Room-level ventilation in schools and universities

Abstract

Ventilation is of primary concern for maintaining healthy indoor air quality and reducing the spread of airborne infectious disease, including COVID-19. In addition to building-level guidelines, increased attention is being placed on room-level ventilation. However, for many universities and schools, ventilation data on a room-by-room basis are not available for classrooms and other key spaces. We present an overview of approaches for measuring ventilation along with their advantages and disadvantages. We also present data from recent case studies for a variety of institutions across the United States, with various building ages, types, locations, and climates, highlighting their commonalities and differences, and examples of the use of this data to support decision making.

Keywords: HVAC; Indoor air; Schools; Sensors; Ventilation.

Fig. 1

Data from a controlled-release CO₂ experiment at North East University A. The decay is exponential as shown in the corrected concentration profile ($C-C_{background}$) on the left. ACH_T is calculated either from the e-folding time or via a linear fit to the log-linear plot. See the Results section for more details.

Fig. 2

Data from the South East University demonstrating the use of in situ data for calculation of ACH_T. (a) Room A CO₂ profile and exponential decay fit (in red) during class time. This room has mechanical ventilation (recirculating indoor air and outdoor air, with a MERV 13 filter between circulation cycles). ACH_T was calculated to be 2.3 h⁻¹. (b) Room B PM_2.5 profile and exponential decay fit (in red) during a fogging event. This classroom has mechanical ventilation and a portable air cleaner with a high-efficiency particulate air (HEPA) filter. ACH_T was calculated to be 2.8 h⁻¹. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 3

Summary of ACH_T as measured via CO₂ decay rate in controlled-release experiments in classrooms for a university in the Northeastern U.S. (North East University A) with a mix of mechanical and natural ventilation.

Fig. 4

Comparison between the controlled release CO₂ decay rate and balometer air change measurements for North East university A.

Fig. 5

The results of CO₂ decay measurements in Coastal California K-12 Schools under different room conditions. ACH_T is binned and color-coded following the ventilation categories of Jones et al. (2020). Dataset represents approximately 460 measurements in 50 classrooms (3 sensors per room, 3–4 measurement conditions per room). (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 6

(a) CO₂ decay measured in three different locations of a Coastal California classroom, under three conditions: closed doors and windows (black), open doors and windows (red) and open doors and windows plus HVAC (blue). The dashed line indicates ambient CO₂ level, measured separately. (b) Log-linear plot of data from panel (a); ACH determined from slopes obtained from linear regression fits (dashed). No difference appears between different locations within the room. (c) Room with non-exponential decay and positional lag. (d) Classroom with an ‘alcove’. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 7

Ventilation test results using CH₄ tracer and a Picarro GasScouter sensor for two different labs at Southern California University. Lab 1 is typical for spaces with isolated ventilation, but in Lab 2 we observed rapid dilution of the tracer (red line) into an adjacent room before stabilizing into a normal decay. Note that each test lasted <30 min total. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 8

Violin plot of ACH_T calculated from CH₄ tracer measurements in 279 rooms from 21 buildings across the Southern California University campus, binned by decade of last HVAC system replacement. Each vertical symbol represents the probability distribution function (PDF) of ACH values among rooms from buildings of that age class. Number of rooms in each age class is given at the top in red. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 9

Comparison of measured outside-air ventilation rates for 11 classrooms and offices in the Southern California Secondary School with windows closed, HVAC on (triangles) versus with windows open (circles).

Fig. 10

Summary of ventilation test results for 22 classrooms and offices of the Southern California Secondary School, binned by year of construction. All spaces were tested with windows open using the CH₄ controlled-release method.

Fig. 11

Passive CO₂ measurements showing relative fraction of time observed in each concentration range from the Mountain West University. (a) Data reported for four different types of classrooms, averaged over in-use periods during three academic terms of observation. (b) Nighttime data for music classrooms and practice rooms for Fall academic term comparing before (left columns) and after (right columns) building ventilation was optimized to run at night and on weekends. See the text for additional details of the averaging periods.

Fig. 12

CO₂ concentrations shown for a single ensemble music classroom at the Mountain West University before (a) and after (b) ventilation schedule optimized to run also at nights and on weekends. Mean value shown for two-week window displayed in each case, where bars show standard deviation of measurements.

Fig. 13

Samples collected from 6 rooms in 6 different buildings with either mechanical (Mech) or natural ventilation over the same 2 week period. Red dots are calculated from in situ CO₂ data during different classes in the same room. Blue triangles are from recorded HVAC air supply rates. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)

Fig. 14

Comparison of PM tracer decay using different filtration/ventilation approaches in a naturally ventilated library test space in North East University B. Left: The plan view indicates the location of two fog generators, the PM sampling locations (S1–S4), and the two supplemental filtration/ventilation devices that were individually evaluated: an air filtering ventilation unit and a commercially-available air cleaner. Right: Ventilation test results for various scenarios are shown with the measured ACH_T.

Fig. 15

Summary of ventilation measurements reported here. NE: North East, SE: South East, CoCA: Coastal California, SoCA: Southern California.