Where a pool's heat actually goes

The one that dominates: evaporation

Water leaving the surface as vapour carries about 1,050 BTU with every pound. Nothing else on the list moves heat that efficiently. A pool does not need to be hot for this to happen — it needs the vapour pressure at the water surface to be higher than the vapour pressure in the air above it, which on most days it is, whatever the air temperature is doing.

That is why an 84 °F pool in 84 °F air still loses heat rapidly. There is no temperature difference at all, so convection is zero, and the pool is still shedding energy as fast as the air can carry vapour away. Pool owners who think of heating in terms of "the air is warmer than the water so it should be fine" are reasoning about the wrong mechanism.

The rate used here is the Carrier correlation for evaporation from a water surface, in the form that appears in the natatorium literature:

evaporative loss (Btu/h) = area × (95 + 0.425 × air velocity in ft/min) × (P_water − P_air)

with the two pressures in inches of mercury. The saturation pressure at the water surface comes from the Buck equation; the air-side pressure is that same function at the air temperature, multiplied by the relative humidity. Note what the wind term does: at 3 miles per hour, the bracket is 207 instead of 95. Wind alone roughly doubles the loss.

The gallons number is the one people believe

Heat loss in BTU per hour is an abstraction. Water disappearing from the pool is not. At the default conditions above the pool is evaporating 281.2 gallons a day, which is where the 102,676 BTU per hour of evaporative loss is physically going. If your autofill runs and you have no leak, that volume is your heat bill made visible.

The conversion is direct: divide the evaporative heat loss by the latent heat of vaporisation near pool temperature, about 1,050 BTU per pound, to get pounds of water an hour, then divide by 8.35 pounds per gallon. Nothing else is involved. A pool losing ten gallons a day to evaporation is losing roughly 88 thousand BTU a day whether or not a heater is switched on.

Convection: small, and sometimes negative

Air moving across warmer water carries away sensible heat. The film coefficient used here is derived from the Bowen ratio rather than assumed, which ties it to the same wind term the evaporation equation uses:

h = 0.010138 × (95 + 0.425 × air velocity in ft/min), in Btu per hour per square foot per °F

Deriving it this way avoids a trap. Evaluating the Bowen ratio directly means dividing by the vapour pressure difference, which approaches zero on a humid night and sends the answer to infinity. Substituting the Carrier expression cancels that term out before the division ever happens.

At the default conditions convection is 16,025 BTU per hour, about 12% of the total. When the air is warmer than the water this term goes negative — the air is heating the pool — and the model carries the sign rather than clamping it, because on a Phoenix afternoon that is genuinely what is happening.

Radiation: the term nobody expects

A clear night sky behaves like a surface at roughly 41 °F when the air is 69 °F. Your pool, at 84 °F, radiates to it. The clear-sky radiant temperature comes from the Swinbank correlation, and cloud cover pushes it back up toward air temperature.

At the default conditions this is 19,304 BTU per hour — 14% of the loss, and larger than convection. It is also why a pool cools faster on a clear night than on an overcast one at the same air temperature, an effect most pool owners have noticed without having a name for it. Cloud fraction here is not asserted; it is derived per location and month from the ratio of measured all-sky to clear-sky surface irradiance.

Sunlight, the one term working in your favour

At 7.28 kilowatthours per square metre per day, the default pool collects 1,004,042 BTU a day, or 41,835 BTU per hour averaged around the clock. Compare that to 138,005 BTU per hour of loss and you can see immediately why an unheated pool in June sits comfortably at swimming temperature and the same pool in April does not.

This is also the answer to why solar pool heating works at all, and why it stops working exactly when you most want it: the resource and the demand move in opposite directions across the shoulder months. The solar page turns that into a required collector area rather than a rule of thumb.