Comparison of different methods of estimating the mean radiant temperature in outdoor thermal comfort studies

Correlations between outdoor thermal indices and the calculated or measured mean radiant temperature T_mrt are in general of high importance because of the combined effect on human energy balance in outdoor spaces. The most accurate way to determine T_mrt is by means of integral radiation measurements, i.e. measuring the short- and long-wave radiation from six directions using pyranometers and pyrgeometers, an expensive and not always an easily available procedure. Some studies use globe thermometers combined with air temperature and wind speed sensors. An alternative way to determine T_mrt is based on output from the RayMan model from measured data of incoming global radiation and morphological features of the monitoring site in particular sky view factor (SVF) data. The purpose of this paper is to compare different methods to assess the mean radiant temperature T_mrt in terms of differences to a reference condition (T_mrt calculated from field measurements) and to resulting outdoor comfort levels expressed as PET and UTCI values. The T_mrt obtained from field measurements is a combination of air temperature, wind speed and globe temperature data according to the forced ventilation formula of ISO 7726 for data collected in Glasgow, UK. Four different methods were used in the RayMan model for T_mrt calculations: input data consisting exclusively of data measured at urban sites; urban data excluding solar radiation, estimated SVF data and solar radiation data measured at a rural site; urban data excluding solar radiation with SVF data for each site; urban data excluding solar radiation and including solar radiation at the rural site taking no account of SVF information. Results show that all methods overestimate T_mrt when compared to ISO calculations. Correlations were found to be significant for the first method and lower for the other three. Results in terms of comfort (PET, UTCI) suggest that reasonable estimates could be made based on global radiation data measured at the urban site or as a surrogate of missing SR data or globe temperature data recorded at the urban area on global radiation data measured at a rural location.