It is well documented that heat-stress burdens sheep welfare and productivity. Peak heat-stress levels are observed when high temperatures prevail, i.e. during heat waves; however, continuous measurements inside livestock buildings are not usually available for long periods so as to study the variation of summer heat-stress levels for several years, especially during extreme hot weather. Α methodology to develop a long time series of summer temperature and relative humidity inside naturally ventilated sheep barns is proposed. The accuracy and the transferability of the developed linear regression models were verified. Temperature Humidity Index (THI) was used to assess sheep's potential heat-stress. Τhe variation of THI inside a barn during heat wave and non-heat wave days was examined, and the results were comparatively assessed. The analysis showed that sheep were exposed to moderate, severe, and extreme severe heat-stress in 10, 21 and 66 % of hours, respectively, during heat wave days, while the corresponding values during non-heat wave days were 14, 33 and 43 %, respectively. The heat load on sheep was much higher during heat wave events than during non-heat wave periods. Additionally, based on the averaged diurnal variation of THI, it was concluded that extreme severe heat-stress conditions were prevailing between 1000 and 2400 hours local time during heat wave days. Cool off night periods were never and extremely rarely detected during heat wave and non-heat wave days, respectively.
Keywords: Heat wave; Heat-stress; Regression model; Sheep; Sheep barn; Temperature Humidity Index.