The Climate Chip database contains meteorological temperature readings and calculated heat stress indicators (WBGT and UTCI). There has been significant research on how to model the wet bulb globe temperature (WBGT), with the agreed upon approach being to iteratively relax WBGT estimates to within a suitable degree of accuracy. Lemke and Kjellstrom published a comparison of various models, with the conclusion that outdoor WBGT is best estimated using the method published by Liljegren et al(1), and indoor WBGT using Bernard’s published model.(2) Below I will discuss the model used by Matthias to calculate the WBGT values in the climate chip database.
The Climate Chip database has historical climate data going back 30 years. This historical data may be used to depict global and local trends in climate change and more importantly, the Climate Chip group aim to predict future trends based on the historical data. Predicting future trends for outdoor WBGT is notoriously difficult as the two biggest variables, cloud cover and wind speed, are difficult to accurately predict. As a result, a number of critical assumptions or approximations must be made:
Predicting outdoor WBGT:
- Define a constant level for solar radiation (reflects level of cloud cover).
- Define a constant wind speed.
For the purposes of the Climate Chip project, the solar radiation is set to 0 and the wind speed is set to 1 (assume worker movement approximate to a slow walk). In essence, this calculates the indoor WBGT. Lemke and Kjellstrom compared the accuracy of Liljegren’s model (outdoor WBGT, solar radiation = 0, wind speed = 1) to Bernard’s model (indoor WBGT) and found that Liljegren’s model consistently underestimated WBGT under these conditions. Thus, they settled on Bernard’s model. Simplification of Bernard’s model (where Tg = Ta and wind speed = 1) reduced the calculation on WBGT to:
And this is the model used in the C# script I inherited from Matthias.(3)
So in summary; my initial thoughts were that the Liljegren model had been used, but it wasn’t immediately obvious that the calculation in the C# script and the model published by Liljegren were equivalent. In fact they are not, the C# calculation is based on Bernard’s model with the assumptions that there is no solar radiation, a constant wind speed and Tg == Ta.
Thank you very much to Bruno Lemke for providing the background to their work, and clarifying all of the above.
(1) Liljegren, J. C., Carhart, R. A., Lawday, P., Tschopp, S., & Sharp, R. (2008). Modeling the wet bulb globe temperature using standard meteorological measurements. Journal of Occupational and Environmental Hygiene, 5(10), 645-655.
(2) Bernard, T. E. (1999). Prediction of workplace wet bulb global temperature. Applied occupational and environmental hygiene, 14(2), 126-134.
(3) Lemke, B., & Kjellstrom, T. (2012). Calculating workplace WBGT from meteorological data: a tool for climate change assessment. Industrial Health, 50(4), 267-278.