Evaporative coolers in dry heat, and when they stop working
A swamp cooler's limit is set by moisture in the air, not by temperature, which is why June in Phoenix cools better than March, and August is the worst month of the year.
Short answer
An evaporative cooler cools by evaporating water into the air, so what limits it is how much moisture the air can still take up, not how hot it is. In Phoenix the typical June afternoon has a 67.8 °F gap between the air temperature and the dew point, and the typical August afternoon only 48.5 °F, so the monsoon is the hard part of the year.
The sentence “swamp coolers do not work in Phoenix” is one of the most repeated claims about living here, and it is half true in a way that gets the mechanism wrong. An evaporative cooler does work in Phoenix (well, for part of the year) and the part of the year it fails in is not the part people usually name.
The confusion comes from treating the machine as a weak version of an air conditioner, which works harder as the outside air gets hotter. An evaporative cooler is not doing the same job by a worse method. It is using a different physical effect with a different limit, and the limit is the moisture in the air rather than its temperature.
The claim: an evaporative cooler is limited by how hot it is outside
The reason this is wrong is worth stating plainly. A direct evaporative cooler pulls outside air across wet pads. Water evaporates into that air, and evaporation takes energy out of the air, so the air that arrives in the room is cooler than the air that went in. How much cooler is set by how much water the air can still take up, which falls as the air gets more humid and is unaffected by the air being dry and extremely hot.
So the honest quantity to look at is the gap between the air temperature and the dew point. The dew point is the temperature at which the air is holding all the moisture it can, and the wet bulb (the coolest the air could be made by evaporating water into it) always sits between the dew point and the air temperature. A wide gap means room to work. A narrow gap means very little, however hot it is.
That is the figure below, and it is drawn for 4 p.m., the hour when the load is heaviest.
The typical 4 p.m. temperature and dew point, by month
The shaded gap between the two lines is the room evaporation has to work in. The wider it is, the more cooling a wet pad can deliver; the figures under the months are that gap, in Fahrenheit degrees.
Typical 4 p.m. temperature and 4 p.m. dew point, NOAA NCEI U.S. Climate Normals 1991–2020, hourly summaries, Phoenix Sky Harbor (USW00023183). The solid line is the air temperature and the dashed line the dew point; both are published normals for the same hour of the same month. The gap between them, printed under each month, is this publication's subtraction: the wet bulb lies somewhere between the two, and no wet bulb is computed here.
The claim: a cooler works worse the hotter it gets
The chart settles this one by putting two months side by side that most people would not think to compare.
In a normal March the 4 p.m. temperature is 76.4 °F (pleasant) and the 4 p.m. dew point is 29.7 °F, a gap of 46.7 °F. In a normal June the temperature is 102.6 °F, twenty-six degrees hotter, and the dew point is 34.8 °F. The gap is 67.8 °F, the widest of any month in the year.
June is the best evaporative-cooling month in Phoenix and it is also nearly the hottest. March is comfortable and it is one of the worst. The temperature is not what the machine responds to.
| Month | 4 p.m. temperature | 4 p.m. dew point | Gap |
|---|---|---|---|
| March | 76.4 °F | 29.7 °F | 46.7 °F |
| April | 83.8 °F | 26.8 °F | 57.0 °F |
| May | 92.9 °F | 30.9 °F | 62.0 °F |
| June | 102.6 °F | 34.8 °F | 67.8 °F |
| July | 104.8 °F | 51.6 °F | 53.2 °F |
| August | 103.3 °F | 54.8 °F | 48.5 °F |
| September | 98.6 °F | 48.2 °F | 50.4 °F |
| October | 87.9 °F | 37.4 °F | 50.5 °F |
The claim: August is the second-hottest month, so it is the second-best
August is the second-hottest month and the worst summer month for an evaporative cooler at the same time, and the table shows why. The temperature is within a degree of June’s, but the dew point has climbed twenty degrees, from 34.8 °F in June to 54.8 °F in August. That is the monsoon, and it is the whole story of the year for anyone running one of these machines.
The counting version of the same fact is the overnight one. The normal overnight low stays at or above 80 °F from 20 June to 11 September: 84 days. Those are the nights when the outside air is not carrying much of a cooling job for anything, and they coincide with the months when the air is also holding the most moisture.
The good news, and it is genuine, is that the monsoon is a season rather than a condition. The gap widens again to 50.5 °F in October and 57.0 °F in April, and the normal high in October is 89.2 °F, which is hot enough for the machine to be useful and dry enough for it to work.
The claim: a cooler and an air conditioner are interchangeable
They are not, and the difference shows up in the shape of the year rather than in a specification. A refrigerated system’s capacity is essentially fixed: it removes a certain amount of heat per hour, and it does so regardless of the humidity outside. An evaporative cooler’s output rises and falls with the weather by a large factor, which is why the same house can be comfortable in June and uncomfortable in August without anything having changed except the dew point.
The practical consequence is that a household comparing the two is not choosing between a good machine and a worse one. It is choosing between a machine with a flat output and a machine with a lumpy one, and the lumpy one is at its best exactly when the Valley’s peak load is at its worst. That is also why the Peak-Hours Habit Planner is worth looking at alongside this: a cooler’s heaviest afternoon is the grid’s heaviest afternoon.
What these numbers do not say
Three limits are worth stating. First, the two lines on the chart are two normals paired by the hour and month they are published for, not a single observed reading; a real afternoon in June is one particular temperature with one particular dew point, and the averages do not capture the spread around them. Second, the gap under each month is a subtraction this publication performs, not a wet-bulb temperature; the wet bulb sits between the two lines and this page does not calculate it.
Third, and most importantly for anyone shopping: this page carries no efficiency rating for any evaporative cooler. When this page was written, this publication could not open a current U.S. Department of Energy or ENERGY STAR page that publishes one, so there is no figure here to quote, and inventing a plausible one would defeat the point of the page. What is published is the climate, and the climate is enough to know when the machine has something to work with.
The maintenance side
An evaporative cooler’s wear items are not an air conditioner’s. There is no refrigerant and no compressor; there are pads that load up with the Valley’s mineral-heavy water, a water distribution system that scales, and a reservoir that collects whatever the water leaves behind. Those are the parts that decide whether the machine delivers what the physics allows, and they are the reason a cooler that worked acceptably last June can be disappointing this one.
That is a maintenance schedule rather than a repair, and the Home Cooling Task Scheduler exists to put pad and service dates on a calendar alongside the filter and coil dates a refrigerated system needs. If the question is why the evening is so hard on any cooling equipment at all, Keeping a house cool at 115 degrees takes the building itself, and A cooling system maintenance calendar takes the rest of the year.
- NOAA National Centers for Environmental Information: U.S. Climate Normals 1991–2020, hourly summaries, Phoenix Sky Harbor (USW00023183)
The typical 4 p.m. temperature and the typical 4 p.m. dew point for every month on this page, and therefore the gap between them that the chart draws.
- NOAA National Centers for Environmental Information: U.S. Climate Normals 1991–2020, monthly summaries, Phoenix Sky Harbor (USW00023183)
The normal monthly high for October, 89.2 °F, the only monthly normal quoted here. Every other temperature on this page is an hourly normal.
- NOAA National Centers for Environmental Information: U.S. Climate Normals 1991–2020, daily summaries, Phoenix Sky Harbor (USW00023183)
The 20 June to 11 September window, the stretch on which the normal overnight low stays at or above 80 °F, which is the period an evaporative cooler has least to work with at night.