Daily mortality data from the period 1968-1993 (ca. 2,3 mill. cases) from south-west Germany have been investigated with regard to possible impacts of the atmospheric conditions of heat exchange. Differently from the approaches in most preliminary studies which simply are using air temperature in order to cover the thermal conditions, a state-of-the-art complete heat budget model of the human being (Klima-Michel-model) is applied. The parameter for the thermo-physiologically significant assessment of the atmospheric environment is a so-called perceived temperature PT which takes all mechanisms of heat exchange into account with due consideration of well adapted clothing. Thus air temperature, water vapour pressure, wind velocity and the short and long-wave radiant fluxes are the meteorological input variables.
During comfort conditions, when the demands on the thermo-regulatory system of the body are minimal, mortality data show the lowest rates. With increasing cold stress death rates increase as well, predominantly due to seasonality. This seasonal behaviour is assumed being mainly caused by a higher probability for infections rather than by the direct effect of cold stress. However, it is remarkable that during the seasonal mortality minimum in summer, death rates sharply increase with increasing heat load, reaching the highest values during pronounced heat waves even in the moderate climate of SW Germany.
The relationships between thermal loads and mortality are closer in summer than in winter. The time series of the mortality data show a 2 to 3 day phase shift towards the atmospheric data. Atmospheric changes towards "warmer" conditions (in terms of PT) result in adverse effects while changes into "colder" conditions provide relief. Cold spells, however, lead to excess mortality which lasts for weeks. On the contrary, the excess mortality during heat waves is compensated by lower than average values in the subsequent weeks which means a "harvesting" effect, i.e. a shift of the death event of risk people in advance by some days or weeks. It is assumed that below the level of the extreme events of death, the thermal conditions are significant for morbidity, well-being, and efficiency of the human being as well.