An Analytical Look at Survivable Submersion Times

After the tragic entrapment of a canoeist on the Little River last weekend, I found myself wondering about the odds of surviving a long submersion.  The canoeist’s peers, a team experienced in swiftwater rescue, persisted in their rescue efforts for 37 minutes by repeatedly swimming into a rapid that had just ensnared their friend in an unseen feature and despite initial orders from responding authorities.(1) Thus is the camaraderie and commitment within the boating community; few rescue personnel would have given the same concerted effort at such great risk to themselves. That a pulse and spontaneous respirations were restored following resuscitation is further testament to their efforts. Unfortunately, the victim succumbed to his injuries later that evening in the hospital.

The decision to continue rescue efforts is personal and based on careful consideration of the risk to oneself and the likelihood of good outcome for the victim.  Many boaters would place themselves in harms way given the slightest chance of successful rescue, and I’m sure some would even risk harm to recover the body of a friend. Nobody can argue with those decisions, as long as they are based on an understanding of the chances for successful recovery. While most boaters have an appreciation of the risk involved in a rescue, few understand the relationship between survival and time submerged. I have heard everything from scaling back efforts following the “1-minute window” to pursuing rescue up to an hour in cold water. Medical professionals were present for last week’s rescue, and their decision to continue the rescue was based on a variety of factors.  While contemplating what I would have done, I realized my knowledge was limited to a recollection that children can survive extended periods submerged in cold water, and that Rod Baird survived 6 minutes submerged under Hydroelectric Rock on the Chattooga(2). The following is an exploration of the subject to aid myself, and others, in similar situations.

The Natural Course of Prolonged Submersion

Death or severe disability by drowning is caused primarily by lack of oxygen to the brain. Fatal neurological injury normally occurs within 5 to 7 minutes of submersion, and almost always occurs following 12 to 14 minutes.(3) Survivors may suffer a spectrum of disability ranging from memory loss to persistent vegetative state, as damage to the brain progresses inward from the cortex (higher brain functions) to the brainstem (heartbeat, respirations, reflexes). Challenges following resuscitation include fatal brain swelling, damage to the kidneys and lungs, and electrolyte imbalances which can cause cardiac arrest. Several studies have demonstrated the relationship between submersion time and survival (5,6), including a case series of children that found the risk of death or severe neurological disability to be 10% for 0 to 5 minutes, 56% for 6 to 9 minutes, 88% for 10 to 25 minutes, and 100% for greater than 25 minutes.(7)

Submersion time over 5 minutes makes intact survival unlikely; however, there have been rare cases of survival following prolonged submersion, including the longest case ever documented, a 2.5 year old girl submerged for 66 minutes in 5°C (41°F) water.(8)  One physician writes, “Reports of such ‘miracle’ cases in the medical literature, although fascinating, can readily introduce a false optimism because of the limited reporting of the dismal outcome in the majority of prolonged submersion victims.”(9)  There are 500,000 fatal cases of drowning per year worldwide(10), and exposure-adjusted, person-time estimates for drowning are 200 times as high as such estimates for deaths from traffic accidents(11). Despite this frequency, a 2011 review of medical and news reports with documented submersion time and age found only 43 cases of survival with near-normal functionality following prolonged submersion (> 4 minutes).(4)  Two-thirds were children less than 12 years old, and the remaining adolescents and adults were noted to be small in size. Only 4 survived prolonged submersion in water greater than 6° C (42.8° F), and all were submerged less than 30 minutes. The authors state that “this is likely to be a reflection of the fact that such survival is extremely rare in water warmer than 6 °C, rather than indicating that we have missed a large number of incidents in our search of the literature”, although the possibility of undocumented cases has been raised (12,13).

Cold water lengthens the survival time by two mechanisms. It triggers the mammalian diving reflex, which halts breathing and conserves oxygen by slowing the heart rate and moving blood to vital parts of the body. This response is stronger in children than adults.(14) An opposing “cold shock response” may predominate, which leads to a faster heart rate with potential fatal rhythm disturbances(15). This response also causes immediate aspiration and swallowing of water, which quickly cools the heart and carotid arteries leading to “selective brain cooling”.(4) A reduction of brain temperature by 10° C decreases energy consumption by 50% and doubles the duration of time the brain can survive without oxygen.(16)  This “therapeutic” hypothermia is accelerated by surface cooling in children and small adults with higher surface-area to body mass ratios and less subcutaneous fat.  Panic by the victim (breath holding and vigorous attempts at escape) and protective gear worn in cold water work against these principles and may prevent therapeutic hypothermia.

Relationship between water temp and submerged survival time for the rare instances of survival with full recovery outlined above. From Tipton 2011

Technical Guidelines for Rescue Attempts

There is no universal consensus on rescue efforts in prolonged submersion. A group of experts published the following based on the cases outlined above: “if water temperature is warmer than 6 °C (42.8 °F), survival/resuscitation is extremely unlikely if submerged longer than 30 minutes. If water temperature is 6°C or below, survival/resuscitation is extremely unlikely if submerged longer than 90 minutes.”(4)  They made no differentiation between children and adults given that there have been rare cases of adults surviving prolonged submersion. The possibility that cases of survival longer than 30 minutes in water warmer than 6°C  exist, but have not been identified, has some promoting between 60 minutes (US Lifesaving Association) and 90 minutes (The Joint Royal College Ambulance Liaison Committee) of rescue efforts regardless of water temperature.(12,13) The original authors point to a lack of evidence supporting such guidelines, adding that “when conditions are extreme, rescuers may be put at risk without foundation.”(17) Regardless of differing views, all agree that it is the responsibility of the commander to tailor efforts to the situation at hand, and specified timeframes are simply guides “likely to be of most use when rescuers are placed at high risk by continuing a search and subsequent rescue attempt.”(17) As another medical professional writes: “It is important to emphasize that the victim first needs rescuing and it is the decision to continue these attempts beyond the ‘likely’ survival time that is important for the commander. If the casualty is still awaiting rescue and is beneath unstable ice, in large seas or in the depths of a cave then we would hope rescuers would think carefully about the likelihood of survival versus the risk to those whom we know to be alive right now – the rescuers.”(18)

Practical Guidelines for Rescue Attempts

In dangerous swiftwater environments, the likelihood of survival should continuously be reassessed by the trip leader or individual leading the rescue. For the average person trapped underwater, intact survival is most likely if rescued within 5 minutes, and unlikely following 10 minutes. Cases of survival longer than this are rare, but efforts may be extended in controlled environments with acceptable risk. Guidelines vary between 30 minutes for water exceeding 6°C (4), to 90 minutes regardless of water temperature (12). The one consensus is that likelihood of survival decreases greatly with time submerged, and this should be considered in light of the risk to the rescuer.

Situational awareness is key, and each rescue will be unique. Accessible location, experienced team, and low risk to the rescuers make prolonged efforts more reasonable. Water colder than 6°C and small size of the person entrapped may increase the length of survival. Rivers in both rainfed and snowmelt regions may be colder than 6°C in the winter, but typically exceed this temperature in the spring, summer, and fall (see figures below the Conclusion section for average temperatures based on USGS data from a variety of popular whitewater runs(19)).  For example, from March 10th to 16th this year, the Little River fluctuated from 4.4 to 10 °C (NOAA).  Of course, rivers vary widely in temperature due to a variety of factors (distance from source, air temperature, reservoir release), and each river should be considered independently for the implications on rescue efforts. Lastly, whitewater is dynamic, and the possibility of air pockets should be factored in to any consideration of submerged time.


The rescue attempt that spurred this discussion was certainly conducted within accepted timeframes for possible survival, and it is admirable that the team persisted through difficult conditions to give their peer a chance, however small. Nobody can fault such selflessness, and I hope a similarly skilled crew is present should I ever become entrapped on the river. I also hope that each rescuer would be able to make an informed decision given the circumstances and consider their risk against my chance of survival. I would never wish for heroics that are not founded in purpose and reason.

Frequency of kayaking and seasonal differences between snowmelt and rain-dependent regions, based on averaged USGS data, from Moore 2010.


  1. American Whitewater Accident Database. Accident #3693. Accessed March 19, 2013.
  2. Chattooga River Fatalities and Near Fatalaties Since 1970. U.S. Forest Service.
  3. Orlowski JP. Drowning, near-drowning, and ice-water drowning. JAMA. 1988; 260: 390-1.
  4. Tipton MJ and Golden FS. A proposed decision-making guide for the search, rescue and resuscitation of submersion (head under) victims based on expert opinion. Resuscitation. 2011; 82: 819-24.
  5. Szpilman D. Near-drowning and drowning classification: a proposal to stratify mortality based on the analysis of 1,831 cases. Chest. 1997; 112: 660-5.
  6. Manolios N and Mackie I. Drowning and near-drowning on Australian beaches patrolled by life-savers: a 10-year study, 1973-1983. The Medical journal of Australia. 1988; 148: 165-7, 70-1.
  7. Quan L, Wentz KR, Gore EJ and Copass MK. Outcome and predictors of outcome in pediatric submersion victims receiving prehospital care in King County, Washington. Pediatrics. 1990; 86: 586-93.
  8. Bolte R and Black P. The use of extracorporeal rewarming in a child submerged for 66 minutes. JAMA. 1988; 260: 377-9.
  9. Suominen P, Baillie C, Korpela R, Rautanen S, Ranta S and Olkkola KT. Impact of age, submersion time and water temperature on outcome in near-drowning. Resuscitation. 2002; 52: 247-54.
  10. Peden M MK, Sharma K. The injury chart book: a graphical overview of the global burden of injuries. Geneva: World Health Organization, 2002.
  11. Mitchell RJ, Williamson AM and Olivier J. Estimates of drowning morbidity and mortality adjusted for exposure to risk. Injury prevention : journal of the International Society for Child and Adolescent Injury Prevention. 2010; 16: 261-6.
  12. Perkins GD. Rescue and resuscitation or body retrieval—The dilemmas of search and rescue efforts in drowning incidents. Resuscitation. 2011; 82: 799-800.
  13. Perkins GD. Reply letter: Rescue and resuscitation or body retrieval. Resuscitation. 2011; 82: e5.
  14. Gooden BA. Why some people do not drown. Hypothermia versus the diving response. The Medical journal of Australia. 1992; 157: 629-32.
  15. Shattock MJ and Tipton MJ. ‘Autonomic conflict’: a different way to die during cold water immersion? The Journal of physiology. 2012; 590: 3219-30
  16. Szpilman D, Bierens JJLM, Handley AJ and Orlowski JP. Drowning. New England Journal of Medicine. 2012; 366: 2102-10.
  17. Tipton M, Golden F and Morgan P. Drowning: guidelines extant, evidence-based risk for rescuers? Resuscitation. 2013; 84: e31-2.
  18. Ramm H and Robson B. Reference editorial – Rescue and resuscitation or body retrieval. Resuscitation. 2011; 82: e3.
  19. Moore RD, Schuman TA, Scott TA, Mann SE, Davidson MA and Labadie RF. Exostoses of the external auditory canal in white-water kayakers. The Laryngoscope. 2010; 120: 582-90.
“Screaming Meanies” on the Little River. Photo by author in 2007.

By Ryan Moore

Ryan Moore is a kayaker currently completing his residency training. His other projects include the Surfer's Ear Tour and Paresthesia Productions.

33 replies on “An Analytical Look at Survivable Submersion Times”

Good Article, but I would note that the rescuers did not “swim the rapid”; they swam in or were pulled in from downstream. Furthermore, the rapid has been safely swum regularly. The canoeist’s failure to “ball up” resulted in him going from flat on the surface to nearly vertical as he washed over the pourover. In essence, he was “planted” in rocks at the base.

Thanks for the clarification. Strong swimming “through” the rapid was used by a rescuer I spoke with, and also appears on the AW incident report, which says the group switched from swimming through the rapid to live-baiting into it after one of the swimmers felt the undercut as he went over the drop. As you point out, the persistent efforts were from below, I’ve edited the 2nd sentence to avoid confusion. The article is not meant to pass judgment or second guess any rescue efforts. The team was clearly skilled and within their comfort zone. It was the safety they had set, large support team, and confidence in mitigating the risk that made the continued efforts reasonable.

Thanks for the great article and thanks to the boaters who made such heroic efforts.

The link below is an account from a miraculous recovery after several minutes Inderwater on the Arkansas in very cold snowmelt. It is accurate as far as I know except that I remember the submersion being more than 4-6′, the rescuer named “Greg” is actually named Craig and beaching a non responsive swimmer in that section of the Numbers is a heroic feat.

Amazing story. In these specific cases, the continued efforts were definitely worth the risk. And to play devil’s advocate, how many people have been seriously injured or killed during a whitewater rescue? I haven’t heard of any. Perhaps that is indicative of the inherent understanding whitewater boaters have of risk, and their skills in navigating rescue situations safely. The point is to understand your environment and be comfortable with your risk given the timeline for potential survival, and if you are, then prolonged efforts are absolutely reasonable.

Good, accurate material. The temp stats are helpful and the survivability data is interesting.Lends credence to an on scene rescuer being cognizant of elapsed time by wearing and using a timepiece. Time is short especially in a low O2 environment.

All I can say is… Never give up. Let the hospital MD make the decision about when it is time to quit resusitation efforts. Make educated decisions on the risks to the rescuers, and if the potential of a rescue outweighs the risks, then go for the rescue. I am here today because of good decisions made by a wonderful group of well-educated SW rescuers who did not give up.
And last of all… pray! It does make a difference!

Thanks so much for putting time into this! I am going to share this with our people and instructors.

agree! (see last sentence of 2nd to last paragraph), and if that is a possibility then all time constraints are off

Ryan, Thanks for an excellent, well researched article. I hope to see more of this kind of valuable information as data continues to be collected and our knowledge base grows – it will be of tremendous benefit to all of us in the boating community who have been, or may be forced to make difficult rescue decisions.

Good article, but at least one part of it is fairly misleading. You state that there are 500,000 drownings per year, but fail to mention that this statistic is the worldwide figure for all drownings. In the USA, there are somewhere in the range of 3500 per year (less than 1% of the world total) for all drownings, not just cold-submersion deaths where the person was retreived between 5 and 14 minutes. I can’t tell where the 43 “miracle” survivals are sampled from (one year, multiple years, USA, worldwide) but I would say it’s a pretty safe bet that they are at least limited to first world scenarios where top-flight medical care is readily available.

A good statistic would need to sample persons submerged in similar circumstances, in other words, the number of extended cold-submersion survivals over the number of extended cold-submersion drownings where submersion times were within a given range. Even though the percentage will undoubtedly remain low and the core ideas presented here are sound, the outlook isn’t nearly as bleak as the numbers presented indicate.

The 43 cases were worldwide with no specific timeframe, collected from a review of “University of Portsmouth library, Royal Society of Medicine library, Cochrane Library, Medline, PubMed, Google Scholar, Internet-based press-cuttings and other media or news websites”. They only included cases with a good outcome (“near normal/normal neurological and other functionality”).

You make a great point that these cases are most likely within the population with access to advanced medical care, which would narrow the “500,000” pool of anual drownings significantly. I’ve added “worldwide” to clarify. There are surely more cases that were not identified in the search, or had an acceptable quality of life without near-normal functionality. Certainly something to consider.

Also, conclusions should be drawn from the cited articles. The intent is to share with the whitewater community previously published studies and guidelines, rather than present any novel ideas.

Yeah, any way you cut it, the incidence of *miraculous survival* is going to be very low, and certainly limited to small-framed persons. I do appreciate the fact that, in the absence of hard numbers, you don’t try to hammer out hard stats. The implication of the existing numbers as presented is sufficient to clearly illustrate the point you’re making.

It would be interesting, although very difficult or nearly impossible, to study the incident rate and effect of other factors such as air-pockets, excessive exertion (struggling in a recirculation vs. pinned in a rag-doll position), clothing, etc. Frankly, I hope we never have enough data to be able to perform that kind of analysis.

I came away with the understanding that after 10 minutes, the chance of survival is unlikely but in general paddlers will continue to try to rescue and revive the victim.

I remember seeing a paddler on the tygart with a snorkle inserted into the nect of his dry suit. I think it was for insurance but who knows?

Thanks for the article.

I’ve copied some links below for replies to the Tipton MJ and Golden FS paper that you site for the 6 degree strategy.

With Wilderness Medical Associates and Boreal River Rescue we provide our students with the guideline of submersion greater than 1 hour for when to switch from rescue mode to recovery mode.

You are correct that there is no definitive ‘consensus’ in the medical community (there rarely is for anything) however, I would point out there that one hour of attempting rescue techniques (within the rescuers’ level of acceptable risk) before switching to recovery is the current best practice recommendation by most rescue training providers.


Your input is much appreciated. I have added the information from the responses that you cite. This blog is intended to point readers to the primary sources and hopefully summarize some of the major guidelines. For that, all differing viewpoints are helpful.

As a follow-up, I traded correspondence with the authors of the 6 degree strategy (2011), who maintain their position to this date that there is lack of evidence to support rescue efforts past 30 minutes in water exceeding 6 C. I’ve added a quote under “Technical Guidelines” from Tipton’s response ( to the articles by Perkins that you cite, while also including the rational that some organizations use to support 60 or 90 minute rescue attempts regardless of temperature.

He countered my statement that “few rescue personnel would have given the same concerted effort at such great risk to themselves”, saying that “part of the ‘problem’ is that many would have made the same effort” [after 30 minutes in conditions that lack evidence to support those efforts]. “In all other areas of medicine decisions are made on the balance of probability and likely outcome; this includes denying people treatment. For some reason in the case of submersion, rescuers are supposed to search for longer than anyone has ever survived, and they must do this in warmer, salty water – the balance of probability in most of these cases is that the rescuer is more likely to be injured than the casualty be found alive.”

One of the concerns Perkins expressed was that these guidelines would be used as a threshold for ceasing rescue attempts, and thus should err on the side of caution (theoretical possibility of survival past 30 minutes in water exceeding 6 C). In addition to the arguments above, Tipton’s published response clarifies “what we have produced should only be regarded as a guide, and local circumstances and/or clinical signs may dictate an alternative course of action to the senior medical responder at the scene. It is likely to be of most use when rescuers are placed at high risk by continuing a search and subsequent rescue attempt.”

Great article.

One clarification: Rod Baird had no air pockets. I pulled into the eddy immediately as he was comming to a stop under there and he quickly saw me and grabbed the nose of my boat, pulling his face into the calm, clear, bubble free eddy. He made many attempts to pull his face out of the water, but with one possible exception, he never got closer than 2 inches. I don’t believe he got any air during that time.

There is footage in the video’s that show the tail of my boat as this was happening. I don’t know why the television show was talking about air pockets. There were none because I was there instantly and he immediately used the nose I gave him to pull his face away from there as he fought hard to reach for the air.

Lee Belknap

I have removed “possibly with air pockets”, along with the National Geographic reference that speculated as such. Thank you.

for non-professionals in a rescue scenario, no-one is keeping time and time becomes a blur. No-one realizes they’ve past the point that survival is likely until shear exhaustion (kept at bay by adrenaline & concentration) finally kicks in (or until they look back afterwards). Guidelines are great for professionals, and for helping put things in perspective for those involved in an unsuccessful rescue, but won’t likely change how boaters strive to help another in trouble.

I’ll try to do this delicately. I’m a professional firefighter at a busy water rescue station with plenty of swiftwater in the area. I am a shift paratrainer for ice water rescue and I would like to acquire any photos and video footage of this or any other incident for training and education purposes. This rescue clearly went as well as it could have, so first hand accounts from rescuers who were present and prepared to discuss it would be great. I will keep this in house and use it for training purposes only. Perhaps knowledge gained from this tragedy may help save a life in the future. Post here if you’ve got something and we’ll figure out how to get in touch.

This is my account of the rescue written the day after. I have a few photos I would share for training purposes with a rescue agency. Verification required. They are copyrighted and I forbid their publication on line or in print.

My son is one of the canoeists that was with Dr. Senior on the Little River, and I was present on the riverbank shorly after the rescue effort started. Dr. Senior came out of his boat on the first drop of “The Screaming Meanies”above the sinks rapid. He went over the second drop and became trapped by the rocks in the hole at the base of the drop. An expert, determined, and sustained rescue effort was mounted. After dispatching someone to drive into cell coverage to call 911, the canoeists, many trained in swift water rescue, began “live bait” attempts. Rescuers tethered by a line jumped into the hydraulic at the base of the drop in and attempted to free the doctor. He was held hard by the rocks, and several rescurers were washed back by the swift current before they could pull Steve free. A three line harness was rigged and rescures, held in the hydraulic, fought to exsustion to pull Steve free. As one man tired he was replace by another to continue the effort. When the local rescue squad arrived on the scene they, by protocal, demanded the recuers get out of the water. The rescuers refused and continued their fight. The local rescue squad quickly reconized the skill set of the rescuers and did not make any other desist demands as they moved to take over. One of the canoiest rescuers was able to get a line on Steve and he was pulled free. CPR was started by the rescuers and continued until the local rescue team could make their way to the riverside. CPR was continued as Steve was placed on a litter and carried to a waiting ambulance. The effort by the canoeists was extraordinary, determined and valiant, I do not call it heroic because it is what they expect of each other. They practice a dangerous sport, but they are skilled, and rely on each other. Each man on the rescue team gave their all.

Leave a Reply

Your email address will not be published. Required fields are marked *