Water Intake Calculator
There is no official per-kilogram water formula — so this calculator does not invent one. It shows the NASEM 2005 Adequate Intake reference values for total daily water, the share that typically comes from drinks, and a clearly labeled rough addition for sweaty exercise, along with the two things every hydration page should say: thirst is a reasonable guide for most healthy adults, and overdrinking carries real risk.
Daily water reference
Reference values: men 3.7 L/day, women 2.7 L/day total water (NASEM 2005).
What the Adequate Intake actually says
The NASEM 2005 Dietary Reference Intakes set Adequate Intakes for total water — everything consumed, drinks plus the water already in food — at about 3.7 L/day for men and 2.7 L/day for women aged 19 and over. Those two figures are the whole of the published content on this page. Both are published for a population, and this page shows them to one reader without that changing what they describe: putting a figure in front of an individual does not make it a figure about that individual.
The distinction changes what the output is for. An equation such as Mifflin-St Jeor was fitted to measurements and returns an estimate that moves with every input you give it. A reference value moves with nothing, because it was never a function of anything you could type. No control on this form is a measurement of the reader: the sex selector chooses between the two published figures, the optional exercise duration leaves those figures untouched and prints a separately labelled band beside them, and the conditions checkbox adds a note and changes no number. Nothing on the page observes the reader.
The companion guide — protein and water: what the reference values actually say — is the longer read: how a reference value differs from a formula, why the protein RDA is a floor rather than a target, and the misreadings that quietly turn either one into something it is not.
Why there is no per-kilogram number here
The one thing this calculator cannot produce is a millilitres-per-kilogram figure. There is no official per-kilogram water requirement, so a rule of that shape has nothing behind it to be traced back to, and the calculator does not manufacture one. Manufacturing one would be easy: pick a plausible coefficient, multiply a body weight by it, print a personalised total to one decimal place. The result would look more specific without being more true, and the decimal place would be the misleading part — precision borrowed from arithmetic rather than from a published source.
The contrast with the neighbouring tool is exact. The protein intake calculator does multiply body weight by a per-kilogram figure, because per-kilogram protein figures were published and it can print which one it used next to the answer. The water calculator multiplies nothing by body weight, because there is nothing published to multiply. Same site, same engine, opposite answers to "can this be personalised?" — and the answer was decided by what exists in the literature, not by which output would look better.
The same reasoning disposes of the glasses-a-day maxim and its relatives. No body defined those rules, and there is no document to check them against. Leaving a gap where a personalised number would sit is the honest way to represent a gap in the sources.
Total water, and the share that comes from drinks
"Total water" is the load-bearing phrase in the reference value. It counts everything consumed, and roughly 20% of it typically arrives in food rather than in a glass. The engine derives the beverage share from the report's ~80% figure, so the totals above correspond to about 2.96 L/day from drinks for men and 2.16 L/day for women — computed by the same function the form calls, not retyped.
Printing both lines is the only way either is unambiguous, and the ambiguity is where a great many hydration claims quietly go wrong. The same reference value for the same person can be quoted as 2.7 L or 2.16 L depending on which quantity is meant, and nothing in the number itself tells a reader which one is in front of them. A total-water figure repeated as though it were a drinks target is the same number asked to mean something it was not defined to mean.
The reference value also draws no distinction between kinds of drink, so this page does not draw one either. Total water against the beverage share is the only split the calculator reports, and it is the only split the source supports.
Exercise: a labelled rule of thumb, kept separate
Sweat losses vary too much for a formula, so the tool does not pretend to one. The add-on band is built from typical sweat rates of roughly 0.5–2 L per hour, per the ACSM fluid-replacement position stand, multiplied by the hours entered and labelled as the rule of thumb it is. One hour of sweaty training therefore adds somewhere between 0.5 and 2 L to the day.
Set that band next to the value it would be added to. At 2 hours the addition runs 1–4 L, a spread of 3 L, against an Adequate Intake of 2.7 L/day for the same reference person. The uncertainty in the addition exceeds the entire quantity it is being added to. Folding the two into a single total would publish one number whose error bar is larger than the figure it started from, which is why the calculator keeps them on separate lines and names the second one a rule of thumb rather than a result.
The input is bounded at 6 hours a day, where the addition already spans 3–12 L. Past that the engine stops rather than extrapolating: asked for 7 hours it returns an explanation instead of a wider band — "Exercise (hours/day) must be between 0 and 6." Refusing an input is a more honest output than stretching a rule of thumb until it covers anything.
The hot-or-humid checkbox works on the same principle. Conditions are part of why the sweat-rate band is as wide as it is, but no multiplier for heat or humidity is published on this site, so the checkbox marks the case with a note and changes no number. A control that silently scaled the result would be inventing precisely what the rest of this page declines to invent.
Thirst, and the risk that runs the other way
Two statements belong on any hydration page, and they point in opposite directions. The first: thirst is a reasonable guide for most healthy adults. The figures above are published for a population, so reading either of them as a personal daily quota reads more into the number than the number contains.
The second: overdrinking during long exercise can cause dangerous exercise-associated hyponatremia. More fluid is not automatically safer, and the assumption that it is has a name and a risk attached to it.
Neither statement is an instruction about what to drink. Together they explain the shape of this tool's output — a published reference value, a beverage share, and a band that announces itself as a rule of thumb — instead of a target with a tick next to it. A plan that tells someone to override thirst and drink to a schedule is making a claim that no reference value on this page supports.
Who the figures describe, and what the page computes
The Adequate Intakes above are published for men and women aged 19 and over. That boundary is part of the reference value as stated rather than anything this page derives, and it is the limit on who the numbers describe at all. The sex control selects between two published figures and is not a model of a body; no weight, height, age, or health context enters the calculation, because none of them enter the reference value. A tool that measures nothing about a reader cannot produce a requirement for one, and this page does not claim to.
Every figure in the narrative above is either a constant declared in the engine — the beverage
share, the sweat-rate band, the 6-hour ceiling on the exercise input — or a
value the same waterIntake() function the form calls returned at build time, with the
refusal message quoted from the engine itself. If the copy and the code ever disagreed, the site
would fail to build rather than publish a stale sentence. When you use the form, everything runs
in your browser: nothing you enter is transmitted or stored.
Frequently asked questions
Is there an official "per kg" water formula?
No. A rule of the form "so many millilitres per kilogram of body weight", like the glasses-a-day maxim, has no reference value behind it to be traced back to — no major body publishes a per-kilogram water requirement for healthy adults. The citable anchor is the NASEM 2005 Adequate Intake for total water: about 3.7 L/day for men and 2.7 L/day for women aged 19 and over, including roughly 20% that arrives in food.
Why does the calculator not ask for my body weight?
Because there is no published per-kilogram water figure to multiply it by. The protein intake calculator does ask for weight, since per-kilogram protein figures exist and it prints which one it used. Asking for a weight here would imply a relationship the reference values do not contain.
Do I have to hit these numbers every day?
The figures are reference values published for a population, not targets this page computed for you, so they are not a quota to be met. Thirst is a reasonable guide for most healthy adults. Treat the figures as context for what was published, not as a daily score.
Do coffee, tea, and food count?
The Adequate Intake is stated for total water, which counts everything consumed — every beverage plus the water already in food, roughly 20% of the total. The reference value draws no distinction between kinds of drink, and neither does this calculator: the only split it reports is total water against the beverage share.
How much extra do I need for exercise?
It varies too much for one number: typical sweat rates run roughly 0.5–2 L per hour, per the ACSM fluid-replacement position stand. The calculator shows that band as a clearly labelled rule of thumb beside the published figure rather than folding it in, because a band that wide would swamp the value it was added to.
Why is the exercise input capped?
The engine accepts 0 to 6 hours a day and rejects anything outside that range with an explanation instead of an answer — ask it for 7 hours and it returns "Exercise (hours/day) must be between 0 and 6." A refusal is a more honest output than an extrapolation from a band that is already wide at one hour.
What does the hot-or-humid checkbox change?
It adds a note, not a number. Conditions are part of why the sweat-rate band is as wide as it is, but no figure on this site is published for heat or humidity, so the calculator applies none. A checkbox that quietly scaled the result would be inventing exactly what the rest of this page refuses to invent.
Can I drink too much water?
Yes, and it can be dangerous. Drinking far beyond thirst during long exercise can cause exercise-associated hyponatremia. More fluid is not automatically safer, which is why no output on this page tells anyone to drink past thirst on a schedule.
Not medical advice: this calculator reports published population reference values and a labelled rule of thumb. It cannot account for your individual health context and must not be used to diagnose or treat any condition — consult a qualified healthcare professional, particularly with illness, medications, pregnancy, or any medical condition, none of which the reference values on this page account for. Values are processed locally in your browser and never transmitted. See the methodology page.