Skip to main content


By December 1, 2020No Comments

Monitoring and improving indoor air quality in naturally ventilated classrooms in relation to COVID-19 contamination risk

Rationale: In confined indoor spaces(classrooms, workplaces, …), especially those with a high occupancy and a low ventilation rate, bio-effluents from occupants considerably affect the indoor environmental quality(IEQ). Especially for schools, there is an urgent need to understand the actual risk and to offer immediate enhancements. Although numerous studies looked at mechanically ventilated spaces, naturally ventilated spaces such as most classrooms are widely understudied. Also, the influences of different architectural typologies together with particular user activities of classrooms are mostly unknown. An immediate understanding and practical tools are needed.

Expired air is such a bio-effluent, well-known to contain amongst others CO2 and potentially infectious droplets/aerosol particles, sized from 0.01 to 100 µm. Coughing or sneezing produces a considerable number of particles (reaching distances up to 6 m, Leung et al. 2020), but also breathing and talking emits particles (Aasadi et al. 2019). Whilst particles >5 µm are known to deposit timely due to gravity, smaller particles will remain suspended for a longer period(up to 3 hours). Personal exposure to potentially infectious particles like COVID-19 consists of a close contact and a distant contact mode(Zhang et al., 2020). To mitigate closecontact risk, i.e. exposure to larger particles(> 5 µm) at short distances(< 1.Sm), social distance and protective masks have been introduced(Zhang et al; 2020). The smaller particles(< Sµm) on the other hand may cause a potential distant contact risk, the so-called airborne transmission. Adequate ventilation is a crucial mitigation measure to reduce the particles’ air concentration and the risk of airborne virus transmission. CO2 and particles are both expired and therefore CO2 is considered a valuable indicator of the potential indoor infectious particle load.

The vast majority of Flemish classrooms relies on natural ventilation(read: window opening). Data

on classroom IEQ in Flanders indicate that only 8% of the classrooms reaches an average indoor CO2-level below the recommended 800 ppm (Codex Welzijn op het Werk, Vlaams BiMi Besluit) during teaching hours (BiBa, 2010; Clean Air Low Energy 2012). Data underline that ventilation is worst in wintertime, when cold outdoor temperatures affect indoor comfort. The COV/0-19 pandemicand the cumulative scientific evidence of its aerosol transmission route (Morawska et al 2020, Morawska and Cao 2020}, urge us to provide effective, feasible and dedicated solutions to increase ventilation – especially in densely occupied indoor space, like classrooms. The resulting research questions are: (1) What is the actual IEQ in terms of ventilation during the COVID-19 pandemic in different classroom typologies with natural ventilation? (2) Which practically and economically attainable mitigation measures are most effective to optimize natural ventilation, avoiding excessive heating costs or reduced thermal comfort?

Objective: Guide school authorities in selecting the best-fit solution to improve natural ventilation of classrooms, thereby reducing the airborne transmission risk of COVID-19, by means of a practical toolbox with customized, low-cost and effective mitigation strategies. Assist schools in defining a “degree of urgency” for fresh air, based on IEQ data and the current pandemic situation of the school environment.

Specify the necessity and urgency of the study, its contribution to the intended solution and societal implementability/impact in the short to medium term for managing the societal impact of the COVID-19 pandemic.

Urgency and necessity: The last months saw comprehensive research into potential airborne transmission route of COVID-19 (Morawska et al. 2020, Morawska and Cao 2020). After WHO’s confirmation of this risk (07/20), national COVID-19 strategic cells adopted new prevention guidelines such as the use of protective masks in enclosed indoor environments, as well as increased ventilation rates. Consequently, guidance has been requested in how to increase ventilation (1) when limited or no natural ventilation facilities are available (2) whilst respecting indoor thermal comfort (3) using limited budget. Governmental initiatives guide schools in their search for a suitable sensor, monitor classrooms and manage data (htt ps:/ / www.zorg-en­ covid-19-en-de-binnenluchtkwa liteit ). However, quantifying indoor CO2 in classrooms requires tailored tools to mitigate solutions for schools, also reflecting the particular typology of their rooms. Today, no such tools are availa ble. Contribution to the solution: (1) By characterizingthe ventilation efficiency and the indoor comfort in naturally ventilated classrooms using low-cost sensor technologies, data and insights on the current IEQ under pandemic conditions will be provided. Problematic situations or conditions will be identified; (2) By connectingroom typologies in school buildings and indoor air flows using CFO modelling, we will increase the general understandingof the built school environment and its specific aeration facilities and uses hereof, including occupant behavior. (3) On the basis of 15 case studies, mitigation measures for an increased natural classroom ventilation will be demonstrated and quantified, comprising interventions on the building envelope, interior transfer openings and the classroom’s layout and its use to allow quantification of the effectiveness as well as ranking the measures’ effectiveness, impact and cost. (4) A toolboxwill be developed from the results with feasible and cost-effective measures to increase ventilation in classrooms during wintertime, respecting the ventilation needs, comfort as well as the typology of a specific classroom. Societal: A set of schools will be used reflecting representative typologies of the Flemish school building st ock. Measures will thus affect ca 1,1M pupils and ca 200K staff. Given the urgency, the toolbox shall be distributed during the study. Once validated, the proposed measures can be rolled out outside Flanders and provide useful guidance for comparable naturally ventilated classrooms in similar climate zones; daycare centers, health care sector and offices may also benefit.

Describe the added value to the current-state-of-the-art and the complementarity with relevant national and international initiatives.

Added value: COVID-19 airborne transmission is a multidisciplinary field but has been mostly coordinated by medical sciences. This practical study will contribute eminent knowledge by connecting building design, air movement and ventilation as well as indoor air quality and aerosol sciences, using the expertise from UA, UG hent and VITO. 

Complementarity:This project adds to a set of projects coordinated by VITO IAQ, on classroom IEQ, ventilation and communication to schools and school building designers, under the authority of the Flemish Government as well as the EU, such as Digitale Tool voor scholen en Technische fiches voor scholenbouwers (dept. Omgeving, 2020), Partnerorganisatie Milieu en Gezondheid, team BiMi (Agentschap Zorg&Gezondheid, 2017-2000), lndoor@Box (dept. Omgeving 2019), Renovair (dept. Omgeving, 2016), Clean Air Low Energy (dept. Omgeving 2012), Sinphonie (DG Sanco, 2013).

Research methodology and work plan.

Step 1: Preparatory phase (VITO). A set of 15 low-cost CO 2/ RH/ T sensors, also connected, at their request, to the online platform owned by department Omgeving of the Flemish Government, for real-time data-transmission and sharing, is developed and calibrated; 5 of which equipped with PMx sensor for insights on the impact of the intervention on indoor PM. All sensors initially “blind” (no display). Risk: sensors will not be delivered in time ordertimely; back-up: sensors from VITO. 

Step 2: School selection phase and predictive modelling (UA, UGhent, VITO}: Selection of 15 classrooms according to pre-defined typologies, allowing interventions on natural ventilation facilities . This will be based on walk-through data (already being collected) as well as

predictive modelling. Risk: school turns to ‘red’ and locks down definition of back-up classes.

— Milestone 1: target schools selected, sensors prepared

Step 3: Case studies in schools (UA, VITO, UGhent): Case studies in occupied classrooms in Flanders where different natural ventilation possibilities are demonstrated and quantified.

Experimental design:

The main independent variable: type of natural ventilation

Type 1: one sided ventilation, door closed; Type 2A: one sided ventilation, door open; Type 2B: transversal ventilation, door open; Type 3: stack ventilation, door open.

The dependent variable consists of the accumulated indoor CO 2 level (per student, per m2 ) during teaching hours. Analysis will consider the difference between initial and final CO 2 as well as its time profile (slope and decay) to assess the efficiency of the remedial ventilation initiative.

Control variables: Occupant number/Classroom volume/Surface of ventilation opening/ Temperature, indoor, outdoor/Humidity/Windpressure

Hypothesis: Ventilation type will differentially impact the CO 2 increase above and beyond the control variables, magnitude of the impact of dooropeningwill depend on the type of ventilation A priori analysis using GPower3.1, and an expected medium size (Cohen’s d 0,5), indicates that a total sample size of n = 352 is needed to be able to note a difference between 4groups with a power of 95%. 1) experiment in 15 classrooms; each classroom 4 manipulations will be repeated 6 times totaling 24 sessions each (24*15=360, exceeds the desired sample size. Replicated up to 6 times, each time using 15 different classrooms.

Risk: school turns to ‘red’ and locks down identify back-up projects; sensors or network

malfunctioning intervention plan/ strategy ready. Milestone: desired sample size achieved. 

Step 4: Ranking the interventions (UA, VITO, UGhent): identify most effective remedial interventions and design toolbox for immediate use in function of possible ventilation typology and usage, as well as the pre-defined “degree of urgency” for fresh air (first draft 12/20 for general use, updates end 03/21 and end 06/21, final 08/21).

Step 5: Reporting and communication (VITO, UA, UGhent): the toolbox and its format are evaluated by the participating schools, feedback collected. Communication is organized via the consortium’s networks, as well as participating schools. VITO assures to communicate outcomes to all leading policy actors, such asAgentschap Zorg en Gezondheid, dept.Omgeving, dept. Onderwijs, as well as educational networks (Go!, Stedelijk Onderwijs, Agion etc.). Participating schools may keep their sensor unit at the end of the project, redesigned with an indicator LED light.

Risk: toolbox instructions are too technical for school authorities evaluation by participating schools. M ilestone : feedback from participating schools implemented

Step 6: Extrapolation. Extrapolation to other sites: record and keep all data and indicate how other typologies of spaces may profit from the data.

COST PLAN: See input at e-loket
Describe shortly the scientific competences, experience and resources and/or achievements on socio-economic utilization of research findings.

VITO Health Unit, Team Indoor Air Quality is specialized in IAO/IEQassessments in various indoor environments, significant experience in classroom environments (IAQdata of ca 200 Flemish classrooms). Policy support as well as communication to building professionals, develops high­ level as well as low-costsensortechnologies, Flemish reference lab for air.

Ghent University, Building Physics group has an international reputation in design and control of ventilation in buildings with publications on performance metrics, optimization of design, demand control and large scale monitoring studies. It is currently leading the IEA EBC annex 86 on smart IAQ management

University Antwerp, Faculty of Design Sciences (UAFDS) connects architectural and technological design; strong focus on building and functional typologies research including schools. Henry van de VeIde Research Group provides interdisciplinary design experts in school design from conception to construction including renovation.

Test population: KITOSvzw, a school group (Mechelen area), will provide the case study objects from 09/20 to 06/21. They will also communicate throughout the educational field in Flanders.


Inter-Agency Standing Committee Unicef, WHO, IFRC, March 2020: GUIDANCE FORCOVID-19 PREVENTION AND CONTROL IN SCHOOLS March 2020, Unicef, 2020

L. Morawskaa, Airborne transmission of SARS-CoV-2: The world should face the reality, 2020

The influence of contaminants in ambient air on the indoor air quality – part 1: exposure of children- interpretation and policy recommendations, part 3: Katleen De Brouwere, et al., VITO, 2007,

PM AND CO2 variability and relationship in different school environments, I. LAZOVIC et al, 2015 ASHRAE Position Document on Infectious Aerosols, ASHRAE, April 14, 2020

Viruses and Indoor Air Pollution, R.B. COUCH, 1981

Airborne spread of infectious agents in the indoor environment, J.Wei et al, 2016

Internet ofThings and Enhanced Living Environments: Measuring and Mapping Air Quality Using Cyber-physical Systems and Mobile Computing Technologies, Gonc;:alo Marques et al.,January 2020

Review and Extension of CO2-Based Methods to Determine Ventilation Rates with Application to School Classrooms, Stuart Batterman, Environmental Health Sciences, School of Public Health, University of Michigan, Ann Arbor, Academic Editors: Alessandra Cincinelli and Tania Martellini, 2017

Indoor air quality assessment in naturally ventilated school buildings: Carrer Pet al, 1994 Vandyck, Frederik and Inge Bertels. “Typology & Mixity: An Approach to Retrofit Production in the City?”, Acta technica Napocensis, 2018.

Leave a Reply