By Roly Nyman : Convenor

1) The Enquiry was held according to the agenda and protocol attached, which was issued immediately prior to the start of the meeting. - Attachment One.

2) The Enquiry was attended by many of the dive team and a broad cross section of divers in the Technical Diving Field. - Attendance Register available on request.

3) The following members of the dive team gave reports - listed below in order of appearance:

4) A written report from Frik Henning, Standby diver on boat, was received on 10/07/98 and has been incorporated into this report.

5) The Proposed Dive and Dive Plan

The Dive was one of a series of Trimix deep dives held during the week to further knowledge of the layout and marine life in Wright Canyon, a steep depression in the sea bed. The bottom on the edge is sandy and virtually flat at about 90m depth. Rehan and his buddy were to make an excursion to the 140m mark, to do a visual comparison of the sea life at that depth as opposed to higher up and collect a few specimens if possible.

Johnny decided, prior to the dive, not to descend to that depth and it was agreed he would remain near the edge at a maximum of 120m and act as a deep cover diver instead. The Dive Planning sheet is available on request.

Decompression was to be done while drift diving. Each deep diver was to inflate a long vertical tubular marker buoy at 40m depth and release it, which would be the signal for a dedicated cover diver to descend and check that all was well. He would then accompany the deep diver as he reeled himself up through his decompression schedule. If the divers drifted apart during decompression, there was a dive boat available to follow each of them individually.

Everyoneís first stop was at 40m. Thereafter, the Deep Teamís next stop was at 33m and Johnnyís stop was at 27m. Allocation of the standby divers was as follows:

Rehan - Andries; Peter - Johan; Johnny - Frik.

6) The dive commenced at 13h50hrs on a clear sunny day. A cold Force Four NW wind, which had not been present on previous days, was raising a 1m chop on the surface and creating a 1,5 - 2,0 knot current. Surface temperature was 24oC but viz had been reduced to 8m.

On the bottom there was no current, only a barely perceptible drift. Although light levels were low, visibility was in excess of 30m. As each diver had torches, keeping track of each other was not a problem.

As Rehan bailed out into the water, he apparently bumped his head on Johnnyís bottles, but on enquiry claimed he was alright. The deep team then submerged.

7) Rehanís Equipment

2 x 12 liter 220 bar cylinder, connected by a Genesis-type manifold (independent taps, but no isolation valve), filled with bottom mix, Dräger Nemo + Spiro Arctic + inflator hose + BC hose to backup BC + contents gauge

1 x 7 liter 200 bar cylinder, strapped to twin, for Nitrox 32, Spiro Arctic + contents gauge + BC hose to main BC.

1 x 10 liter 200 bar cylinder, filled with Nitrox 80, Sherwood? + contents gauge. No helmet, one hand-held torch, no camera, small reel + AP Valves marker buoy, crayfish bag, wetsuit + chicken vest.

Note : Position of power inflators not able to be confirmed. See Free-flow theory.

8) What happened? A "best fit" collation of the eye witness reports.

The descent and most of the bottom time went off as per plan and were uneventful. Due to the not quite vertical walls of the canyon, the deep team had been displaced sideways about 7 - 8 m from the vertical. In addition the deep pair had drifted a few metres apart while collecting their specimens.

Near the end of the bottom time, Peter saw Rehan swimming towards him, when he stopped, turned and started swimming vertically upwards. Before Johnny could react Rehan had passed him, looking upwards in a fixed direction. As he was sideways on to Johnny, he did not notice him, but could easily have seen him if he had been looking.

Johnny went deeper down the cliff face to check on Peter and they met at 126m, exchanged signals and commenced their ascent on their planned schedule. They changed from bottom mix (12:60) trimix) to travel mix (Nitrox 32) and deployed their buoys at 40m. They were soon joined by Johan, one of the dedicated cover dives. Johnny soon moved to his planned stop at 27m where he began drifting away from Peter who had stopped at 33m. They soon lost sight of each other as the visibility deteriorated above the thermocline at about 50m. Both he and Peter completed their schedules and surfaced without incident.

Back on the boats, everyone was keeping an eye out for the marker buoys, which were expected any minute. Thus when Rehan himself surfaced he was spotted immediately. All agreed that it was a controlled ascent - no rush of bubbles, no rising to his waist out of the water and no unduly inflated ABLJ. Distance estimates from the various boats vary from 30 - 50m He signalled in a calm manner to be collected and shouted for more oxygen and "gas". Witnesses disagree as to whether he spoke English or Afrikaans and whether he called for air, gas or nitrox. All agree he was calm and controlled. Frikís report states that Rehan shouted "I have had a free-flow. Go to the beach and fetch oxygen.

The boat arrived within seconds and two fully kitted dedicated standby divers jumped in the water, one of whom, Frik Henning, carried an extra side-slung bottle of oxygen apart from his set filled with Nitrox 32. It was for this reason that Frik jumped in as well as Andries, as Andries was not carrying oxygen. Rehan then disappeared beneath the surface in an apparently controlled descent. His surface interval was estimated at between 30 seconds and two minutes. No-one has a clear recollection of whether he submerged before Frik reached him or whether they went down together. Frikís report simply states that at 6m he offered Rehan oxygen but Rehan chose the Octo and signalled for them both to descend.

Andries Prinsloo, the other dedicated standby diver who had baled out over the other side of the boat, swam under it and was hit by it due to the choppy wave action. The blow was enough to distort the plastic frame of his mask, which created a copious leak. Nevertheless he descended to 20m where he saw Rehan and Frik holding onto each other and descending rapidly. Andries then aborted his dive and returned to the surface.

Rehan was hanging on to Frikís BC with both hands, and by keeping his elbows bent, had minimised the distance between them, presumably to relieve any strain on the octopus hose. Unfotunately this position prevented Frik from reading Rehanís pressure gauges. Rehan and Frik repeatedly exchanged "are you OK" signals on descent.

They rapidly reached 40m, at which point Frik showed Rehan his depth gauge. Rehan then handed over the marker buoy and reel and signalled to Frik that he should inflate it.

Frik then gripped Rehan round the waist with his legs, in order to free his hands, and started trying to inflate the buoy. This of course prevented Frik from finning and would have accelerated their descent rate. This buoy is an open ended tube (with a butterfly valve in the open end) about 2 metres long which is normally filled from an accessory filling hose attached to one of the regulator first stages. This option was obviously not available to Frik. While busy, he became aware that he had suddenly become dizzy, stopped what he was doing and checked his dive computer which read 65m. He showed Rehan the reading. Rehan looked shocked and immediately signalled that they should ascend. It is possible that Rehan started finning and of course exertion would precipitate unconsciousness. Frik immediately commenced inflating his BC. Frik remembers Rehan staring at him wide eyed before he (Frik) passed out. His next recollection is waking up on the dive boat.

From the medical teamís viewpoint Frik surfaced in a highly agitated state, tangled in Rehanís marker buoy, and virtually incoherent when questioned. He was helped into the boat, put on oxygen and checked out by Dr Brits. A check on his decompression meter by the medical team showed that his maximum depth was 65,2m.

9) What went wrong? - An analysis of the eye witness reports

The trigger incident that caused Rehan to abort his dive and shoot to the surface could have been either a free-flowing regulator or (less likely) an out-of-gas situation. The free flow would not have been spectacular as there was very little gas left anyway at the end of the bottom time. It would however have dumped very quickly for the same reason.

From previous discussions it was clear to others (inter alia JvdW) that Rehan was well aware that one of his regulators was near its limit, and that he was anticipating a failure.

His apparent abandonment of his buddies is no mystery. Rehan had often expounded his twin philosophies that a) each deep diver should be self-reliant and not dependent on his buddies and b) that you could in practice switch to travel mix at any depth because you reached its safe depth before its toxic effect could become apparent.

What is certain is that he did not panic but put into action a pre-planned escape plan. From records kept by boat personnel, Rehan took at least six to seven minutes to surface, where he arrived calm and controlled. But the early change-over obviously ate heavily into his reserves of travel mix. It is not impossible that he deliberately slowed his ascent, using all his reserves in the process, so that he could out-gas as much as possible before being forced to the surface.

Having made a text book ascent and made his needs known, the next logical step in his revised dive plan would be to get down to his first stop (40m) as soon as possible to resume his decompression. So probably without realising the consequences, he must have vented the entire contents of his double BC. Only after he had descended a few metres would he have discovered that he had absolutely no breathing gas left, apart from his oxygen. This would be more than sufficient to fill a BC even if he had only started inflating at 10m depth. The evidence shows that he did not inflate either BC which implies he could not.

Free-flow theory

If we assume a free flow at depth towards the end of bottom time, the cylinder pressure would be fairly low so there would be no spectacular rush of bubbles. Neither would there be a cooling of the cylinders due to pressure drop and therefore no tendency for water to be sucked back past the damaged seat. In fact the opposite would be true. As ambient pressure fell, so air from the cylinders would bubble gently out to equalise the pressure, and Rehan would arrive at the surface with the cylinders effectively empty.

What is more relevant is that due to the valves up configuration of his set, Rehan could not reach the cylinder valve to shut off the faulty regulator.

Once he started descending however, the whole picture would change. It is now possible that water could leak into the cylinder in ever increasing volume as the pressure differential increased. Ultimately this negative buoyancy would grow to the point where even two standby divers would not have been able to support him. Remember that only one standby diver was allocated to each deep diver.

There was some conflict of opinion as to how Rehanís BCs were inflated. Photographic evidence was inconclusive and different witnesses were adamant that either both BCís were inflated from separate cylinders or that one was inflated from the travel mix. It seems clear that he was completely out of both travel and bottom mix and therefore unable to inflate from either source, which makes the point somewhat academic.

Rehan did not wear a weight belt because he was already slightly negative in full rig without one. It was no doubt for this reason that he handed the buoy over to Frik for inflation. Rehan was now totally dependent on his support diverís Octopus and BC.

In all probability Rehan would not have been disturbed by an initial rapid descent as this fitted in with his emergency dive plan. However he could not have been happy at being totally reliant on his buddy. It is probable that neither realised quite how rapidly they were sinking, although they may have realised they had not achieved neutral buoyancy. Inflating the buoy may have also been an attempt to gain additional buoyancy. Again it is probable that neither diver immediately realised the implications of being at that depth on Nitrox 32, and how rapidly their options were being limited.

Sinking from 40m to 65m could not have taken more than 25 - 30 seconds, and this was also the first time that the divers could have realised that things were deviating from the revised dive plan. This was all the time they had to assess the situation, analyse the problem and take any possible corrective action. Tragically, this wasnít nearly enough. At 65m, the pO2 of Nitrox 32 is 2,4 bar and unconsciousness would be inevitable within a few minutes. Frik was strongly positive after he inflated his BC, while Rehan remained negatively buoyant. Both divers would have lost consciousness virtually simultaneously, lost their grip on each other and promptly parted company.

10) How could this accident have been prevented? assuming no changes to the existing dive plan, equipment and personnel.

The accident sequence we have postulated above has an awful inevitability about it - and in truth Rehan had very little room to manoeuvre. There are three possible courses of action that may have helped :

11) Conclusion

It would seem "obvious" then, that the direct cause of the fatality was breathing Nitrox 32 below the safe depth. How this situation was arrived at can never be determined for certain despite being satisfied with constructing a logical "best fit" scenario that does not clash with observed "facts. This scenario suggests that the trigger incident was probably regulator failure (free-flow), which was compensated for in text book fashion as shown by Rehanís controlled arrival at the surface. The indirect cause of the accident was Rehanís inability to attain, and maintain, neutral buoyancy at his desired deco stop.

It is also clear that this was not due to simple failure of equipment or personnel at that stage of the dive. This means that future Technical Dive Planning must be thought through from first principles to take the circumstances outlined above safely into account.

12) Note

This is the final report on this accident.




Reason : To analyse the circumstances of the disappearance of Rehan Bouwer

Status : This committee has no official status at all, neither legally, nor within the officialo theofficial Technical Diving Training Organisations.

Convenor : Roly Nyman - at the request of the diving team.

Nominations for Chairman for the meeting:

As Roly has no official status, being an ex instructor and having no formal technical diving certificates, nominations for a Chairman for the meeting may be called for.

Format : The enquiry will follow the format evolved by the now defunct TUU accident investigation committee, which worked extremely well for over 10 years. (See below).

Purpose : The purpose of the enquiry is not to apportion blame in any way but merely to try and establish the facts behind the accident so that we can all learn from it, and thus minimize the possibility of it ever happening again. Where hard facts are not available, then the most probable scenario will be reconstructed.

Format as per TUU protocol :

1) Each available witness will be interviewed in isolation, starting with those closest to the victim, then moving through the backup teams up to and including casual shore witnesses if necessary.

2) Please note that initial questioning will be limited to clarifying what the witness said, not querying his policies or his decisions.

3) Only after each witness has had his turn to speak will the discussion become more general.

4) Only once we are reasonably sure of what happened will we draft a statement for general information.

5) In very broad terms there are only four possible outcomes to this enquiry:

    a) Insufficient information available - no recommendation

    b) "Procedures not adhered to" a motivational, and disciplinary problem

    c) "Procedures inadequate" on crucial points - need revision to keep up with progress, a training problem

    d) Freak circumstances unlikely to happen again.

6) Our findings - and any recommendations based on them, will be issued as soon as possible to all those present who have given in their contact details; made available to anyone who requests them; and forwarded to anyone suggested by this committee now, eg SAUU officials.

7) In recording our findings please remember that hindsight is an exact science and it is very easy to be wise after the event. Our findings - and recommendations (if any) must rise above our emotions and prejudices and must be based on the facts, where irrefutable, and if not on the most logical deductions.

Remember "Occams razor" : If two scenarios fit one set of facts, the simplest one is the most likely to be true.