25m down, Maureen Coulthard was exploring the wreck of the Breda when, without warning, her drysuit hose ruptured. As air gushed from her first stage, there was insufficient pressure to send any gas to her second stage, leaving her unable to breathe. Her buddy was five meters away with his head in a hole. Coulthard made a split second decision to make a free ascent. She survived unscathed. Whether a less experienced BSAC trained diver faced with such an emergency would have made the ascent successfully is questionable.

Although Coulthard has been diving for 26 years and is an Advanced instructor and First Class Diver she had not been trained to make a free ascent. Had she been even an entry-level diver from almost any other major training agency she would have made a practice controlled emergency swimming ascent before being qualified to dive. However as a BSAC diver she is banned from practicing this particular self-rescue skill. That the BSAC has remained out of step with much of the international diving community on this safety issue is a matter that needs to be debated.

The controlled emergency swimming ascent, or free ascent, is a fast and efficient way for an out of air diver to reach the surface. It simply involves finning up while exhaling to keep expanding air in the lungs from causing lung expansion injuries. There is no question that a risk of serious injury or death exists in carrying out a CESA. However it is a technique used only to avoid drowning from which recovery is a matter of reincarnation. So all training agencies discuss CESA as a valid self-rescue skill. The difference is that BSAC only discuss it as a theoretical exercise wheras most other leading agencies require their instructors to teach it and students to practise it in open water from a depth of 6 to 9 metres.

In challenging the Club's long standing policy on refusing to sanction CESA training we have to look at a host of safety issues. Currently BSAC divers, like all divers, are provided with a range of options for successfully surviving an out of air emergency. These include using a redundant air source, a buddy's alternate air source, buddy breathing, buoyant emergency ascents and the CESA. Each of these methods has pros and cons. Each also makes certain assumptions about the diver's equipment and abilities and sometimes that of their partner's. Incident reports suggest that oftentimes these assumptions are optimistic.

In a perfect world of course divers would not run out of air at all. But they do. When we surveyed divers during a study into alternate air source use for Dive, half our participants had shared air. This may indicate that training agencies need to implement more stringent procedures that train divers to check their air more regularly, much as some drivers are taught to make 10 second checks of their rear view mirror. Certainly it raises cause for concern.

Clearly if a diver does run out of air or suffer equipment failure then the best solution is a fully redundant or Independent Alternate Air Source (IAAS). Following our study, co-author Matt Crowther and I, recommended as a minimum that divers operating to approximately 40 m and working within the no-stop times should carry a 3 litre pony and regulator and that divers operating deeper or making stage decompression stops should select either a twin set made up of two independent single tanks and regulators or a twin set with two regulators and an isolation manifold. Using an IAAS avoids reliance on another diver and putting that diver at risk, does not deplete the assisting diver's own air supply, does not compromise the donor's own regulators ability to supply air to two heavily breathing divers and should eliminate any surprises which using unfamiliar equipment can spring. But in reality few divers own this equipment. While IAAS remain the safest option for solving an out of air situation they are also expensive, bulky and heavy. Many entry-level divers will not buy them and for the travelling diver baggage and filling restrictions make them impractical.

In the early days of the Club, the BSAC did indeed require BSAC members to make CESAs as part of their training. According to the 1961 edition of the Diving Manual, "On the grounds that it is as well to be totally prepared for any eventuality it is right that the principles of emergency ascent are known and taught to trainee divers, though its dangers must also be stressed. The ability to carry out a calm and deliberate ascent from depth calls for confidence and control". But by 1966 the rules had changed. Writing in the new manual, Lieutenant Commander Mathew Todd, then Commanding Officer at HMS Dolphin Submarine Escape Training Tower stated:

"As the practical training is likely to be more dangerous than the actual emergency drill no training should be carried out except under the closest supervision and with a compression chamber on the spot and at immediate notice. Training in free and buoyant ascents is for divers like training in the use of ejector seats for pilots - the final resort. There is a great danger that the lungs will be overloaded during this type of ascent, particularly if the diver is supporting any weight, perhaps unknowingly."

It has been suggested that the Royal Navy was put in an impossible position by the BSAC seeking its professional advice. Had the senior service agreed to free ascent training and an accident occurred they might well have been open to criticism. But Todd, himself the survivor of a night time free ascent from under the hull of an aircraft carrier after his breathing set failed, denies this. He maintains that the RN genuinely believed more amateur divers would be injured or killed practicing the ascents than would be hurt doing it, untrained, for real. The Club's concern, then, was that in training divers to make free ascents the skill might execute the diver rather than the diver execute the skill. On the Royal Navy's advice BSAC banned CESA training along with buddy breathing ascents. Some time later buddy breathing ascent training was reinstated. CESA was not.

However it seems likely that confusion existed between the Navy and the BSAC over what, exactly, the skill involved. At HMS Dolphin the Navy trained submariners to make an ascent from 30m. According to professor Adam Curtis of the Scottish Sub-Aqua Club, "Surprisingly the RN does not appear to have carried out any detailed research into the incidence of free ascent barotrauma amongst amateur divers in arriving at its recommendations and may not have been aware of the type of training that was in fact being used. In the late 60s and early 70s, the RN had a small number of cases of fatal barotrauma during submarine escape training, mainly among the trainees. The training requires the use of very fast highly buoyant ascent with speed of 2 metres per second or faster, with buoyancy in excess of 10 kg. Clearly these very fast ascents do have a relatively high risk of barotrauma, perhaps particularly amongst those who have little or no previous experience of being underwater." Curtis' point is well made. An ascent speed of 120 metres per minute is in line with that of a diver using a BCD to make a buoyant emergency ascent. The BSAC during the time they sought the Royal Navy's advice instructed divers making a free ascent to ditch their weight belts. Today this would be considered a buoyant ascent, not a CESA. Dropping belts would make the diver positively buoyant and make it almost impossible to control the ascent rate. A free ascent made for real with weights in place is likely to be made at around 0.5 m per second or 30 metres per minute. NAUI recommend 6-12 metres per minute in training. The modification to the skill justifies re-examining the issue of teaching CESA again.

Figures published by the late Surgeon Captain Stanley Miles RN in Underwater Medicine and repeated in Carl Edmond's Sub Aquatic Medicine estimate Navy fatalities at one in 50,000 (fifty thousand) for submarine escape training. This is despite the provision of an on-site medical team and recompression chamber into which a casualty can be placed in seconds. It is understandable that the RN might have feared for the safety of sport divers who would not have the luxury of such medical support, and this is a concern that continues to disturb Mathew Todd.

In the USA they faced a similar dilemma. Jim Stewart is one of the icons of American diving. A former member of the crack Underwater Demolition Team, he was investigating the results of the H-bomb blasts at Enewitick just three days after the detonation and is a veteran of a multitude of diving expeditions to both poles. The former diving officer of Scripps Oceanographic Institute, Stewart developed the diving program adopted by scientific divers and university diving programs throughout the States. He recalls the aftermath of abandoning CESA training:

"A sport diver diving in very clear water lost contact with his buddy. Rather than exhale and come up, which he had been taught was dangerous, he tried to swim to his buddy who was some distance away. Divers who were entering the water saw him from the surface as he passed out on route. Before help got to him he was dead. The distance from the diver to the surface was thirteen meters and to his buddy it was thirty."

The Americans reintroduced CESA training and stand by that decision.

So how dangerous are controlled emergency swimming ascents? As with so much related to diving safety little hard information exists. However we can assemble a picture from the few studies that have been made.

Reg Vallintine, former BSAC Chairman, discover of a Estruscan shipwreck from 700BC and founder and archivist of the Historical Diving Society presented a paper at the 1972 Diving Officers' Conference. Entitled "Which Way Up", Vallintine's figures were encouraging. Using statistics gleaned from research into free ascents made between 1956 and 1962 Vallintine estimated that between 15000 and 30000 ascents had been made without injury by BSAC divers.

In 1977 the Undersea Medical Society convened to discuss emergency ascent procedures. Evidence was conflicting. Opposing, John McAniff of University of Rhode Island National Underwater Accident Data Centre cited evidence of some 28 fatalities that had occurred Stateside during emergency ascent training between 1970 and 1976. 10 Were a result of buddy breathing and 9 a result of CESA training. McAniff estimated the fatality rate at 0.0000077 and 0.0000141 respectively. For, the Scottish Sub-Aqua Club Adam Curtis countered with figures showing that at that time SSAC divers had made 2800 free ascents during training without incident.

During the period referred to by both Vallentine and Curtis CESA training was not conducted as it is today. It would appear that more opportunity existed for accidents. Divers ascended with their regulators out of their mouths and were not necessarily restrained by an instructor who had physical hold of them and could brake the ascent at any time using an anchored shotline. Yet the combined safety record seems impeccable.

In 1993 the South Pacific Undersea Medical Society revisited emergency ascents. PADI, who certify more divers than any other agency, showed a single fatality as a result of CESA training between 1989 and 1992. This provides an official fatality rate of one for approximately 2,000,000 (two million) training CESAs. In reality even this figure is probably overestimated: often CESAs are halted part way through the ascent when the student fails to make the surface on the first attempt. Even a partial ascent places the student at some risk of barotrauma. Its fair to say that the potential for accidents is far greater than the PADI figures, implying the real accident rate is even lower. Certainly CESA training involves some risk to the student---but so do many other skills which all agencies teach, such as mask clearing and buoyancy control. In the litigious environment of American diving it seems unlikely that major US training agencies would include a skill they considered open to serious challenge in the courts. In fact the number of US training agencies teaching CESA is increasing with the announcement that IANTD, so far a technical diving agency catering for qualified recreational divers from other agencies, will introduce it's own entry level diver program that will also require a controlled emergency swimming ascent. Dr Art Bachrach, co author of "Stress and Performance in Diving" has called for more, not less, CESA training, stating "practicing emergency ascents while exhaling from moderate depths is not emphasised enough in courses."

In Britain, where BSAC had been the dominant training agency for many years, but had forbidden CESA training for most of them, it seems reasonable to assume that were CESA training a genuine safety issue then a noticeable accident rate would have become apparent as PADI, NAUI and SSI moved in. This should have skyrocketed as these agencies gained market share, especially during the time when the NVQ scheme was in full force and PADI was literally training tens of thousands of new divers, each performing a controlled emergency swimming ascent.

But according to Dr Phil Bryson of Diving Diseases Research Centre they've seen no evidence of this. Bryson told me, "this is really a training issue, but I agree the potential risks may also make it a medical issue. As far as incidents occurring during training, we haven't seen a major problem. DDRC has not been involved in treating cases of decompression sickness or air embolism resulting from a controlled emergency swimming ascent. When making ascents from 6 to 9 metres, the risk of DCS is far less than that of air embolism. At greater depths the risk of DCS is increased, especially for instructors who make multiple ascents." This raises a legitimate concern than while the student who may make only one CESA is at little risk of DCS, instructors teaching the skill to a group of divers may put themselves at risk of a hit.

Marjorie Baldwin, spokesperson for Stoney Cove concurs. Thousands of divers train every year at the inland National Diving Centre, many instructors choosing it for the safety its on site chamber affords. Yet Baldwin has never recorded an incident relating to CESA training.

Some may argue that with the widespread adoption of alternate air sources the value of CESA training has been diminished. But other industry insiders strongly refute this. According to one accident investigator the mystery of out of air divers who have commenced an assisted ascent breathing from their buddy's octopus, then inexplicably broken away and drowned may be a result of poor regulator performance. ASSET (Association of Scuba Service Engineers and Technicians) technician trainer and owner of STATS (Scuba Technical And Training Services) Stewart Meinert offers an explanation that supports this view, "it's certainly possible theoretically for two divers to beat the lung, even with a modern regulator. Even if the regulator can supply adequate air flow, its performance may be sluggish. This could give the out of air diver the sensation that he's running out again and cause him to panic." Another concern is that many second stages used for octopuses are detuned by their owners to prevent free-flows. According to Meinart "a diver hitting a second stage that's been locked down could easily feel air starved." During our study of AAS use we hinted at this after a highly experienced instructor had problems using an octopus whose venturi assist was turned off. An additional problem is that rental regulators are often not only bottom of the range, but also have a dubious service record. Stan Ellis of ANSTI, explains that a poorly serviced regulator could easily fail to support two heavily breathing divers in an sharing situation. Another concern is that in the past some cylinders may have been fitted with cylinder valves that cannot themselves supply adequate air flow under high demand. This has serious implications even for the responsible diver who fits his own high performance regulator to a rental tank.

The value of CESA is that it does not require the assistance of another diver. During research into alternate air source use and misuse for Dive Magazine my co-author Matt Crowther, himself a BSAC and PADI instructor identified problems with more than 60% of the exercises despite the fact that the survey should have been loaded by the level of training of the participants, the regular refresher exercises they attended and that they had been briefed beforehand removing the element of surprise from the exercise. It has to be accepted that many divers today dive only occasionally, do not periodically review their skills, may be diving abroad using unfamiliar rental gear of questionable quality and in poor repair and diving with divers from other agencies who may not share the same skills or even use the same signals. The huge combination of alternate air source types in use, how they are stored, released and donated has made standardisation on a AAS ascent procedure quite unworkable. The resulting confusion in an out of air emergency can easily lead to tragedy.

Buddy breathing is also a controversial choice. Long considered to have a poor track record, some agencies, including PADI, have made it an optional skill. While BSAC continues to teach it, an out of air BSAC diver may find himself reliant on a diver from another agency who has not been trained to do it at all. As divers move away from the Branch environment of the past towards diving on holiday with people they have never met before, let alone trained with, problems with differences in training become exacerbated. Writing in "Stress and Performance in Diving" recognised dive safety experts Bacchrach and Glen Egstrom underscore the problems: "in the case of buddy breathing, a study conducted by the UCLA Diving Safety Research Project determined that 17-21 successful trials of the skill resulted in performance without errors in a group of basic students. It was also noted that that following three months of diving without reinforcing the skill, there were were degraded performances involving errors in procedures. It seems obvious then that not only should the skills be well learned, but they should be periodically reinforced, especially in circumstances where the buddies are diving together for the first time".

Another fear is that divers getting into difficulties may rapidly slide into a panicked state of mind from which they are unable to extricate themselves. At this point their only chance of surviving lies in the hands of their buddy who may or may not be present and may or may not have been trained in the rescue skills required to save them. In an out of air scenario, rather than fumbling an attempt to secure an AAS or buddy breathe which requires a large number of skills to be completed perfectly while stressed which may plunge the diver deeper into the incident pit, it may be safer to immediately initiate a CESA while the diver still has some control.

James Stewart also strongly raises another issue, "having dived for more than sixty years of my life I can state that no matter how competent and aware a diver may be, there will be buddy separation. The lower the visibility, the more probable."

This is a point made time and again in fatality reports. Divers die while separated from their buddy. Though many divers expect to buddy dive, separations do occur. In the event of an out of air situation a diver without an IAAS must get to his buddy in order to breathe. Not only is the risk of separation increased in the low viz typical of British conditions, in clear water divers often drift apart to such an extent that an out of air diver has little chance of reaching his buddy. For our AAS study most divers managed a horizontal swim of only 5 or 6 meters having exhaled before they needed to breathe again.

Accident investigators also have concerns about the honesty of some survivors' accounts, implying that they may not have carried out the correct rescue procedures and are covering up for their failings.

So having a buddy in no way guarantees a diver's safety. They simply may not be there to assist even if they have the essential skills needed to help in the first place. Teaching self reliance rather than buddy dependence is crucial to protecting divers.

This strengthens the argument that all divers need to learn self-rescue skills that can halp them survive an accident while diving alone---either by intention or default. In the absence of an Independent Alternate Air Source this leaves the CESA or buoyant ascent as the only options; other than drowning.

"Scientific Diving - A general Code of Practice" also supports CESA training for the scientific diving community, the professional divers whose style of diving and training most closely approaches that of the sport diving community.

"The routine training of divers in the practical experience of free ascent training is controversial. It does not however seem reasonable to advise divers to rely in emergencies on physical skills and psycho motor coordination for which they have no practice. It is unsatisfactory that many safety authorities presently advise students not to do training, but to use the methods in emergencies. Book learning cannot provide the conditioning of reflexes and instill the necessary confidence and relaxation. A number of national diving authorities have devised safe training methods. Most divers quickly manage to learn the technique, after which they usually wonder what all the fuss was about."

The program Jim Stewart developed for scientific divers enabled him to get scientific diving excluded from the heavy legislation OSHA (the US equivalent to our HSE) placed on commercial diving in 1981. Stewart explains, "our Scripps Research Diving Program was cited in the Federal Register as the model 'of Safety and efficiency'. Open water (emergency) ascent training has been and remains a crucial part of our diving training."

What happened to Maureen Coulthard could happen to anyone. She now dives with twinned independent singles. Emphasising that she was not speaking on behalf of the Club for whom she acts as Treasurer, she told me: "When my hose ruptured my tank drained in seconds, high and low pressure came into force and the ruptured hose won against my DV. I received no air at all. I had to make an instant decision. The whole thing depends on what the situation is and who your buddy is, but a free ascent is certainly one way out of trouble provided it is undertaken properly. Personally, if given the choice I would rather have been trained to undertake the action I took. I am a diver with an opinion and that opinion was gained through actual experience."

Incorporating CESA training into entry level training compels every diver qualified by the BSAC to leave with an effective self-rescue skill to deal with the eventuality of air failure. We cannot force divers to buy or use alternate air sources, nor can we control with whom they dive or even stop them running out of air. If we cannot solve the underlying causes then we must, at least, offer solutions to the problems.  

Steve Warren is a former BSAC instructor and branch training officer and a PADI Master Scuba Diver Trainer. For this article he completed a 30 m CESA without incident.