Estimating ventilation rates in school classrooms from time series CO2 measurements

This study evaluates the applicability of two analytical (steady-state) and one numerical (transient) method for estimating ventilation airflow rates based on time-series CO ₂ measurements. Over the course of a year, CO ₂ concentrations were monitored in two Austrian school classrooms-one with natural ventilation and one with a decentralised mechanical ventilation system. The results show that in mechanically ventilated environments with stable CO ₂ levels and continuous ventilation, all three methods produce comparable results. However, in cases with intermittent ventilation events, such as naturally ventilated classrooms or mechanically ventilated classrooms supplemented by window ventilation during warmer weather, significant differences were observed between methods. The steady-state method using daily mean CO ₂ concentrations yields ventilation rates up to 35% higher than the steady-state method based on the 95th percentile. The transient model produces similar estimates to the steady-state method using daily mean concentrations but captures short-term fluctuations more effectively. The findings indicate that steady-state methods based on the 95th percentile effectively estimate baseline ventilation rates provided by air handling units but fail to reflect dynamic air exchange variations. In contrast, transient and steady-state methods using daily mean CO ₂ concentrations offer a more accurate assessment of user-driven ventilation behaviour. Further work is needed to improve the numerical limitations in the transient solution including the identification of transient occupancy changes.