Unofficial transcript: Day 4, Session 7
Choose a session: Day 1, Session 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8
Day 2, Session 1 | 2 | 3 | 4
Day 3, Session 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8
Day 4, Session 1 | 2 | 3 | 4 | 5 | 6 | 7
Day 5, Session 1 | 2 | 3 | 4
Session 7 March 8, 2001 3:45 p.m.
DIRECT EXAMINATION BY CAPT MACDONALD:
Q Why is that, sir?
A Because it records -- it archives on a hard
drive within the sonar system, one-second data from
basically all the ship's parameters -- course, speed,
pitch, roll, several other parameters that are
relevant to the Greenville's track -- in this case,
the submarine's track, as well as a good deal of
tactical data that is processed on the ship --
specifically, any contact that is being tracked by
the sonar system is logged there, and fire control
solutions are logged there, and at fifteen second
intervals.
As far as reconstruction effort, that data has
heretofore never been available.
We've had to go with a lot more
less-often-recorded data, and not as -- not a copious
amount. So we were able to download this information
off this hard drive, and it really aided in the
reconstruction of USS Greenville's track, and the
contacts that the Greenville was tracking on the day
of February 9th.
Q Captain, did you bring a series of slides with
you to describe the reconstruction effort?
205
A Yes, I did. I believe I loaded it on.
Q Cmdr Harrison, could I have you start up our
Power Point machine, please, and could I have this
series of slides marked as court exhibit next in
order, please?
THE WITNESS: Yes, sir. This will
be marked as Exhibit 40.
(Exhibit 40 marked.)
CAPT MACDONALD: And that's all 16
slides as Exhibit 40?
THE WITNESS: Yes, sir.
Q And copies of Exhibit 40 have been provided to
the parties and counsel. If you could wait just a
minute until the machine wamrs up.
Commander Harrison
VADM NATHMAN: Could you bring that
further out? See the "overview of incident" I don't
think we need that. Let's give more magnification,
so can you go out out out to stay focused? Well,
don't know if that is helpful or not.
THE WITNESS: You can probably lower
the projector, the front leg on the projector.
VADM NATHMAN: Can you see it,
Captain, okay?
THE WITNESS: Yes I can, sir.
206
Q Captain Kyle, could you please describe for the
court what we're seeing in the first slide?
A Yes, I can. This is our best overall depiction
of the tracks of the Ehime Maru, and the Greenville
on the day of February 9th and they are based on a
couple of things -- I'll tell you how this is
basically generated.
This purple line or pink line coming down, as
it indicates, is the Ehime Maru.
That track was generated based on the
statements of the master of the Ehime Maru, and the
times that he said he left port, and the times that
he passed Buoy Hotel, and Honolulu Harbor, the speeds
and courses that he set.
He said he set his auto helm to course 166,
that he left at about 12:00 o'clock, he had trouble
stowing his anchor, the speed was lower at the
beginning, and so we took that into account and we
drew that track.
It was verified by some NTSB data that came
back -- the fact that the air traffic control radar
at Honolulu Airport and an Air Force radar, a similar
type of radar actually had track of a vessel coming
south-bound out of Honolulu that seemed to mesh with
what the master had said. It was on course 166,
207
looked like speed about 11. We verified that was on
166. So this track we feel is fairly accurate.
It was anchored right here at the collision
point, based on the Greenville's reported position of
the collision.
Q So for this reconstruction, you used
Greenville's reported position as your anchor for
both the Ehime Maru and the Greenville?
A Yes. That is that point right there. To the
degree that the position logged by the Greenville was
accurate, that point is accurate geographically.
I must point out that even if this is not
exactly the right position geographically, the
relative tracks between Ehime Maru and the Greenville
remain anchored to this point wherever it may be and
I think the relevant tracks are absolutely very
tight. It may be off a few hundred yards, based on
the position reported by the Greenville.
Q Could you briefly describe how the Greenville's
track was reconstructed?
A The Greenville's track on this particular
effort was taken -- again anchored at the collision
point using the sonar logger data the one-second data
which as I said a moment ago includes course speed
and depth.
208
Basically, that was back -- back dead recon,
using that data. At one-second intervals, it becomes
very precise in here, as we go back and we basically
and notated this track, although you can't read these
very well -- annotated it with key information from
other logs that were maintained by the Greenville or
data that was in the -- in the sonar logger itself.
And so, this was backed out based on the sonar
logger data. It is not tied to any other geographic
points. Basically, the only tied point on the
Greenville was right there.
Q Captain, you've had a long opportunity to take
a look at this reconstruction. How comfortable are
you with the fidelity of the reconstruction effort?
A I am very very comfortable.
I think this is one of the best products we've
developed, solely because we're using this high
intensity data from the sonar logger at every
one-second. We've never had the privilege of having
that kind of recorded data before.
Q Could we have the next slide, please?
Captain, could you describe what this track
reconstruction is?
A Yes, I can. This is -- this visual aid assist
was designed to present the difference between my two
209
organization's independent efforts. This dark blue
line right here represents the track from the
previous slide.
The green track is the reconstruction effort
from my -- ah -- at sea training team's efforts.
And as is this red line to the west, slightly
to the west of this purple line.
The N 70, reconstruction is a little bit west
in all regards and the reason that is is they not
only anchored, the two tracks at this location --
they also anchored the USS Greenville's track to
their last logged inertial navigation position rather
than just back -- they had two points anchored and
they did a best fit analysis between those two anchor
points.
But I know the ESGN on the Greenville --
although I don't know the exact performance on that
day or that minute -- can typically be off from a
five hundred yards to a thousand yards, commonly, so
I would not -- I think for all practical purposes,
these points are roughly the same based on the
accuracy of the inertial navigation.
The key thing is though the area right before
the collision this point for the last five to ten
minutes, are almost on top of each other -- they are
210
very very tight in this area.
So if I were to tell you where the most
accurate -- where they are in agreement, they are all
in agreement around this location here around the
collision point.
Q Could we have the next slide, please?
Captain, would you please describe this third
reconstruction?
A This is a product we just received this week
from the NTSB, we had at their request we had sent
them a copy of the raw data from the sonar logger.
Basically, it comes off the -- the hard drive,
and the sonar logger is converted to a digital tape.
We sent the digital tape data to the NTSB
headquarters in Washington to their Data Analysis
Group which does similar effort with flight data
recorders and so forth on commercial or aviation
accidents.
And their technical team did a similar process
of reconstructing the track, and as you can see here,
this is old data, but from coming northbound here,
the two tracks are absolutely on top of each other,
and on their display, this is actually their
presentation that they sent us -- these green track
or circles right here represent the air traffic
211
control radar information that they received from the
FAA.
So again, you see a very tight agreement, and
the NTSB did this completely independently in
Washington from our effort here.
Q So essentially what we're seeing is the NTSB
track overlaid on top of --
A -- yes.
Q -- Two tracks?
A There are two tracks over here, you see a dark
blue and a light blue track, but they are actually on
top of each other -- absolutely on top of each
other.
This red track is the track from the
reconstruction you saw on the first slide.
The only track they provided for the Ehime Maru
was the green FAA information.
Q Could we have the next slide, please?
Captain could you describe this
reconstruction?
A This product was received just yesterday.
This reconstruction was done by the Real World
Analysis Division of a Commander Submarine
Development Squadron 12, which is home-ported in
Groton, Connecticut.
212
They have a cell similar to hours, a Data
Reconstruction Group at that location. They do all
the Atlantic tactical development reconstruction
products, but they again had absolutely no knowledge
of our product before we sent them the data.
We sent them the raw sonar logger data and
said, we asked them to do two things, reconstruct the
two tracks, and then speculate or come to any
conclusions as to any of the sonar contacts that were
included in that sonar logger data that may have
correlated to the Ehime Maru.
They came back with this answer. Our
reconstruction again is in blue, theirs is a little
bit to the west, because they did much the same as my
other team did -- they anchored -- we didn't tell
them how to do the reconstruction, they chose to
anchor the position here that correlated to the
ship's last log inertial navigation position.
So again, it shows sort of a western set in
comparison to our reconstruction effort, and they
also picked a point, one of those ATC Air Traffic
Control radar points, and decided to anchor the
collision point at that location instead of the
position logged by the Greenville.
So you can see that that anchor point is a
213
little bit to the left, but if you follow along, you
can see that the -- in a relative sense -- there is
no difference in the tracks, it's just a different
anchor point for the collision.
So based on these four independent efforts, I
am very confident that this depiction of the overall
track of the two vessels is very accurately depicted
at this point.
Q Could we have the next slide, please?
Captain, would you describe what the data on
this slide is telling us?
A The data on the left side here is the -- this
pink line -- this plot right here is a plot of the
bearing from the Greenville to two different contacts
or two different items, over time.
The pink line is the bearing to the
reconstructed track of the Ehime Maru, as depicted on
the first slide that I showed up here.
The little dots, the blue dots along the track
are the log sonar bearings to contact Sierra 13, and
was also logged on the sonar logger.
You can see that the fit through most of the
track is very very close to the reconstructed track
of the Ehime Maru.
It falls apart a little bit here, but it's
214
noted that the ship's speed -- Greenville's speed
during that green banded area was greater than 20
knots, and there is some significant maneuvers down
in there -- high speed turns and so forth -- and I
believe the reason these dots are over here and not
on the pink line is that the tractor had tracked off
during those high speed maneuvers and required to be
set onto the target which is not uncommon for the
sonar trackers, the high speed, the signal from the
track ship kind of gets lower in relation to the
noise around the boat, and the tracker has a tendency
to drift off and I think that is just poor track
data.
But that end here doesn't quite line up, but
that's very very close range. And so it's -- that's
not -- that's a very tight reconstruction by
comparison to most reconstruction efforts.
So what that does is reconfirms the
reconstruction -- one added element of confidence
that the reconstruction evidence depicted on slide
one is very, very accurate, because we took those
independent track generations and we checked it
against the sonar bearings and it matched.
And by default or corollary here, I pretty much
conclude that Sierra 13 was the Ehime Maru contract
215
track. The Sierra 13 was the Ehime Maru.
RADM SULLIVAN: There are two
periods -- captain, I have a question.
There are two periods that are fairly long
periods of contact prior to the high speed maneuvers
and afterwards.
THE WITNESS: Sir.
RADM SULLIVAN: Can you tell me what
the ranges of signal-to-noise ratio were during both
these periods?
THE WITNESS: The signal-to-noise
ratio during these periods were fairly low.
I don't -- I just don't remember them right
now. I can look it up I have that data available,
and can follow-up with that information. It's in the
sonar logger. It's one of the items logged in the
sonar logger. But it's fairly low -- and depending
on Greenville's speed it's -- it's minus 5, minus 7
minus 10. But it's a lot of data and it varied a
little bit.
Up at this point the S&R in this phase right
here -- as the Greenville came out of this turn right
here, it was reducing speed, and the S&R built
because the speed of the Greenville was slowing
down. I should explain for everyone's benefit.
216
S&R means signal-to-noise ratio, and that's
really a comparison of how much signal a contact is
putting out relative to the noise around the
submarine sonar system.
At high speeds, the noise around the sonar
system gets higher, because the boat is running
through the water at higher speed. The signal from
the target remains the same, the contact.
So the S&R goes down effectively even though
the strength of the signal from the -- the true
strength of the signal remains the same.
And then the S&R built very strongly, up to
plus-7, plus-13, and I think I even saw a plus-20
there -- very much stronger. That's the strongest
track they had on the Ehime Maru, or Sierra 13.
RADM SULLIVAN: Thank you.
Q (By Capt. MacDonald) Captain, again, and it's
your N 70 or N 72 group that prepared this?
A Well, both -- both teams prepared these
plots -- plots very similar to this.
This particular plot right here is one produced
by the N 72 group.
The right-hand side of this slide is a
different plot. It has time across the bottom.
And it has range along here. And the pink line
217
again depicts the reconstructed range using Slide 1,
between Ehime Maru or Sierra 13, actually, or Ehime
Maru and -- and the Greenville.
The green dots here are -- reflect the fire
control solution data logged in the sonar logger for
Sierra 13, in terms of range. It's only range only.
I should point out a couple things about that.
These look like on this depiction here as bar,
they are very close together, and they look like a
bar.
Secondly is an anomaly of the SAR logger, that
it only logs ranges close to the nearest thousand
yards. That's why you see a streak up here, and then
it shifts up to the next thousand yards as the range
varies between, say -- well it looks like probably
8,000 yards and 7,000 yards. So it makes a step
change as the contact moves from range to range.
It shows up that out in this early time that
the fire control solution was not particularly
accurate in terms of range, but at this point there
is pretty good set.
And at this point right before the collision,
the fire control range looks very good in comparison
to what the actual range was.
RADM SULLIVAN: Captain, is that
218
the system solution that is recorded, or is it one of
the various sundry methods of doing TMA on the --
THE WITNESS: Sir, as I explained to
the court when we were over at the training center,
there are really three possible solutions that are
portrayed in the fire control system.
One is the trial solution which is what the
operator is looking at up on his screen there when
he's working the solution.
The second is the mate solution -- sort of a
place-holder solution, that he can come back to to
revisit if he has to move away.
And one is the system solution, and that is the
system of record solution for that target at that
time.
These dots portrayed the system solution.
In other words, what was accepted, if somebody
had set as the program or the system -- the solution
of record for that particular target at the time.
And that is a physical action you have to do on
the console, to promote that solution that he's
looking at to the system solution, is a physical
button push saying, basically, I believe this is the
best answer right now, he pushes that button to make
that happen.
219
So someone on the ship had to believe that that
was the range at that time, and so on and so forth.
RADM SULLIVAN: I believe you
mentioned that prior to about this time, where the
range is -- the system range is 15000 yards and
starts coming in, that prior to that time, it was not
very good, and it was really good after that.
What do you base that on?
THE WITNESS: Based on the fact that
I feel very confident that the pink line reflects
what really happened between the two ships.
It is a depiction of the range between Ehime
Maru and the Greenville, from basically 12:30 to the
collision point.
I also, from this plot over here, believe that
Sierra 13, because the bearings match so closely, I
believe that Sierra 13 was Ehime Maru.
These are the ranges for the fire control
solution of Sierra 13 over time. Before this time,
you can see that if in fact this is an accurate
depiction of the range, that the fire control
solution range does not accurately follow the
reconstructed range.
So I would say that in this period of time,
solution was fairly rough. As I described over at
220
the training center, earlier in the week, by
developing the solution is an intra-active process,
you put up a possible answer, you let it generate for
a while, you come back and revisit and reassess.
It is not that unusual in an early development
of a solution to have errors, significant errors,
until you recognize until you maneuver the ship and
restrict the number -- limit the number of
possibilities for that particular solution.
It's pretty clear in this area that that was a
pretty good assessment, but the range did not draw
down, so I would say that was a good range estimate
at that point, but perhaps not such a good course or
speed estimate. Because the range didn't follow the
track in.
But in this area, you can see that there was a
set there, and a set here and down, it looks like a
pretty good -- pretty good solution was set just
prior to the collision point.
VADM NAHTMAN: Captain, can you
infer anything about the operator looking at that?
I mean, just from my standpoint here, and not
having any experience with these kinds of systems,
you've got a fairly steady sonar contact or a very
predictable path from about 13:00 up to -- what --
221
13:25, 1323 -- somewhere in there.
I mean, there is a very good correlation of
sonar.
THE WITNESS: Right.
VADM NATHMAN: You've already
mentioned that it's a low signal-to-noise ratio, and
my understanding, listening to testimony, means that
it could be a very distant target or it could be a
small target -- a small ship, in terms of its ability
to generate noise, and that's why it was a small
signal.
So is there an indication here about the fire
controlman's technique that it takes a while for him
to figure out, or is he relying more on the system,
and then suddenly he finally gets it right?
Because I notice at 13:00 -- around 13:25 -- he
finally gets the system and the operator together to
get the range right.
And then, we immediately get a -- what looks
like a disconnect about where that target is going.
So -- and then down to the bottom.
So is there -- what do you see in there in
terms of technique or skill?
THE WITNESS: Yes, sir. As I tried
to point out in the demonstration at the training
222
center, the program that the fire control operator is
using is -- is a computer-assist mode.
And the greater the change in bearing rate over
time -- the change in bearing over time -- the more
quickly the solution will converge to an answer.
And you can generate that bearing by the
contact being close, and he's just going by you, or
you can generate that bearing by maneuvering the
submarine to generate bearing rate.
But once you start generating bearing rate, you
can start getting answers.
If you will notice on this bearing plot --
VADM NATHMAN: -- There is no
bearing, right?
THE WITNESS: There is no bearing,
right. So he's working with a low S&R contact, and
coincidentally, whatever the ship was doing, it was
not generating between the two ships much bearing
rate.
So, I would not expect to see a very accurate
solution with this kind of bearing change over time.
You could have many different possibilities.
You could have answers up here, up here, that would
probably look fairly good on display, it would look
reasonable, it would fit but -- it's not very well
223
defined yet, it hasn't been narrowed into a unique
answer.
What I think happens here if I can just -- this
is more surmise -- but you can see this is a time --
it looks like about 13 -- this is 10, this is 20 --
so 1320, right up in here, we pick up a bearing rate
and he comes to a answer.
And generally speaking, what he saw before here
was the ships -- he's matching the bearing rate with
a solution, and generally speaking, they are two
possibilities, initially, that would match bearing
rate -- one with a closing, and one with an opening
close that matches the same amount of speed going
perpendicular to the line of sight between the two
vessels.
You match those, you match the bearing rate,
and that's what he probably did.
But there could be an opening and a closing
aspect. And I'd say based on the fact that his range
continued to generate here did not follow the track
in, he selected an opening aspect.
Just on analysis, I happen to know what that
solution was that he logged in there. It was in fact
an opening solution and that's again not
unreasonable.
224
The general technique is you assume a closing
aspect for conservative sake, if you don't know one
or the other, you would tend to pick a closing --
want to pick a closing one and evaluate that one
first, if it doesn't work out right, then look at the
flip course refer to his flip course which is the
opening aspect.
RADM SULLIVAN: When you say you
would "assume" are you talking about just the fire
control total operator or the sonar or anyone else
who may be involved in this problem -- who would --
when you say "they" would normal assume a closing --
THE WITNESS: Mentally normally
trained to all of our parties is to start with a
closing solution, because that could generate to a
more tactically challenging position, start with that
first, if that doesn't pan out, if you don't know
which one it is, then evaluate the other.
In fact, there is a button on the fire control
screen that just says, flip course, and it will
portray the opposite course and lets you do a quick
very easy analysis to look from closing to opening
and see which one you like better, which ones tends
to fit better. It's designed in. It's recognized as
often being the case of two different possibilities.
225
So I would say the whole party goes to work
initially with a closing presumption, and then tries
-- if that doesn't pan out, then you go look at the
opening.
RADM SULLIVAN: Just a couple quick
questions.
This last dot -- can you explain what that
means and what is -- is that an accurate range?
THE WITNESS: That is not an
accurate range. I know a little bit about that dot.
It's back out at 9,000 yards.
It would indicate -- there is a couple of
interesting things about the timing here.
For instance, this update to this close range
position happened after the ship was ascending to
periscope depth.
And this one was done -- this update was done
after the collision.
RADM SULLIVAN: Which was updated?
THE WITNESS: This one, the one that
comes from this range, down to this range, that
update happened when the ship was already making its
ascent to periscope depth. And this update was done
after the collision was done.
And I've thought about that quite a bit -- how
226
could that happen, or why is that in that time
frame.
And again, I don't know precisely, but I could
be through a scenario which sort of explains all that
a little bit.
As I mentioned, the system solution is a
discrete action to hit a button on a screen that
says, I buy what I am seeing on my screen, I want it
to be the system solution, I think this is good.
And he pushes a button, and he makes it
system. That's after a period of time of evaluation
and looking at the process, and watching the solution
generate, and evaluating that bearing difference dot
stack that I showed you over at the training center.
So actually, the good solution was probably
portrayed on that screen before the ship began its
ascent to periscope depth, because the fire control
man was sitting there evaluating that solution
probably while the ship was one hundred fifty feet
making preparations to go up.
Once the ship ascended, began it's ascent, he
got around to saying, I'll update system solution at
103 feet going up, because I really believe this is a
close fit. That is a possible scenario.
I don't know -- that's purely speculative in my
227
view.
RADM SULLIVAN: I understand.
THE WITNESS: This dot, in
discussions, interviews with some of the personnel,
particularly the fire controlman during the NTSB
interviews, he indicated that the ship got to
periscope depth, back in this area, somewhere right
before that time, the scope was -- he looked around
for the contacts, no contacts at this range.
Two to three thousand yards were reported or
seen by the scope operators, so the fire control man
assumed that the contact could not be that close he
needed to be farther out, otherwise you would have
seen him because it's so close.
And he, in his mind, thought well, I have to
make something work farther out. This solution is
not accurate. It must be farther out.
Now the rest of the data for this particular
dot shows a solution that is not possible. It showed
a 99 dot target which indicates to me that he
couldn't make it work it didn't fit.
And that's typical of trying to make the range
go up, but the rest of the parameters don't fit to
make this dot stack stay vertical and zero.
So he's trying to make it work but it's not
228
working, the collision happens, he's distracted with
the SAR efforts or further on duties, for some reason
he updates the system on something that is farther
out. I can't explain exactly why, but that is a
surmise.
RADM SULLIVAN: He might have had it
in trial?
THE WITNESS: He might have had it
in trial, trying to make it work, assigned some other
job, and then ended system.
But the trial had to be up for a period of time
before he promoted the system.
RADM SULLIVAN: One more question,
please.
In your learned experience at sea, doing this
job, how would this sonar supervisor describe the
type of track he has on Sierra 13, both prior to the
high speed operations, and then after the high speed
operations?
In other words what type of report would he
have made? Again, this is your opinion, but what
sort of -- if you call the track, what would he have
made?
THE WITNESS: I mentioned earlier
this is lower S&R, but that's good sonar contact.
229
The sonar operator would say, I am tracking a
contact, bearing O10, medium S&R contact.
If he saw a contact with plus-15 to plus-20
S&R, I would expect him to say that that would be a
fairly -- indication of a close contact.
It's fairly loud.
RADM SULLIVAN: Would you expect to
-- again, this is typical in your speculation -- to
be able to classify that sort of contact, other than
the fact it's a service contact?
THE WITNESS: Yes, he would probably
classify it -- you could classify it even by nature
of sound as to heavy ship, merchant, warship, light
merchant. I would not be surprised -- Ehime Maru to
be classified on a light merchant, based on nature of
sound, and there should be some classification data.
He should be able to get some classification
data on that trying to indicate how many screws it
had, what speed it was going, and make a more
definite classification.
RADM SULLIVAN: Thank you.
VADM NATHMAN: I have a question.
My understanding was from earlier testimony
that if you had a low signal-to-noise ratio target,
you could make a couple of assumptions without
230
knowing much about the target. It could be a target
that was along the system, or it could be a
relatively small target close in.
THE WITNESS: Yes, sir.
VADM NATHMAN: I am saying -- the
fire controlman seems to sense that it's close in.
His first, you know, trials or -- about where
this target is -- he's sensing, well, I'll try close
in. So he's assuming that he's got a small type of
target, or a quiet kind of target, I guess, that's
close in. And he goes there for a while until he
decides, ah, I know what it is this is here.
THE WITNESS: You are referring
from the first part to there?
VADM NATHMAN: Then he says, well,
maybe it's going away. But there doesn't seem to be
much correlation and collaboration between sonar and
fire control on signal-to-noise ratio.
So who is getting the information on the
signal-to-noise ratio on Sierra 13?
THE WITNESS: It's presented on the
fire control operator's screen that is part of the
data that is transmitted from sonar in addition to
bearing -- the signal-to-noise ratio is presented
there, and it's available for his own personal
231
analysis. It doesn't have to be conveyed by voice,
or anyway, it's sent with the data coming from
sonar.
But I wouldn't -- I wouldn't put too much
conclusion on a low S&R -- lower S&R target.
S&R on these are not really low, I mean, they
are sort of medium. And that could be, as you say,
due to a distant contact, it could be to acoustic
shadowing, if you have sounds being bent to the
bottom. It could be due to a quiet contact -- one
that is isolated.
So the normal practice for a fire control or
what the guidance says to him is to select the
predicted range of targets of that nature, based on
the acoustic conditions, and start with that range --
just work that first, if you know nothing else about
it.
Now, sonar in some cases can provide range
information and get them in the ball park.
In this case, I don't think it was available,
and then starting at 10,000 yards, it's probably a
little too close, honestly, based on the sound
conditions available that day -- I have data that
shows that the sound conditions were great, and they
could hear 30,000, 15, 30 to 40,000 yards, all the
232
way to land, basically, from where they were.
So I -- if -- I was inspecting this individual,
I would say he's not following the guidelines, he
should be starting out here someplace, probably where
these dots are, a reasonable start.
VADM NATHMAN: But it goes to my
next question.
Is anyone backing this individual up?
I mean, signal-to-noise ratio, it seems to me,
is an important consideration right now, i.e.,
wouldn't someone look over his shoulder -- the
officer of the deck -- is there some interest here?
It all goes back to, you know, what I think maybe is
this point about the CEP, you have a low
signal-to-noise ratio to target that day, you've got
someone trying to guess in trials -- is there any
backup for the fire control technician on watch to
say, I think you may be off here a little bit because
my correlation says it ought to be here.
So that information goes to what -- the officer
of the deck? Help me with this one, so I
understand.
THE WITNESS: Yes, sir.
In a normal tracking situation, sonar --
although they have no processing tools there to
233
develop the solution, they will try to do mental
estimation of what the range is, and they'll come up
with their answer sort of -- it's sort of a
competition.
The sonar man likes to try to come up with the
answer independently, and derive it by doing mental
analysis using mental power to come up with an
answer. And they will share that answer with the
fire control man. They will get on the phone and
ask, what are you holding for solution on Sierra 13?
And if there is a great disagreement between
the two parties, they will come to some resolution
they will talk back and forth. That's at the
operator level.
Additionally, the office of the deck, who is a
direct overseer of the fire control man on the watch
has an opportunity to provide feedback.
He says, based on the conditions today, I think
the range is farther, he's the fellow that would be
back, you know -- the team concept between sonar,
fire control, and the officer of the deck in
processing the contact is how that feedback is
supposed to occur.
RADM STONE: Could I just follow-up
on this. Because for me, it's a very important point
234
in understanding the internals here in the control
room on that team work together.
If in fact on any submarine the FTOW makes a
mistake, and he breaks down on that job, there are
mechanisms in how our nuclear submarines operate so
that you don't have a single point failure if the
FTOW doesn't do his job.
THE WITNESS: Correct.
RDML STONE: And your answer to that
question is, what happens if the FTOW doesn't do his
job properly? What are the mechanisms that are
supposed to kick in to keep that from turning into a
major incident?
THE WITNESS: We can go back a
little bit, and say that early on a tracking phase
like this, it's not uncommon to have inaccurate
solutions.
We'll back up a little further.
As I said a minute ago, when you are focused on
contact analysis, that is the focus -- the team work
between the fire control and the sonar and the
officer of the deck is behind them.
If the contact challenge is significant and we
are tracking many contacts, or we are in a hightened
condition of readiness, forward engaged, we might
235
have more parties out there that we -- maybe a second
fire control operator, a couple more officers
directly overseeing the development of these
solutions.
In a day steaming situation, much as Greenville
was involved in, the watch is reduced because while
the Greenville is submerged at 400 to 600 feet,
essentially, no threat to Greenville or to the other
ships around there. So it's looser track. It's not
-- that is not -- we're not focused on contact
analysis during those maneuvers.
We're trying to keep an idea where everybody
is, we still don't like to run under even a ship that
is just steaming by -- I mean, that's just not a good
practice, we try to avoid that.
But during the periods of time before going to
periscope depth over here, is what you refer to is a
sort of loose tracking going on of this contact.
The exact location of that contact is not of
critical importance to the submarine at that point,
so it's pretty much being done between sonar and fire
control, but the moment the decision is made to go to
the interface, we're going to go up to periscope
depth, we're going to bring the submarine up near the
surface, now contact management becomes centrally
236
important, and that's when it's obligatory that the
officer of the deck becomes directly involved in the
over site, and this three-way team concept kicks in
-- between fire control, sonar, and the officer of
the deck -- to manage the contacts and understand the
contacts with much more resolution than when we're
just steaming around at 400 or five hundred feet on a
trip somewhere else.
Does that answer your question, sir?
RDML STONE: Thank you. That's
helpful.
RADM SULLIVAN: You talked about the
fire control system in sonar. But where else in the
control room is it displayed?
THE WITNESS: It's normally --
normal conditions for general tracking, it would be
displayed on the remote sonar repeater, which on this
particular day was out of commission.
But that would be available on the officer of
the deck. It's a repeat of what is on sonar.
RADM SULLIVAN: You refer to that as
the AVSDU?
THE WITNESS: The AVSDU. It's also
properly maintained contact evaluation plot. The
plotter is supposed to maintain a plot, so you can
237
see rising S&R information.
So it is on each of the fire control screens.
It's on sonar display. It's on the plotted data.
RADM SULLIVAN: Is it on some other
displays -- I don't know what they call it on
Greenville -- but a display of contact that you are
tracking -- displays that shows S&R bearings
different contacts?
THE WITNESS: As I said, it's right
on the fire control data that is coming in to that
contacts page. He can look right on the screen for
that contact, get the bearing, the time, and the S&R
of that particular target.
RADM SULLIVAN: Okay, thank you.
THE WITNESS: You know it is
displayed elsewhere, there is a stand-alone computer
in there that takes sonar data for independent
analysis, in which case was not on the day in
question, was not employed, not being employed.
CAPT MACDONALD: Mr. President, I
recommend that given the lateness of the hour that we
recess for the evening.
VADM NATHMAN: Very well. This
court will be recessed in 0800 tomorrow morning.
(Court adjourned at 4:30 p.m.)
Choose a session: Day 1, Session 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8
Day 2, Session 1 | 2 | 3 | 4
Day 3, Session 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8
Day 4, Session 1 | 2 | 3 | 4 | 5 | 6 | 7
Day 5, Session 1 | 2 | 3 | 4
|
