View the original post on the IUSSCAA message board here

Rebutting Conjecture SCORPION Reversed Course Just Before Being Lost

By Bruce Rule - September 22, 2013


in 1968, the SCORPION Court of Inquiry (COI) was provided with significant misinformation about the event including the conjecture that SCORPION had reversed course just prior to being lost in order to deactivate a torpedo that had be come active in its tube. Although this conjecture was refuted on 18 July 1968, that evaluation was never acknowledged by the COI. This monograph provides the basis for the assessment that the course-reveral conjecture was erroneous.

The position of the SCORPION wreckage derived from a comparison of signal arrival times by a Canary Island hydrophone and SOSUS arrays 3131 and 3141 at Argentia (a time difference solution) was 1.1 naiutical miles from the actual position verified by an imaging survey of the bottom. This accuracy was the best that could be achieved when uncertainties in knowledge of the velocity of sound are considered over the involved ranges and is consistent with accuracies derived for other events in the Atlantic and Pacific at equal or greater ranges. Note; however, that, as discussed in the following article, the accuracies of relative positions (the position of one event in a sequence of events relative to the position of another event in the same sequence) can be determined with possible errors of less than 100 feet. This was the case because whatever velocity of sound is involved, those values do not change within the duration of the sequential events.

In one case, the Navy spent approximately $300,000 on what they believed to be be ever more sophisticated (and; hence, more accurate) analyses to derive a time-difference solution.When the actual position was verified, it was found that the original position determine by the Office of Naval Intelligence (ONI) using the same approach that would have been used by IUSS - then COSL - was the most accurate and that the progressively more sophisticated approaches had provided progressively greater errors which were as much as 150-percent more than the ONI solution. The bottom line to this technical misadventure was that the Navy should have trusted their own analysts.

Analysis of Acoustic Data Used to Support the 1968 Assessment that the

USS SCORPION Reversed Course and Was on an Easterly Heading When Lost.


There was not in 1968, nor is there now, any acoustic data known to have existed that could provide reliable information on the heading of SCORPION before, during or after either the precursor acoustic events at 18:20:44 Greenwich Mean Time (GMT) (ZULU) (attributed to explosions of hydrogen outgassed by the main storage battery) or pressure-hull collapse and subsequent acoustic events during the 18:42:34-18:45:53 GMT period. These are actual event times onboard SCORPION.

This monograph provides expanded discussions of analysis (refined time measurements) of the acoustic data used in 1968 as the basis for an assessment that SCORPION had reversed course and was on an easterly heading when lost on 22 May 1968.

These refined measurements, made on 18 July 1968, were reviewed in March 2009. The review confirmed SCORPION was nearly horizontally stationary (within less than 100-feet) during the 111.6-second period when it was claimed in June or early July 1968 that SCORPION had reversed course to an easterly heading.

During that 111.6-second period, SCORPION fragmented wreckage was sinking nearly vertically following pressure hull collapse.

This conclusion eliminates (refutes) the basis for the 1968 conjecture provided to the SCORPION Court of Inquiry that SCORPION had reversed course to disarm a Mk-37 torpedo that had become active in a torpedo tube.


If a surface ship or submarine produces (radiates) strong pulses of acoustic energy while operating in the open ocean, these pulses (signals) can be detected at significant ranges by underwater sensors (hydrophones) in fixed (stationary) known positions on the sea floor.

The delay between the time when such signals are radiated and the time they are detected by hydrophone sensors is a function of the distance (range) between the radiating ship or submarine and the hydrophone, and the speed of sound in water.


SCORPION produced the first of more than 15 high-level main-sequence event acoustic signals at 18:42:34 GMT on 22 May 1968. This strong pulse of acoustic energy was detected at a range of 821 nautical miles (nm) by the Columbia University Hydroacoustic Station Canaries (CUHSC) hydrophone A at 18:59:31.7 GMT, 16 minutes and 58 seconds after it was radiated by SCORPION. This delay was the time required for the signal to travel 821 nm through the ocean at a speed of 4905 feet per second.

This signal was detected at 19:02:43.2 GMT by a Sound Surveillance System (SOSUS) hydrophone array 3141 located near the Grand Banks, Newfoundland. The delay of 20-minutes and 9-seconds between the time of origin of this signal and the Agentia SOSUS array detection time was three-minutes and 11-seconds longer than the delay associated with the CUHSC detection.

This additional delay, relative to the CUHSC detection time, was produced by the greater range of the Argentia array from the SCORPION position: 976 nm compared to 821 nm for CUHSC hydrophone A.

If SCORPION continued to radiate acoustic signals from the same position during the 111.6-second period, this delay relationship would not change, i.e., the Argentia detections would always occur three minutes and 11- seconds after the CUHSC detection.

If; however, SCORPION changed position between the first and subsequent acoustic events during that 111.6-second period, the value of three minutes and 11-seconds would also change. Since the CUHSC hydrophone was located to the east-southeast of the position where the SCORPION wreckage was located and the Argentia array was located to the northwest, movement of SCORPION to the east between the time of the first acoustic event and any subsequent event would increase the three minute and 11-second difference while any movement to the west would decrease that difference.

Argentia array 3141 detected four SCORPION-associated acoustic events that spanned a total time of 111.6-seconds. The time spacings (separations) among these four signals within that 111.6- second period, discussed in greater detail below, provided unambiguous correlation of these four Argentia detections with four specific CUHSC detections.


The conjecture that SCORPION reversed course to an easterly heading during this 111.6-second period was based on the belief that there was a change of two-seconds in the delay of signal detection times between the Argentia array and the CUHSC hydrophone. If valid, such a two-second change (increase of the time delay difference from 3 minutes and 11-seconds to 3 minutes and 13-seconds) would have required a movement by SCORPION on an easterly heading for a distance of about 4900 feet in 111.6-seconds for an average speed of 26 knots.

In response to this course-reversal theory, advanced in 1968, high-time resolution VisiCorder paper displays were reproduced on 18 July 1968 from magnetic tape recordings of the CUHSC hydrophone detections of these acoustic events and measurements were made from the original Helicorder displays of the Argentia array data. The respective accuracies of these measurements were 0.01 seconds and 0.1 seconds.

The very accurately measured change in signal arrival delay times at the two sensors of 0.04 seconds is consistent with movement by SCORPION a distance of not more than 100-feet to the east (toward CUHSC hydrophone A) over the 111.6-second period which equates to a maximum horizontal speed of only 0.5 knots. Since the 0.04-second measurement is within measurement error of the array 3141 data, it appears probable the horizontal speed of SCORPION was not more than one-half knot and could have been significantly less than that value.

The results of these refined time-delay measurements were provided to the originator of the course-reversal theory at 2130 EDT on 18 July 1968. These measurements indicated the total time difference of the four SCORPION signals measured for the Argentia detection was 111.6-seconds compared to 111.56-seconds for the CUHSC detection. This change of time difference of less than one-tenth of a second compares to the original claim of a two second difference and indicates the original estimate of the change of time-delay measurement upon which the course=reversal conjecture was based was not accurate.

To have averaged the 26-knot speed required by the claim of a change of two-seconds in the relative delays of signal arrival times at the CUHSC and Argentia acoustic sensors over the 111.6-second period, SCORPION would already had to have been operating on an easterly course at 20-22 knots and to have ordered flank-speed no later than about 60 seconds before the first acoustic event which marked the beginning of the 111.6-second period. It was impossible for SCORPION to have reversed course during the 111.6-second period and to have averaged 26 knots on a final easterly heading during that period even if the submarine had been operating continuously at flank speed.

Finally, no acoustic detection of SCORPION at those speeds - or any other speeds - was made by the CUHSC hydrophone despite the fact that during this period the sensor made numerous, continuous long range detections of transiting surface ships for periods at great as 3-4 hours.


The highly-refined Visicorder and Helicorder time-of-signal detection measurements made on 18 July 1968 indicate there was no horizontal movement greater than 100-feet by SCORPION on a NW-SE axis during the 111.6-second period when the same four acoustic events were detected by the CUHSC and Argentia hydrophones.

This conclusion disproves any claim that SCORPION reversed course. As previously noted, such a course-reversal was originally proposed as a maneuver by SCORPION to deal with a problem with a Mk-37 torpedo.

The conclusion that SCORPION was nearly horizontally stationary during the 111.6-second period also is consistent with analysis that confirmed the first of the four high-level acoustic events detected by both the Argentia and CUHSC hydrophones was SCORPION hull collapse at a depth of 1530-feet. This event generated a strong bubble pulse frequency of 4.46 Hz, a value so low as to be almost unique to submarine pressure-hull collapse events. The forces associated with pressure-hull collapse events are sufficient to not only destroy a submarine but also to cancel all pre-existing directions of motion. As confirmed by imagery of the SCORPION wreckage and the dimensions of the debris field, the major hull components sank nearly vertically from the point of collapse at 1530-feet to the bottom at a depth of 11,100-feet.

Thus It would have been impossible for SCORPION to have changed position horizonally during the 111.6-second period between the first and fourth acoustic events detected by both the CUHSC and the Argentia hydrophones because the submarine was destroyed by the first event.

SCORPION was lost because two explosions contained within the pressure-hull were produced by the ignition of hydrogen outgassed by the main storage battery within a one-half second period at 18:20:44 GMT. As a result of these events, SCORPION was unable to maintain depth-control and collapsed at a depth of 1530-feet 21-minutes and 50-seconds later, at 18:42:34 GMT on 22 May 1968.

All information provided by this monograph has been derived from unclassified documents and acoustic data that have been in the public domain for between 15 and 40 years.


The four SCORPION-associated acoustic events discussed above were detected by CUHSC hydrophone A during the 111.56-second period beginning 18:59:31.7 GMT on 22 May 1968.

The same four SCORPION-associated acoustic signals were detected by the Argentia Sound Surveillance System hydrophone array 3141 during the 111.6-second period beginning 19:02:43.2 GMT on 22 May 1968.

Not only did the CUHSC and Argentia events span the same time period within measurement error of 0.1-seconds, 111.56-seconds and 111.6-seconds respectively, the intermediate events had the same detection delay intervals.

The intervals from the first event to the second event were 101.22-seconds for CUHSC hydrophone A and 101.5-seconds for Argentia array 3141. The intervals from event two to event three were 3.49-seconds for CUHSC hydrophone A and 3.4-seconds for Argentia array 3141. The intervals from event three to event four were 6.85-seconds for CUHSC hydrophone A and 6.7-seconds for Argentia array 3141.

This exact agreement in total elapse-time and the component interval times for all four SCORPION-associated signals require all four signals to have followed the same acoustic transmission path toward the southeast to CUHSC hydrophone A and the same acoustic transmission path toward the northwest to Argentia array 3141.

Accordingly, all four SCORPION-associated acoustic signals originated at the geographic position established for the first event by a time-difference of arrival analysis which also used detection of events two, three and four by Argentia array 3131 to provide the triangulation. This was the position used to direct the search for SCORPION and it is where the SCORPION wreckage was found.

Further, as discussed above, the same total elapsed- and interval-times for detections of events one through four by CUHSC hydrophone A and Argentia array 3141 eliminates the possibility that any of the signals arrived by a tranmission path other than the direct path demanded by the elapsed sound-travel-time from the wreck-site to the two sensors. Thus, none of these four signals represented “reflected” energy or echoes from bathymetric features such as seamounts (guyots), i.e., all four represented “original” events.

As also discussed above, the source of the first event was implosion of the SCORPION pressure-hull. The sources of events two through four detected during the 111.6-second period were the implosions of internal structures that survived hull collapse to implode at greater depths. Based on the acoustic energy peaks below 400-Hz of two of the four events that occurred during the 111.6-secpnd period, the source of these events is assessed to have been the collapse of two SCORPION torpedo tubes at depths of 3370- and 3750-feet. The four other torpedo tubes collapsed later at depths of 3810-, 3950-, 4510- and 4570-feet. Note that the SCORPION torpedo tubes were designed not to warp at test-depth and were capable of impulse-launching weapons at test-depth at flank-speed.