ML20053A512
| ML20053A512 | |
| Person / Time | |
|---|---|
| Site: | Big Rock Point File:Consumers Energy icon.png |
| Issue date: | 05/21/1982 |
| From: | Betourne G CONSUMERS ENERGY CO. (FORMERLY CONSUMERS POWER CO.) |
| To: | |
| Shared Package | |
| ML20053A496 | List: |
| References | |
| NUDOCS 8205260166 | |
| Download: ML20053A512 (17) | |
Text
..-
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD
- l In the Matter of
) Docket No. 50-155-OLA
)
CONSUMERS POWER COMPANY
) (Spent Fuel: Pool
)
(Big Rock Point Nuclear Power Plant)
)
SUPPLEMENTAL TESTIMONY OF LT. COL. GARY P. BETOURNE ON O'NEILL CONTENTION II-D This testimony addresses the questions posed by the NRC Licensing Board in its February 19, 1982, Memorandum and Order, and it supplements my testimony which was provided at oral deposition on July 13, 1981.
My educational background and work experience are set forth in the transcript of the deposition a t pages 11 to 15.
My answers to the Licensing Board's questions follow:
1.
(Q )
A Bayshore plane overflew the plant on July 5, 1979.
We infer that the error that occurred resulted from the use of the Big Rock Plant as a navigational offset or reference point.
Deposition at 56-57, 101-102.
The error apparently was a failure to remember that the Plant was being used as a reference point and that navigation was not " direct mode," causing the plane to i
fly directly over the plant instead of 5.7 miles away.
Deposition at 57.
Consequently, SAC has prohibited 8205260[(o(7
. further use of the plant as an offset.
- However, the plant is an obvious, visible landmark and no effort has been made to verify compliance with the off set prohibition rule.
(A)
A radar offset aiming point is a highly accurate range and bearing from a specific radar target used in solving the weapon release equation.
It is programmed into the aircraf t bombing computer.
These aiming points are not established by the individual crew members, but instead by a senior staff officer at the headquarters of each operational wing.
The aiming points are furnished to the aircrew in a mission information folder by this senior officer.
The power plant was removed as an offset aim point by HQ SAC from the present route in August 1979.
While the power plant still remains a suitable
(
radar return for cross-checking enroute navigation, i
the air crews obviously no longer have the detailed range and bearing information about the power plant radar target which would make it useful as an offset aim point for their training missions.
. Further, if there is any lingering doubt as to the probability of using the plant as an offset, there are safeguards to help preclude its use should any radar navigator choose to develop the power plant as an off-set for his own use.
A camera is attached to each radar set to monitor the radar position of the air-craft for each training mission.
The camera is set to take scope-photos at pre-set rate.
These photographs tend to discourage the use of personal offsets since these photographs' would reveal the use of an illegal offset.
The film is examined af ter each training sortie.
Finally, compliance inspections are conducted at each operational unit.
This inspection includes the target study folder developed for each training route.
2.
(Q)
There was an actual crash of a B-52 in 1971 near the plant.
Deposition at 4 7.
(Contrary to intervenor's suggestion, there was only one crash on the Bayshore route, Deposition at 4 6-4 8. )
Major Betourne did not include pre-1971 crash data because of dissimilarities in aircraft and training practices.
We accept that decision, howeve r, Major Betourne did not have a suf-i ficiently thorough knowledge of the 1971 crash to dis-cern its possible implications for his risk analysis.
Deposition at 4 8.
. (A)
The 1971 crash of a B-52C (no longer in the inven-tory) while on a low altitude training mission on the Bayshore route occurred while flying over Lake Michigan.
The aircraft was on a simulated bomb run.
It is my understanding that the crash occurred about 10 nautical ndles north of the Big Rock Point Plant.
At the time of the crash the Bayshore site observed (by radar) that the aircraf t was on the centerline of the established rcute.
Since it is my understanding that the aircraft was on course, I do not believe that the 1971 crash impacts my risk analysis.
3.
(0)
The sample was inadequate.
It apparently consisted of 10 gross navigational errors occurring during a 2-month period.
Deposition at 52.
These 10 errors apparently were extrapolated into a rate of 6 0 per year.
Deposition at 53.
The method of
(
collecting data and determining the number of l
l gross navigational errors has not been discussed.
Furthermore, the number of runs during the period l
j was considered to be representative as the result of visual inspection of additional data, but the l
r I
- data inspe cted were not produced for the record or s tatistically analyzed to support the representative nature of the number of runs.
Deposition at 54.
What we do know is that the sample has 354 scored bomb runs during a year in which there were 2,986 runs, suggesting that the annual occurrence of navigational error should be projected to be over 8 (not 6) times the number occurring in the sample period.
Compare deposition at 51.
Furthermore, the number of gross navigational errors was considered to be representative as a result of unreported interviews.
Deposition at 55.
Considering tha~t the deponents are SAC employees and that Bayshore run is of very great importance to SAC (deposition at 31), it would be helpful if additional data and s tatistical verification were made available.
4 (A )
Additional information is not available.
As previously stated, the data is retained - by the 1st Combat Evaluation Group (1CEVG) for a brief period of time, then it is destroyed.
(1CEVG is the Air Force Organization that operates the cites; they also functioned as the analytical agency for all the training results at the sites.
This is now a Headquarters function.)
The worksheets containing this data used in the development of the risk of navigation error were also destroyed.
As I recall at the time the data was requested, I asked 1CEVG if this sample represented a normal volume of training at Bayshore.
I was told it did.
I also asked if the sample represented an expected amount of navigation errors (all causes).
I was told it did.
Finally, while the number of navigation errors (exceeding corridor) is tied statistically to the amount of activity at the site, it is not a fixed ratio of errors from month-to-month.
The number of errors (60) is considered reasonable for the site and the aniount of activity of the route during the time at which the analysis was conducted.
(
I Final ly, there is no recollection of a navigation 1
error in the target area as great as 12NM -- the approximate distance from the current Bayshore route's centerline to the containment facility onshore.
l I
. With regard to the importance of the Bayshore route to SAC, the changes in the route in reaction to the 1979 overflight of the containment facility have caused the route to be less desirable.
As a result, much of the training equipment at the site has been removed and sent to other training ranges where it can be better utilized.
The equipment removal has resulted in a drop 1743 vice 2986 in the site's activity in relation to the time period of my risk analysis.
This reduction will further reduce the risk to the containment facility.
It is now the Air Force's desire to close the site at Bayshore as a direct result of the loss in training potential on the route.
Steps have been taken in the Fiscal 1982 and 1983 Defense Budgets to remove all equipment and personnel from Bayshore and close the route.
4.
(Q)
There is some ambiguity about whether the number of
" scored runs, " used in the Analysis, equals the total number of flights on the Bayshore run or whether sub-stantial additional use is made of the corridor.
(A)
There is other activity in addition to the total number of " scored runs" at Bayshore.
My analysis only covers USAF activity.
It is exclusive of Air Force Reserve, Air National Guard or other military aircraf t.
USAF aircraf t nay fly at high altitude (35,000 ft) on simulated high altitude bombing missions without benefit of radar scoring.
Under these circumstances, tracking (scoring) is achieved through the use of radar scope photographs.
Unscored activity represents less than 30 percent of the route's_ total activity.
It was factored into that total number of runs.
5.
(Q)
There is no discussion about the nature of accidents to which the relevant aircraf t are subject when flying low-altitude missions.
Indeed, there is no discussion of the nature of the accidents to which these aircraf t are subject, during either regular or low-altitude missions.
Hence, it is impossible to know whether the crash area of 3 square miles around the plant, used in l
l the Analysis, is conservative or not.
Deposition 108-111.
In particular, there is no discussion of whether an error in altitude might accompany an error in course so that no further event need intervene before a crash occurs.
There also is no consideration l
h i.
,. s 3
^
t i,
l~
i of whether there are crddible accident events that could lead to long, uncontrolled glides that might commence even while a, plane was on-course.
(A)
The gliding distance of a B-52 from 1000 (slightly higher than specified) feet to impact is 17,000 feet.
In the analysis it is assumed that the crew ejects s
s.
shortly after the B-52s eight engines simultaneously s-1 fail (the probability of all 8 failing simultaneously is also very low).
Af ter the crew ejects the aircraf t '
glides in the direction of the power plant, crashes.:,
and F 1ris from the crash hits the plant.
It was s
lassumeu that the containment facility was anywhere s
within the 3 square miles and that the aircraf t would s
hit it during its glide or that debris from the crash would reach the plant.
It is assumed that the debris that hit the plant has I
sufficient kinetic energy to cause serious damage to the containment facility.
No probability of damage calculations was made considering the mass and velocity of the debris or the hardness / vulnerability of the structure.
In practice, this probability would be less than unity.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ It should also be noted that the crash scenario used in the analysis, while furnishing the greatest vulnerable zone, is also the least likely to occur.
The larger the vulnerable zone, the greater the probability that a catastrophic event could occur in it.
A more realistic approach, but nonetheless conservative, would be to use an expected debris area caused by a crash on relatively smooth terrain (rough terrain or water would greatly reduce the debris area although the containment facility situation actually is a combination of smooth terrain and water).
This would reduce tl:e vulnerable area from 9 square miles to an expected debris area, 600 x 6,000 feet or about 0.09 square miles.
Still the assumption is that the containment facility is within the vulnerable debris area.
The probability of damage to the facility, given these tighter constraints, is still less than unity.
We can, therefore, conclude that we are conservative (overstating risk) by a factor greater than 100 in this portion of the analysis.
Finally, flight altitudes selected for each route is based on a safe clearance from the highest obstacle or
\\
- or point along the route.
The B-52 has four crew positions that monitor altitude; both atmospherically and electronically; the FB-111 has two.
To strike the plant, a 3-52 would have to fly at an altitude below 120 ft. which is well below the required minimum altitude of 400 ft.
I have not calculated the likelihood of an altitude error, but in my j udgment it would be very low for the. Bayshore route as its current ruaneuver (bombing run) has no altitude change requirements other than the initial IFR or VFR altitudes of 2000 Mean Sea Level or 400 f t.
above ground (water) Level respectively.
Only visual (VFR) conditions permit the lower altitude to be flown by the flight crew otherwise, the instrument (IFR) altitude must be flown.
6.
(Q)
The experience with crashes of the relevant aircraf t appears to have been applied to the Bayshore Range without recognition that low-level missions may be more hazardous than other flight activity.
(A)
Low-altitude training, while challenging to aircraf t i
and crew, is not the highest risk of all flight activity -- landings and takeof fs represent the
?
-- highest risk.
It is my understanding that the total number of B-52 crashes (models B thru H) during low-altitude training represent about 6 percent of all B-52 crashes, while the terminal area phase of flight represents about 49 percent.
The assumption that low-altitude flight is more hazardous is unfounded.
7.
(Q)
There is no allowance for failure of communication other than through complete loss of power, assigned a
-3 value of 1.7 x 10 Yet, pilot inattentiveness or mishearing-would appear to be a credible source of tailure of communication.
Indeed, one overflight is known to have occurred despite radio communication to the crew that they had exceeded the Western limit of the corridor.
Betourne/ Thomas Exhibit 2 at Telegram by Richard J. Camp, Operational Analyst.
(A )
I am unable to explain why 6 or more individuals on the aircraf t failed to respond to the Bayshore systems call to avoid over-flight of the power plant.
This documented case item represents one case in 60 calls where communications failed and the aircraft was going out of its corridor.
i I
1 w
7-w
. But a navigation error and communication failure are not dependent events and therefore one must take into consideration all communications attempts to determine the probability of failure.
There are numerous radio attempts during the course of the 7-8 minute bombing run.
He can only assume that this " failure" to commu-nicato represented a random failure albeit untimely, but it is the precise nature of our risk analysis --
to determine the probability of an event-to occur and to properly fix appropriate probabilities to each event taking care to treat dependent and independent events correctly.
If we examine the 1980 analysis with the view that a radio warning would be issued to only 70 percent of the aircraf t exceeding the corridors (30% of the activity is unscored) by the site as a result of naviga-tion error, then the probability of a crash at Big
-6 Rock would change to 1. 64 0447 x' 10 as a result of adding this increased risk.
We assume that radio f ailure (for that activity which is scored) would occur at the precise moment in time when the aircraft was starting its simulated bombing run.
Clearly, this assumption is conservative as the radio failure could have occurred anywhere along the entire route not just the bombing run.
Further, if I replace some of the conservatism reflected in the vulnerable area calculations and replace the 1980 calculations that used nine square miles with the revised vulnerable area of 0.09 square miles, the resultant calculations show the probability of crash at the facility of 1. 6 4 0 4 4 7 x 10".
This is a net reduction in risk that includes the lack of communications opportunity for 30% of the activity.
8.
(Q )
There is a need to explain the circumstances which led Col. James M.
Campbell to reassure Consumers Power Company in May 19, 1971, letter, that, "We trust our analysis will.
provide you a basis for reaching an agreement with your insurers."
Betourne/
l Thomas Deposition, Exhibit #2.
Is there any relevant correspondence with the insurers?
Is there an actual computation of risk which could be made available in l
l this proceeding?
l l
. (A )
There is no correspondence between Col. Campbell and anyone connected with the previous risk analysis.
We can only assume that the Consumer Power Company was motivated to ask the Air Force for a risk assessment at the request of their insurers.
The Ja nuary, 1971, crash in Lake Michigan undoubtedly precipitated the question.
9.
(Q)
Did SAC perform any other analyses for the purpose of
)
deciding whether to move the Bayshore Range?
What do 4
those analyses show?
Why was the Bayshore Range moved?
How was the decision reached?
(A)
The manuever area was altered from that which was depicted in 1980 risk analysis to that which was pre-l sented in July 13, 1961, deposition.
He conducted no risk assessment of this new route although our exper-tise tells us if there was any element of risk before it has been virtually eliminated now by the route change and the prohibition of the containment facility offset aimpoint which caused the one USAF overflight.
The decision to change the route was reached by HQ SAC.
It was moved to preclude accidental overflight
. of the power plant.
This new maneuver area route passes about 12 NM from the containment facility.
The route of flight through the maneuver area is over Lake Michigan.
Precise radar position is furnished by the land / water contrast of the route.
Visual flight conditions also furnish excellent cues to avoid navigation error, i.e.,
if the aircraft is moving too close to the shoreline.
10.
(Q)
Is it proper to assume that the probability of a navigational error is independent of the probability of a crash or is it more likely that a crew which would make a gross navigational error would become involved in a crash?
Is independent probability correct if the navigational error was induced by an equipment failure?
(A)
Navigational error can certainly be a contributing cause to any crash.
However, it is my understanding that navigational error has not been a contributing cause t'o aq/ B-5 2 crash.
11.
(0)
Is the method of notifying pilots of the Big Rock Point no-fly zone adequate?
If the Bayshore Route is e
,y,,
v,-
_17_
used for non-scored runs, is the method of notifying such users of the no-fly zone adequate?
(A)
I assume that the reference in the question to the "no-fly zone" means the area outside the corridor of the training route.
The corridor established by 1CEVG at Bayshore is protected via their tracking radar at the site.
The radar continuously monitors the exact position of the aircraf t as it passes by the site along the maneuver area.
The aircraft's track is plotted in real time on a chart which delineates the corridors.
If the corridor is exceeded the site calls the aircraf t with a warning to return to course.
Naturally, this protection is not available when the site is not open f or scoring.
Non-scored activity currently comprises about 2 8 percent of Bayshore site total activity.
Total site activity now averages about 140 simulated 4
bombing runs per month including the non-scored activity.
i l
In addition, HQ SAC has reduced the width of the corrid or, on the side nearest to the Big Rock Point Plant, for the current Bay Shore route from 4 to about 2 nautical miles.
This provides for a further reduction in risk to the Big Rock containment facility.
-..