter based on satellite
photos from Google Earth. The ground distance between the tank and the camera is
calculated using the latitude and longitude of the tank and of the aircraft. The aircraft's
location in this frame is 18 31' 21 N and 67 06' 42 W, while the tank is located at 18 29'
02 N and 67 08' 29 W. Using the haversine formula, the distance between the two points is
calculated as 17,441 feet. The air to ground distance from the camera to the tank can be
calculated since it is represented by the hypotenuse of a right triangle. The aircraft's altitude
89
Appendix G
is 1760 feet so the actual distance between the camera and the tank is 17,530 feet. Now that
we have the size of the tank and its distance, we can calculate the tank's angular size based on
the trigonometric properties of a right triangle. The tangent of the angle (angular size) is
equal to the opposite side the angle (width or diameter of the tank) divided by the adjacent
size (distance to the tank). Solving tan = 108 / 17441 gives an angle of 0.35299. Image
processing software developed at the National Institute of Health and known as ImageJ was
used to analyze the image and calculate the diameter of the tank in pixels. The diameter of
the tank was equal to 238 pixels. The angular size represented by each pixel is therefore
equal to 0.35299 / 238 or .001483 degrees.
The angular size of the unknown object in the videos will vary with distance. But with the
value of .001483, we can determine the the object's angular size in any video frame, even if
the zoom factor (focal length) changes since the pixel size will be proportional to the zoom
factor.
Figure 1: Frame 0892. Known tank and unknown object.
90
Appendix G
2.0 Size of the object

As one watches the video at regular speed the object appears to tumble and if one watches
carefully, it even appears to change shape. Viewing of the object frame by frame makes it
much easier to see the sometimes rapid changes in shape and apparent size. Any change in
apparent size of the object will add an error into the calculation of the object's distance when
using a known angular size and a hopefully constant actual object size. Later in this
appendix, an error factor will be used in calculating the object's actual locations during the
video.
The size of the object can be determined when its distance is known. The angular size of the
object is known based on the discussion in Section 1.0. The distance can be accurately
calculated whenever the object is at very low altitude. This occurs towards the end of the
video when the object passes behind a telephone pole, behind trees, and then finally enters
the water. During these periods of time the crosshairs of the camera, via its laser range finder,
are providing an accurate distance measurement because the object is at a very low altitude
and there is no longer any error due to the crosshairs actually focusing on an object that is
potentially far from the camera. For example, in the right triangle shown in Figure 2 the
camera is at point 'C.' The crosshairs of the camera are pointing towards 'A'. Any object in
the crosshairs (represented by point 'D') of the camera could be at any location along line
'AC'. However, when foreground objects such as trees or a telephone pole or the water
surface itself interact with point 'D' then one knows that point 'D' is close enough to point 'A'
(point 'A' is on the ground) to allow for a reasonably accurate determination of the distance
from the object to the camera.
Figure 2: Right triangle
An example of the calculation will be discussed utilizing Table 1, which reflects information
from Frame 5085 in Figure 3. In this frame the object is just above the water as three seconds
previous to this frame the object was in the water. The distance to the object using the
system's laser range finder is 5.2 nautical miles, which equates to 31,595 feet. This value is
accurate to within 304 feet because the distance in nautical miles on the screen is rounded to
the nearest .1 nautical miles. The use of the latitude/longitude coordinates that the thermal
imager displays for the object and for the aircraft provides a distance that is accurate to
91
Appendix G
within 70 feet of the object's location and probably much closer. Using the haversine
formula, the distance between the locations of the aircraft and the object is 31,469 feet. The
object's angular size in video frame 5085 consists of six pixels, which at .000149 per pixel
as described in Section 1.0. is an angular size of .00894. With a known distance and a known
angular size, the actual size of the object can be calculated. As described in Section 1.0, the
tangent of the angle (angular size) is equal to the opposite side the angle (length of the
unknown object) divided by the adjacent size (distance to the unknown object). In this frame,
the calculated size of the object is 4.9 feet.
Table 1: Information and calculations from Frame 5085
92
Appendix G
Figure 3: Frame 5085. Object in cross-hairs and just above the water.
Calculations of the object's size were done on multiple frames whenever the object was a
known distance from the ground which allowed accurate values of the object's distance and
its angular size. These values varied significantly from a minimum size of 3.0 feet to a
maximum size of 5.2 feet. The variation in size is believed to be due to either varied angular
sides of the object as it is tumbling or temperature variations that are reflected in the shape
that the object presents to the IR camera. Calculations done on known objects in the video
such as water tanks, aircraft, cows, and moving automobiles eliminate issues with the
accuracy of the IR camera as a significant source of the variations in size. We can
conclusively say that this object is between 3.0 feet to 5.2 feet in length.
3.0 Path of the object during the video

The path taken by this object during the video cannot be ascertained simply by plotting the
latitude/longitude coordinates that are displayed by the thermal imager based on the crosshairs. Those coordinates are driven by a laser range finder, which is not necessarily striking
the object itself but the ground and other large objects in the background. As a result, when
the object is at altitudes above about 40 feet there can be significant differences in the actual
distance between the object and the camera. This was ascertained by careful observation of
the latitude/longitude values displayed on the thermal video as the object moved and
93
Appendix G
sometimes the cross-hairs in the thermal video were stationary, which resulted in
latitude/longitude values linked to the cross-hairs and not the object itself. It was clear that
the latitude/longitude measurements correlated to the farthest ground-based location that was
at the center of the cross-hairs. Referring back to Figure 2, the line AC represents the distance
as measured by the thermal imaging system while line DC represents the true distance
between the object and the camera.
The actual distance between the camera and the object can be determined using the angular
size of the object and the object's true size. The one exception is when the thermal video
system was at minimal magnification and the object consisted of only a 3-4 pixels in size.
The errors in the calculated distance values using angular size were gross and did not match
up with the quality of the system's information obtained at medium and high magnification.
The exact cause of this error is not yet known but is suspected to be related to minimal pixel
displays of object's in the infrared. Another method to determine the object's actual distance
is with a known altitude of the object. Whenever the object is near the ground or passing
between known objects then the actual distance can also be determined using the object's
known altitude and the azimuth of the object relative to the camera and aircraft. This method
is the most accurate because any errors in the object's size and angular size are eliminated.
The best determination that could be made of the object's actual path is shown by the brighter
of the three light blue lines in Figure 4 of this appendix. This figure is a Google Earth image
of the northwest coast of Puerto Rico. The airport that is seen in the image is the Raphael
Hernandez Airport. It is a joint civil-military airport located in Aguadilla, Puerto Rico. This is
the airport that is seen in the video. The top of the page faces west and the right hand side of
the page faces north. The dark blue aircraft icons are the actual location of the aircraft with
the camera as verified by both the thermal video system's latitude/longitude values and
locations as supplied by radar from the Pico Del Este radar site. The numbers next to the
aircraft represent the time in Zulu (aka Greenwich Mean Time) hours. A corresponding UFO
(UFO represents the unidentified flying object and is not meant to indicate any other quality
about the unknown craft.) location is on the map for the same time. The UFO locations
marked in red are exact locations of the object at those times due to accurate altitude values
being available. The UFO locations marked in orange represent approximate locations of the
object to within about 500 feet. The UFO locations marked in yellow with a time value next
to them and the darker light blue line connecting them represent a best guess of the object's
location based on the previous path of the object and its known direction from the aircraft.
This blue line ends at a question mark that represents that uncertainity and also is a possible
point of origin of the object. The object's route does raise the question of the possibility that
its origin could be the same as its final destination or its origin could be up to one mile
farther to the west as is shown in the other two light blue lines. Those lines connect UFO
locations that are also possible routes taken that are more westernly. The higher level of
uncertainity in the yellow colored UFOs is believed to be due to the thermal video system
being in operation at its lower magnification level.
The object's path is one that approaches the island of Aguadilla from the ocean. Its exact
origin is unknown. It crosses the airport runway once it comes over land and then re-crosses
the airport runway on its way back out to sea.
94
Appendix G
4.0 Speed of the Object

What is the energy source that propels the object in this video? The movements made by this
object require some type of power source. The object transverses about four miles during the
video and during that process changes direction from a southward direction to a northerly
direction. No type of propulsion is evident from the infrared video yet some form of
propulsion is required for the object to maintain and vary its speed, change directions
multiple times, and move in and out of the water. The source that propels this object is not
evident.
The speed of the object is most accurately determined during the latter half of the video when
the object's location can be more accurately determined as discussed in Section 3.0. The
calculation of the object's speed is straightforward as distance/time. The time between
measurements is provided by the thermal video system's clock and the location of the object
and its distance traveled is determined by the latitude/longitude locations provided by the
video system. The speed of the object was measured every eight seconds. The main error is
that the latitude/longitude values are in degrees, minutes, and seconds so that the location is
rounded to the nearest second, therefore the accuracy is to within 0.5 seconds (maximum
rounding error) of a degree. With eight seconds between speed measurements, the error due
to rounding could equate to 51 feet from the object's true location which could result in an
error in the object's speed of up to 4 mph.
95
Appendix G
Figure 4: Possible Paths taken by Object
96
Appendix G
Table 2 shows the time of the latitude/longitude measurement, the distance traveled since the
last measurement, and the calculated speed of the object. Although the speed of the object is
fairly constant and normally varies from 70 mph to 110 mph, it is clear that the object slows
and speeds up during this portion of the video, which again indicates some type of power
source should be present. Some of the speeds shown in the table are noted as being through
water.
ZULU TIME
DISTANCE
TRAVELED
SPEED
01:23:21
.3753 km
105 mph
01:23:29
.3448 km
96 mph
01:23:37
.2735 km
76 mph
01:23:45
.2459 km
69 mph
01:23:53
.2623 km
73 mph
01:24:09
.3211 km
90 mph
01:24:17
.3409 km
95 mph *
01:24:25
.3179 km
89 mph*
01:24:33
.3072 km
86 mph*
01:24:45
.2141 km
60 mph**
01:24:53
.1784 km
50 mph**
01:25:01
.2459 km
69 mph
* Speed under water.

**Speed through water and air.
97
APPENDIX H
Modeling of Object
98
Modeling of Object
Appendix H
Introduction
The amount of data provided in the IR video this study is predicated upon is extensive. In
most situations that is a desirable feature; however, when faced with a complete unknown it
tends to hide aspects that may have been obvious if there was less data. Combining data by
bundling multiple items into one is a feature provided by modeling. The data reduction
obtained generating the model permits a more efficient utilization of the data and allows
some aspects of the object to become obvious. Additionally, the thought that goes into
determining a model tends to highlight aspects not otherwise noted. All aspects determined
will be listed in the conclusion of the appendix.
Although it is philosophically possible to define a subjective 1 reality where anything dreamt
of by the observer is equally real, scientists define existence as objective. Objective reality
defines reality as that which does not require the participation of the observer. Effectively
that means that science is the end result of observation and measurement.
In essence this appendix is attempting to define an object conforming to what is seen in the
video while also obeying the laws of Physics; therefore, taking the physical world as real, the
appendix is also asking if the object is real. One of the characteristics of a real "object" is its
mass or more accurately its "invariant mass"2. As was stated in Appendix H, shape can be
defined as the characteristic surface configuration of the invariant mass. Therefore a defined
mass also implies an invariant shape. The difficulty in determining these objectives in the
video supplied is, it does not show mass; it shows heat.
Another problem encountered when looking at unknown objects in an IR video is the
difficulty in "seeing" all shades of the hot and cold areas of the object and in differentiating
them from the background. The images provided are what is termed 8-bit grayscale.
Grayscale indicates that the images are provided in various shades of gray. An 8 bit depth
indicates that each pixel will have one value out of 256 ( 2 8 ) possible shades (intensities) of
gray. More will be said about this later in this document. Although the shades are difficult to
distinguish using only the eyes, the computer has no problem distinguishing them. That
differentiation is provided by the "Surface Plot" function in the "ImageJ" program. That
function will be used extensively in this document.
Various views of the object will be considered in the document. The most obvious way to see
both hot and cold areas of the object is to look at the object in front of something else that
was also warm but not hotter than the hot portions of the object. In that case an outline of the
cold portions could be seen as a white shadow over the background heat while also seeing an
outline of the hot portions. In a sense this is similar to seeing a shadow of the object, al la
Lamont Cranston when he is in his Shadow identity. It is also possible to infer portions of an
outline from its effect on its environment. Different views will be presented in an attempt to
obtain clues to its shape.
It should be noted that the best this appendix can provide is a model of the object. It will be a
model that fits the aspects seen; but it is a model and not the object. As such, it can only have
validity in the subspace of reality that is described by the aspects used to create it. Expanding
the model beyond that subspace would be highly speculative and wasteful of time and
energy.
99
Modeling of Object
Appendix H
1.0 Reality of the Object

The first step in defining the shape of the object in the video is to determine if it is a real
object or is it just some odd juxtaposition of heat rays. That is accomplished by determining
if it has mass. Since the
Figure 1 - 1: Frame 3769 - Surface Plot
object in the video can't be weighed to determine its mass, a less direct indication has to be
used. Happily one such indication was described in Section 2 of the "Water Transit Appendix
J". In that document it was shown that in Frame #3769 a splash can be observed when the
object enters the water. That splash is the effect of an external mass displacing some of the
water mass. It was also noted in that section that the video does not show a normal (to us
rephrase?) picture of the world. It is IR and shows a heat picture. Therefore a determination
had to be made as to what a splash would look like in terms of heat rather than mass. A
splash is taking a volume of water and drastically increasing its surface area. Since both
evaporative and radiative heat transfer are proportional to the volumes surface area, a splash
provides a means to allow that volume of water to become a "little" cooler. Little is in quotes
because the change is very small and is basically invisible to the viewers eyes but not to the
FLIR system and to the computer.
The "Surface Plot" tool in "ImageJ" provides a three-dimensional view of the intensities of
pixels. It therefore converts the heat variations in the IR frame to height variations with (in its
default operation) the lighter (cooler) pixels being represented as hills and the darker pixels
as valleys. Figure 1 - 1 is the surface plot of a small area around the object in Frame #3769.
The cooler areas representing the splash are seen as raised areas around the upper right
corner of the plot. Although only an indirect indication, this figure indicates the object in the
video has mass and is therefore a real object with some constant characteristic shape or
surface configuration seen from various angles in various frames. It is also known that the
shape may not be possible to infer from the views seen in the video.
100
Modeling of Object
Appendix H
2.1 Initial View

Although not seen often in the video, in this view the object is seen to present itself in front
of a building window in an oblong or slightly triangular shape in Frame #2616. This is shown
in Figure 2.1 - 1. It is agreed that this figure is not particularly sharp but the general two
dimensional shape can be seen. A red circle was placed around the object to outline its
location.
A better view of the object can be seen in the Surface Plot view in Figure 2.1 - 2. It shows a
much more detailed picture of this view of the object. In the IR pictures the object in both of
the frames seems to be moving to the left with the hot area leading the colder area. In the
surface plots it can be seen that although there is a hot area in lower right, there seems to be a
cooler area behind it. Additionally, in terms of heat, the trailing cold area shows an almost
cone shape getting cooler as it goes back but with circular warmer waves around the cone.
There also looks like there may be a hot area above the cone to the front. Although the "cone"
shape is interesting, there are two items that must not be forgotten: this is heat not mass; and
this is only one view of the object.
Figure 2.1 - 1: Frame 2616
Figure 2.1 - 2: Frame 2616
There is an additional item of interest seen in Figure 2.1 - 2 that will have to be checked in
later views. There does not seem to be any heat being transferred from the unknown object to
the air surrounding it. That indicates a lack of turbulence surrounding and trailing the
unknown object. It almost looks like the object is slipping through the air with minimal
friction.
2.2 Second View

The most common view presented in the video is something that looks spherical. Frame
#1240 was chosen to represent this shape. It and the corresponding surface plot can be seen
in Figures 2.2 - 1 and 2.2 - 2. It should be noted the Figure 2.2 - 1 has been magnified to
~600% over the camera's basic magnification. Although this picture is a still, the hot (black)
and (white) portions of the object can be clearly seen. As stated the figure shows a roughly
101
Modeling of Object
Appendix H
spherical shape surrounded by a colder spherical portion. Additionally, the object seems to
have spokes of heat radiating out from the center to the outer edge of the colder section.
Although the object is unknown, when seen from this angle it almost looks like it is banking
in a turn. A quick look a the surrounding frames shows the object seems to be moving to the
left. It also should be noted that what looks like a dark outline around the object is an artifact
resulting from the magnification and the video compression. Although it took a long time for
this author to notice it, there is also a definite similarity between Figures 2.2 - 2 and 2.1 - 2.
Basically Figure 2.2 - 2 is the inverse of Figure 2.1 - 2. This is easily seen in Figure 2.2 - 3
where the elevations of light and dark areas have been reversed.
Figure 2.2 - 1: Frame #1240
Figure 2.2 - 2: Frame #1240 Surface Plot
Figure 2.2 - 3: Frame #1240 Reversed Surface Plot
As in the previous section, there doesn't seem to be any heat being transferred to the air
surrounding and trailing the object.
2.3 Third View

The third view to consider is one which is totally black. This can be an effect of a longer
distance between the object and the camera or of the object itself. Frame #1194 is an example
where the blackness of the object is not due to distance. It is observed in Figures 2.3 - 1 and
2.3 - 2. This view shows an almost dumbbell shaped object with very little (if any) cool
102
Modeling of Object
Appendix H
areas. The subject view also shows basically nothing outside of the black area. Unless there is
a portion of the object that has almost exactly the same heat signature as the background, the
shape shown is the outline of the object at this angle. Interestingly the object seems to be
moving directly to the left and not in the direction of either its long or short axes. Also again
no heat seems to be transferred from the object to the surrounding air.
Figure 2.3 - 2: Frame #1195 Surface Plot
2.4 Possible Shape

Although not yet complete, it is possible to use the views in 2.1 through 2.3 as a start to
determine a shape for the unknown object. Since it is hard to draw in three dimensions Figure
2.4 - 1 has been provided as a two dimensional view of the side of the object. The number at
the base of the figure labels the location along the side in degrees. Additionally the label
"Black" indicates a warmer area and the label "White" a cooler area. It is easily see that if
one looks at the half labeled 0 - 180 degrees, the object will appear to be black in the middle
with white at the bottom and going up the sides. Similarly, if one is provided with the half
labeled 90 - 270, the object will appear as white in the middle with black across the top and
going down the sides. Those 2 views therefore invert the object and reverse the colors. It
should be noted that the size of the undulating curve is only meant as illustrative.
Figure 2.4 - 1: Expanded View of the Object Side
To obtain the object seen in Figures 2.1 - 2 and 2.2 - 2, requires shrinking line B-0, B-360 to
almost a point and attaching the resulting sheet to the long side of an ovoid (acorn shape).
103
Modeling of Object
Appendix H
The result is illustrated in Figure 2.4 - 2. In this figure the dotted lines are only included to
show the 3 dimensional nature of the object. Although this shape is indicative of some of the
object's aspects, it is certainly not complete. At best this is a gross model of a completely
unknown object.
Aside from the angles seen in many frames (particularly in Figure 2.3 - 1), the object shown
in Figure 2.4 - 2 easily replicates the temperature outlines seen in Sections 2.1, 2.2 and 2.3.
As indicated above, observed at an angle from the right replicates the temperature outline of
Frame #1240. Rotate it a quarter turn around the B axis and it replicates the temperature
outline of Frame #2616. Finally seen from the left it would replicate the temperature outline
seen in Frame #1195. At this point the above is all that this model was created to do. The fact
that it duplicates the temperature outlines is not sufficient to consider it the model desired in
this appendix.
Figure 2.4 - 2: Ovoid
As many undergraduates have discovered, any attempt to fit a continuum to a finite number
of points is very dangerous. That is basically due to the fact the mathematically there are an
infinite number of solutions to any continuum when only discreet points are known. It also
does not matter how many discreet points are used to fit the solution to. There always remain
an infinite number of other solutions. Mathematicians call these problems, "ill-posed."
Fortunately a Russian mathematician named Andrey Tikhonov provided a regularization
procedure3 that gave a iterative process for obtaining a particular solution. The fact that this
appendix provides a pictorial model rather than writing an equation for it, does not eliminate
the ill-posed nature of the problem. Unfortunately it does however eliminate the possibility of
utilizing the above iteration procedure. Additionally there is no equivalent pictorial analog.
The only available process is to demand that all transitions from solution to solution be
accomplished by a continuous rotation of the object. Although that seems easy it necessitates
a viewing consistency which due to reticule and background interference, is not completely
provided in the video.
The lack of a viewing consistency is somewhat mitigated by the frame rate per second
(approximately 29 - 33 fps) of the camera. That means each frame is approximately 30 m-sec
in length. Although it is not known if the viewing angle of the object is changing due to a
rotation of the object or the varying relative locations of the object and the camera, there
seem to be long periods where the object presents views very similar to that seen in Frame
#1240. Therefore the rotational speed is either intermittent, very close to zero, or
104
Modeling of Object
Appendix H
synchronous with the frame rate. The problem with accepting the coincidence of
synchronicity is that there are areas where this view is seen to change quickly and the other
views do not seem to have the same unchanging long stretches. It is therefore believed that
the change is due to varying sighting angles possibly combined with transitory rotations of
the object. This will be checked later in this section.
Figure 2.4 - 3: Frame #0775: Cropped & Lightened
Although Frame #2616 provides what is probably the best view seen of this angle of the
object, it is not useful in looking for transitions to and from it. In this frame the object is
viewed in front of a building window. The background in and around this frame is rapidly
changing, making it very difficult to determine which effects are due to the object and which
are due to the background. To solve this problem it was determined that Frame #0775 (Figure
2.4 - 3) is basically the same view as #2616 but with a more constant background
Looking backward in time from Frame #0775, Frame #0760 represents a time approximately
450 m-sec prior to that of Frame #0775. A cropped version of that frame is shown in the
following Figure (2.4 - 4). Along with it, Figure 2.4 - 5 is a surface plot of the same view. As
can be seen this view shows an object which is entirely black or hot. (This is similar to Frame
#1194.) Therefore in 450 m-sec the object has transformed from one of the basic forms to
another. Although not shown here there is no intervening frame that shows anything other
than a continuous transformation.
Figure 2.4 - 5: Surface Plot - 0760
Looking forward approximately 1 second from the time represented by Frame #0775 to
Frame #0810 shows a second conversion to a view similar to that seen in Frame #1240. This
105
Modeling of Object
Appendix H
shows that the object seemed to rotate between the 3 basic views in approximately 1.5
seconds.
Figure 2.4 - 7: Surface Plot - #0810
Although it was determined above that that a rapid synchronous rotation cannot be occurring,
that did not eliminate rotations entirely. It was left open that it is possible that in addition to
varying sighting angles at least some of the changes could be attributed to slow and/or
transitory rotations. Figure 2.4 - 8 has been provided to check the possible effect of sighting
angles as the source of the above viewing changes. It shows the transformations discussed
above in terms of the locations of the aircraft containing the camera and the target the camera
is aimed at for Frames; #0810, #0775, and #0760. The lines connecting the targets and
aircraft have been provided to show the approximate viewing angles for each frame.
Although it is known the object is not located at the target location in any of these frames, it
is believed the targets provide a reasonable approximation for checks such as this. It is easily
observed that in this case the viewing change cannot be a attributed to variation of viewing
angle. This therefore proves that the object does rotate.
106
Modeling of Object
Appendix H
Figure 2.4 - 8: Relative locations of Aircraft and Target
2.5 Object Angularity

The problem with the shape discussed in Section 2.4 is that it assumes a completely smooth
object. Since the video is entirely IR, that isn't a particularly surprising assumption. As has
been emphasized, in IR one sees heat, not mass. Seeing an angle in an IR object does not
necessarily indicate the object has an angle in that location. It indicates the heat source has an
angle. In Section 2.4 a reference was made to angles seen in Figure 2.4 - 2 (the totally hot
view of the object). It is seen in that figure that the hot area looks sort of like two offset
overlapping squares. Even if there were no proof countering the assumption of a smooth
object, it has to be questioned
There are, however, a few locations in this video where angularity can be seen directly. They
are the times the object enters or exits the water. Concentrating on this entry period, each
frame that shows that entry effectively provides a horizontal slice of the object. If the camera
were looking directly downward, each of those slices would show the outline of the portion
of the object which is located at the water level. However, since it is known than in all frames
the aircraft is a distance off to the side of the object being viewed the camera never is looking
107
Modeling of Object
Appendix H
straight down at the object. It is aimed at an acute angle off of the horizon. Because of this,
even the frames entering or leaving the water can only show an outline of the side of the
object closest to the camera.
2.5.1 Entering the Water

Although Frame #3769 was used to illustrate the splash generated when the object entered
the water, the last frame that didn't show any indication of a splash at all was #3758. Figure
2.5.1 - 1 is a cropped portion of that frame showing the object magnified to 400% over the
camera's basic magnification. As can be seen, this view is similar to that seen in Figure 2.3 2. The object is almost all hot (black) and it shows a similar angularity. It remains to be seen
if the angularity is real or an effect of the extreme magnification.
Figure 2.5.1 - 1: Frame #3758

Cropped and Magnified
Figure 2.5.1 - 2 provides a composite of Frames #3762 - #3768. It has been provided in an
attempt to answer the above question. As previously stated, only the side closest to the
camera in these plots should be considered as indicative of the object's shape. The back
portion is heat from that portion of the object still above water. In these plots, the front of the
object is found in the lower left corner and the back in the upper right. Since although the
object may not be round, it has to be assumed that only the lower half of the left side can be
seen directly. Additionally, in order to see the totally hot (black) area of the assumed ovoid,
it has to be entering the water at an angle. In this case, the lower left should be entering the
water first. Since the frame labels can be considered as labeling time the figure shows an
object initially hitting the water in #3760 and sinking into the water as the numbers go up.
108
Modeling of Object
Appendix H
Figure 2.5.1 - 2: Surface Plots - Object Entering the Water
It is initially seen in #3760 that the lower left shows a sort of stair-step shape. Thus the
angularity discussed above is the actual shape of the object and not an artifact of the
magnification. Although there is no way to prove the following assertion, it will be assumed
that the object is symmetric and that the stair-step shape also occurs in the back. It should not
be forgotten that this is an assumption.
The second hint about the shape of the object is seen when looking at how the stair-step
shape changes as the surface plots move forward. As time moves forward and the object
sinks into the water, the edge moves in toward the middle and steps become less distinct.
Basically this is exactly what would be expected if the general shape of this portion of the
object was a dome.
Since the section of each frame shown in the surface plots was chosen by eye, the area
covered by the plots is not constant. Due to this the writer checked the sizes by magnifying
each object 4000x over the base magnification of the frames and counted the black pixels
running along the center line. The result was that for each increase by 2 in the frame number
there was a corresponding decease by 1 to 2 pixels in the pixel count.
2.5.2 Traveling through and Exiting the Water

There is a partial exit from the water in Frame #3912. There is very little information that can
be obtained during this period. In addition to emerging under the frame reticle, due to
molecular vibrational and rotational excitations, electromagnetic (EM) radiation is strongly
attenuated by water. Although it is well known that visible light can penetrate water, the same
is not true for the longer wavelengths in the EM spectrum. The attenuation of
109
Modeling of Object
Appendix H
electromagnetic radiation in water4 for the longer wavelength portion of the IR spectrum is
seen in log-log graph in Figure 2.5.2 - 1. Since the wavelengths used by FLIR lie in the
ranges 3 - 5 and 8 - 12 micro-meters, it is easily seen that for any frequency in those ranges,
the absorption coefficient is over 100 cm-1. (A similar graph5 specifically for seawater can be
found in the Notes section of this paper.)
Figure 2.5.2 - 1
The usefulness of the absorption coefficient is seen in calculation the intensity 6 of the
transmitted EM wave in water.
I ( x ) = I0 e- x ,
Where "x" is a distance in cm in the water, "" is the absorption coefficient (assumed
constant over the range), and I0 is the initial intensity. With a coefficient of 100, the intensity
drops by over 44 orders of magnitude in 1 cm and over 5 orders of magnitude in 1 mm).
Essentially this means these frequencies do not penetrate water.
Although in essence, this means that a layer of at most a mm around the object will absorb
virtually all of the IR generated, it doesn't answer the question of how it will affect what is
seen in an IR video. Since the object is pumping heat into the surrounding water, it would
seem reasonable to expect to see a sort of heat shadow of the object above the object and
trailing it. Due to the depth of the top of the object while underwater and its speed that is not
observed to occur. The speed obviously spreads out the heat but it is believed the depth is the
larger reason. If the depth of the object were less than the wave size (peak to trough), it
would break the surface as it moved and become visible as the water moved away from it.
Therefore its minimum depth must be greater than the wave size.
While the object is underwater, there is additionally a competing thermal effect which can be
seen in the video evidence as motion but not in the individual frames. It is a slightly cooler
area which seems to remain over the object while it is underwater. It is believed by the writer
that this is what is termed a Bernoulli Hump. It is basically a slight bulge in the surface due
110
Modeling of Object
Appendix H
to displaced water caused by the objects size and speed. The surface disturbance from a
Bernoulli Hump is approximately given by7:
Y = W(d, v, h) S
where "W" is a scaling function and "S" is a shape function. In these equations: "d" is the
objects diameter; "v" is the objects speed; and "h" is the objects depth. The shape function is
a function of front to back location along the object and for this appendix will be assumed to
have a value8 of 0.8. The equation7 for the scaling function is:
W = d2 v2/(8 g h2) ,
where g is the acceleration due to gravity. Therefore the Bernoulli Hump increases with
speed and cross sectional size and decreases with depth. Using a speed 9 of 82.812 mph and a
diameter of 3 ft., the following graph for the surface disturbance (Y) was obtained.
Figure 2.5.2 - 2: Bernoulli Hump Height as a function of Object Depth
The above figure shows that the Bernoulli Hump decreases with the depth of the object.
Since we are assuming it is the expansion of the surface due to this bump in the water that is
causing the decrease in temperature that we see in the video, the temperature change due to
the surface expansion also decreases with the depth of the object. Additionally as has been
said before, the temperature change is not visible in any single frame but its consistently
coordinated movement is easily followed in the video evidence.
As per Weather Underground (www.weatherunderground.com), the average wave height
(crest to trough) just off of the northern coast of Puerto Rico is 1-3 feet. Since it is possible to
see some of the waves in the both the video and in single frames, the difference between the
waves and the Bernoulli Hump provides a way to estimate the depth of the object.
Specifically it indicates that the Bernoulli Hump height is less than 1-3 feet in height. Thus
from Figure 2.5.2 - 2 for a height of <2 feet, the objects maximum depth is between 3 and 5
meters.
111
Modeling of Object
Appendix H
2.5.3 Second Entry

This water entry occurs after the initial object has divided into two objects. Figure 2.5.2 - 1 is
a surface plot of Frame #4677 of the two objects prior to the water entry. The object on top is
the one that will begin to enter the water 2 frames later. As indicated in this figure, the
existence of the cooler (white) areas show that these objects bear a closer resemblance to a
long end of the ovoide shown in Figure 3.4 - 2 than to the end view that was indicated in
Section 2.5.1. Since the viewing angle is different this section may provide additional clues
as to the objects shape.
Figure 2.5.2 - 1: Surface Plot - Frame #4677
As in Section 2.5.1, Figure 2.5.2 - 2 provides a composite of Frames 3762 - 3768. It has been
provided in an attempt to answer the above question. As previously stated, only the side
closest to the camera in these plots should be considered as indicative of shape information.
112
Modeling of Object
Appendix H
.
Figure 2.5.2 - 2: Surface Plots - Object Entering Water Second Time
The difference between this and that observed in Section 2.5.1 is easy to see. The most
obvious difference is at this point there are 2 objects. Their shape is smooth with none of the
angularity seen previously and they seems to be thinner in one of their lateral dimensions.
Finally the last portion entering the water is offset from the middle. The sum of these
statements seems to indicate that at this angle the objects looks very much like the side or
end of the long side of the ovoid chosen in Section 3.4 but showing differing lateral
dimensions.
2.6 Final Shape

It has been shown that the ovoid shape discussed in Section 2.4 was relatively close to what
has been determined. The changes that occurred started at the increased temperature end of
the object. Seen end-on (Figure 2.3 - 1) the object seems to be angular and have unequal
lateral dimensions. The angularity was shown to be a result of the object's shape and not an
artifact of the magnification in Section 3.5.1. The unequal lateral dimensions was then shown
to carry through the entire ovoid in Section 3.5.3.
There is however a conundrum in what has been found. Seen from one side the shape or
shadow of the object is an oval. At least when entering the water, the other side of the object
is angular. Obviously these two sides cannot be along the same dimension. They also cannot
be directly along either of the lateral directions the would be seen when the cooler or white
sections are observed. This only leaves an angle including both the long axis and one of the
lateral dimensions. From that angle, the angularity can be observed on one side with the far
113
Modeling of Object
Appendix H
side being smooth. This, of course, means the assumption near the end of Section 3.5.1 that
the angularity would be symmetric is incorrect. It also means that when seeing the angularity,
one is also seeing a small section of the right hand side (long portion) and that when seen
directly from the left hand side, the object would be an oval. This is shown in Figure 2.6 - 1.
Figure 2.6 - 1
In this figure the back portion (possibly aside from the exact center) is black or warmer. The
right hand image is being viewed in the negative y direction. The left hand image is being
viewed in the positive z direction. A view at a 45 degree angle between the y axis and the
positive z axis would angular on one side and round on the other.
3.0 Heat Transfer

It is obvious from the model that most of the object is hot or warmer then it's surroundings.
That should raise the question of whether there is any heat being transferred to the
surrounding air. Heat transfer essentially occurs via two different mechanisms; radiation and
conduction. Normally a third mechanism, convection, is also stated. It however is just
conductive heat transfer between two objects with a relative velocity between them.
In its simplest form convective heat transfer for a system where the wall temperature does
not change is governed by Newton's law of cooling10.
dQ/dt = H A T .
In this equation: "Q" is the thermal energy transferred from the unknown object to the air;
"H" is the heat transfer coefficient11 (assumed independent of both temperatures); "A" is the
unknown object heated surface area; and "T" is the difference in temperature between the
unknown object and the surrounding air. Although this equation is easy to apply, we do not
know the surface temperature of the unknown object or even its area, thus limiting the
equation's usefulness.
In the present situation, since the fluid (the air) is forced to flow over the surface by the
movement of the unknown object, the heat transfer is what is termed forced convection. The
central concept used in forced convection is that of a boundary layer12. Any flow bounded by
114
Modeling of Object
Appendix H
a surface will develop a region adjacent to the surface, in which the flow properties are
different from that seen an infinite distance from the surface. The primary cause of the
boundary layer is friction.
The boundary layer is an important concept because it is the region in which heat transfer
between the fluid (gas) and the surface takes place. It is known that the boundary layer will
include both velocity and thermal layers. Since we have no knowledge of the surface
conditions of the unknown object, this appendix will concentrate on its thermal properties. In
a thermal layer, the temperature varies from a temperature To at the wall to the equilibrium
temperature T at the outside edge of the layer.
The simplest type of convection is that which takes place in a laminar flow. Laminar flow is
easy to predict and has very little fluctuation in it. Most situations begin as laminar flows and
then later transition to turbulent flows. Laminar flow develops an insulating blanket around
the object and restricts heat transfer. Conversely, due to the agitation factor, turbulent flow
develops no insulating blanket and heat is transferred very rapidly. Also due to the same
factor, turbulent flow is less structured and predictable than laminar flow. Structures called
eddies dominate the flow. Since the driving force for heat transfer is the difference in
temperature between the fluid and heat source, moving that fluid away from the source and
replacing it with cooler fluid will carry off more heat. Turbulent flow therefore tends to carry
off more heat than laminar flow.
It is noted in Appendix J that no wake was seen for the unknown object while it was traveling
underwater. As in that appendix this lack of a wake was attributed to the unknown object
being able to maintain laminar flow around it during that period. The effect being looked at
in air is similar but not quite the same. In water turbulent flow produces pressure and shear
waves that are termed wakes. Since the video in question is IR, those waves would have been
seen as slight differences in heat. Neither they nor any convective heat transfer were seen
during the water transit. While still looking for difference in heat when traveling in the air,
the lower density eliminates the possibility that waves in the air could themselves produce
heat that could be seen. Therefore the only heat to look for in air would be a trail following
the unknown object and quickly dissipating into the bulk air temperature.
Since neither the unknown object surface composition nor the temperature are known, it is
not possible to know if the air flow around it is laminar or turbulent. It is however known that
due to expected conduction from the unknown object surface to the surrounding air
molecules the air should carry off some heat and that heat should be noticeable in surface
plots. Interestingly none is seen in any surface plots looked at. The reader is invited to inspect
Figures 2.1 - 2, 2.2 - 2 and 3, and 2.3 - 2 in this appendix. It is therefore assumed that,
although surprising, the air flow is probably laminar and that the temperature of the unknown
object warmer areas is probably not exceptionally high. Thus any heat being carried away
would dissipate so quickly as to not be noticeable in these plots. To check this, the writer
used the "Transform Image to Results" function in Image J. The result of this function is a
spreadsheet of the pixel values. Figure 3.0 - 1 is small portion of that spread sheet for Frame
#0760.
115
Modeling of Object
Appendix H
Figure 3.0 - 1: Heat Trail - Frame #0760
In this figure heat is denoted by lower numbers. It shows the unknown object traveling to the
right and the pink area is basically the outline of the unknown object. It should be noted that
the author arbitrarily chose a cutoff value of 120 to distinguish the unknown object from the
background. It is easily seen, there is no absolute outline for the unknown object. The
numbers over 100 and below 120 and along the edge and represent areas which are partially
unknown object and partially background. The interesting areas are those shown in orange.
They are areas slightly warmer than the background and represent the heat trail of the
unknown object as it moves through the air. It can be seen that there are also some areas
around the unknown object that show a slight warming. The long axis of the object is
contained in 8 pixels. Assuming the unknown object is 4 feet long, each pixel would
represent about 6 inches and the heat trail only lasts 1 pixel length (about 6 inches) beyond
the unknown object.
The reader is invited to compare Figure 3.0 - 1 with 2.4 - 4. In the latter figure it is seen the
entire object is black with no specific area distinguishable. That isn't so in the above figure.
In this figure it is easily seen the maximum heat (single digit values) is clustered near the
center and falls off in all directions. It is believed that the smallness of the very hot area is the
reason that overall the heat produced is relatively low.
4.0
Conclusion
This Appendix started by using various views of the unknown object to compile a model of
it, During that process it was determined that:
Since a splash was seen in Frame #3769, the object is physically present;
In most of the frames the unknown object displays a smooth exterior shape;
In some frames while entering the water some angularity was observed;
In addition to its forward motion, the unknown object is observed to rotate;
In relation to the forward motion, the rotation looked slow and variable;
116
Modeling of Object
Appendix H
Assuming the hot and cold areas remain fixed on the unknown object, no relation was
seen between the hot areas and the lead area (front) of the unknown object;
The lack of a relation between heat and direction of motion eliminates most terrestrial
propulsion systems;
Only a minimal heat trail could be found following the unknown object.
The air flow around the unknown object is likely laminar;
With less than a mm of water covering the object, no IR radiation will escape;
The wave size shows the object's minimum depth underwater to be between 1 to 3
feet.
The Bernoulli Hump shows the object's maximum depth underwater to be between 3
and 5 meters; and
The areas of the unknown object are not as hot as would be expected from a
conventional aircraft or as cool as a balloon or plastic bag.
This is a surprising amount of information to be obtained from consideration of a simple

model. Since these results are also all independent of the model, they could all have been
(and some were) obtained without the model.
117
Modeling of Object
Appendix H
Notes and References

1. Subjective Realism is a world viewpoint in which the world is in the mind of the viewer.
2. Invariant mass is the portion of mass that is not a function of velocity. It is the portion of mass that
defines shape. There are field quantizations which have mass but no invariant mass and there
are mathematical constructs such as energy which is equivalent to mass. Those objects,
however, do not define a shape.
3. Tikhonov, A; Arsenin, V; "Solutions of Ill-Posed Problems", John Wiley and Sons; 1977.
4. Hale, G.M., Querry, M.R.; " Optical Constants of Water in the 200 nm to 200m Wavelength
Region"; Applied Optics; 12, Issue 3; 555-563 (1973): Plot obtained from data table provided in
paper.
5. Wozniak, B,; Dera, J.; "Light Absorption in Sea Water"; Springer - Atmospheric and
Oceanographic Sciences; (2007); Chapter 1; Page 4; Figure 1.2a
6. Electromagnetic rays interact with particles via discrete scattering. When energy is deposited
onto the scattering material, the process is called absorption. The number of photons that
experience an energy loss is proportional to the differential thickness of the scattering material
and the number of photons incident on the material. The equation provided in this appendix is
basically the number of photons left undisturbed after a distance x.
7. Stefanick, T.; "Strategic Antisubmarine Warfare and Naval Strategy"; Institute foe Defense and
Disarmament Studies; (1987): Appendix 3; "Non-acoustic Means of Submarine Detection";
Equation A3-1. The assumptions used to create this equation were:
The ocean is infinite in depth and extent (the location of the object is far from the ocean's
bottom and boundaries);
There is no surface tension at the oceans surface (the effect wavelength is large compared to
surface tension wavelengths);
The ocean water has no viscosity (the boundary layer is small compared to the object's
diameter);
The object can be approximates as an ovoid moving parallel to the ocean's surface.
8. ibid: Figure A3-2
9. Average underwater speed calculated in Table 3.2 of the Water Transit Appendix (Section 3).
10. Burmeister, LC; Convective Heat Transfer, 2nd ed.; Wiley-Interscience,; 1993; p 107
11. The heat transfer coefficient is normally defined by inverting Newton's law of cooling. Since that
would result in a circular system, it is better to define it using the Nusselt number (Nu):
H = k Nu / L
In this equation "k" is the thermal conductivity of the fluid and "L" the characteristic length of the
problem.
12. Anderson, J.D.; "Ludwig Prantl's Boundary layer"; Physics Today; 58; No.12; pg42-48; (2005)
118
APPENDIX I
Alternate Speed Calculation Using Background Objects
119
Appendix I
OBJECT GROUND SPEED MEASURED WITHIN FRAMES 3769 THRU 3851

Abstract: During the specified frame range, the object was over the ocean and underwater
for a subset of these frames. The speed was measured using the distance provided by the on
screen video data, X (horizontal) pixel positions of the object in each frame and the X pixel
positions of waves in the background which provided an angular rate of camera panning.
These measures not only provide the ground speed but also evidence of object ground speed
accelerations while underwater.
Method of Speed Measurement
First, the changes in the horizontal (the X) pixel positions from one frame to the next were
derived by subtracting the current frame X position from the previous. Thus if the object
moved to the left relative to, say the reticle, then the resulting difference would be positive.
Consequently movement to the right would be negative. The background motion was always
toward the right for all of these frames. To obtain the incremental change of the background,
the X pixels of a selected background location were subtracted; previous frame minus the
current frame opposite to the objects incremental measure. This ensured that the
background incremental measures were always positive since the background motion relative
to the screen was always to the right.
Next, the incremental measures of both the object and the background can be converted to a
horizontal angular rate by multiplying the degrees per pixel by each of those increments. If
the object moved to left, relative to the screen, then its angular rate is faster than the angular
rate of the background (i.e. a fixed location appearing to move to the right on the screen)
thus the angular rate of the object, relative to the fixed background, should be added to the
angular rate of the background. This gives the angular rate of the object relative to a fixed
background location which will enable a ground speed measure of the object. Figure 1
illustrates the method.
120
Appendix I
Figure 1 (Frame 3823)

Figure 1 is frame 3823 showing the object circled in red (difficult to see but present where
indicated) and its relative direction of motion as the camera pans left. Waves, seen in the
background, track right as the camera pans left. The full horizontal field of view at this
magnification (2024) is about 0.3602 degrees. For 704 horizontal pixels, this gives about
0.000512 degrees per pixel. The background wave and the object can be seen in frames 3823
and 3824. The wave has moved about 16.33 pixels (an average) while the object has moved 5
pixels. Thus the angular increment for the wave is 16.33 pixels * 0.000512 deg/pix = .0084
deg change while the object angular increment is 5 pixels * 0.000512 deg/pix = 0.0026 deg
change. What is the objects angular incremental change relative to the background? Since
the objects relative motion to the left implies that the object is moving faster than the camera
is panning to the left then we should add the angular changes; 0.0026 + 0.0084 = 0.011
degrees change relative to the fixed background.
Thirdly, we use the aircraft to target distance indicated at the upper right of Figure 1 to derive
the ground speed of the object. This distance given in frame 3823 is 6938 meters (although
not very legible in this frame, it is clear in immediately previous frames.) However, in frame
3824 the distance was updated to 6951. Under this circumstance, we took the average:
(6938+6951)/2 = 6944.5 meters as the distance to the object from the camera. The object
ground travel during frames 2823 to 3824 was 2*6944.5*tan(.011/2) = 1.33 meters or about
4.34 feet.
121
Appendix I
Fourth, we use the time increment from the start of one frame to the start of the next which
is, essentially, the time increment for 1 frame. Because of the variation in frame rate, the on
screen clock was used to count the number of frames in 1 second for specific sets of frames.
In this case the frame rate was 32 frames per second. Thus the ground speed of the object
based on movements measured from frame 3823 to 3824 was determined to be 4.339756
ft/0.031 sec or about 138.87 fps or about 94.69 mph.
Figure 2 depicts the moving average of objects ground speed, based on the above method,
for frames 3769 thru 3851; 83 frames and 320 data points. That moving average includes sets
of 5 frames.
Figure 2 (Frame pairs 3769-3770 thru 3850-3851)
122
Appendix I
The difference between pixel locations for adjacent frames were used to determine
incremental movement. Consequently pairs of frames are indicated instead of individual
frames. Additionally, the frame index indicates the frame pair with less clutter across the base
of the plot. The frame index represents frame pairs as follows; 1:3769-3770, 2:3770-3771 ...
82:3850-3851. Because each point represents an average of 5 frames, the plot does not start
with 1 and end with 82. The methodology did have a fair amount of noise since tracking a
wave across a sequence of frames did involve some estimates of location. Some frames had
as many as 5 data points, tracking 5 different waves, while most others had 4, 3 and at least 2.
The average standard deviation was 3.34 pixels per frame an average noise level that is not
too bad.
Viewing the specified frames of the video, the burst of acceleration has been determined to
be real by this analysis even though the panning rate did slow which could have created the
illusion of object acceleration.
123
APPENDIX J
Water Transit
124
Water Transit
Appendix J
Introduction
Although most of the video concerns the object moving through the air, there are portions
where the object interacts with the ocean. These include approaching the ocean in preparation
to enter it; entering the water; motion underwater; exiting the water; object division; and the
transition period from water transit back to air transit. This appendix will examine each of
these periods individually. After determining what is shown in the video for each portion and
what is implied by each item, some discussion will be provided of the present level of
Earthly science and engineering in these fields. A final section will also be provided continuing the results of each of the previous sections.
The basic difficulty in interpreting infrared (IR) pictures is our lack familiarity with them. We
expect pictures to show the subjects mass and shape. IR pictures do not do that. They only
show heat variations. Since we do not know what the nature of the object being filmed, we
also cannot know how its heat would vary in any normal operation. However, although the
object is unknown, its environment is not. The environment is a known; therefore, clues
about the object can be derived from seeing how the environment reacts to it. Where possible
this appendix examines the known to determine the unknown.
1.0 Preparation for Entering the Water

Since it is known that water is approximately 50 times more viscous1 than air, a calculation
of the speed in air before entering the water and its speed while in the water was made to see
if there was any difference. It may be noted that, if the unknown object is piloted, it can
expected that it may slow down some prior to entering the water to increase the time the
reaction force from the water is applied thus reducing the instantaneous force. This
calculation was done for a set of frames a short distance away (# 3700 - #3750) from that
specific frame (#3769) in order to reduce any effect introduced by the period where the
unknown object was actually entering the water. Since this 19 frame difference only
translates to slightly more than 1/2 second, it is known that it can only be partially effective
in eliminating any slowing down period.
The basic data for the frames chosen is shown in Table 1.1.
Frame
3700
3750
Latitude (DMS) 18 30 49 18 30 51
Longitude (DMS)67 7 23 67 7 25
Time (H:M:S)
1:24:10
1:24:12
Table 1.1
The problem with the data given above is both coordinates and time are stated to within 1
second.
Since it is implicitly assumed that the unknown object will travel in a straight line during this
period, the time and coordinates to use in calculating this speed are totals. Since any internal
error between cells in this series will cancel, corrections have only been applied to the end
cells. Those corrections were calculated using linear scaling. A check of the frames showed
125
Water Transit
Appendix J
that the first frame that showed the time of 1:24:12 was # 3734 and the last frame to show
that time was # 3765. Therefore as shown below, frame # 3750 occurred at a seconds value of
12.52125 seconds.
X =12 + (3750 - 3733) / (3765 - 3733) = 12.53125
Similarly the first frame that showed a time of 1:24:10 was # 3676 and the last frame was #
3701. This gives a seconds value of 10.96154. The total time is therefore 1.5597 seconds.
The interesting result with this calculation is that if one had blithely used the 2 second
difference determined by subtracting the given times the resulting velocity would have been
approximately 28% lower than what will be calculated here.
Before considering a repeat of the above calculation for the coordinates, it is instructive to
determine the possible maximum error that could have occurred due to their truncation. By
truncating the coordinates of each frame the actual location is only known to a half second of
degree for both Latitude and Longitude. Using 3963.191 miles for the equatorial radius of the
Earth and 3949.903 miles as the polar radius, the radius of the Earth at latitude (L) of
18.5225 degrees is calculated as:
R1 = { RE4 cos2( L ) + RP4 sin2( L ) } / { RE2 cos2( L ) + RP2 sin2( L ) } = 3961.86
miles .
Therefore the distance equivalent to a half second of latitude is
D(1/2 sec Latitude) = {( (R1 / 90) (5280) / {2 x 3600} = 50.71 feet .
A similar equation exists for Longitude, but the radius used has to be (R 2), the perpendicular
distance from a line connecting the poles to the specific Latitude location.
R2 = R1 cos( L ) = 3756.63 miles
This results in the distance equivalent to a half second of longitude being
D(1/2 sec Longitude) = {( (R2 / 90) (5280) / {2 x 3600} = 48.08 feet
Therefore every frame can be off by a maximum of 69.88 feet. Since the distance per frame
is approximately 5 feet, the above possible maximum errors present the possibility of
introducing spurious results. This could cause the distance between frames to vary from
frame to frame. Therefore the velocity will also have to vary. In particular there will be
instances where the object will stop for a few frames or even move backwards while instantly
accelerating to high velocity values to account for other changes. Since it is known that with
multiple frames internal errors always cancel twice the 69.88 feet is the maximum error
introduced by the truncation of the seconds term in the coordinates regardless of how many
frames are included. In the present case where the distance is 50 frames long, the total length
is approximately 250 feet and the error bars would be plus and minus 140 feet. Therefore the
correct values for the two ends must be calculated in this case.
The 49 second latitude seen on frame # 3700 runs from frame # 3690 to frame # 3706. The
23 second longitude value seen it runs from frame # 3685 to frame # 3700. Therefore by
linear interpellation the coordinates of frame 3700 should be { 49.67, 23.94 }. Similarly the
51 second latitude seen on 3750 runs from frame # 3727 to frame # 3765. The 25 second
longitude value seen on 3750 runs from frame # 3740 to frame # 3759. Therefore by linear
interpellation the coordinates of frame 3700 should be { 51.62, 25.55 }.
126
Water Transit
Appendix J
Using the radius { R1 } calculated above and denoting the coordinates of location 1 as { L 1 ,
Lo1 } and location 2 as { L2 , Lo2 } the distance between 2 sets of coordinates is calculated
from: d = R1 cos-1 { sin (L1 ) sin ( L2 ) + | cos ( Lo1 - Lo2 ) | cos ( L1 ) cos ( L2 ) } .
This equation with the above coordinates and radius of the Earth yields a total distance of
251.16 feet and results in an air-speed prior to entering the water of 109.72 miles per hour.
2.0 Entering the Water

In today's science, it is impossible to enter, leave and move through a fluid and not affect it.
However, that seems to be the case in the video. Although the effect may be less than
normally encountered, it is this author's opinion that the lack of a visual effect is basically
due to our difficulty in translating a heat signature into the more normal mass picture.
Frame 3769 and those around it seem to show an object larger than 3 feet moving at over 100
miles per hour hitting and enter the ocean seemingly without creating a splash. Although
present science knows ways to almost make the object almost invisible to the water and thus
minimize the splash, eliminating it is not possible. Effectively a splash is taking a volume of
water and drastically increasing its surface area. Since both evaporative and radiative heat
transfer are proportional to the volumes surface area, a splash provides a means to allow that
volume of water to become a "little" cooler. Little is in quotes because the change is very
small and is basically invisible to the viewers eyes. This is seen in figure 2.1. In this figure,
the red circle outlines the unknown object that has just hit the surface of the ocean and the
red arrow indices the direction it is traveling in. As was stated above, even zoomed in no
splash can be seen in this figure.
Figure 2.1: Frame 3769 - 300X zoom

The "Surface Plot" tool in "ImageJ" provides a three-dimensional view of the intensities of
pixels of a non RGB or grayscale image. It therefore converts the heat variations in the IR
frame to height variations with (in its default operation) the lighter (cooler) pixels being
represented as hills and the darker pixels as valleys. The red outline in figure 2.1 was
provided to allow a direct comparison of that picture with the surface plot shown in figure
2.2.
Although still small, the cooler areas representing the splash are seen as raised areas around
the upper corner of the plot. It is believed these represent a splash rather than simply cooler
areas of the unknown object since they do only show up in these plots where the unknown
object is entering the water. By comparing the 2 figures it can be observed that the object in
figure 2.1 is moving to the right and slightly down. This raises an interesting observation.
127
Water Transit
Appendix J
Rather than the splash being in front of the unknown object, it is trailing it and to the right.
Since a front located splash cannot be found in this plot or any of the later frames, it is
believed the splash has been caused by a portion of the unknown object more to the middle
or back and the front of the unknown object is angled such that it sliced into the water with
little or no splash.
Figure 2.2: Frame 3769 Surface Plot

Although the discussion of the entry portion is almost complete, there is an additional piece
of information that can be gleaned from the above surface plot. While it is understood that
this plot is not showing a real hole in the water, it does allow a view of 1 slice of the
unknown object. Since the slice isn't exactly flat, it isn't quite the same as the 3-d printer
slices but given many slices it would be possible to reconstruct the top portion of the
unknown object's warm sections outline. In particular the above slice show that in the
unknown object contains a warm section shaped vaguely like a dumbbell. It also shows the
forward portion of the warm area is not rounded but has some sharp corners. It may also be
possible to outline cold portion of the top since they would rise up higher than the water. It is
difficult to distinguish them from splashes. The end of the splash can be seen in Figure 2.3.
At this point, only a very small remnant of the unknown object heat signature remains along
with four low remnants of the splash.
Figure 2.3: Frame 3777 - Surface Plot
128
Water Transit
Appendix J
3.0 Transiting the Water

During the underwater journey there are a couple of places where a dot can be seen
indicating the unknown object is quite near the surface. Additionally, although the video does
not show the object for most of its underwater period each time the object appears, the
camera is found to be pointed almost directly at it. Since one of the witnesses specifically
rejected the idea that the camera was locked onto the object, this implies that the object has
remained visible to either the camera operator or the pilot or both. It further implies that
while traveling underwater the unknown object has remained relatively close to the surface
throughout. Since these assumptions essentially mean the unknown object is at an altitude of
sea level and is placed at the target location printed on each frame, we can assume the target
location as the unknown object location throughout this period.
If close to the surface it is possible to determine the object's speed while underwater. Since
frame lengths are measured in the tens of milliseconds, it is understood that arguments can be
made as to the exact frame number to use for the start and end of the underwater period.
Frame 3769 was chosen as the start frame where the object can be seen entering the water
and the end frame to be Frame 4560 where it starts to emerge. Although other choices would
change the central results given below, any difference would be relatively small. The basic
data from these 2 frames is provided in Table 3.1.
Frame
3769
4560
Latitude (DMS) 18 30 52 18 30 53
Longitude (DMS)67 7 26 67 7 56
Time (H:M:S)
1:24:13
1:24:39
Table 3.1: Underwater Frame Span
Although it would be possible to correct these coordinates the same as was done in section 1,
the frame span here (791 frames) is a lot longer than the 50 frames previously considered. As
in that section the maximum error introduced by the truncation of the seconds portion of the
coordinates is twice 69.88 feet over the entire span. Where that error was approximately plus
or minus 56% of the total length, this error is only plus and minus approximately 3.5% of the
total length. It is therefore not reasonable to search for the exact start and end coordinates.
However, the time is a smaller number and has more of an effect on the result and should be
determined exactly.
A check of the frames showed that there were 26 frames that showed the time of 1:24:13 with
the first frame being # 3766. Linear scaling therefore indicates that frame 3769 occurred at a
seconds value of 13.15385. Similarly there were 32 frames that showed a time of 1:24:39
with the first frame being # 4544. Linear scaling therefore indicates that frame 3769 occurred
at a seconds value of 39.84375. The total time underwater is therefore 26.6899 seconds.
Since reality is most likely to be a constant underwater speed but not necessarily a constant
direction, the most reasonable method is to calculate the total distance by adding calculated
129
Water Transit
Appendix J
sections of frames and determining the velocity from it with the time calculated above. The
sections used were each 50 frames long except at the start and end. The initial section had 31
frames and the final section had 60 frames.
The error bars were determined first assuming the coordinate for those frames were each
shortened and then elongated by a half second. The results for the above calculations are
shown in Table 3.2.
Start
Frame
3769
3800
3850
3900
3950
4000
4050
4100
4150
4200
4250
4300
4350
4400
4450
4500
End
Frame
3800
3850
3900
3950
4000
4050
4100
4150
4200
4250
4300
4350
4400
4450
4500
4560
Total
Speed (mph)
Distance Dis. per

(feet) Frame
96.170
3.10
217.43
4.35
279.52
5.59
217.43
4.35
217.43
4.35
192.33
3.85
192.33
3.85
217.43
4.35
192.33
3.85
217.43
4.35
217.43
4.35
217.43
4.35
192.33
3.85
217.43
4.35
217.43
4.35
139.76
2.33
Shortened Elongated
End Frames End Frames
152.904
69.881
217.43
217.43
279.52
279.52
217.43
217.43
217.43
217.43
192.33
192.33
192.33
192.33
217.43
217.43
192.33
192.33
217.43
217.43
217.43
217.43
217.43
217.43
192.33
192.33
217.43
217.43
217.43
217.43
209.543
69.880
3241.69 <4.10>
3368,21
82.812
86.044
Table 3.2: Speed in Water
3145.39
80.352
As has been, it is believed that while underwater, the unknown object travels close to the
surface. In our science, when an object travels on or close to the surface of water, it produces
a wake trailing it. Although the wake equations will not have to be solved in this document, a
small discussion of the mathematics of wakes should be included. Wakes are threedimensional. They include the scalar problem of pressure waves and the two-dimensional
shear problem. Shear refers to a material deformation that occurs due to movement of
internal surfaces parallel to each other. Basically, water shear refers to the extra water being
dragged along as an object moves through it. Effectively what is actually being moved is
much more massive than the object itself. The equation used to describe this situation is the
vector Helmholtz2 equation.
130
Water Transit
Appendix J
As described above shear is basically a friction and like all friction it opposes motion and
generates heat. In an IR image, the effect of the shear portion of a wake would be a dark "V"
whose apex originates at the object. Additionally the pressure wave would add lighter lines
following the same path to the wake.
The "stack" tool in the program ImageJ was used to look for any sign of a wake from the
unknown object as it moved underwater. This stack included all images from frame 3757
through frame 4272. By animating them and going backwards and forwards it was possible
to follow the unknown object's path while it was underwater. Figure 3.1 is one such still
image during this period. The unknown object is the slightly whiter circle within the red
circle. Although exceptionally hard to see in a still picture, it is easy to follow this whitish
circle when the frames are animated. Since the object shows as slightly whiter than the ocean
around it, it is actually cooler than the surrounding water but does not show any sign of a
wake. It is, however, interesting since the unknown object showed itself as hot when entering
the water and it is now located by a slight cooling effect on the water. This seems to indicate
that not only is it not creating heat via friction with the water, it is also not transferring heat
to the water via contact.
Figure 3.1: Frame 3781

The relative coolness shown is assumed to be a slight "hump" in the water at the location of
the unknown object. This is similar to the same effect that occurs over submarines. The lack
of any wake indicates that once the unknown object is inside the water, it is essentially
invisible to it. As the object moves, water directly in its path flows smoothly around it in a
shell and exits at the same point with the same energy as it had before the object arrived. The
cool area observed would be the top of the shell where the "hump" slightly increases the
surface area. Additionally, all water outside that shell would remain still. The word
131
Water Transit
Appendix J
"invisible" to describe this was chosen for a reason. Although this description considers
particles of water rather than photons, it is completely equivalent to the problem of normal
invisibility (if invisibility can be considered normal). Since the water outside the shell
remains still, there is effectively no shearing force, and since the water at the shell returns to
the location it was prior to the arrival of the object and has the same energy, there is no
pressure wave. Hence, there is no wake generated. This movement does however take energy
that would not normally be expended in movement through the air. That use of energy would
be shown as a reduction in speed of the unknown object while in the water.
4.0 Exiting the Water

As has been discussed, the unknown object exits the water in or around frame 4560.
Normally it would be expected that the act of exiting would bring water up with the object
thus increasing its surface area and showing in the video as a white area surrounding the
unknown objects hot section. However due to interference by the reticule none of this was
visible.
Although it was impossible to "see" the unknown object exiting the water, it is possible to
replicate the work done when investigating the preparation undergone before entering the
water. As in that section an effort was made to reduce the effect of the period where the
unknown object was actually emerging from the water, the calculations were done for a set of
frames a short distance away from the exit frame. The results from these calculation are
shown in Table 4.1.
Start
End
Frame Frame
4570
4620
Distance Time
(feet)
(Sec)
Speed
(mph)
4620
53.99
1.781
36.81
4660
187.83 1.169
109.55
Table 4.1: Speed in Air upon Exiting
As was expected, near the exit location, the speed is significantly less than its normal air
value. It was rather surprising to find the speed immediately upon exiting the water to also be
less than that in the water. If this effect is real, it would seem the object has altered its path at
this point to be almost completely upwards while exiting the water. Additionally, since the
calculations occurred approximately 10 frames following the start of that period, the period
must have extended for over the approximately 0.3 seconds it took for the 10 frames to
complete. Although a specific acceleration value cannot be determined due to the lack of
knowledge about the initial velocity or time, the expected strong acceleration between the
initial section leaving the water and the section immediately following it is apparent.
5.0
Object Division
Following its exit from the water the unknown object appears to divide into 2 separate
sections. Since this is confusing and difficult to understand this section will begin with a
132
Water Transit
Appendix J
description of what is seen in the video and the attempt to discuss the possibilities that are
occurring.
The division seems to begin in frame 4627. The word "seems" was specifically chosen since
at the point of division the camera operator has the camera at the long range zoom factor of
625. At that setting the unknown object is essentially a dot over the water. In frame 4627 the
dot is starting to become elongated but two distinct dots cannot be seen. Additionally there is
a period where the reticule hides the unknown object but the first frame in which this viewer
can make out 2 separate dots is 4740. Interestingly in frame 4758, the upper dot (unknown
object) seems to go back underwater and then re-emerges 6 frames later (4764). This
apparently interested the camera operator since he then changed the zoom factor to 2625 two
frames later. It should be noted that when the camera changes zoom factors there is a set of
frames that are completely black.
It isn't until frame 4676 that an image of the unknown objects and the water begins to
reappear. Interestingly the top unknown object appears to enter the water again starting at
frame 4678. Figures 5.1 and 5.2 are surface plots for just before the top unknown object hits
the water (5a) and just after it hits the water (5b). A slight splash in front of the object can be
seen in 5b. The next 20 frames of the video shows the upper unknown object to be skimming
along the top of the water; disappearing and reappearing a couple of times. The lower
unknown object then evidently goes a little lower and begins to copy the first unknown
object starting around frame 4788. Although the 2 objects can be seen to move together for
many more frames, this discussion will be concluded at that frame.
The problem we have with discussing this section of the video is that it is not known if the
unknown object actually divided or if there were simply 2 unknown objects that appeared
from beneath the water and the emergence of the second had been hidden by the first.
Although the second possibility makes more sense both possibilities will be considered.
The first item to look at is speed. We know there is something driving these objects. We do
not know what is providing this motive power but it must exist. If a single object divided into
2 objects it has to be assumed that the "engine" driving the object also divided. That would
seem to imply that after a division both object would travel slower than the single object did
133
Water Transit
Appendix J
prior to the division. A quick calculation was made between frames 4679 and 4990. The
results can be found in table 5.1.
Title
Average
Top Error Loc
Btm. Error Loc
Distance
(feet)
680.77
795.63
577.00
Speed
(ft/sec)
68.08
79.56
57.70
Speed
(mph)
46.42
54.25
39.34
Table 5.1: Speed of both unknown object between frames 4679 & 4990
Remembering that the final speed of a single unknown object after leaving the water was
determined to be 109.55 mph, it is seen these results are significantly lower and seem to
indicate that the objects have indeed divided their engines. It should be noted, however, that
this is only an indication. It is possible but very unlikely that two independent unknown
objects with exactly the same initial speed, decided to reduce that speed to this value at
almost exactly the same time.
An interesting but unexplained oddity occurs during this period. Even though there is a
period of time where one of the 2 objects was moving through both the air and water while
the other was totally in the air, there was no difference in speed between the two seen. It
would be expected that the object partially traveling through the water would be slightly
slower than the object traveling totally in the air. However, the unknown object in the air is
not seen to pull away during that period.
6.0 Comparison to Terrestrial Science

Although most of what has been described above could have been done as individual effects,
We have not yet replicated all of them in a single object. The following two paragraphs
discuss two such terrestrial objects. In each it is seen that in addition to their presenting an IR
signature at odds with the what is seen in the video, neither can duplicate the ability to travel
through the water without producing a wake. The section concludes with a statement of the
present level of terrestrial science in the field of wakes.
Submarine missiles provide an example of objects that can leave the water, fly in the air and
split into multiple flying objects. The Lockheed-Martin Trident II D5 missile 3 has this
capability. It is launched underwater, pops up through the surface, accelerates off and divides
into up to 14 independently targetable warheads. It, however, cannot first dive down into the
water and after traveling a distance underwater re-emerge for the remaining portion of its
travel. It is also much bigger than the unknown object seen and since it is a rocket, would
produce a markedly different IR signature. It also doesn't really "swim" up through the water
as a powered object. It is shot upward 30 to 40 feet through the water by compressed gas
with the rocket essentially creating a "hole" in the water initially filled by the compressed
gas. The movement of the "hole" and the water rushing back into it would easily be seen as a
134
Water Transit
Appendix J
wake behind the rocket. Finally at the surface, the movement of the holes and water rushing
back combine to form a plume of water that follows the missile upward. The plume would be
seen as a splash. At that point, the missile's rockets ignite.
Figure 6.1: Trident II Missile exiting the water

Since they have been in existence for over 20 years, a rocket driven supercavitating torpedo 4
should be considered. Although to the author's knowledge it hasn't been done to date, there is
absolutely no scientific or engineering reason to rule out the possibility of building one that
could be launched and as was seen in the video, fly in air prior to entering the water. Travel
in the water and then reemerge into the air. In both the air and the water this object is simply
a rocket. Supercavitation is only an effect that exists while in the water. It is an effect that
occurs when the water pressure around the rocket is lowered below its vapor pressure 5 thus
creating a bubble of air around the rocket. That allows the torpedo to essentially fly in the
bubble while underwater. There are, however, some major problems with attempting to
equate this concept with the unknown object in the video. The first is all supercavitating
bodies produce compression waves at the front and strong two-phase wakes at the trailing
end. The second is they also require very high speeds to maintain the cavitation bubble. The
second is they are simply rockets that can fly underwater. They are bigger than the unknown
object observed and as rockets, would produce a markedly different IR signature.
In both of these examples it was mentioned that the object under consideration produced a
wake. That is not particularly surprising. It is simply a statement that the water reacts to an
object moving through it. Although that seems like an obvious requirement, recent physics
papers6,7 have begun to question its validity.
135
Water Transit
Appendix J
Figure 6.2: Russia's Shkval Rocket Torpedo

Those papers and others employ the Transformation Optics (TO) procedure initially proposed
at Duke University for electro-magnetic cloaks. That procedure makes use of the fact that
both Maxwell's equations of electromagnetic (EM) theory and the Helmholtz equation are
invariant under a coordinate transformation. Although not as exact as electromagnetic cloaks,
these papers of a part of a flurry of theoretical papers recently published on acoustic and
fluidic (water) invisibility cloaks. The above two referenced papers form a complimentary
pair of views of invisibility cloaks in water. In the first one, by Farhat et. al., the water is
invisible to the object6 (the water doesn't affect the object - protection against Tsunamis, etc.).
In the second, the object is invisible to the water 7 (the object doesn't affect the water). The
latter paper envisioned using a meta-material8 shell composed of parallel rows of fiberglass
slats etched with copper to transport the water in a laminar fashion around the object
Although the paper by Farhat6 et. al. was published first, in relation to this appendix the
second paper7 by Urzhumov and Smith is the most important in. Similar to the EM cloak, it
"warps" the water around the object such that it attempts to preserve the streamlines of flow
and the pressure distribution that would have existed in the absence of the object. Since it
was known that simply steering the water around an object would tend to slow down the
water thus causing a frothy wake, it was proposed that small piezoelectric pumps be placed in
the shell to offset the energy loss. Consequently, the structure cancels the viscous drag force
and prevents the onset of turbulence. However, the paper was published in 2010 and only
shows that scientifically the shell is possible. Although it is believed that such a shell would
be exceptionally interesting to the navy, as far as this author knows, to date there hasn't even
been a proof of principle engineering design.
An additional advantage of the cloak envisioned by Urzhumov and Smith is that it makes the
object essentially hydrophilic9 It was experimentally shown by Truscott, Aristo, and Techet 10
that since there is no void (bubble) produced when a hydrophilic ball is dropped into water, it
makes a much smaller splash than the more normal hydrophobic ball. There is, however, a
relatively large vortex wave that spreads out from the ball. The splash for the hydrophobic
object is a result of the bubble collapsing due to hydrostatic pressure and forming an upwards
jet or water.
136
Water Transit
Appendix J
7.0 Conclusion
This appendix began with a short discussion of the difference between infrared pictures and
visible light pictures in air and water. Specifically it was shown that in water, it can be very
instructive to look at how the environment (the water) reacts to the external objects and how
its heat may change due to that interaction.
It was determined that the first frame that showed the unknown object entering the water was
# 3769. To provide a base or point of departure, the air speed prior to that frame was
calculated using frames 3700 through 3750. The speed calculated was 109.72 mph.
Following that, the period of entering the water was investigated. It was seen in figure 2.1
that no splash can be seen by eye the figure. However as is seen in figure 2.2, the computer
was able to see the difference in heat between the ocean and a splash as the unknown object
hit the water.
The period of moving underwater was covered in section 3.0. It was shown in that section
that the unknown object traveled at about 82.812 mph or approximately 75% as fast as it had
traveled in air prior to entering the water. It was also seen in this section (figure 3.1) that the
unknown object produced no visible wake as it moved through the water. The unknown
object however did seem to slightly raise the water level immediately above it.
The exit of the unknown object from the water (frame 4560) was discussed in section 4.
Interestingly immediately upon exiting (frames 4570 - 4620), the unknown object moved at a
speed of 36.81 mph which is less than half of its speed underwater. This may, however be
due more to direction of motion than an actual loss of speed. The speed then increases to
109.55 mph in frames 4620 - 4660.
Section 5 then discusses a division of the single unknown object into 2 unknown objects
starting at frame 4627. After the division, the speed of each was calculated to be 46.42 mph
indicating a true division of one object into two rather than the emergence of a hidden object
from behinds the first object.
The following table shows the speed variation though the entire period covered by this
appendix.
Frames
1.
2.
3.
4.
5.
Speed in air
Speed underwater
Initial Speed in air
Later Speed in air
Speed after division
Distance
.
.
(feet)
3700 - 3750
251.16
3769 - 4560
3241.69
4570 - 4620
53.99
4620 - 4660
187.83
4679 - 4990
680.77
Table 7.1: Speed Table
Time
(Sec.)
1.560
26.690
1.781
1.169
14.665
Speed
(mph)
109.72
82.81
36.81
109.55
46.42
The interesting result of this appendix is as previously stated, although most of what has been
described could be done as individual effects, a capability of doing all of them has not yet
been demonstrated.
137
Water Transit
Appendix J

1. The dynamic or shear viscosity of a fluid is a measure of its resistance to shearing
flows. In relative units, the dynamic viscosity of air at 20 C is 0.0198. The dynamic
viscosity of 20 C sea-water is 1.08. Therefore sea-water is 50.46 more viscous than
air at 20 C.
2. The vector Helmholtz equation is:
, where F is a vector function, Delsquared is the vector Laplacian and k is a scalar constant.
3. http://www.naval-technology.com/projects/trident-ii-d5-fleet-ballistic-missile/
4. The first super-cavitating torpedo was the Russian Shkval torpedo. Work began on
that torpedo in 1960 and it was deployed in 1990s. It is reported to travel over 200
knots That means even if fired from ~3.5 miles away, the target has less than 1
minute to employ counter-measures. Recently there have also been (unconfirmed)
reports of a new German super-cavitating torpedo capable of ~500 mph.
5. When the water pressure is less than the vapor pressure the water vapor remains in
gaseous form.
6. Farhat, M;, Guenneau, S; et. al.; "Analytical and numerical analysis of lensing effect
for linear surface water waves"; Physical Review E 77, 946308; 2008; 11 pages
7. Urzhumov, YA; Smith, DR; " Fluid Flow Control with Transformation Media"; Phys.
Rev. Lett. 107, 074501; 2011
8. Meta-materials are artificial materials engineered to have properties not found in
nature. They are assemblies of multiple individual elements arranged in repeating
patterns fashioned from conventional microscopic materials such as metals or
plastics. Unlike natural materials, meta-materials are able to reduce the "index of
refraction" to less than one or less than zero.
9. A hydrophilic object is an object with a strong affinity for water. They seem to attract
water Examples are objects with dissolve in water such as sugar cubes. A
hydrophobic object is one which tends to reject water.
10. Truscott, T.T.; Aristo, J.M.; Techet, A.H.; "Dynamics of Water Entry";
arXiv:0810.1888
[physics.flu-dyn]; http://arxiv.org/pdf/0810.1888v2.pdf
138
APPENDIX K
Estimated UAP Temperature
139
Appendix K
Introduction
Since we do not actually have possession of the object seen in the video, determination of the
principles employed to create its motion remain a mystery. That, however, does not mean we
can ignore this subject. The object is seen to move making multiple direction, altitude, and
speed changes. Some source is supplying the energy needed to do that. There should be waste
heat produced. This appendix provides an approximate calculation of the waste heat seen in
the video. In this appendix:
Section 1 provides a pixel based approximation to the objects temperature; and

Section 2 provides a determination of the appropriate heat equations for the object
and integrates them with the results of Section 1.
Section 1.0
Temperature measurements were approximated to estimate the heat distribution of the
unknown object. The range of temperatures of the object, found by the methods detailed
here, were from 10o F below ambient air, 69o F to 70o F, thru 105o F or higher. A single frame,
1251, was used for all the pixel value measurements to avoid frame to frame temperature
range adjustments that may occur with the infrared (IR) or thermal imaging equipment. It
should be noted that among the 5,000+ frames of video containing the unknown object, the
temperature distribution appears to change. A complete study has not been done to include a
determination of any correlation of the presence of cooler areas of the unknown object to the
higher temperatures this could be IR artifacts.
1.1
Pixel Value to Temperature Relationship
In an ideal world, we would have access to an equation such that given a pixel value, 0 to
255, the equation would produce a limited temperature range represented by that pixel value
produced by the IR hardware and software. However, analysis of the various frames indicates
that the temperatures associated with the 256 different gray shades of pixels is periodically
recalibrated to the temperature range in each frame. Without the software algorithms used in
the thermal video processor, the temperature of the 256 different gray shades must be
determined by using objects of known temperatures in a given frame. Frames were selected
where known objects such as roads, pasture and animals, were used to associate pixel values
to their estimated temperatures.
Frame 1251 was used because there were three known temperatures that could be used to
establish the temperature of the unknown object based on the relationship of temperature to
the 256 pixel shades. An assumption was made that there is a linear relationship between
pixel values and object temperatures within a given frame.This is represented in Figure 1-1.
140
Appendix K
Figure 1-1
The average temperatures for asphalt roads, cows and pastures were found to be 99.6 F 1,
101.5 F2 and 79.34 F3 respectively. These temperatures were plotted against the average
pixel value for the road, cows and an area of pasture seen in Frame 1251 (see Figure 1.5). A
regression line, seen in red in Figure 1-1, is the best fit for all three points. Although
regression is not usually used for extrapolation, a linear relationship between the endpoints of
the temperature range is reasonably assumed here for estimation purposes. We can see,
apparently for this frame using a linear assumption, the IR equipment used a temperature
range of 56 F to 105 F to adjust the pixel values to temperature. All temperatures colder
than 56 are mapped into pixel 255 while all temperatures hotter than 105 are mapped into
pixel 0. The equation for the regression line provides an approximation of temperature with t
being the temperature in degrees F, p the pixel value used to obtain the temperature estimate
t:
t = -0.1939 p + 105.385
141
Appendix K
It can be seen that each pixel represents an incremental change of 0.1939 degrees F given our
stated assumptions.
1.2
Unknown object Temperature Distribution
Using Equation 1-1 we can take every pixel value seen to comprise the unknown object and
convert them to temperatures. Figure 1-2 is a 3D false color representation of the unknown
object temperature.
Figure 1-2
This 3D image was produced using the ImageJ (version 1.45 S) Interactive 3D Surface Plot.
The temperature ranges are given in between each z axis pixel value. Red represents the
hottest, temperatures 101 F and up, and blue the coldest at 74 F and below. Note the
smoothing is set fairly high to remove pixelization.
Figure 1-3 depicts the hottest and coldest locations of the unknown object.
142
Appendix K
Figure 1-3
It is usually the case that the hottest (blackest) locations are in and about the center of the
unknown object while the coldest are usually around the edges. It has not yet been
determined whether the cooler (white) pixels are possibly thermal imager artifacts due to
high differential temperatures. The red rectangle, non unknown object pixels, are in the 79 to
80 degree range and corresponds to the pasture in the background.
Figure 1-4 is profile of temperatures along the pink line seen in the figure that cuts through
the center of the object. This graph demonstrates how the hotter part of the object is usually
in its center zone although not completely symmetrical. It also shows how the infrared
system detects the heat of the object as a contrast against the ambient temperature.
143
Appendix K
Figure 1-4
144
Appendix K
Figure 1-5
Figure 1-5 was created from Frame 1251. Within this IR frame is the object labeled as UFO,
cows in the upper right and upper left quadrants, trees/grass throughout the frame, and a road
that runs along the upper part of the frame.
The average (21.07 on a scale of 0 to 255) and standard deviation (4.73) of pixel values of
the road was obtained from all 24 pixels along the yellow line seen on the road. ImageJ
provides the average and standard deviation of all selected pixels in a profile selection. A
Google Earth view of the area pictured above, shows what appears to be an asphalt road so
asphalt temperatures were assumed for the given time of day. These temperatures can be
determined based on the maximum temperature during the day, the amount of cloudiness,
and the length of time since sunset.1
Each of ten cows, as indicated, were selected within a rectangle and the lowest valued pixel
was chosen to be representative of the temperature. An average (29.5) and standard deviation
(16.94) of the pixel values were then derived.
145
Appendix K
All pixels within the red rectangle were selected to be representative of pasture. The
average(133.59) and standard deviation(1.49) of the 242 pixel values within the rectangle
was provided by ImageJ as a matter of course.
The unknown object seen in Figure 1-5 was magnified in Figures 1-2, 1-3 and 1-4.
1.3 Sources
1.
American Concrete Pavement Association;

http://www.pavements4life.com/qds/environment_1heatisland.asp; last accessed 11/10/2014
2.
"Animal Heat." Encyclopedia Britannica. Chicago: Encyclopedia Britannica, 1965:A 965.
3.
Remote Sensing of Environment 89 (2004) 467483
Section 2.0
Forced convection is the heat transfer mechanism occurring when a fluid is forced to flow
over a hot surface. Although it sounds different, there is actually no difference between the
above definition and the heat loss of the object seen in the video. In this case the motion used
to transfer the heat is motion of the object itself through a quiescent volume of air.
In this case, the heat transfer will be a function of many variables. The only information
source available for this investigation is the video itself. Therefore everything is predicated
on that video. It should be stated at the outset that convective heat loss depends greatly on
source shape. The object's shape is believed to be spherical. It is possible that this assumption
may not be correct, but a shape has to be used.
As stated the information source is a video. That means we are seeing individual static
snapshots at different times. There are a couple bits of information that are immediately
obvious in this video. The first is, the object does not seem to be getting any hotter as it flies
around. Therefore if the object is generating heat, it must also be expelling it as it moves. As
the object moves we can see a heat trail following it.
2.1 Pixels and Distance

Since the frames are basically showing the same picture (albeit at slightly different locations)
over and over again, we really have only one picture. Since the video runs at 30 frames per
second, each frames is approximately 33 milliseconds long. It has been estimated that the
object is approximately 4 feet long (See Appendix G). Since it has been shown that it is also
8 pixels long on the average, we can say that each pixel covers approximately 6 inches. If
the object is traveling at X mph, in one frame, at any point in the frame it will travel a
distance in feet shown by the following forumla:
(D / Frame)obj 0.033 (5280 X / 3600) .
Therefore the approximate number of pixels moved by any point on the object per frame is:
N (D / Frame)obj / 0.5 .
146
Appendix K
2.2 Heat Equation

The heat equation describing the diffusion of heat through the air is:
= 2 T .
Although we are assuming a sphere, we are also seeing a tail directly following the object.
Therefore the above equation can be simplified to one spatial dimension.
Tt - Txx = 0 .
In the above equation, the subscripts are denoting partial derivatives with respect to the
subscript. The initial and boundary conditions for this equation are:
IC:
T ( t=0 ) = T0
BC1: -k Tx| x=L = h ( T - T1 )
(The vert. line is for "evaluated at")
BC2: T ( x=0 ) = T0 t
( means "for all")
There is also a boundary condition at infinity but it would be superfluous.
This equation can be solved by separation of variables, but it is easier to just use a source 4,5
that has already solved it. (For any interested reader the complete solution methodology can
be seen in reference 5.) These references use an electrical analogy to obtain the following
thermal response function:
T0 - Tair = ( T - Tair ) exp { - h 6 t / ( D Cp ) .
In this equation:
D is the diameter of the sphere;
V is the flow velocity of the air;
Cp is the specific heat of the air at constant pressure;
is the density of the air;
h is the Heat Transfer Coefficient
2-3 Application to Video Frame

Solving Equation 2-4 for "t" yields an expression that gives the time needed to reduce the
temperature in the trail to some set value (higher than the air value).
t = { D Cp / (6 h) } ln { (T0 - Tair ) / ( T - Tair ) }
t = {Time Coef} ln { Tinit / Tfinal }
Interestingly this result is not simply dependent on the bulk air temperature and the source
temperature. It has a has a third temperature. This is a result of the resulting temperature
being an exponential starting at the source temperature and falling to the bulk air temperature
"at infinity". Therefore a solution ending at a finite distance is defined as a value that is close
enough to be considered correct. That is the variable T with no subscripts in the equation.
To determine numeric results, the density and specific heat are treated as constants. The
Engineering Toolbox values6 at 300 K is used.
It is obvious that the convective heat transfer must be dependent on the fluid removing the
heat. Since convective heat transfer is governed by Newton's law of cooling:
dQ/dt = h A ( Tobj -Tfluid ) ,
147
Appendix K
That functionality must be found in the heat transfer coefficient ( h ). An approximate

relation for it if found in the Engineering Toolbox7 pages and is shown in Figure 2-1.
Figure 2-1
As stated above, the heat equation (eq. 2-5) requires a definition of an intermediate
temperature close enough to be considered acceptable. Stated in this manner, that definition
would be a complete guess. However in the present case, there is another way to look at this
definition. It is how many pixels show the trail following the object. That distance was
discussed in section 2-1. The results are shown in Figure 2-2 and Table 2-1.
V(mph)
50
60
70
80
90
100
110
120
130
140
150
160
170
180
190
200
h
25.5988
26.6824
27.6232
28.4509
29.1862
29.8441
30.4360
30.9707
31.4552
31.8953
32.2956
32.6600
32.9918
33.2939
33.5687
33.8184
Time
Coef
0.0098
0.0094
0.0091
0.0088
0.0086
0.0084
0.0082
0.0081
0.0080
0.0079
0.0078
0.0077
0.0076
0.0075
0.0075
0.0074
Dis in 1
frame (ft)
2.4444
2.9333
3.4222
3.9111
4.4000
4.8889
5.3778
5.8667
6.3556
6.8444
7.3333
7.8222
8.3111
8.8000
9.2889
9.7778
Ln( X )
1 Cell
3.4002
3.5441
3.6691
3.7790
3.8767
3.9640
4.0427
4.1137
4.1780
4.2365
4.2897
4.3381
4.3821
4.4223
4.4588
4.4919
Table 2-1
148
1/X (%)
1 Cell
3.3368
2.8895
2.5500
2.2846
2.0720
1.8986
1.7551
1.6347
1.5328
1.4458
1.3710
1.3062
1.2499
1.2007
1.1577
1.1199
Ln( X )
2 Cells
6.8003
7.0882
7.3381
7.5580
7.7533
7.9281
8.0853
8.2274
8.3561
8.4730
8.5793
8.6761
8.7643
8.8445
8.9175
8.9839
1/X (%)
2 Cells
0.1113
0.0835
0.0650
0.0522
0.0429
0.0360
0.0308
0.0267
0.0235
0.0209
0.0188
0.0171
0.0156
0.0144
0.0134
0.0125
Appendix K
Figure 2-2
It is obvious from the above that the result for two cells is very close to zero. However the
result for one cell is also quite good if the average temperature of the object is not greatly
different from the bulk air temperature. Since it is possible to look at pixel values in each
frame there is a means of obtaining an approximation to these temperatures.
The "Transform Image to Results" function in ImageJ allows the investigator to see the pixel
values throughout any picture. Figure 2-3 is small portion of Frame #0785. The numbers
depict the relative heat of each pixel. The higher the number the cooler the area represented
by the pixel and the lower the hotter. Although the object is hotter than its surroundings, it is
easily seen that the large distance between the object and the camera makes the heat outline
difficult to determine. To help the reader, color was used to provide an approximation of the
object. The cells with backgrounds that are various shade of blue are the colder areas of the
object (lighter blue areas are warmer); those with a violet background are the hot area of the
object; and the cells with yellow backgrounds are the heat trail being discussed in this
document.
Figure 2-3
149
Appendix K
The outside box of numbers represents the background at the location of the object in this
frame. Although many frames show a background value of approximately 150, this frame has
the object within a warmer rectangle (average pixel value pf 113.58).
One of the problems with Figure 2-3 is it is difficult to see the shape of the object. Figure 2-4
is provided to help with that problem. This figure was generated using the "Surface Plot"
functionality in ImageJ. In this figure colder areas are higher and hotter ones lower.
Figure 2-4
It is easy to see the coldest area of the object is in the front; it is a little cooler on the sides;
and the hot area of the object is in the center. There is a small heat trail immediately to the
right and a larger one near the bottom. This correspond to the yellow colors in Figure 2-3.
2.4 Temperature distribution

Most IR systems use some equalization variant to overcome the problem of distinguishing
low contrast targets in dynamic scenes. The most common methods used are variants of
histogram equalization. The problem faced in this investigation is that these equalization
systems are inherently non-linear. They provide enhancement by increasing contrast in the
dominating temperature range in a scene and decreasing it in the non dominating range.
Additionally the histograms used to describe the scene in 256 levels of gray are functions
which are unlikely to be straight lines. Therefore Equation 1-1 is at best a rough estimate of
150
Appendix K
the pixel-temperature function. It does however provide a realistic approximation. Applying

it to Figure 2-3 yields Figure 3-1.
Figure 3-1
Although this figure is identical to the previous one, it now gives the results in terms of
degrees Fahrenheit. Using the temperatures around the outside of this figure the average
temperature of the background air is found to be 83.37 and the standard deviation is 0.55
degrees. Additionally the hottest point in the object is 105.39 degrees. The figure therefore
shows an average maximum temperature differential between air and object of 22.02 degrees.
It was noticed in Figure 3-1 that there are locations in the air (pixel values of <80) where the
temperature is less than the background air. It was initially thought that these cooler pockets
may be examples of low pressure zones that always follow objects moving in fluids. The size
of these zones are a function of the relative velocity of the object and the fluid, and the shape
of the object. If the object is moving slowly and has a streamlined shape (canoes etc), the
zone will be small but it will still exist. In the present situation although we do not know the
object's shape, we do know the speed is relatively high. We therefore can expect that this
zone may be sizeable enough to have an effect on the results found. Rather than make a
molecular argument for the temperature-pressure relationship it is simpler to just state the
Gay-Lussac gas law:
Pressure = k * Temperature
where k is a constant. Therefore a region of lower pressure is also a region of lower
temperature. With very short heat trails this effect would be a slight cooling below ambient
just before returning to the ambient air temperature. It could then also be the cause of some
of the cooler pixels trailing the object.
2.5 Conclusion
The problem with assigning the temperature variation to low pressure zones is that variations
with the same order of magnitude are seen to be occurring at areas away from the unknown
object. This was particularly noticeable when the author looked at an area twice the size of
the matrix shown in Figure 3-1 (28x36 verses 14x18). With the larger matrix the average
151
Appendix K
ambient temperature (outside rectangle) was calculated to be 82.99 F and the standard
deviation is 0.5 F. Both values are slightly cooler than the ones calculated using the smaller
matrix. That is to be expected since exponentials never actually reach zero. The interesting
result is that around the large matrix, the maximum temperature value was fond to be 83.86
F and the minimum value to be 81.73 F. That results in a delta temperature of 2.13 F at a
distance where it cannot be attributed to the unknown object. Additionally there is nothing at
all seen in the picture outside of the unknown object (see Figure 3-2).
Since it is known that the maximum and minimum temperatures and related pixel values
are:
Temperature Pixel Value
Maximum
105.39
0
Minimum
66.41
201 .
each integer pixel value equates to:
T / p = | (105.39 - 66.41) / (0 - 201) | = 0.198 { F per pixel } .
Therefore the temperature variation occurring along the periphery (2.13 F) is equivalent to
10.76 pixels and no source for this variation is known. Since the 2 degree variation with
unknown source in the "ambient temperature" is approximately four times larger than the
assumed low pressure zone delta temperatures, that assumption cannot be defended and must
be discarded. Since the delta temperatures in the tail are approximately 3 degrees the
assumption of a heat tail falling off exponentially remains.
A statement has to be made concerning the long heat trail that occurs at the bottom back of
the object. Initially this looks like a much longer tail than what has been suggested. Since this
tail begins approximately 2/3s to 3/4s back from the start of the object and not directly at the
rear of the object, it is assumed the heat is not a single pixel location. It is believed heat is
being exhausted along the object. Therefore although heat at any particular location is falling
off backwards exponentially, that loss is being made up for by more heat being exhausted.
2-5
4. Bahrami, Majid; "Forced convection Heat Transfer"; Simon Fraser University; ENSC 388 (F09)
5. John H. Lienhard IV & John H. Lienhard V; "A Heat Transfer Textbook"; Cambridge TJ260.1.445
2000, 3rd Edition; Chapter 5, "Transient and Multidimensional Heat Conduction"
6. "Dry-Air Properties"; http://www.engineeringtoolbox.com/dry-air-properties-d_973.html
7. "Convective Heat Transfer"; http://www.engineeringtoolbox.com/convective-heat-transferd_430.html
8. The result is a value of 6.8054 for "c". This would indicate a very rapid heat dissipation.
152
APPENDIX L
Line-of-Sight Evaluation
153
Appendix L
Introduction
Some arguments have been advanced that the object could be a balloon and that the motion
of the balloon relative to the background is an illusion created by the motion of the plane
circling the balloon. This is incorrect. There are some frames of video having no background
motion at all and the UAP can be seen changing locations relative to the background.
Consider Frames 711 and 712 of Figure 1:
Figure 1
Figure 1 clearly demonstrates the intrinsic motion of the UAP and that the motion of the
plane from one frame to the next contributes nothing to the difference in the location of the
UAP. This, in itself, does not eliminate the possibility of a balloon since a balloon could have
drifted into the frames due to its own motion.
The wind was out of the east at 8-13 mph.1 Upper wind speeds were measured out of San
Juan, which is 50 miles to the east of Aguadilla. At 8 pm local time the upper wind speeds
from 400 feet to 3200 feet were similar and were out of the east northeast at 12 to 18 mph. 2
Given that eighteen miles per hour was likely the fastest speed of a balloon in Frames 711
and 712 then the distance from the plane could be no more than 1,250 feet considering the
angular distance the object traveled within these two frames. (The calculation of this distance
is detailed later in this appendix.) This 1,250 feet from the plane creates serious discrepancies
with the aircraft-to-target azimuth readings3 given by the on screen data. The discrepancy is
illustrated in Figure 2.
http://www.wunderground.com/about/data.asp
University of Wyoming, Department of Atmospheric Science.
http://weather.uwyo.edu/upperair/sounding.html
3
This would be the compass direction in which the IR camera was pointed.
2
154
Appendix L
Figure 2
The first position (P1) of a possible balloon is established along the line oriented at 227 o
azimuth given by the aircraft-to-target screen data seen within Frame 0711 (See Figure 3) and
established at a distance from the aircraft position of 1,250 feet. The Frame 0711 on screen
data indicated the plane position (Fr0711 ACFT in Figure 2) to be 18 o 31 13 N and 67o 06
22 W. The second balloon position (P2) was given by the distance a balloon would have
covered in 4.2 seconds at 18 mph; 110.88 feet. The next frame, 0836, was arbitrarily chosen
to allow comparison of aircraft and balloon travel distances. The 2 nd plane position, specified
by the yellow pin annotated (Fr0836 ACFT), is the plane location 4.2 seconds after the
Fr0711 ACFT plane location. During this time the aircraft traveled a distance of 1504 feet
along a WNW (295o azimuth) path which subsequently created a line of sight (LOS), from
plane to balloon, of 164o. The actual azimuth to the target (UAP) for Frame 0836 can be seen
as 219o (See Figure 4). This is a discrepancy of 55 o (219o 164o) between the actual target
azimuth and the one calculated for a balloon.
155
Appendix L
Figure 3
The 227o bounded in red is the plane-to-target azimuth; the compass direction in which the
infrared (IR) camera was pointed within Frame 0711.
156
Appendix L
Figure 4
The 219o bounded in red is the plane-to-target azimuth; the compass direction in which the
IR camera was pointed within Frame 0836. There is a time difference of 4.2 seconds between
Frames 0711 and 0836 (1/30th of a second per frame). Over that time, the plane tracked the
UAP over an 8o degree azimuth change (227 o 219 o). The plane would have passed an
object as slow a balloon creating the much greater azimuth change of 68 o.
How it was established that the balloon was 1,250 feet from the plane in Frame 0711 does
need some attention. See Figure 5.
157
Appendix L
Figure 5
Figure 5 establishes that the UAP seen in the video moved about 17 pixels from Frame 0711
to 0712. The Field Of View (FOV) of the IR Camera was determined to be about 1.07
degrees at magnification 675 which amounts to 0.001483 degrees per pixel (see Appendix G
for details on degrees per pixel). The UAP having moved 17 pixels implies that the UAP
traversed 0.025211 degrees in 1/30th of a second. It was noted earlier the weather at the time
had maximum winds of 18 mph winds thus inferring that a balloon, going with the winds out
of the ENE, would have moved 26.4 fps * 1/30 = 0.88 feet. However, the trajectory of the
UAP over Frames 0711 to 0836 was almost due south (188 o azimuth). The fact that the image
is a two-dimensional projection of a three-dimensional area changes the path length of 0.88
feet as seen from the IR camera perspective. The angular change of 0.025211 degrees is not
over 0.88 feet but rather 0.55 feet. See Figure 6.
158
Appendix L
Figure 6
The angular change of 0.025211 degrees, from the IR camera perspective, must be applied
using 0.55 feet. Consequently the distance a balloon, traveling at 18 mph, from the aircraft
would have been 0.55/(2*tan(0.025211/2)) = 1250 feet.
Line-of-Sight measurements, angular size, and speed preclude a balloon from being a possible explanation.
159

2013 Aguadilla Puerto Rico UAP Final Draft v5s

Uploaded by

Copyright:

Available Formats

2013 Aguadilla Puerto Rico UAP Final Draft v5s

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

2013 Aguadilla Puerto Rico UAP Final Draft v5s

Uploaded by

Copyright:

Available Formats

2013 Aguadilla Puerto Rico

The detailed analysis of an Unidentified Anomalous Phenomenon

EXECUTIVE SUMMARY.......................................................................................................... 1-4

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

Thermal Imaging Video

Significant deceleration as an unpowered object, such as bullet, striking

2013 Aguadila Puerto Rico UAP

Splitting Into Two Parts

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

III. ANALYSIS AND CHARACTERIZATION

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

Radar AnalysisVerification of Witness Testimony of Visual Sighting The witness

2013 Aguadila Puerto Rico UAP

Radar AnalysisVerification of Unknown Targets

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

FIGURE 5: Frame 0141. Unknown object near center of cross-hairs.

2013 Aguadila Puerto Rico UAP

FIGURE 6: Right triangle

ImageJ 1.47v. National Institute of Health, July 8, 2013.

2013 Aguadila Puerto Rico UAP

FIGURE 7: Frame 5085. Unknown object in center of cross-hairs.

FIGURE 8: Frame 2697. Unknown object visible.

2013 Aguadila Puerto Rico UAP

FIGURE 9: Frame 2705. Unknown object disappears behind trees.

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

2013 Aguadila Puerto Rico UAP

< 1 ft 18:30:46N 67:07:19W

117 +/- 3 mph

TABLE 1: Speed of object at known positions.

**Speed through water and air.

2013 Aguadila Puerto Rico UAP

Graph 1: Speed measured every 50 frames

2013 Aguadila Puerto Rico UAP

ImageJ 1.47v. National Institute of Health, July 8, 2013.

2013 Aguadila Puerto Rico UAP

FIGURE 13: Object's < 1/6 of a second later in Frame 3773

2013 Aguadila Puerto Rico UAP

FIGURE 14: Frame 3769 - 300X zoom

FIGURE 15: Frame 3769 Surface Plot

2013 Aguadila Puerto Rico UAP

Haffa and Patton, Analogs of Stealth, Northrup Grumman Analysis Center

2013 Aguadila Puerto Rico UAP

FIGURE 16: Cooler signature of object in frame 3781

Graph 2: Bernoulli hump; height vs depth based on a 3 ft object moving at 83 mph

2013 Aguadila Puerto Rico UAP

FIGURE 17: Heat of object indicated by dark color in frame 3937

2013 Aguadila Puerto Rico UAP