What is RSPEC!
Fact Sheets
Energy Smart Pools Software
RSPEC Presentation
Manufacturer Lists
Links to Other Related Sites

Measurement and Analysis of Evaporation from an Inactive Outdoor Swimming Pool
Randy Jones
U.S. Department of Energy
Denver Regional Office
1617 Cole Boulevard
Golden, CO 80401
Charles C. Smith
Solar Energy Applications Lab
Colorado State University
Fort Collins, CO 80523
George Löf
Solar Energy Applications Lab
Colorado State University
Fort Collins, CO 80523


Evaporation rates and total energy loads from an unoccupied, heated outdoor pool in Fort Collins, CO were investigated. Pool and air temperatures, humidity, thermal radiation, wind speed, and water loss due to evaporation were measured over twenty one test periods ranging from 1.1 to 16.2 hours during August and September of 1992. Data were analyzed and compared to commonly used evaporation rate equations, most notably that used in the ASHRAE Applications Handbook.


Measured evaporation was 72 percent of the ASHRAE calculated value with near zero wind velocity, and 84 percent of the ASHRAE value at five MPH wind. A modified version of the ASHRAE equation was developed as shown below:
Two overnight tests showed energy loss of 56 percent by evaporation, 26 percent by radiation, and 18 percent by convection. A correlation between radiation loss and temperatures was also found for the range of test conditions.


There are over 5.9 million heated swimming pools and spas in the U.S.1, consuming billions of dollars of energy annually. Because of this significant energy use, the U.S. Department of Energy (DOE) has launched a nationwide campaign to Reduce Swimming Pool Energy Costs, or RSPEC. Market ready energy efficiency and renewable energy products such as pool covers, solar hot water systems, and windbreaks will be supported through information and technology transfer to pool owners.

Pool owners must have reliable information about the cost effectiveness of potential energy efficiency investments. This requires accurate engineering methods to calculate pool energy loads before and after measure implementation. Evaporation is the chief component of energy loss in pools, so its prediction is very important. Energy analysts rely on methods presented in the technical literature, such as the American Society of Heating, Refrigeration, and Air Conditioning Engineers (ASHRAE) handbooks. However, there is significant disagreement in the results of various evaporation rate equations when applied to swimming pools.

Evaporation rate calculation disparities are primarily due to a lack of good experimental results based on pool direct water loss measurement. Therefore, DOE has sponsored a series of tests to measure evaporation and total energy loads in swimming pools. Tests have been conducted by the Solar Energy Applications Lab at Colorado State University. The background, procedures, and results of tests on a quiet, outdoor pool follow.


Measurements have been conducted by several investigators under controlled laboratory conditions of evaporation from pans or small tanks inside of wind tunnels2,3. Measurements of evaporation from lakes and reservoirs have also been performed to estimate losses to irrigation users3,4. These offer some information applicable to pools. However, water body geometry's and surroundings differed from those of swimming pools, and they were unheated. In addition measurement accuracy's were limited.

The widely used ASHRAE Handbooks5 have, for many years, contained an evaporation rate equation based on evaporation experiments, primarily those of Carrier2 as shown below:


W - evaporation rate, Lb/Hr-Ft.2
V - air velocity over water surface, MPH
Pw - saturation vapor pressure at the water temperature, in. Hg
Pa - saturation vapor pressure at the air dewpoint, in. Hg
Y - Latent heat at pool temperature, Btu/lb

The ASHRAE evaporation equation for undisturbed and actively used pools has been questioned on the basis of German investigations6,7,8. These reports show condensate collection from air dehumidifiers as functions of vapor pressure difference between the pool water and the air over a pool (the same form as the ASHRAE equation). They indicated that the quiet pool evaporation rate was about one half of the calculated ASHRAE values. Two concerns arise however with this assessment; the condensate collection from the dehumidifiers may not represent the total evaporation from a pool, and there are different coefficients used with the equation. Two of the papers use a coefficient 21 percent higher than in the ASHRAE equation.

Evaporation tests were conducted at Colorado State University during April of 1992 on an inactive indoor pool9. Evaporation rates during controlled conditions for eleven hours and more were 74 percent of that predicted by ASHRAE.

Outdoor pool energy estimates are available from Florida10 and Switzerland11, however they are based on few measured results. These studies took the total input energy to pools and attempted to analyze how this total would be broken down using heat transfer theory.

Outdoor pools are subject to greater evaporation, radiation, and convection heat losses than occur with indoor pools. A major difference is due to wind or air velocity as observed from the (A+BxV) component of the ASHRAE/Carrier equation.

Rohwer and others have developed wind velocity coefficients (B) from laboratory measurements. Figure 1 illustrates graphically the projected wind effects upon evaporation taken from six references. Even considering the lowest wind coefficients (the lowest slopes of the lines in Figure 1), it is clear that evaporation is some multiple of the static air condition due to normal wind levels.

Figure 1
Fig. 1. Evaporation vs. Wind Speed From Non-Pool References


The CSU study attempted to provide, for the first time, an evaporation rate prediction method based on accurate measurement of direct water loss, vapor pressure differences, and wind velocity, in an outdoor swimming pool.

3.1 Test Site

A neighborhood association pool with 4125 square feet total surface area and 144,000 gallons volume was studied. Buildings, trees, and fences were set back twenty or more feet such that the pool was well exposed to wind and solar radiation. Also most outward radiation from the pool was to the sky.

The pool water was circulated by conventional means through a sand filter, chlorinator and gas fired boiler for heating. The pool is maintained at 84oF by a thermostat in the return water line from the pool. Natural gas billing records indicated energy input to the pool was approximately 8 million BTUs per day without covering, and 5.5 million BTUs per day when covered for about 12 hours over night.

3.2 Methods and Procedures

Air and water conditions were monitored at six-minute intervals by a data acquisition unit, supplied to a desk-top computer. The computer provided real time output of the predicted mass evaporation using the ASHRAE equation and the corresponding drop in pool level.

Pool water temperature and air temperatures were measured with T-type thermocouple welded junctions. Thermocouples agreed to within .5 oF with a precision scientific mercury-in-glass thermometer and within .2 oF with each other. This precision includes the electronic signal conditioning and was repeatable during the testing period. Pool water temperature was measured throughout the pool volume several times during testing. There was no variation in temperature.

Air humidity was determined by monitoring the dew point temperature with an EG&G Model 660 dew point hygrometer. This instrument was calibrated just prior to testing to +/- .2 oC against a secondary dew point temperature standard. The combined accuracy of air temperature and dew point temperature measurement translates to approximately +/- 1.5 percent relative humidity.

The evaporation mass flux from a water surface cannot be measured directly by practical means. It must be measured by the liquid volume loss during the test period under consideration. At typical conditions in outdoor pools, evaporation rates of about .1 lb. of water per hour per square foot of water surface may be expected. At this rate, the water level in a pool will decrease about .02 inch per hour. With suitable equipment, water levels can be measured to an accuracy of +/- .001 inch, so the measurement of water level over a four-hour period can yield an accuracy within about three percent. Water additions and discharges to the pool circuit were prevented during the test period.

The pool water level was determined by a microtector gauge rigidly mounted to the pool side. This gauge has a high precision adjustment and senses electrical contact with the water surface. The level measurements were observed inside of a stilling well submerged below the surface, designed to suppress wave motion.

Wind speed was obtained from a rotating cup anemometer located at the edge of the pool one foot above the water surface. Two radiation sensors; an Eppley solar pyranometer, and a net radiometer were located on an extension arm over the pool. Radiation losses were computed as the difference between net radiation and the incoming solar radiation.

3.3 Outdoor Pool Testing Conditions

There is no practical way of controlling the conditions over an outdoor pool such that a single parameter, such as the wind, could be investigated. Consequently, numerous observations were recorded for entry into a spreadsheet format, so that the trends of interest would appear. Obviously more observations improve the confidence level of the results from this approach.

There was a limit to the number of observations possible because evaporation water loss is slow relative to the measurement techniques available, and also because there were no more than fifteen days available for testing. Most of the testing was conducted at night when the pool was inactive and the atmospheric conditions were steadier than in daytime. Night testing also improved the accuracy of total energy budget measurements by avoiding the influence of solar gain.

Testing periods of three hours duration and longer were desired. Winds were never steady for three hours, however, and the resulting averages seldom exceeded two miles per hour. Since the wind term in previous evaporation equations was linear, it appeared that averages were sufficient for the purpose of obtaining a wind speed coefficient. Higher wind speed points would add confidence to the results, however. Thus a method of greater precision in water loss measurement was sought to reduce the time period requirements for testing.

3.4 Evaporation Pan Measurements

Floating evaporation pans were introduced midway into the testing program to permit short term evaporation measurement. Two shallow aluminum pans, eight inches in diameter, were maintained approximately ten feet from the pool side by line and anchor. The evaporative water loss from the pans could be accurately determined at the pool site, by transfer of the water to a graduated cylinder. Typical water losses of 50 milliliters or more per hour could be measured to +/- one milliliter; hence an accuracy of within two percent.

The thin aluminum material of the pans insured temperature equilibrium between the contents and the pool water. A least squares straight line comparison of the pan measurements to the water level measurements showed a slope of 1.04, which means that pan evaporation rates were four percent higher than those in the pool. The cause of the small difference was not known, however it was consistent enough that pan evaporation could be used in short period tests by applying the four percent correction.

3.5 Evaporation Results

Twenty one evaporation test conditions provided sufficiently reliable results. The test periods ranged in length from 1.1 to 16.2 hours; the results are presented as hourly values for consistency. Evaporation water loss during short time periods under high wind conditions were obtained from the two pan measurements.

Winds for the outdoor test ranged between 0.3 and 7.2 miles per hour. Because there was no practical means to control either the wind or water vapor pressure difference under outdoor conditions, the test points are multivariant with respect to these parameters. To present evaporation as a function of wind, the form of the evaporation equation can be rearranged as follows:


Figure 2 presents a straight line fit to the test data for water evaporation rate per unit of water vapor pressure difference vs the wind speed. Figure 2 also presents a comparison of the test results with the two most applicable evaporation references; ASHRAE/Carrier and Rohwer. The test results yield coefficients of "A" equal to .068 and "B" equal to .032; or in the ASHRAE form of the equation:




Also indicated in Figure 2 is the evaporation rate found in previous indoor pool tests9; .071 Lbs/Hr-Sq Ft-In Hg at .06 MPH "wind" speed. This is nearly identical to the .070 Lbs/Hr-Sq Ft-In Hg at .06 MPH from the outdoor test results.

Figure 3 presents the effect of wind velocity upon evaporation as determined by the above equation (V as MPH) for a set of lines of equal vapor pressure difference. Values of vapor pressure difference in the figure were selected arbitrarily. This form of presentation is intended to show the relative importance of wind in different climates or with hot water spas, for example.

Figure 2
Fig.2. Evaporation vs. Wind Speed Test Data Comparison with References

Figure 3
Fig. 3. Evaporation vs. Wind Speed At Selected Vapor Pressure Differences

3.6 Radiation Measurement

Long wave radiation emitted from the pool surface to the surroundings was monitored each six minutes by subtracting the short wave solar radiation from the net radiation. Most of the testing was conducted with near zero solar radiation. The field of view from the pool water surface was almost entirely sky. Thus the equation for radiation heat loss from the pool surface is:


The sky temperature (Tsky) ranges from about 10 oC below ambient for hot humid conditions to 30 oC below ambient in a dry cool climate12. For a specific climatic region, an approximate linear correlation exists with air temperature, thus:


For the range of temperature difference experienced in these tests (approximately 10-30 oF), the radiation heat loss fit the linear relationship:

Qr = 25.1+34 x (Tpool - Tair)

Data scatter suggested that this relationship was valid to within about ten percent. The site conditions were relatively low humidity with clear skies, and thus may represent higher radiation energy loss than average for the US.

3.7 Total Energy Loss

During two periods of continuous cool down, after the pool was closed for the season on September 8, the total energy loss was measured. Under these conditions, the combined losses of evaporation, radiation and convection were equal to the change in energy content of the pool water over the time period of observation. These measurements were made at night with no incoming solar radiation. Evaporation and radiation were determined according to the methods discussed earlier. Total energy loss was found from the change in pool temperature without heating. Convection loss was obtained as the residual difference.

Energy losses were 56 percent by evaporation, 26 percent by radiation, and 18 percent by convection.


The rate of evaporation from an unoccupied outdoor swimming pool is 16 to 28% lower than that predicted by the ASHRAE equation. In 21 closely monitored tests, at a pool temperature of 84 oF, air temperature of 58 to 82, and relative humidity of 27 to 65%, the evaporation rate was 76 percent of the ASHRAE value at no wind velocity, and 85 percent of the ASHRAE value at five miles per hour wind speed. A linear fit to the test data yields an equation in the ASHRAE form but with adjusted coefficients:


This relationship also conforms closely to results found earlier for the inactive indoor pool.

Radiation loss was approximately one-half as much as that due to evaporation. A correlation between radiation energy loss from the pool surface and temperature difference between the pool and air temperature was found for the range of conditions during testing. Heat loss by convection was approximately eighteen percent of total losses.


The success of this project required the technical and administrative assistance of several individuals. Among them are Mr. Randy Martin, and Ms. Sigrid Higdon of the Denver Regional Office of the US Department of Energy, Mr. Craig Christensen of the National Renewable Energy Laboratory, and Mr. R. Norman Orava of Associated Western Universities.


1 "Swimming Pool and Spa Industry Market Reports", National Spa and Pool Institute, 1987 and 1988.
2 Carrier, W. H. "The Temperature of Evaporation", ASHVE Transactions vol. 24, p. 25, (1918)
3 Rohwer, D. "Evaporation from Free Water Surfaces", Tech. Bulletin no. 271 US Dept. of Agriculture, 1931.
4 Meyer, Evaporation from Lakes and Reservoirs, Minnesota Resources Commission, June 1942.
5 1991 ASHRAE Handbook HVAC Applications, p.4.7
6 Biasin, Von K., and Krumme, W., "Evaporation in an Indoor Swimming Pool", Electrowarme International, p. a115-a129, May 1974 (Germany).
7 Reeker, J., "Water Evaporation in Indoor Swimming Pools", Klima & Kalte Ingenieur, no. 1, p. 29-32, January 1978 (Germany).
8 Labohm, G., "Heating and Air Conditioning of Swimming Pools", Gesundheits-Ingenieur, p. 72-80, March 1971 (Germany).
9 Smith, C. C. "Measurement and Analysis of Evaporation from an Inactive Indoor Swimming Pool", Report submitted to the U. S. Dept. Of Energy, Denver Regional Office, June 1992.
10 Root,D. "How to Determine the Heat Load of Swimming Pools", Solar Age, , November 1983.
11 Guisan, O., "Thermal Analysis of Five Outdoor Swimming Pools heated by Unglazed Collectors", (submitted August 1992 for publication in Solar Energy)
12 Bliss, R.W., "Atmospheric Radiation Near the Surface of the Ground" Solar Energy, Vol. 5, Nr. 103, 1961


Home | What is RSPEC! | Fact Sheets | Energy Smart Pools Software | RSPEC Presentation
Manufacturer Lists | Links to Other Related Sites

This web site is posted here as a service by R. L. Martin & Associates, Inc.. This is a former U.S. Department of Energy program that no longer has any funding so the information is provided here as is.  There is no one to answer any questions anymore, but the site still contains very useful information.