Unofficial transcript: Day 5, Session 2
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
Feb 21, 1994
SESSION TWO --- MARCH 9, 2001. DAY 5 8:40 a.m.
BY RADM SULLIVAN:
Q It's my instinct, I think through this is when you press
the clock in the back of my head it was always you run the
risk of pure solutions are not going to be as good as they
could be?
A That's correct. You lose, as I just said, you lose
precision, you may make a improper conclusion.
Q Doesn't mean it's not the right thing to do, but it's
just something you have to weigh?
A You have to keep that in consideration, that's correct:
That's -- you know, the -- there's always, I think Admiral
Nathman was talking about that, your thresholds go up.
You know, operating a submarine in any condition under
any circumstances under water is a risky event. You got a big
ship, lot of steel, lot of people under water. If you ask the
average public person you'd say is that a risk free event,
absolutely not. That is risk involved in going to sea.
When you decide to go to periscope depth, the risk of --
goes up a notch. We're going up toward the interfacing, we're
in our own environment while we're deep and it's pretty safe
down there. There's not many things -- there are hazards but
it's -- relatively speaking it's safe compared to going to
periscope depth.
The risk factors go up, the obligations to mitigate
those risks go up as well and you have to span the time
required to make sure that the risk is under control before
you go up there.
Q Just a -- for background question, when you discuss or
you talked about mental analysis, it's been a long time since
I've had to do this, but for my other Court members, can you
describe in great detail what you're talking about?
A I'll try to do it so it's not too mental, but it's
sometimes difficult. What we do, we -- we're trained on the
principals of relative motion. And it really goes down to
line of sight analysis. You're trying to look at -- draw
conclusions.
If we could go forward two slides I can give you an
example to answer the Admiral's question. These hill pictures
on the side is what you kind of visual or mentalize. You --
you're trained to do this sort of mental analysis, draw these
little pictures, start to form those in your mind as what is
the contact doing.
For instance, on this situation where the contact is
drawing -- let's take a hypothetical situation. I look up, I
see a contact drawing right at a fairly good rate, this little
diagram here is what you construct in your mind. I'm on
course 340, the bearing to the contact is 007, and I don't
know this arrow. I'm trying to formulate this in my mind.
Where could this arrow be?
In this case we know the solution, but this is a
hypothetical case. I don't know this arrow, but if I see this
high bearing rate of right six, I would pretty much
automatically say, well, it's -- there is a possibility he can
be coming in this direction and I'm drawing that bearing rate
but it's pretty low and I would probably say most likely the
arrow is coming in this direction. I don't know if it's
pointing down or pointing this direction or pointing to the
right, but I know probably he's going opposite direction.
And you go through this process. There are actually
(indiscernible) that can say based on a right six you can
makes assumptions with respect to my speed and his speed and
you come up with ideas of what the range could be. There are
formally -- we talked how to make those calculations simple,
simple division problems.
The second maneuver, after I changed the course, if this
contact has not changed course and it's new zero bearing rate,
you say that eliminates this possibility, he could not have
been going this direction. If he was, my bearing rate would
be going to the left. And all of a sudden I say he has to be
in this direction and my speeds have to be the same, this
component and this component have to be the same to have the
bearing three.
So you go through this, you say I now know he's either
this way or that way, speed's matched and I have a pretty good
idea of what the course is and I can do some range
calculations and say the range is about this range. That's
what we do, and that is taught in basic submarine school.
It's taught to the FTOWs, the sonarmen, that's how we do
mental analysis. Does that answer your question, sir?
Q Yes, it did.
BY VADM. NATHMAN:
Q Captain, as a follow up, I'm not a submariner but I do
understand your calculi and your -- I think I understand what
-- obviously you have a way of training that builds in your --
your receptors go up, your thresholds change. You talk about
thresholds changing for going to periscope depth, you have to
be more careful.
One of the things that I want to understand is I do
understand what I call constant bearing. Constant bearing to
a Captain on a surface means you could be in real trouble.
Constant bearing decreasing range means you got a problem, you
got a collision if you don't change it. You've got to change.
And what I see in this one here is -- is constant bearing.
A Yes, sir.
Q I don't see a decreasing range, but what I do know from
the analysis is this implication of increasing signal to noise
ratio.
A Sir.
Q So, does that become -- is that something -- in other
words, if you knew that you had constant bearing and you had
increase in signal noise ratio, is that the same analogy to --
A Yes, sir. In fact, we even hold constant bearing now
can be as being a trip wire as well, trip wire to potential
close CPA, closest point of approach.
We're looking for that as an indicator -- indication of
trouble of close quarters.
The same as -- same principals apply under waters above
water, same things that you're looking at. In a normal
encounter, long range, you could have a zero bearing rate
situation under two possible conditions: One, where they are
enclosing -- a closing aspect like this one where this
situation indicates collision is imminent -- is inevitable if
you keep this orientation, these two vectors will end up at
the same point. Very close quarters.
The other possibility is he could be very, very distant,
maybe 40 thousand yards and his bearing rate is just very slow
and it's slow to develop. But, in those situations long
bearing rates like long range contacts, there will be a
bearing rate over time. It may be very slight but he'll draw
away eventually. You have to look at it over a longer period
but you'll recognize that he's moving in and he's a long
distance contact.
Furthermore, the long distance contact, as I maneuver my
ship....
Q He won't change.
A He won't change. In this case it will change and you'll
see that the contact is close. So the combination of SNR, the
reaction of the bearing rate, all those things would indicate
zero bearing rate.
In fact, if they have a zero bearing rate situation and
someone calls it out and says Sierra 13 has got a zero bearing
rate, had a zero bearing rate for ten minutes, he may be
closing contact. I'd expect that's a good condition, a good
stimulus to say let's take'em across the line of sight and
check him for range to see how far away he is. That would be
a good approach. That would be good target motion analysis.
Q This is why I went -- I want to go back to my question
about who else we should scrutinize.
We have this thing about this late fire control solution
and that being kind of a trip wire and we heard a lot of
comments on this, but this is all on sonar. I mean, you've
got sonar -- you got a supervisor of the watch in there who I
assume he's very skilled, he's got a lot of experience, you
got two sonar technicians, ones that are on the watch, one is
on one panel, one is under instruction not properly
supervised, we're still figuring that one out. Certainly
between the two qualified guys they know they've got a
constant bearing rate for a period of time that looks to me
like two minutes, three minutes, four minutes, somewhere in
there, but for sure in their analysis they're seeing segments
of this. And they know, I think -- I believe they know
because they're the sonarmen they've got a decreasing signal
to noise ratio. Is this a signal that would make light bulbs
go off?
A Yes, under normal conditions it would. Please
understand that this -- this is like magnified like 20 times,
it's blown up, significantly in larger scale than would be on
his displays.
This would look -- this would be just a few minutes of a
tail down at the bottom and this is why I come to say the trip
driving here did not help the situation. If this had been
more than just a minute of data steady course here, this is
really only a minute and a few seconds, and they're turning
again, it would be very apparent. It may just be a great time
to go --
Q Before you do, --
A They would have seen that. The only thing I can say,
sir, is that that's the real critical thing, how much of this
did they really see unfolding in front of their eyes. It's
the combination of the two of them. If they had really been
on their toes could they have seen it? Yes, maybe, if it was
right there. I don't know.
I've tried to replay this in that same simulator and
it's 20/20 hindsight, you know, the scenario. You can see it.
But it's not -- it doesn't leap off the screen at you and say
oh, my gosh, it's a real close contact because this steady
part was so short.
Q Yes, sir. Let me -- let me -- I want to understand this
from your sense, 'cause I -- I agree, this is blown -- I mean
this looks very clear, you can drill this to death and that's
not reality. Reality is we saw the displays, so the reality
is this is a very small display and then it's aggravated by --
not aggravated, but it's complicated by the small amount of
time. But what I was trying to focus on here was from the
time 36 to 38.
A Right.
Q That looks like it's a clear -- you're not in the turn,
you've stabilized, so in a sense their antennae, their
sensitivity should be elevated. Now, their sonar's got a
constant bearing contact, although it's for a relatively short
amount of time, but very, very critical phase here, I believe.
And they know they have decreasing or increasing signal to
noise ratio.
A Yes, sir.
Q Now, -- now, put it in perspective for me. Now take me
back to that room and say --
A Those are all key things. I'll be just -- just to take
the other side for a second. Let me put some other mitigation
in there.
Commander, if we could go to number 10, slide number 10.
Just keep going. Keep going. This one.
This is a plot of all the contacts that the -- this is
a contact evaluation plot, it shows time along the left side.
It shows all the contacts being tracked by the sonar system on
the ship at the time of the day. It's reconstructed, we took
the sonar logger data and basically back generated contact
evaluation plot.
This is Sierra 13, and you see that zero bearing rate,
zero bearing rate and then you have some tracked off time and
then this little segment of right bearing drift, go to the
next slide, please, this picks it up here and then this little
right bearing rate in the back is steady again.
This more or less replicates what they would have seen
on their display in the long term history portion of the sonar
display that I showed at the training center. He -- you could
make a case and say, well, I don't know, maybe that was
tracker drift, maybe it didn't -- maybe the tracker tracked
off a little bit but it's back on its normal zero bearing rate
solution. It looks like it may be a distant contact.
We've maneuvered across the line of sight, it's back to
zero, it's almost the same bearing rate as it had. Remember,
they don't have this part yet. So, it's not that
inconceivable based on this very short leg that they could
kind of dismiss that as being bad track during this maneuver.
And -- and that's what I'm trying to get at, is that
time would have helped tremendously here. A little bit longer
time on that 340 leg would have made it clear as can be.
That in combination with the zero bearing rate follow on
would have locked the solution immediately, we would have
known everything there is no know about Sierra 13.
So, to say that the sonarmen -- that's something you'll
have to come to grips with honestly is should the sonarmen
have picked up on the fact the guy is close and there's
indication and the bearing rate is changing, you could make a
case and say yes, but you could also say there was -- this
other data displayed that would say, well, maybe he's far
away, should they raise their hand and called more attention
to this, this may be a close contact? In hindsight you'd say
obviously you should have spent more time doing that. But I
can kind of understand also why it didn't leap off the screen
here.
BY RADM SULLIVAN:
Q Just to follow up on the Admiral's question. You are,
again, the force training officer.
A Sir.
Q Is it fair to say your knowledge of watchstanding in the
sonar space is good?
A Yes, sir.
Q Knowledgeable.
In prior testimony or discussions there was some
discussion about the common waterfront practice of having one
of the watchstanders in work share on (indiscernible) broad
band being under instruction watch. I assume his oversight
watch is either the supervisor or the other operator. Can you
comment on that?
A Yes, I can.
I very -- this came to light during our interviews with
the sonarmen during my National Traffic Safety Board -- role
-- investigation role, and as I was very upset by that, kind
of bothered. Let me explain what I know about what really
happened in the sonar room so it's clear.
The fact is that the Petty Officer or the seaman on --
the operator on the workload share was not qualified operator.
But, the fact of the matter is there was a fourth person in
sonar, STS1 Reyes had come into sonar and picked up his
jacket. He is a qualified operator and he came into sonar
just prior to the time the ship was getting ready to go to
periscope depth and recognized that factor, and de facto
stationed himself as a watchstander behind the workload share
operator.
And in his testimony to the investigators at the NTSB,
he described the situation where he became very engaged with
the contact analysis. In other words, it was not just a
casual stay behind, he did engage himself in the analysis of
the contacts. And I -- I think was an up-check for this young
fellow. He recognized, hey, this guy sitting here is not a
very experienced operator, I'm going to stand behind him and
make sure that this goes right.
And the reason I know that he was engaged is because in
the process of going through the reconstruction and analysis
it became I think clear to him that Sierra 13 was in fact the
contact. He didn't believe it when he came into the
interview, and at the end when he kind of came to the
conclusion on his own that Sierra 13 was the Ehime-Maru, he
actually lost his composure. He broke down and felt very bad,
obviously, that he had missed that contact. So I'm absolutely
convinced this this Petty Officer was engaged in the
situation. He was a player in there.
So, technically there were two qualified operators, plus
a supervisor through no fault of the plan or the watch bill,
or the situation, just because Petty Officer Reyes happened to
be coincidentally in sonar.
Now to get to your question --
Q Before you leave that, he wasn't directed to take
station?
A No, he wasn't, he did that on his own.
Q By anybody?
A My assessment of the interview was he did that basically
on his own volition. His own sense of obligation.
BY VADM. NATHMAN:
Q Captain, would he logically, then, or -- have knowledge
of the changing signal to noise ratio?
A He probably did not. He did not have the long time
history. He kind of stepped into this problem underway, in
the middle of the story. But he is -- he picked it up while
they're doing the baffle clears and steering around, so he was
looking at the contact motion.
And then I got into this issue about talking to the
sonar supervisor in his interview he said, oh, yeah, this is
common practice, we have these unqualified guys sitting in
here. That's how everybody learns. And I said to myself, I
was very -- I was not happy about that answer because I said,
I am responsible for this area of submarine force training and
maintenance of the sonar watch stations. And so I did some
independent investigation.
First of all, I found out just to reassure myself that
there's nothing written about this that would allow this to
occur. Both the NWP, Naval Warfare Publication for operation
of the sonar system, and the standard submarine operations
regulation manual both specifically say that no unqualified --
only qualified personnel are allowed to be stationed on a
watch station. It's very clearly spelled out. There's no
ambiguity there, there's no footnote except for any of that
for sonar, only qualified operators should stand on the
consoles.
So, I wanted to find out if there was sort of a
waterfront practice that was going on in this vane. I called
two -- several different people, two different groups of
people really, the -- the Command Master Chief level folks
that assign and write the watch bills and fill in the names so
the people who are supposed to stand watch develop the watch
bills, and asked them folks to -- is there any practice -- I
didn't tell 'em the background, I just wanted to, you know,
kind of -- I didn't introduce the background, sort of an
unprompted question. Is there any practice which, you know,
the watches can switch themselves in sonar at the discretion
of the team. Can you put unqualified people on the consoles.
And none of the Master Chiefs, Command Master Chiefs that I
talked to say absolutely not, you have to have a qualified
watchstander in that station.
Then I talked to my, one of my sonar inspectors. I
mentioned I have this underway evaluation team, that code 70
group that I discussed yesterday, and in there are some senior
sonarmen who do underway evaluations of sonar watch sections.
And I asked one of the inspectors that I have a great deal of
confidence in what's the status on this, do you ever find
situations where unqualified people are sitting on the
consoles when you do your underway evaluations? And he said
yes. He said I've probably found out about 20 percent of the
time, I pointed out immediately as a problem, and we need to
correct it.
So, it's -- it is an issue that I have to come to grips
with, 20 percent is not adequate in my mind, it should be zero
percent, but there is apparently some sense among some of the
ships that it's okay to have a non qualified watchstander.
But there's nothing that condones that policy in any things
that we have written. It's not a state of policy of submarine
force, that's for sure.
BY RADM SULLIVAN:
Q When you said 20 percent, was that biased toward a given
squadron.
A No, he said --
Q Configuration?
A He couldn't, you know, he said I -- this is an viceral
calculation, I don't have -- he didn't have, you know, any
distinction on any particular squadron or, you know, any
unique boats of any kind. But he says I probably caught that
about 20 percent of my rides, and he rides many ships on all
squadrons, both here, Pearl Harbor, San Diego and up at
Banger, so he rides across submarine force wide.
BY VADM NATHMAN:
Q Let me ask you a follow up question, Captain.
You've got an experienced guy that's part of the
training team that senses that there's members of the force
using an improper method for manning watch stations. Did he
get off his circuit, you know, his chief circuit, did he go
back to the chiefs of the boat, did he go back to senior sonar
watchstanders to provide feedback to the squadrons? Did he
try and close loop this all, or just now it became an issue
his reaction was I think maybe we got 20 percent of the force
out there maybe what they're doing is wrong?
A We, I -- I don't know -- I didn't ask that question
about how far did you take this issue. I was really more
interested in the immediate answer at that time, but it's --
the way that's processed, I mean that's brought to the senior
team leader right there on the ship immediately as it happens.
And they -- this group holds, just tell you what they do, I
don't know whether this issue was really brought up at this
thing, but they hold seminars and group training exercises
that disseminate common problems that they see on the various
ships they ride. There is a process by which that -- those
issues they find are disseminated. Whether or not this
particular issue was disseminated to any of that training at
any of those training sessions, or he they also put out
messages that talk about common problems to all the boats.
Whether those have been discussed I have to get back to you.
I don't know.
Q I would appreciate that.
A I'll have to follow up on that. Those two -- both my
sonar team people are underway this week doing training. As
soon as they come back from sea I'll ask them those questions.
BY RADM SULLIVAN:
Q Captain, I know I'm kind of drifting away from the
testimony, but why -- we've had some discussion here about the
use of the sonar work tape, the one quarter inch tape recorder
I believe. Can you tell me what that is really used for?
A Well, it's -- it's no longer -- it's a different than a
one quarter inch work tape. I hate to tell you, sir, but --
the system that you're familiar with is long gone.
Q Thank you for that comment.
A We did -- we do have a work tape in sonar, and this work
tape is used for if something of interest occurs. If you --
if something happens that's of interest you would like to have
ability to replay and listen to the event again. And to
capture that event on tape so that it could be used for
further analysis.
In this case the work tape system that was used on the
-- normally be used on the ship, best of my understanding
again through the NTSB investigation, what this work tape
system was being used to play back ocean sounds for the
visitors on the ship that day. And although there are no
visitors in while they were going to periscope depth or during
the actual period right prior to the collision they had
stopped the tape but had forgot to reload follow-on work tape
to start the work tape process again.
Q But as a senior submariner, if you walked into a sonar
control space and saw sonarmen using this tape recorder as a
demonstration for sounds of biologics, whales, what would be
your reaction?
A I'd say why do we have this distraction going on in the
sonar room, number one; and what are you using as a work tape.
We need to have a work tape going. It's -- I mean there are
periods of time when you clean the heads on that a quarter,
you don't stop the ship if you have to take that system down,
it's not a critical -- you don't -- the system goes down you
cease operations. It should be running, it's a standard watch
standing practice but there has to be maintenance done on the
tape recorder and everything else and it's not unusual to have
it off-line for short periods of time, but you should not be
operating for long hours without the work tape going.
Q Okay. Thank you.
BY RDML STONE:
Q Follow up on the issue related to the 20 percent or the
anecdotal number for the watchstanders that may not be fully
qualified. If in fact a command makes a decision to go down
that road and not meet the requirement of a fully qualified
watchstander, do you think it's fair to say that making that
decision by the command incurs increased risk to the operation
of the ship when you make that decision?
A Yes, it does. Clearly the rationale behind him only
having qualified people on watch is you want to make sure that
they know all the information they're supposed to know to
operate that console.
Now, I want to put this in proper context, that all the
watchstanders on the panel, the panel operators in sonar are
under the direct supervision of the sonar supervisor.
If at any place, and I'm not condoning this at all, if
at any place the risk is less of having a nonqualified person
there, I suspect -- I'm not really saying this right, but all
the operations of those four panel operators are under the
direct observation of a direct supervisor.
It's a different situation if you had somebody operating
the diesel by himself and he was not qualified at all. I
mean, the seriousness in my mind of an independent operator
around the ship being nonqualified is higher because there is
direct supervision here. But I -- I'm not condoning it, it's
not right. I'm just telling you that there is inappropriate
action by this operator would be caught by that supervisor and
if he's not doing his job right the supervisor would get 'em
out of the way or move 'em out, get somebody else in there.
So it's sort of a -- this is a directly supervised watch by a
senior sonarman.
Q With regard to target motion analysis, the submarine
force in fact talked to the surface Navy when they received
our total rays or 19 rays, a lot of lessons about target
motion analysis because the submarine force I would think it's
safe to say is one of the world's leading experts in the art
and science of target motion analysis because of the medium
you operate in. Would you not agree with that?
A Yes, sir.
Q One of the lessons that was frequently being reenforced
on board our ships was this issue that was raised earlier by
Admiral Sullivan is the relationship between time spend on a
TMA leg and the quality that you would get. In other words,
if you cut the time short, the lesson that was constantly
enforced was that you would be affecting the quality of the
product. Could you say a few more words about that
relationship between time on a TMA leg and quality?
A There is a need for both the mental back up and the
machines that do the target -- machine assisted target motion
analysis to make sure that you have consistent tracking data.
In other words, you need to be able to look and say that the
data that I'm receiving for analysis is consistent and honest.
It's not subject to excessive data scatter.
For instance, if we go back to the SNR comment, if the
SNR is low, the ability for the tracker to stay on the target
is degraded. And it may hunt back and forth across that
target.
So, if you just looked at two sonar bearings, two dots
and this tracker is hunting back and forth across the contact,
you're -- you could make an extrapolation between those two
dots for bearing rate that would be totally inappropriate
because they're wrong. They're on the edge of either side of
the sonar contact. But if I have a string of data, maybe 10
dots or 15 dots, you can fair through with your eye or the
machine can fair through with a cursor the real trend of those
dots.
You take out the scatter, the noise of that tracker.
And clearly in total ray systems that's even more of a
problem, the trackers are not as accurate. So, I mean, we're
not dealing with total rays here. This is a spherical ray
with very good trackers in it. The longer you have the more
assurance you have, the data you're looking at is consistent
and reliable and high quality. And time is required to make
that assessment.
The amount of time required is dependent upon the
situation. You need to make enough data there to convince
yourself that the data you're looking at is real and accurate
and that the bearing rates are real and accurate.
If you have strong SNR you might make that conclusion in
just a couple minutes. If you have a weak SNR it may take
five or six minutes to get a good bearing rate. That's kind
of what we say with total ray analysis. Spherical ray you've
heard the number three minutes and we talked about it earlier
today. It's a minimum of six minutes because the bearings are
not as stable.
Total rays are not in effect here in this particular
incident, but it really it dependent on the sensor you're
listening to and the conditions you are encountering.
BY CAPT. MACDONALD:
Q I'd like to back up a few slides to the USS Greeneville
parameter slide.
A Yes. Here we go.
Q Captain, would you describe for the Court what this
diagram, what these charts depict?
A Sir, again, this slide is generated from our
reconstruction analysis equipment. Basically once we settle
on a reconstructed track, which we discussed basically slide
one from yesterday, once we have that data in the machine and
this -- you can ask the machine to print out slides like this.
So this is reconstructed date based on the reconstruction that
we've -- that I have pretty much accepted as being very good,
the first slide I showed yesterday. And it showed there's
three different plots here obviously, time across the bottom,
time scales are consistent. And this basically shows you a
picture of USS Greeneville's course over that time between
time 1330 and time 1344. This shows its speed over that same
time interval. And this shows the depth of the Greeneville.
This is really taken from the ultimate source of all this data
again is the sonar logger data.
You can see that on this -- that's basically what this
slide depicts.
Q What I'd like you to do, sir, is if you would, could you
take us through the 340 leg, the time Greeneville spent on the
340 leg and discuss course speed and depth?
A Yes, I can. The 340 leg is depicted from here, you see
000, this 350 and 340 is right in here. That's the 340 leg
right there.
You see it came down and it looks like about time 1331
and 40 seconds, and lasted the time almost 1333 and maybe 20
seconds.
Q So she was steady on course 340 for how much time?
A A minute and 25 seconds, something like that.
Q Okay.
A Steady on course.
Q Would you now move down to the -- your speed slide and
discuss, again for that same time period, was she -- what was
her speed?
A Her speed was ever decreasing. You see it never really
stabilized during any of this period of time that she was
steady on course 340.
Q And from what speed, from her highest speed to lowest
speed during that time?
A She starts out at about 18 knots and drops down to
somewhere around 10 knots during the time she was on 340.
Q Okay.
With respect to speed, Captain, what would be optimum,
or the acceptable speed for conducting TMA?
A Generally you'd like to go at a steady speed, actually
the -- the best speed you could make and still track the
contact is optimum, because you can drive the highest bearing
rates with the highest speeds.
So, you know, if you could -- 10 knots is generally a
speed we try to go with, you go a little faster, maybe that's
a little bit better but we usually go 10 knots.
There's another factor that comes into play. You see on
the depth scale they're coming up to 150 feet, which is the
normal launching point for going to periscope depth, as the --
as you come up to 150 feet the faster you go the more likely
you are to cause cavitation, only own ship's noise and that's
considered bad practice. We try not to cavitate to make
unnecessary transits in the water. Typically good speed is 10
knots, maybe 12, maybe nine, something like that is a good
speed to steady at.
Q And speed is important because?
A It aides in your target motion analysis.
Q And why does it do that? Why does a 10 knot speed, why
is that better than 15 knots?
A Actually 15 knots would be a better speed for target
motion analysis, per se. The greater the speed the more
you're going to be able to drive the bearings and assess -- do
your ranging and so forth. The better it is for target motion
analysis, but 15 knots at periscope depth -- I mean at 150
feet is kind of a high speed. You're fairly close to the
surface, you're going pretty fast at that depth, you try to
change course you'll probably cavitate, you'll make a lot of
noise.
The -- your margin for error in depth control is less,
you know, have a problem with your planes or surfaces. You'll
-- everything happens a lot faster at 15 knots than it does at
10. So, 10 or 12 is sort of the normal upward bound. If you
go to 15, probably, but you're not comfortable at that speed
at 150 feet.
Q So, with respect to all three of these different
parameters, course speed and depth, when you start your TMA
leg do you want to be -- is it a good thing to be steady on
course?
A Okay. This is a very important question, because from a
standpoint of the machine assisted algorithms, as I said
earlier the course speed and depth did not need to be
constant. TMA can go on nonstop through own ship's course and
speed maneuvers.
The fact is, if you look on the previous slide the fire
control men eventually came to a fairly good solution probably
somewhere on this leg while the ship was not steady on speed
and course, because the machine can work through those
problems.
But, from a standpoint of mental analysis, as Admiral
Nathman was asking, why didn't the sonarman see this or
understand it, the fact that the ship's speed was coming down
this whole time and that the course was steady only for a
little over a minute, and the ship was changing depth which
has some impact on ability to sonar track, for such a short
period of time degraded the ability of the operators to do
independent mental analysis of the target motion analysis.
So, it is an important plot, there is data on the ship
that could go and come up with a conclusion, but it's -- the
ability to independently verify the accuracy of that TMA
solution presented in that fire control screen is degraded by
the fact that the ship's parameters were continuously changing
through that entire maneuver. And I think that's an important
point.
Q So let me understand this. The way you're saying this,
it takes away that interaction, independent operators coming
to their own conclusions, and you basically are putting all
the reliance on what the fire control system has generated?
A That's correct.
Q You're down to a single point?
A That's correct.
Q Single point solution or single point answer?
A That's right.
We did not like to just trust -- as a submariner I don't
trust just -- I need to verify that fire control solution. I
want to know that that makes sense, it does -- it correlates,
it makes sense, it conforms with my mental analysis, back it
up, look at the time bearing display and looks at lots of
information to confirm that this contact is close.
You know, I -- I would certainly expect the fire control
operator who thinks -- I have a solution that's tracking at 25
hundred yards or four thousand yards, I would expect him to
raise his hand and say come over here and look at this and let
the officers take a look and get the team playing -- he's got
an important piece of data here for the team. I would
certainly expect him to raise his hand and announce that fact.
But, the -- that would just incur further delay, I mean
further length of time. Not delay, but it would require more
analysis to say, geez, 25 hundred, this is potentially
serious. That's how you'd like it to work. And then you
would stay a little longer, look at the leg a little longer,
watch it develop and come to your conclusion, yeah, it is that
close or, no, that was just bad data, it's just a bad set of
bearings, you know, it's not. It doesn't indicate it's close.
BY VADM. NATHMAN:
Q Captain, you mentioned the fact that the TMA occurs
continuously. Sometimes the quality is not very good and
that's varying on speed, the noise you're creating, the
contact quality.
A Your maneuvers.
Q Those maneuvers, et cetera, et cetera. But -- but
Admiral Griffiths in his testimony was very clear that he felt
there was only one -- I think he used the words one good TMA
leg, but I took from that that there was one TMA leg is what
he saw, that would be the next leg, so is your evaluation of
all this stuff, is that consistent with what Admiral Griffiths
arrived at in terms of a good TMA leg?
A Sir, if I was to classify this 340 leg, I can only
assume in the mind of the folks driving the Greeneville on
that day that they considered that leg one. I would consider
that leg at best marginal, only because it's so short and it's
not steady.
Is there TMA being done? Yes, and as evidenced by the
fact that someone came up with an answer that's pretty good.
Is it a good TMA leg? Is it sufficient? I would say it's not
sufficient. It's not sufficient to do an independent review
or analysis to understand the contact motion.
BY CAPT MACDONALD:
Q What would you have have handed it to be to be
sufficient?
A Several more minutes. And if I --
Q Is that several more minutes steady on depth?
A Steady on depth, course and speed to make the situation
obvious.
Q All right, Captain. Then my next question is, I'd like
you to take a look at that 340 leg and tell us how long
Greeneville was steady on course at approximately speed 10, 12
knots and at depth 150 feet?
A It was probably -- there's 33, 3240, 30 seconds. Thirty
five seconds.
Q And those are parameters that you would have wanted to
see more time spent at in order -- in order to get a solution?
A Not only.
-- three minutes, somewhere in that area, plus or minus,
depending on -- around three minutes is what we say is a good
value and that is a good value to start with, does it have to
be exactly three minutes dot zero seconds? No. But in the
order of three minutes.
Q Could we have the next slide, please?
Captain, I believe you've already talked through this
slide?
A Yes, I have.
Q Could we have the next one, please?
A This next one shows that -- what I did here on this
particular chart is I blew up the reconstruction of the last
few minutes prior to the collision. The collision happens
obviously where the orange and blue lines cross. This is the
120 leg, and here is the 340 leg and we're coming off of a
high speed transit here, kind of difficult to see on this
depiction but you can see this particular mark and that
particular mark and this particular mark here, there's very
small on this, but those are one minute intervals. You can
see that the space between the one minute intervals is getting
smaller which indicates as the previous slide did the ship is
slowing down.
What I did was just extend this leg for three more
minutes, one, two, three, and drew bearings to the
reconstructed track of the Ehime-Maru and came up with this
bearing distribution.
And if they had stayed on this leg for the three minutes
and just steadied out, the 340, 10 knots, they would have
developed 11 degree permitted bearing rate over that three
minute period to the right. That's significant. That would
be, as we looked on the display, if you remember when you were
over at the training center we showed a contact at 4 thousand
yards away that showed like a 7 degree bearing rate that was
very obvious on the sonar display. It was very clear. The
contact was breaking over to the right and -- and very
apparent to all the operators, all the sensor people who have
an easy time with 11 degree right, I think it was four
seven over there, this would be 11 which is higher.
If you kept going this would go to a maximum of about
14, which is a fairly high bearing rate and I don't think
there's any submariners would not recognize that as being a
close encounter. Contact is inside a mile. Mile or so.
Q Next slide, please.
A What I did on this slide was take the slide that you're
familiar with already, the blue dot slide, expanded time
bearing and I basically plotted those orange dots. What this
would have looked like on the display that we've already seen,
if I continued on that 340 leg, projected bearings. You could
have seen this thing would have continued to draw at the right
at a very large rate, and I think it would have been obvious
to all the players on the ship that the contact was fairly
close.
BY RADM SULLIVAN:
Q Again, just sort of boil this down, the ship (inaudible)
fairly good contact, especially the minutes leading up to the
collision. And you alluded to, when you talked about half
hour, 45 minutes ago about how a big portion of TMA is -- how
good it is is how you drive your ship. How you position it to
generate bearing rates to change bearing rates?
So, that first leg, if they had just stayed longer it
would have been a great leg to see what they needed to see,
which was --
A From a course standpoint, sir, it's excellent. It's an
excellent leg to deduce if they just stayed with it a little
longer it would have clearly shown the contact at close range.
The problem is they went then to a 120 leg which is a
zero bearing rate leg and didn't really add much information
in view of the long history, as we discussed a little while
ago, the long history. It seems it's a been consistent over a
long period of time and they missed out on this opportunity
right here to see the really relevant information.
Q So the course selection, 340, whoever that might have
been, the Officer of the Deck or with the helping of the
Commanding Officer or whoever might have selected it was a
great selection. In other words, they had a great plan, they
just didn't execute it as you'd like to see it?
A Yes, sir. If you see, they were doing 10 knots. The
contact bearing when they went -- was roughly, you know, it
was over at the 010020 leg, we're over here at 020 he's taking
a course that's 40 degrees from the bearing of a contact. He
has a 10 knot speed. That's a significant amount of speed
going across the line of sight which would exactly what I was
talking about in target motion analysis, drive the bearing of
the contact to the right and would -- would clarify the
picture very quickly.
I can guarantee if this had happened the fire control
systems solutions would have locked up on a unique solution
very quickly. There would be an absolutely no doubt in your
mind, everybody would concurred right off the bat and said
this contact is close, we need to stay down, we need to go to
some other location. Made an alternate decision. I don't
think they would have gone to 120 to go to periscope depth,
that's not a safe course to go to periscope depth on based on
this analysis.
Q What about the axial of the middle (inaudible), that is
you alluded to as being done?
A That would be enhanced as well, high bearing rates,
large numbers like 11s and big numbers like that make that
formulis work better to come to a more accurate answer.
Q Go to the next slide please.
I believe, Captain, you've already talked through the
CEP plot slide. Do you have anything additional you'd like to
add?
A Just -- the -- this does show all the contacts that were
being tracked by the Greeneville on that day. It does show
the SNR values on here. You see the minus -- this is a
question from yesterday, minus 10, minus 8, minus zero, minus
three. It all kind of depends on the speed the ship is
having. Those are moderate, not weak SNR but moderate SNR
contact.
Contact does go up in SNR right prior to the collision
clearly. I'll show that in the next slide. This again is the
top half of that plot. You see the Sierra 14 contact, through
my interviews with the sonar team become a center of some
focus just prior to going to periscope depth.
He emerges after what appears to happen, the 340 leg
does drive Sierra 13 a little bit to the right and I would say
that most likely Sierra 14 was being hidden, masked by Sierra
13 during some period of time during these maneuvers. In
other words he was behind a weaker contact behind Sierra 13
which is the closer contact. And he came out and needed to be
evaluated.
Now, this evaluation of Sierra 14 was not very good.
They had this one dot here on a 340 leg, this one X, maybe a
couple sonar bearings and then they went to 120 to go to
periscope depth. This put this contact right on the edge of
the sonar's baffles, making it very difficult to track that
contact.
So, you know, I -- he's being tracked here, but he's
very -- that's a very marginal position to place Sierra 14
from a standpoint of tracking and getting further data on the
second contact.
Q Again I'm probably showing my age, but putting the
contact on the edge of the baffles or in the baffles, is that
a good idea?
A It's not a good idea. It's because, as explained over
at the training center, there is -- there's basically 120
degree slough directly astern that the sonar spherical ray
system cannot accurately track the contacts. The closer they
are to that 120 quadrant, 630 degrees on either side of the
stern is a rough number, depending on the elevation angle.
The accuracy of the track, the tracker becomes less and
less accurate as it's listening further and further behind so
the track data you're getting is less likely to be accurate,
an accurate depiction of the bearing.
BY CAPT. MACDONALD:
Q Captain, just to clarify a point. This slide and the
one previous to it are the reconstructed CEP plot, correct?
A Yes, they are. We took -- we generated these by taking
the sonar logger data, which -- (indiscernible) generated this
black line reflects what Greeneville was doing. It shows what
courses and speeds this -- CC means change course to 340.
This is a depiction of Greeneville's track as the speeds here,
the speeds that are going on, right here, raise number two
scope.
This is basically a scrolled chronology of Greeneville's
actions taken from the sonar logger data and the deck logs,
and it also took sonar logger data and plotted each contact
that was logged on the sonar logger, Sierra 12, 13 and 14
during the period of concern here.
Q And this -- the reconstruction was done by your N70?
A No, this is -- N72, the data analysis group. They took
the spreadsheet generated by the sonar logger data and just
put this plot together.
Q Captain, after your reconstruction efforts, is there any
doubt in your mind that contact Sierra 13 was the Ehime-Maru?
A No. No doubt in my mind whatsoever. This Sierra 13
reacted, it shows all the indication of a close contact. You
see that it -- actually we could look at this on the next
slide after that is even better blown up. The only two
contacts really being tracked at the time were Sierra 13 and
14. Go to the next picture.
This is just a blow-up, easier to read of that same
period of time. This is Sierra 13 over here, the blue or
purple lines and the orange lines are the Sierra 14 lines.
You can see that -- although again this contact is
starting out way -- just on the edge of the baffles and you
see the SNRs are low, that's an indication of bad track, comes
back up consistent with this bearing drift once we get the
tracker back on it.
This contact, on the other hand, Sierra 13, remember own
ship has come down from -- this is a zero bearing rate leg we
were talking about at -- while own ship was on 120. Boat now
goes deep to four hundred feet and starts a turn to the left,
as he increased speed going deep to 400 feet from periscope
depth that increase in speed alone is causing the bearings to
drive to the left, just by the fact we've increased speed.
This contact is reacting to own ship's maneuver. That's an
indication that the contact is close. And the SNR is going
up, that this contact is, again, another indication that close
contact, but must be kept in mind that as we're going from the
interface deeper we may have better sound conditions as we go
deeper and that could cause the SNR to go up in its own right.
And then it's very high bearing rate to the left as we
go by the ship at very close quarters and actually have the
collision, which means we're driving right by. So there's no
doubt in my mind that this Sierra 13 tracked here the last few
minutes was the Ehime-Maru.
Now, we go back one slide. It is very possible that in
this phase back in somewhere in here that Sierra 14 was in
fact behind, or masked by the sonar display by Sierra 13, that
they were on the same bearing and presented Under the same
trace on that sonar system? Laura checking sonar system.
Q If they had detected Sierra 14 behind Sierra 13 what
would you have expected them to do?
A They could -- until they had broken apart like that,
treat Sierra 14 as brand new contact and you need to do the
same analysis, as I said before, before going to periscope
depth, you're obligated to understand all the contacts that
you have around you, even if you picked one out, one became
one mass, you have to figure out is this contact close, far
away, where is he, what's the relationship to own ship before
we go to periscope depth.
Q Okay.
Captain, you mentioned the acoustic conditions in your
previous testimony. Can we go ahead a couple of slides?
A All right.
Q Are we able to determine the acoustic conditions for the
9th of February?
A Yes, I was. We took this data off the ship's recorded
data, we record the sound velocity profile. The ship actually
made a fairly deep dive during the day so they had pretty good
data then what happens is this bottom part that is not in the
area, the very deep data on this chart are merged historical
data for that particular area. But this top part was actually
measured by the ship's sensors, and indicates near the surface
between zero and 400 feet a fairly, we referred to
isovelocity (phonetic). This is sound speed on the horizon
axis. And this indicates that the speed is fairly iso
(indiscernible). In other words it's the same speed and sound
is about the same all the way down 400 feet.
And that, when you have that kind of condition, the
sound basically travels straight. There's no bending. If you
have a change in velocity it tends to bend the sound waves in
the direction of the slowest speed and that's what these lines
are trying to depict and the verbiage on here discusses, it
slows down the waves where it's deeper and where it's faster
the waves go faster, stay at a higher speed and kind of bends
them toward the point of minimum velocity.
But up in this area where Greeneville was operating it's
fairly iso-velocity, good sound conditions.
Go to the next slide. We have an acoustic prediction
bottle that we have a great deal of confidence in that
predicts the acoustic performance on a given day. And this
black part reflects land, and it's on the quadrant, own ship
is at the middle here of this diagram and this V here shows
the direction that this plot is depicting, which is the area
due north, area up toward Oahu from where Greeneville was and
that's the area that the Ehime-Maru was coming from.
This black line indicates the bottom, so it's -- up here
is the island itself, this is the very shallow near the
island, it's sort of stylized, it's a rough area. There's a
shelf and then a deep area that falls away pretty quickly.
This scale over here indicates the transmission loss
along this line. Greeneville's operating over in this area,
zero range, and looking up to the north you see that there's
very little -- this is from low transmission loss to high, the
orange is good, even the green is good. Good transmission
loss all the way up to the beach. You lose some, can't hear
quite as well out there at 40 thousand yards as you can close
by, but you're still hearing pretty well.
Let's go to sound conditions.
And this really is the minimum -- this is the -- this is
really the (inaudible) excess required to hear contacts above
the noise in the area, and the noise was not that loud. So
you have good sound conditions all the way, all the way to the
beach. There's no issues with some kind of bending motion or
some bending, some weird sound conditions that would have bent
the sound from the Ehime-Maru away from the Greeneville
sensors.
VADM. NATHMAN: Go back three or four slides.
This was the sonar drift rates.
Q Captain, you mentioned -- this will be the last question
and we'll take a recess.
You mentioned that your boats are normally doing
constant TMA. And we had that one leg on 120 where we had the
steady bearing, and it starts -- can you help me with the
time?
A Right here.
Q Go to the bottom of the chart, the first time. Go to
the left. I think that's time on the left, right?
A Yes. 2334.
Q Okay. 34. And take me through the -- when it starts
sweeping to the left and you see a drift rate, when it starts
there, what time is that?
A That's 2340.
Q Okay. A little bit less maybe?
A 23, yes, 2339 and a half or something like that.
Q All right. But now you're starting to see this analysis
should still be going on, right? You have it -- what -- it
looks to me like you're doing two things, you're diving,
you're increasing speed so you get drift rate based on that?
A That's correct.
Q If I get this right, you're now -- you're going to start
a turn, okay, that changes drift rate?
A That's right.
Q So you've got sonar suit and you've got sonar
watchstanders now that have been able to watch this, what
looks to me right now as a lot more data?
A Yes, sir.
Q What should be the conclusion from that?
A I would say that if you're really paying attention to
this contact you recognize that this is about two minutes of
data drawing to the left rising SNR that would be an indicator
that it's close contact.
Q Okay.
A This would also be depicted on the fire control screen
as well. It shows that bearing drift going to the left.
Q I want to ask you (inaudible) I want to ask Admiral
Griffiths this, but it was the irretrievable nature of doing
the blow. In other words, you're still maneuvering the shift
up until the time you do the blow. This is what I'm
understanding. You put a turn on the ship, you've changed
depth, you're still in control, and at the same time you're
still in control it looks to me like there's a lot more data
now available to the sonar operators in particular?
A Yes, sir.
Q In terms of the knowledge they could gain on this
particular contact.
Once you do that blow, do you have any ability -- you're
going to go to the surface is what I understand?
A That's correct.
Q Do you have any ability to change your course? I mean,
I think you're going to probably increase speed because you're
rising, but do you have any way to influence your -- the
dynamics of where the boat's going to go?
A Not really, sir.
In fact, procedure says for the emergency blow you want
to keep your rudder mid shift for stability. You want -- you
don't want to be turn at the same time you're rising at that
speed for the stability of the submarine itself.
Once -- basically, once the emergency blow is actuated,
it's pretty much you're going to the surface on the course
that you're going to the surface on. And from what I
understand, I just -- my picture of understanding what was
going on at the controls at this point, once we went deep and
they started the speed increase to 12 knots, go to full bell,
going down four hundred feet. The natural focus of the
Officer of the Deck, the Captain perhaps, standing at the ship
control probably would be watching the actions of the ship
control party executing this emergency deep so their focus
will no longer be over at the contact picture. Their specific
focus would not be over there at the contact analysis. Now,
it does stop the sonarmen or the FTOW from doing that work,
but the focus of the officers on the CON would be making
sure --
Q The control is proper?
A The control is proper, yes, sir.
BY RADM SULLIVAN:
Q I agree with what you're saying. I think what drives
that is the fact you've gone to periscope depth.
A Haven't seen anything.
Q Haven't seen anything, you've done a visual very much so
you're very satisfied or you wouldn't be doing the evolution --
A That's correct.
Q -- unless it's clear.
Would it bother, again we're talking not necessarily
Greeneville but just in general, you go deep and start turning
and generate a bearing rate like that, or contact that wasn't
seen at periscope depth in the evolution was conducted, would
that be troublesome the --
A I'm sure that's a factor in his mind. We didn't see
this guy, how can he be this close.
Q I mean, the fact that he kept him as a contact through
an emergency blow to me means it was very locked contact.
A That's right. You can see what happens to this other
guy, they were having a hard time tracking him through this
emergency deep process. That's what caused him to lose track.
He was going down fast and turning course, this is a weaker
contact and it drifted off. This one they tracked solidly
right through the whole process, he's close.
So, I'm -- I would guess, this is surmise, I would guess
that paradox was in the mind of the sonar people. We just
were at periscope depth, we looked around, we didn't see
anybody, he can't be as close but it looks close, I don't know
what -- I don't know what thought process was going through
their mind.
Q It wouldn't just be the sonarman --
A Any -- the whole people doing the contact, continuing
the contact analysis process here.
VADM. NATHMAN: Captain, we've got lots to cover
today. Let's take a recess until 1000. This Court is in
recess.
(Recess taken at 9:47 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
|
