Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Measurement of conditions of human comfort

A convenient standard for thermal comfort is required. Analysis shows that a variety of factors can be involved in situations of discomfort. For example, temperature alone does not determine discomfort. In Athens, 32 °C (90 °F) is quite bearable, but it is generally intolerable in Bahrain. The difference is due entirely to the relative humidity of the atmosphere. In Bahrain the air is very humid and perspiration evaporates slowly, decreasing the body's ability to lose heat. In Athens, with its dry air, the evaporation rate is high and perspiration evaporates quickly lowering body temperature.

The factors that have been identified as standard for thermal comfort within buildings are: air temperature, air humidity, rate of air movement, level of radiation, and rate of heat production by the bodies of people in the building. Extensive studies have established representative physiological scales that take into account all of these variables. An index used in the United States, and which with one limitation appears to provide an adequate measure of environmental warmth, is effective temperature. This takes into account temperature, humidity, and airspeed, but not radiation. Introduced by Houghton and Yaglou, this measure of heat sensation is defined as the temperature of saturated motionless air that would produce the same sensation of heat or cold as the combination of temperature, humidity, and air motion under consideration. An improvement on this measurement by Vernon and Warner uses the temperature given by the globe thermometer instead of the dry-bulb air temperature and thus includes an approximation of the radiation component. This standard is known as the corrected effective temperature and is the most useful scale of thermal sensation now available for the Tropics.

The effective temperature scale is in fact a physiological temperature scale. To establish it, a large number of people were exposed to wide ranges of temperature, humidity, and airspeed, and their sensations recorded. Later it was determined that the physiologically objective reactions of the subjects, such as pulse and perspiration rates, were in agreement with this effective temperature scale. However, it must not be assumed that this scale can be indiscriminately applied throughout the world with equal accuracy. Its American originators were the first to point out the limitations imposed by the fact that the scale was established from experiments on American subjects wearing clothing of American style and material. To establish an accurate, effective temperature scale for, say, Pakistan, a complete investigation using Pakistani subjects and clothing would be necessary.

The physical parameters to be measured and the instruments needed are shown in table 4.

Measurements made using a globe thermometer include the heating effects of infrared radiation emitted by warm flooring, roofing, and walls. The dry-bulb thermometer of a whirling psychrometer permits a nearly accurate evaluation of the basic air temperature; its speed through the air is sufficient to eliminate radiation effects. The Kata thermometer is superior to the usual type of vane anemometer. It indicates the sum of the effects of variable draughts to which a vane anemometer is not sensitive but which are physiologically important.

Table 4. Parameters to be measured for establishing an effective temperature scale and the corresponding instruments required

It also records velocities lower than most anemometers, and it needs no calibration.

Table 5 gives some examples of effective temperatures for different combinations of air temperature, relative humidity, and airspeed. For optimal comfort in air-conditioned buildings, the recommended range of effective temperatures is 22.2-23.3 °C (72-74 °F), corresponding to drybulb temperatures of 25.6-26.7 °C (78-80 °F), at 50% relative humidity.

Such physiological scales are useful when comparing the relative comfort of different sites. It should be remembered, however, that buildings can reduce the free wind speed. Studies in London have shown that wind speed at street level is generally about one-third of the unimpeded wind speed.

To subjectively compare human reactions to various conditions of heat, humidity, and airspeed, several microclimatic comfort sensation scales have been established. An example of such a scale and instructions for its use are given in Appendix 2.

At the London School of Hygiene and Tropical Medicine, a group of 32 students were asked to record their sensations of comfort under precise air-temperature, humidity, and airspeed conditions. They included approximately equal numbers of students from Great Britain and the United States, and from tropical countries. A summary of the student responses at 22.2 °C (72 °F) dry-bulb temperature, 16.1 °C (61 °F) wetbulb temperature, 56% relative humidity, and 0.25-0.38 m/s (50-75 ft/min) airspeeds is given in table 6. Although this is a preliminary, and by no means conclusive, experiment with only a small number of subjects, it indicates some fundamental difference between people from tropical and temperate countries with regard to comfort sensation.

Table 5. Examples of effective temperatures for different combinations of air temperature, relative humidity, and airspeed

Note: All absolute temperatures are in °C (°F).

Table 6. Summary of the comfort sensation of two groups of students exposed to 22.2 °C (72 °F) dry-bulb temperature, 16.1 °C (61 °F) wet-bulb temperature, 56% relative humidity, and 0.25-0.28 m/s (50-75 ft/min) airspeeds

Table 7. The values for the ambient and most appreciated air-conditioning temperatures and humidities in four tropical cities

Note: All temperatures are in °C (°F).

Table 8. Comparison of outdoor and indoor temperature and humidity conditions provided by a continuous airspeed of 0.3 m/s (60 ft/min) over a wet surface

Note: All temperatures are in °C (°F).

Table 7 shows values for air-conditioning that were found to be generally favored by the occupants of buildings in tropical countries. The airspeed was taken to be 0.3 m/s (60 ft/min) in these effective temperature calculations.

Table 8 shows that it may not be necessary to use powered airconditioning, an expensive expedient in places where ambient conditions are hot and dry, as in Delhi or Lahore. The inside effective temperature can be reduced using only evaporation in such climates, merely by ensuring a continuous air speed of 0.3 m/s (60 ft/min) over a continuously wet surface. Thus a reduction in effective temperature of 2.3 C° (4.2 F°) can be achieved.

With this understanding of the physical principles affecting human comfort, it is now possible to examine the applications of scientific concepts to architectural design and town planning in hot arid regions.

Contents - Previous - Next