Occupant window opening behaviour: the relative importance of temperature and carbon dioxide in university office buildings

Occupant window opening behaviour has become of increasing concern because of the role windows play in controlling ventilation and hence in building energy consumption. Previous studies (in different countries, climates, buildings and room types and over different observation periods) have shown a general trend that window opening is related to weather conditions, indoor temperature and some non-environmental conditions. However, seeking to reduce the amount of energy lost through ventilation and infiltration can be conflicting with the need to maintain high indoor air quality (IAQ) levels. Carbon dioxide (CO2) which is an occupant generated pollutant and also a good indicator of ventilation and IAQ is known to have negative physiological impacts on occupants. It is therefore important to seek to understand occupants’ interaction with building controls in response to changes in IAQ as well as changes in thermal conditions. The work described in this thesis is focussed on window opening behaviour of occupants in university buildings. The influence of indoor and outdoor environmental variables on window opening will be assessed. Also the influence of CO2, as an indicator of IAQ, on window opening will be considered. Field observations were conducted in two naturally ventilated office buildings over three different seasons. Window states, window state changing events and environmental data were measured during the survey periods. An experiment with controlled conditions was also conducted to investigate the influence of CO2 concentration in different thermal conditions. This was achieved by observing participants window use in environments with pre-set CO2 and temperature conditions. Results from these studies confirmed that window opening behaviour is heavily influenced by temperature. However, temperature alone did not explain all the variance in the observed behaviour. Differences were found in behaviour at different times of the day and in different seasons as different combinations of variables affected window opening at different times. From the experiment, it was found that perceived environment was also significantly associated with window opening. Based on the observations made in this study, models for window opening were generated for both indoor and outdoor temperature. These were compared with models from previous studies and it was shown that there is a range in the prediction of window opening. The comparison highlighted the disparities between the window opening models and questions the generalizability and reliability of the models, highlighting the need to consider the effects of a wider range of variables.