During the past three decades, indoor air quality (IAQ) has received increased research attention, and a wide range of indoor air pollutants, such as volatile organic compounds (VOCs), particulate matter, inorganic compounds, and radon, have been assessed and characterized in various indoor environments: Offices, schools, supermarkets, houses, and so on [1
]. Over the years, comprehensive knowledge on the mechanisms of, and health effects from, exposure to indoor pollutants has been reviewed and asthma, allergies, and heart attack are highly correlated with indoor air pollution [1
]. Recently, indoor environmental quality (IEQ) has received increasing attention from the public under the context of global climate change and the green building initiative. Aspects of IEQ that directly influence the comfort, health, productivity, and satisfaction of occupants of a building include ventilation, IAQ, acoustic and thermal conditions, and illumination.
Several studies have examined IEQ in homes and K-12 schools with young kids as research objects [17
]. While children and youth represent a potentially vulnerable population, adult students in universities are a large prospective cohort whose institutional environment should be explored. In 2017, there were 1.7 million students in Canadian universities and around 20 million students at universities in the United States [26
]. Unfortunately, in these institutions, classroom IEQ, especially in large lecture classrooms, are scarcely explored. Despite this lack of attention, the effects of deficient IEQ, such as low ventilation, hot or cool conditions, dry or humid conditions, too noisy or too quiet background, as well as too bright or too dark lighting, may adversely affect student and academic staff performance and attendance [28
Excluding comprehensive walkthrough investigations and measurements of physical and/or chemical indoor elements, some IEQ-related research collected occupants’ responses to questions such as health issues, exposure data, and sensations, to a built environment through a questionnaire survey and provided acceptance or satisfaction criteria and health implications of a building’s IEQ [28
]. Several attempts have been made to conduct on-site standardized academic or physical tests, such as cognitive tests, spirometry tests, memory tests, and mathematics and reading tests, to evaluate how IEQ elements affect learning performance in elementary and middle school classrooms or productivity in offices [33
]. Since ventilation is a critical parameter for indoor environmental performance, most research used experimental, modeling or field-testing methods to examine natural ventilation or natural ventilation combined with a simple fan system to observe the impacts of outdoor environmental parameters on IEQ [25
]. Mechanical systems are another prevailing ventilation method, especially in non-mild climate zones or noisy or polluted regions. However, limited research has explored the interactions between IEQ parameters and how mechanical systems and occupant behavior affect IEQ. Although inadequate mechanical ventilation leading to poor indoor air quality was reported in some papers [2
], there were no detailed characteristics of the employed mechanical ventilation systems used in their studies. ASHRAE Guideline 10-2016 is the most up-to-date document to describe the IEQ interactions for achieving acceptable indoor environments [48
]. The current study will provide new scientific evidence to promote the new version of the guideline in the near future.
In order to update professional knowledge on the design and construction of healthy, sustainable, and energy-efficient IEQ for institutional buildings, this study was conducted with three objectives: (1) Providing evidence to define IEQ performance metrics for lecture classrooms in a very cold region, (2) further exploring interactions of IEQ parameters in order to update currently available knowledge of interactions from ASHRAE Guideline 10-2011, and (3) quantifying the impacts of mechanical systems on IEQ for the development of advanced IEQ design guides for institutional buildings. This paper reports on IEQ elements repeatedly measured in four large lecture classrooms over two seasons in a conventional institutional building in a cold climate and provides statistical analysis results to reveal the science associated with the interactions of IEQ parameters and factors impacting on IEQ.
4.1. Within- and Between-Classroom Comparisons
The variance analysis indicated that the within-lecture classroom variations in most IEQ parameters exceeded between-lecture classrooms (Table 3
). In particular, indoor CO2
, sound, illuminance, and pressure had large within-classroom variations. Within-room variation results from multiple factors: Outdoor environment, HVAC system operation, occupant behavior (number, voice frequency, learning activity), use of indoor facilities (microphone and light system). Low between-classroom variations are attributed to the mixed mechanical system of the central AHU with heating units and similar indoor design. Occupant number difference (latent heat is given off by humans) could be a reason for the slightly higher between-classroom variation in RH than the within-classroom one since only the central AHU had a humidification function. Temperature and RH had almost equal within- and between-classroom variations indicating that the central HVAC system, combined with VAV terminal reheat boxes, provided constant conditioned air to the classrooms, resistant to influence from fluctuations from other indoor and outdoor factors across the heating season. When the between-classroom variance is small, it means that the IEQ evaluation from the selected lecture classrooms can represent the IEQ characterization within the building. For all IEQ elements, the differences between classrooms were statistically significant. Hence, classroom characteristics, internal classroom operation (manual light levels and microphone voice volumes), the VAV system, and occupant behavior play a key role in determining indoor IEQ.
4.2. Seasonal Comparisons
Most indoor and some outdoor parameters varied by season. For example, outdoor temperatures at study classrooms averaged −14.2 ± 5.8 °C in winter, statistically lower than the average temperature in fall by both ANOVA and K-W tests (Table 4
). Outdoor RH, wind speed, and wind direction had no statistically significant seasonal differences. Except for acoustics and the number of total persons, all other observed IEQ elements had statistical differences, and the effect sizes of such differences were large (Cohen’s d > 0.5). The below-average ACR in winter compared to fall led to an increase in indoor CO2
concentration in winter. In addition, although the winter heating load was larger than the heat requirement in fall, the mean indoor temperature was higher in winter (23.7 ± 1.1 °C) than in fall (23.1 ± 0.8 °C), indicating that, to save energy, some operation should be adjusted for the thermostats or VAV heat boxes in cold weather.
4.3. Interactions Between Indoor Environmental Quality Parameters
Since HVAC systems work as a bridge to link outdoor and indoor microclimates, IEQ was highly correlated to outdoor environments (Figure 6
). For example, indoor RH increased with outdoor temperature (r = 0.336) as humidity ratio (g moisture/kg dry air) increased with outdoor temperature across two seasons. Outdoor temperature was inversely correlated with outdoor pressure due to the higher density of air at the lower temperature (ideal gas law), while indoor pressure was highly dependent on the outdoor pressure (r = 0.995, 0.50 ± 0.06 kPa higher indoors). Thus, the outdoor temperature was negatively correlated with indoor pressure. ACR was linearly correlated with outdoor temperature, possibly due to the energy-saving strategy discussed in Section 4.4
, leading to higher indoor CO2
levels at colder days.
Indoor temperature was positively and significantly correlated with total occupants (r = 0.403), indoor sound (r = 0.313), ACR (r = 0.320), and indoor light (r = 0.283). Occupants and equipment, i.e., light system and sound amplification system, are the internal thermal sources that contribute to indoor temperature. This observation was in line with Yang’s finding that the effects of room temperature were significant on the room brightness [60
]. Light level was negatively and correlated with RH (r = −0.506) and occupancy number (r = −0.284) because water vapor tended to condense on outer lens of LED lights at high humidity surroundings [61
] and occupants were light absorption source, which degraded optical performance. This study has revealed that light performance interacted with its surrounding thermal conditions, which supported the other researchers’ finding that light affects human thermal perception also [62
]. The stiffness of the acoustic wall panels was temperature-dependent, leading to the lower sound level at lower indoor temperature. In addition, the attenuation of sound in air was affected by RH. Moist air was less dense at a higher temperature as holding more water vapor, thus dry air at low temperature absorbed far more acoustical energy than did moist air at high temperature. Thus, sound passed through hot air easier than through cold air. Different from indoor temperature, indoor RH showed positive correlations with classroom volume and outdoor temperature and was negative to indoor light levels. The ACRs determined for the classrooms were significantly associated with CO2
concentration (r = −0.740). Table S1 in the Supplementary Material
shows all the results of the Spearman tests. It is interesting to note that the complicated correlation of IEQ parameters found here was based on the analysis of local outdoor weather in conjunction with HVAC operations in the building, constrained by indoor operations. For example, the brightness setting of light systems varied significantly.
MANOVA was implemented to indicate that the interaction of outdoor environments (outdoor temperature (p
= 0.000) and RH (p
= 0.000)), indoor behaviors (ACR (p
= 0.000) and illuminance (p
= 0.012)) has significant impacts on a linear combination of the indoor dependent variables (indoor temperature, indoor RH, and indoor CO2
) (Table S2
). Indoor sound levels did not appear to have a significant effect on indoor conditions. In order to further explore the complicated inter-relationships and importance of independent variables, the ANN model was trained 10 times to generate mean and standard deviation of relative errors and correlation coefficients for three indoor dependent variables (indoor temperature, indoor RH, and indoor CO2
) (Table S3
). The trained ANN model had high correlations (R2
= 0.469–0.928) between measured and predicted three variables. Hence, the established ANN model was able to provide the relative importance of five independent predictors (outdoor RH, outdoor temperature, occupant number, daylight lux, and ACR) in estimating three dependent variables. The normalized importance chart (Figure 7
) from 10 ANN model tests shows that the indoor conditions were dominated by the ACR performed (99.0% ± 2.1%), followed by indoor operations (light system (85.1% ± 10.1%), and occupant numbers (79.2% ± 11.9%)), followed distantly by outdoor conditions. Therefore, the HVAC system operation, classroom conditions, and building location play critical roles in determining classroom IEQ.
4.4. Ventilation and Air Change Rate (ACR)
In all cases, ACRs were in the range of 1.3 to 6.5 per hour (4 ± 1 per hour) during lecture periods when the building’s HVAC systems commonly operated. The linear regression model in Figure 8
a presents the increasing trend of outdoor ventilation rates with outdoor temperatures across all classrooms. Assuming return airflow rates remained the same across seasons, it was highly possible that outdoor air intake rates decreased as outdoor temperature decreased. This engineering ventilation strategy was developed for energy saving with consideration of a significant amount of energy required for air conditioning during extremely cold days. Figure 8
b shows that the mean and median ACRs of R1-017 and R1-013 were higher than those of R1-003 and R1-007. ASHRAE 62.1-2016 recommends a minimum outdoor air rate (accounting for both people- and area-related sources) for lecture classrooms of 8 cubic feet per minute (cfm) (4.3 L·s−1
) per person. The calculated outdoor air rates were 15 ± 6 cfm (7 ± 3 L·s−1
) per person with default occupant density 65 persons/100 m2
defined by ASHRAE 62.1 [55
], indicating that the adequacy of the ventilation system in the ETLC building was compliant with the standard. If based on actual average occupant density across all measurements, outdoor air rates were 47 ± 15 cfm (22 ± 7 L·s−1
) per person, much exceeding the recommended ASHRAE values. Hence, smart controls for building systems should be implemented. For example, if occupant sensors are sensitive to the occupant number, they then have the ability to adjust outdoor air intake rates according to the real number of persons in a room to save energy costs.
4.5. Limitations and Future Study
This study repeatedly characterized IEQ in four lecture classrooms housed in the ETLC building for the fall and winter seasons in the climate zone seven. Our results may not apply to other institutional buildings that are older, naturally ventilated, located in other climatic regimes, or accommodated with different HVAC settings. In addition, each lecture classroom was assumed to be a single well-mixed zone with measurements in a single location. Repeated measurements across seasons and random measurement location selection somehow offset the spatial variations. The CO2-derived ACR was not verified by other methods, like the tracer gas CS6 method.
Other indoor environmental parameters, such as volatile organic compounds, particulate matter, reverberation time, sound uniformity, and sound intelligibility, were not measured in phase I but will be included in phase II. Some building features, such as building envelop heat resistance values and exposure areas, may need to be involved in the development of the input units of the ANN model to better predict indoor temperature performance. While the data collected appear sufficient to characterize IEQ in lecture classrooms, future studies might use long-term measurements (workdays), repeated measurements at different times in each season, and shift samples between classes that better characterize IEQ in institutional buildings. Moreover, the unoccupied period is of less interest to IEQ research purposes. However, the quantification of IEQ changes during HVAC shut-off periods would provide more evidence to optimize the HVAC control for energy-saving purposes. These topics will be the focus of further studies.
Due to the amount of time that young adults and faculty members spend in lecture classrooms, they are vital environments in postsecondary education institutions. To date, however, characterizations of IEQ in these institutions have been limited, and studies comparing IEQ in institutional buildings are not available. Impacts of the institutional environment on the academic performance of students and energy costs to create healthy IEQ motivated this study to better characterize environmental conditions.
This study examines IEQ parameters, including pressure, illuminance level, acoustics, CO2 levels, temperature, and humidity, with appropriate monitors allocated during a lecture (duration 50 min or 80 min) in four lecture classrooms 1676 ± 44 m3 repeatedly from October 2018 to March 2019. Outdoor temperature was −25 °C to 12 °C during two-season measurements, while the classroom environment was maintained at 23 ± 1 °C and 33% ± 3% RH. Through a central AHU and VAV terminal boxes, indoor CO2 had mean concentrations of 550–1055 ppm, indicating the CO2 concentrations were mostly acceptable. Mean sound and illuminance levels were 58 ± 3 dBA and 235 ± 112 lux, respectively, which imply acoustic and visual comfort was created. Except for RH, within-lecture classroom variations of IEQ elements during the two seasons exceeded between-lecture classrooms. Using the one-zone steady-state box model, the ACRs were configured at 1.3–6.5 per hour based on continuous CO2 measurements and occupant loads in the lectures. The actual outdoor air intake rates, 47 ± 15 cfm (22 ± 7 L·s−1) per person, were much higher than the recommended value (8 cfm (4.3 L·s−1) per person) for lecture classrooms by ASHRAE 62. High-sensitivity occupant sensors would be an option in conjunction with a dynamic CO2 control sequence to adjust outdoor air intake rates for energy-saving purposes. The results of the IEQ investigation in this study updates the literature by putting the spotlight on the correlations of the building and the HVAC operations, as well as the effects of the outdoor environments on IEQ.