Why heat stress is measured with three thermometers
A body loses heat four ways: convection to the air, radiation to and from surrounding surfaces, evaporation of sweat, and conduction. Air temperature alone describes only the first. WBGT was designed to capture all the routes that matter outdoors and in industry, using three sensors weighted according to how much each mechanism contributes to heat strain.
The natural wet bulb carries 70% of the weight in both formulas. It is a thermometer with a wetted wick exposed to whatever air movement is present, so it falls when evaporation is easy and rises toward air temperature when humidity blocks it. Since evaporation is the body's dominant cooling route in hot conditions, humidity dominates the index — which is why WBGT often flags a humid 30 °C day as more dangerous than a dry 35 °C one.
The globe thermometer is a black sphere with a sensor at its centre. It absorbs radiant heat from the sun, from furnaces and from hot surfaces, so it reads well above air temperature in direct sun and near a heat source. Outdoors it carries 20% of the weight; indoors it carries 30%, absorbing the share the dry bulb gives up.
The dry bulb appears only in the outdoor formula, at 10%. Indoors, with no solar load, the globe reading already reflects the air's contribution well enough that the separate term is dropped.
Two things WBGT does not measure directly are how hard the work is and what the worker is wearing, and both change the answer more than the weather often does. That is why the index is never read alone: it is compared against a criterion selected for the metabolic workload class, after adding a clothing adjustment.
The criteria, the clothing factor and the acclimatisation split
ACGIH publishes screening criteria as a grid: rows for the work/rest allocation within each hour, columns for the metabolic workload class. Two grids exist — a TLV for acclimatised workers and a lower Action Limit for unacclimatised ones. Find the highest work allocation whose criterion your effective WBGT does not exceed, and that is the schedule screening supports.
The clothing adjustment factor is added to the measured WBGT before the comparison, because clothing that resists evaporation makes a given environment behave like a hotter one. Ordinary work clothes and cloth woven coveralls add nothing. SMS polypropylene coveralls add 0.5 °C, polyolefin coveralls 1 °C, double-layer woven clothing 3 °C, and limited-use vapour-barrier coveralls 11 °C. That last figure is not a refinement — an 11 °C adjustment moves almost any working environment past every criterion, which is ACGIH's way of saying that WBGT screening does not apply to encapsulating suits and that physiological monitoring must govern instead.
Acclimatisation is a real physiological adaptation, not a state of mind. Over roughly a week of progressive exposure, sweat rate rises, sweat becomes more dilute, plasma volume expands and heart rate at a given workload falls. It is also lost quickly: about a week away from the heat, including a holiday or a spell of cool weather, substantially reverses it. Because the unacclimatised criteria sit several degrees lower, the returning-worker and new-hire cases are where most serious heat illness happens.
Metabolic class is estimated from the task. Light work is around 180 W — standing or seated work using hands and arms. Moderate is around 300 W, sustained walking with moderate lifting. Heavy is around 415 W, and very heavy around 520 W: intense shovelling, sledgehammer work, climbing with load. Notice that very heavy work has no criterion at all for continuous or 75% work; screening does not support it in any warm environment.
Worked example: an outdoor crew at moderate workload
A site takes three readings at midday: natural wet bulb 25.0 °C, globe 35.0 °C, dry bulb 30.0 °C. The crew is outdoors in full sun, wearing ordinary long-sleeve work clothes, doing moderate work, and has been on site for three weeks in similar conditions, so they are acclimatised.
- Choose the formula. There is solar load, so use the outdoor form with all three terms.
- Weight the readings. 0.7 × 25.0 = 17.50. 0.2 × 35.0 = 7.00. 0.1 × 30.0 = 3.00.
- Add them. 17.50 + 7.00 + 3.00 = 27.5 °C-WBGT, which is 27.5 × 9 ÷ 5 + 32 = 81.5 °F.
- Clothing adjustment. Work clothes add 0, so the effective WBGT is also 27.5 °C.
- Compare against the acclimatised moderate criteria. Continuous work is permitted up to 27.5 °C, and 27.5 ≤ 27.5, so continuous work is supported with zero margin.
Zero margin is the point worth dwelling on. Now change one thing at a time and watch the schedule collapse.
Put the crew in double-layer clothing. Add 3 °C: effective WBGT becomes 30.5 °C. That is above the 27.5 continuous criterion, above 28.5 at 75% work and above 29.5 at 50% work, but within the 31.0 criterion at 25% work. The schedule drops from 60 minutes of work per hour to 15 minutes, from a clothing change alone.
Or bring in unacclimatised workers, leaving the clothing alone. The acclimatised grid gives way to the action limits: 25.0 continuous, 26.5 at 75%, 28.0 at 50%. At 27.5 °C the first two are exceeded and the third is met, so the allocation is 30 minutes of work and 30 minutes of rest per hour for exactly the same environment and the same task.
Or move the same job indoors beside a heat source with the same wet bulb and globe readings. The indoor formula drops the dry bulb term and raises the globe weight: 0.7 × 25.0 + 0.3 × 35.0 = 17.50 + 10.50 = 28.0 °C. Half a degree, and the acclimatised moderate schedule falls from continuous to 45 minutes of work per hour. The environment did not get much hotter; it stopped having a cooler air temperature to dilute the radiant load.
Using the schedule properly
A work/rest allocation only works if the rest is real. Rest means a genuinely cooler location — shade with air movement at minimum, an air-conditioned space where available — and it means the rest minutes are taken within the hour, not banked to the end of the shift. Ten minutes of sitting in the sun is not rest in the sense the criteria assume, because heat storage continues.
Treat the result as a screening decision with three possible outcomes. Continuous work supported means proceed with the ordinary controls: water, shade, rest breaks as needed, and supervision. A partial allocation means the schedule is now a control measure and someone has to enforce it. No allocation supported means screening has failed and you move to a detailed analysis — physiological monitoring of heart rate, core temperature or recovery heart rate on individuals, engineering controls, or rescheduling the work.
Watch for the factors screening assumes away. The criteria assume adequate hydration, workers who are not on medications that impair sweating or thermoregulation, no fever or recent illness, and normal body composition. They do not account for age, pregnancy, or cardiovascular conditions. An individual can suffer heat illness at a WBGT that the screening criteria clear, which is why heat illness prevention programmes rely on buddy systems and symptom recognition as well as on numbers.
Finally, do not confuse WBGT with the heat index. The heat index is a public-weather apparent-temperature scale computed from air temperature and relative humidity in the shade, and OSHA's public heat guidance has used it because it is available from any weather forecast. WBGT requires instruments and accounts for radiant load and air movement, which is why it is the occupational standard. The two are different scales and their numbers are not interchangeable — see the heat index calculator for the meteorological measure.
ACGIH screening criteria in °C-WBGT
| Work / rest each hour | Light (TLV / AL) | Moderate (TLV / AL) | Heavy (TLV / AL) | Very heavy (TLV / AL) |
|---|---|---|---|---|
| 100% work | 29.5 / 27.5 | 27.5 / 25.0 | 26.0 / 22.5 | — / — |
| 75% work, 25% rest | 30.5 / 29.0 | 28.5 / 26.5 | 27.5 / 24.5 | — / — |
| 50% work, 50% rest | 31.5 / 30.0 | 29.5 / 28.0 | 28.5 / 26.5 | 27.5 / 25.0 |
| 25% work, 75% rest | 32.5 / 31.0 | 31.0 / 29.0 | 30.0 / 28.0 | 29.5 / 26.5 |
Dashes mark combinations for which no screening criterion is published — very heavy work is not supported continuously or at 75% work in any warm environment. Compare your effective WBGT, after the clothing adjustment, against the column for your workload class and the appropriate acclimatisation state.
The three sensors, and what each is not
- Natural wet bulb
- A thermometer with a wetted cotton wick exposed to the prevailing air movement. It is not a psychrometric wet bulb, which is aspirated at a fixed air speed and reads lower in still air. Substituting one for the other understates WBGT.
- Globe temperature
- The centre of a 150 mm matte-black copper sphere at equilibrium. It integrates radiant heat from every direction, which is why it can read 15 °C above air temperature in direct sun.
- Dry bulb
- Shaded air temperature, shielded from radiation. Used only in the outdoor formula and only at 10% weight.
- Clothing adjustment factor
- Degrees added to the measured WBGT to account for clothing that impedes evaporative and convective heat loss. It adjusts the index, not the criterion.
- Acclimatisation
- The physiological adaptation to heat built over about a week of progressive exposure and largely lost after about a week away. Acclimatised workers are assessed against the TLV; unacclimatised workers against the lower Action Limit.
Mistakes that produce a dangerously low WBGT
- Using a psychrometric wet bulb. An aspirated wet bulb reads below a natural wet bulb in still air, and the natural wet bulb carries 70% of the weight.
- Using the indoor formula outdoors. With solar load present the dry bulb term belongs in the calculation; whether that raises or lowers the answer depends on whether the globe reads above or below the air temperature.
- Omitting the clothing adjustment. Double-layer clothing adds 3 °C, which in the worked example cuts the supported work time from 60 minutes an hour to 15.
- Assuming everyone is acclimatised. New hires, returning workers and the first hot spell of the season all face the lower action limits.
- Underestimating the workload class. Moving from moderate to heavy work lowers the acclimatised continuous criterion from 27.5 to 26.0 °C.
- Measuring in the wrong place. Take readings where the work is done, at working height, not at a weather station or in the site office.
- Treating rest breaks as optional once the allocation is set. A schedule that is not enforced provides no protection at all.
What is a standard here and what is not
The WBGT index itself is standardised: ISO 7243 specifies the instruments and the two formulas. The screening criteria and clothing adjustment factors used here are the ACGIH TLV for heat stress and heat strain, which is a consensus occupational exposure guideline published annually, not a regulation. In the United States there is no specific federal OSHA heat standard for general industry at the time of writing; heat hazards are addressed under the General Duty Clause, section 5(a)(1) of the OSH Act, and several states operate their own heat illness prevention standards with their own trigger temperatures and requirements. Check what applies in your jurisdiction, and use the edition of the TLV your programme cites.
Where WBGT fits in a heat illness prevention programme
WBGT sets the schedule; the rest of the programme makes the schedule survivable. Hydration is the largest single controllable factor, and sweat rates during heavy work in heat can far exceed thirst-driven drinking — the water intake calculator gives a baseline that heat and workload both add to substantially. Shade, air movement and a cool break area convert rest minutes into actual recovery.
On the measurement side, if you have air temperature and humidity but no globe thermometer, you cannot compute a true WBGT — but you can understand the humidity half of the problem. The relative humidity calculator and the dew point calculator both describe how much evaporative capacity the air has left, which is what the natural wet bulb is sensing.
Heat illness cases are recordable when they meet the general recording criteria, so a heat programme's effectiveness shows up directly in the OSHA incident rate. Severity matters more than frequency here: heat stroke is a medical emergency with long recovery times, so a single case can move a severity rate far more than several minor injuries would.
One practical note about seasonality. The most dangerous days of the year are usually not the hottest ones but the first warm spell, when nobody is acclimatised and the criteria in the lower half of the table apply. Build the acclimatisation schedule into the season's start rather than reacting to the first incident.