People on this site frequently ask about how much water they should drink while doing strenuous outdoor activities such as hiking and running. How broad is the safety range for water consumption in order to avoid both dehydration and hyponatremia?

2 Answers 2


The range of safety is extremely wide. It's not dangerous to get thirsty, nor do you increase your risk of heat stroke by drinking less water. Drinking very large amounts of water causes a danger of hyponatremia, but only if you distrust your body's sensations and force yourself to drink vast amounts of water because someone told you that was a good idea.

A lot of the science on this topic was done by a South African researcher and runner named Tim Noakes. In the 1970's, he pushed to convince long-distance runners to stop ignoring signs of thirst and to drink during races when they felt thirsty (which had been against the rules in marathons). As an overreaction set in, he then tried to convince people not to overdo it, which can lead to hyponatremia.

The evidence from marathon races is that drinking less water does not increase post-race body temperatures, so it probably doesn't increase your risk of heat stroke (Noakes et al., "The danger of an inadequate water intake during prolonged exercise," European Journal of Applied Physiology and Occupational Physiology 57 (1988) 210). It is not true, as often claimed in folk wisdom, that "thirst is too late," so that by the time you feel thirsty you're already "dehydrated." Actually, you will feel extremely thirsty long before you become dehydrated according to any medical definition. It's true that heat stroke is insidious and potentially fatal, but drinking more water doesn't prevent it.

At the other end of the spectrum, there is a great deal of evidence that drinking too much water can lead to hyponatremia, which is potentially fatal. However, the amount of water you have to drink to kill yourself this way is an incredible amount. In one of the best documented cases, a military recruit was told by his drill instructors to drink a huge amount of water because of very hot temperatures (Garigan, T. P., & Ristedt, D. E. (1999). Death from Hyponatremia as a Result of Acute Water Intoxication in an Army Basic Trainee. Military Medicine, 164(3), 234–238. doi:10.1093/milmed/164.3.234 ). But this unfortunate person drank something like 20 liters of water over the course of one morning. Although people do not have a built-in mental ability to recognize hyponatremia, as they do with dehydration, he had severe symptoms indicating a medical problem: "He appeared pale, complained of severe thirst[!], and drank three more quarts. Then he vomited for the first time. [Later,] the soldier complained of dizziness, "throbbing" headache, and nausea."

In summary, the medical evidence seems to be that it's virtually impossible to kill yourself by drinking too little or too much water during a day of physical activity. However, low-level hyponatremia, which is not life-threatening, is quite common among endurance athletes (Almond et al., Hyponatremia among Runners in the Boston Marathon, 2005, NEJM, DOI: 10.1056/NEJMoa043901 ). Among marathon runners, the condition was found to be especially common among less experienced runners who drank more than 3 liters of water and tried to load themselves up with fluids.

  • Maybe off topic, but: does this imply that the idiocy we hear ad nauseam from Oprah and other highly regarded medical specialists that you MUST drink eight 8-oz glasses of water a day is nonsense?
    – ab2
    Commented Dec 28, 2020 at 19:05
  • Great answer! May I add a few bits and pieces? On a short time-scale like 20 l water in a few hours, no, we don't have a hyponatremia alarm. On a more normal/slower drinking time scale we do have sensation for lack of salt though: salt hunger/craving/appetite as reaction to low Na⁺ is well-documented. In addition, I recently came across some findings that suggest that humans can store some Na⁺ independently of water: ncbi.nlm.nih.gov/pmc/articles/PMC4932095 (not sure though whether/how much of this is physiologically normal - they discuss correlationg with various disesases) Commented Dec 29, 2020 at 11:13
  • At a guess (based on what seems to make sense from evolution) I'd expect the salt craving (+ buffer capacity) to work OK on water intake rates up to maybe what our sweating rate is. That would be something like 1 - 2 l/h. Another review with interesting bits and pieces is IMHO dx.doi.org/10.1111/sms.12408 (I summarized a few points in my answer on heat adaptation and thirst outdoors.stackexchange.com/a/24306/2404). They cite some studies finding that heat exhaustion (which is a mechanism to prevent heat stroke) happens at ≈38.5 °C. Putting this together with Noakes may... Commented Dec 29, 2020 at 11:17
  • mean that lower water intake doesn't lead to more heat stroke since the prevention mechanism against heat stroke works as usual - instead, reduced ability to sweat may be limiting the mechanical power output. (I don't have access to the Noakes paper, though). Another interesting point in the heat adaptation review is that heat adaptation also leads to volountary water intake better matching the loss due to sweating, volountary intake was lower than loss before adaptation. All in all, I'm happy to trust my sensations wrt. water + salt requirements. But I'd recommend to not "overwrite" them: Commented Dec 29, 2020 at 11:29
  • apparently one can get into dangerous conditions that way (see the hyponatremic soldier, heat stroke during exercise also happens, and be careful with drugs/medication that may mess around with these). Last but not least some friends tell me they know their personal thirst sensation is slow/weak: they have to consciously think of drinking or they'll end up with the kind of headache that can be cured by 1 - 2 l of water by the end of the day's tour. Also, personal needs seem to vary a lot from what I see. Commented Dec 29, 2020 at 11:37

Practical checks:

Watch your pee, both frequency and colour.

If your pee is clear, and you are peeing often, you are at no risk of dehydration. May be drinking a bit too much.

If your pee is clear to pale yellow, and you peeing at reasonable interval, this is optimum.

If you pee is dark yellow, strong smell, then you should be drinking more.

Background: High elevation requires more breaths to get enough oxygen. Your mouth, sinuses and upper respiratory tract have to bring cold dry air up to 90+% relative humidity at body temperature. This takes a fair amount of water. (About 1/4 of your total metabolic expenditure in cold weather is used to evaporate water from your lungs.)

On winter dog sled treks we found that people routinely weren't drinking enough because we just didn't make it available. We would make up the deficit in the evening around the fire with many cups of tea and hot chocolate.

Water in camp was made from melting snow in the same buckets we used for cooking. This resulted in water that at best tasted like bucket, and at worst tasted burnt.

The solution was to bring 1 gallon thermoses. We would fill these with boiling snowmelt water and juice crystals (ok, flavoured sugar) breaking camp in the morning. Each sled carried one of these (1 sled per 2-3 people) At breaks you would fill your cup with tightly packed snow and add enough of the hot faux juice to melt it all. By the end of the day the juice was cool enough to drink directly.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.