key: cord-0864484-0za41d9q
authors: Aguilar, Antonio J.; de la Hoz-Torres, María L.; Costa, Nélson; Arezes, Pedro; Martínez-Aires, María Dolores; Ruiz, Diego P.
title: Assessment of ventilation rates inside educational buildings in Southwestern Europe: Analysis of implemented strategic measures
date: 2022-02-17
journal: Journal of Building Engineering
DOI: 10.1016/j.jobe.2022.104204
sha: c0171218a046e7ca1f0861c679fd91fd0f344c62
doc_id: 864484
cord_uid: 0za41d9q

The pandemic caused by COVID-19 has highlighted the need to ensure good indoor air quality. Public buildings (educational buildings in particular) have come under the spotlight because students, teachers and staff spend long periods of the day indoors. This study presents a measurement campaign for the assessment of ventilation rate (VR) and ventilation strategies in educational buildings in Southwestern Europe, Portugal and Spain. A representative sample of the teaching spaces of the Azurém Campus (Guimarães, Portugal) and the Fuentenueva Campus (Granada, Spain) have been analyzed. Natural ventilation is the predominant ventilation strategy in these spaces, being the most common strategy in educational buildings in Europe. VR was estimated under different configurations, using the CO2 decay method. Subsequently, the CO2 concentration was estimated according to occupancy and the probability of infection risk was calculated using the Wells-Riley equation. The obtained VR varied between 2.9 and 20.1 air change per hour (ACH) for natural cross ventilation, 2.0 to 5.1 ACH for single-sided ventilation and 1.8 to 3.5 for mechanically ventilated classrooms. Large differences in CO2 concentrations were verified, depending on the analyzed ventilation strategy, ranging from 475 to 3903 ppm for the different scenarios. However, the probability of risk was less than 1% in almost all of the classrooms analyzed. The results obtained from the measurement campaign showed that the selection of an appropriate ventilation strategy can provide sufficient air renewal and maintain a low risk of infection. Ventilation strategies need to be reconsidered as a consequence of the health emergency arising from the COVID-19 pandemic.

Indoor Environmental Quality (IEQ), and in particular the Indoor Air Quality (IAQ), is a 28 crucial aspect to consider in the design of educational buildings. As these buildings are 29 often designed for high occupancy density for long periods of the day, the quality of the 30 indoor built environment is crucial to providing a healthy, safe and comfortable space [1] . 31

In addition, exposure to indoor air pollutants might exacerbate diseases, such as asthma, 32 or allergies [2] and can lead to a risk of short and long-term health problems, including 33 various respiratory diseases [3, 4] , cardiovascular diseases [5], irritated nose and/or eyes, 34 headaches, etc. [6] . Therefore, IAQ is an essential parameter for the well-being of students 35 and teachers, as it can have a direct impact on concentration, productivity and academic 36 achievement [7] . 37 However, previous studies have shown that educational spaces in non-retrofitted 38 buildings in Southern Europe do not have suitable conditions of comfort and IAQ [8] . 39 Moreover, even in those educational buildings that have been retrofitted, the effect of the 40 intervention showed some differences from what was expected at the design stage [8] . In 41 fact, the effects of poor IAQ in these spaces have recently been put in the spotlight due to 42 the global pandemic caused by COVID-19 since they are risk environments for the 43 transmission of airborne viruses such as the Severe Acute Respiratory Syndrome 44

Coronavirus-2 (SARS-CoV-2) [9] . This fact is a critical issue and has resulted in 45 increased concern among building managers about IAQ. Measures to contain the 46 transmission of SARS-CoV-2 constitute a major challenge inside enclosed environments 47 such as classrooms. An asymptomatic, infected teacher or student could spread a virus-48 containing aerosol inside classrooms if the air is not adequately renewed. Factors that 49 gas concentrations [22] are methods used to determine ventilation metrics (i.e. ACH (h -98 1 ), VR per person (dm 3 

Previous studies have used these methods to assess the probability of the airborne spread 100 of SARS-CoV-2 (COVID-19 virus) [24, 25] . Berry et al. [26] reviewed methods to reduce 101 the probability of the airborne spread of COVID-19 virus in ventilation systems and 102 enclosed spaces. They highlighted that ventilating an enclosed space is an effective way 103 to reduce the concentration of airborne particles carrying COVID-19. Li et al. [27] 104 concluded that there is evidence of an association between the transmission and spread of 105 infectious diseases (i.e. measles, Tuberculosis (TB), chickenpox, anthrax, influenza, 106 smallpox, and SARS) and the ventilation and the control of airflow direction in buildings. 107

In addition, Guo et al. [28] analyzed the operation guidelines from different countries 108 (including ASHRAE, REVHA, SHASE, the Architectural Society of China, and the 109 Chinese Institute of Refrigeration). This review concluded that all guidelines emphasize 110 the importance of ventilation (both natural and mechanical), which can effectively 111 decrease the concentration of virus-containing droplets. However, it is still unclear as to 112 the specific ventilation rate that can eliminate the risk of transmission of airborne 113 particulate matter. Pan et al. [29] concluded that Heating, Ventilation, and Air 114 Conditioning (HVAC) system design should be adaptive, in anticipation of the needs of 115 emerging situations, such as the pandemic. The supply of fresh air, in higher volumes, 116 should be considered in the design. Li Portugal) and the Fuentenueva Campus (Granada, Spain) were selected for the analysis. 136

The field measurements and subsequently data post-processing followed the following 137 phases: 1) study of the characteristics of the indoor spaces at the Azurem Campus and 138 Fuentenueva Campus; 2) definition and selection of different ventilation strategies in each 139 teaching space; 3) assessment of the VR using the decay method; 4) estimation of the 140 CO2 concentrations and identification of the probability of airborne virus infection risk 141 as a function of the VR obtained in the field measurements. 142

In order to characterize the ventilation strategies implemented in educational buildings in 144 without the gas source, the rate of decreased concentration of the tracer gas is determined 187 under a ventilation rate strategy. In this study, carbon dioxide (CO2) was used in the tests. 188

The decay method uses CO2 as a tracer gas, and it is based in the Equation (1) . From this 189 equation, the CO2 concentration C(t) in an effective mixed zone at time t is given by: Since the multi-point decay method was applied, the UNE-EN ISO 12569:2017 standard 206

[34] establishes the following procedure to calculate the level of confidence for the 207 estimated airflow rate ( ̅̅̅̅̅̅ ). The confidence intervals denoted as FACH for the estimated 208 ACH can be calculated for a level of confidence of 100(1 − α) as: 209

where t(k-2,1-∝) is the value calculated from the Student's t-distribution table; k being 211 the number of samples, 1-∝ is the confidence level of ACH (set up as α = 0.05 and the 212 results are analyzed at a confidence interval of 95%) and EACH being the predicted 213 standard error for the specific airflow rate ACH (the regression coefficient) which is the 214 standard deviation of the sample mean or the mean variance. 215

Taking into account the previous equations (1-2), the experimental setup to analyse the 216 decay process of the gas concentration was divided into two phases, on the basis of the 217 continuous monitoring carried out by CO2 sensors. During the first phase, five sensors 218 were evenly distributed throughout the classroom (Annex A shows the location of the 219 sensors in each classroom). The outdoor CO2 concentration was then measured before the 220 start of the test. Next, windows and doors were closed, and in those cases where the 221 classroom has mechanical ventilation systems installed, the system was switched off. 222

With this setup, the CO2 concentration inside the classroom was increased using a source 223 J o u r n a l P r e -p r o o f of this tracer gas, in this case dry ice [25] . In order to achieve a homogeneous 224 concentration, two fans were used to mix the generated CO2 into the air in the room. Once 225 the required CO2 concentration level was reached (around 2000 ppm), the CO2 source 226 was removed and the fans were turned off, thus ending the first phase. In the second 227 phase, the ventilation strategy under study was set up (i.e. windows and doors are opened 228 according to Table 2) . Once the CO2 level decayed 37% of its peak concentration above 229 the background, the test ended [32, 33] . This process was repeated to analyze all the 230 ventilation strategies identified for each classroom are analysed. It should be remarked 231 that the room must be unoccupied during the experimental test. 232 HOBO® MX1102 sensors were used to measure CO2 concentration at know times during 233 the experimental tests. The sensing method of this instrument is based on non-dispersive 234 infrared (NDIR) absorption and a measurement range from 0 to 5000 ppm (accuracy ± 50 235 ppm ± 5% of readings at 25°C, less than 90% RH non-condensing and 1.013 mbar). 236

The ACH is a parameter that is regulated in the national ventilation regulations of many 237 to the initial CO2 concentration ( 0 ) and ACH is expressed as ACH = G/V. Equation (5) 269 can be expressed as follow: 270 As can be seen, since the control of virus-containing aerosol concentrations depends on 291 ventilation solutions when the social distance is greater than 1.5 m, the CO2 concentration 292 can be used to assess the effectiveness of ventilation and, thus, the likelihood of infection. 293

The risk of infection can be calculated for different activities and rooms using a standard 294 Wells-Riley airborne disease transmission model, which can be calibrated for viruses 295 such as SARS-Cov-2 by adjusting the correct source intensity, i.e. quanta emission rates. 296

Equation (8) 

where is the probability of infection risk, is the number of cases that develop 299 infection, is the number of susceptible people, is the number of infectors ( =1 has 300 been assumed in this study), is the quantum generation rate by an infected person (h −1 ), 301 is the exposure time (h), ACH is the air change per hour in the room (m 3 /h) and is the 302 pulmonary ventilation rate of susceptible people. In this study, a quanta emission rate 303 equal to 5.0 /ℎ has been assumed and, since students are sitting, has been 304 assumed to be = 0.54 3 /ℎ [42,43]. 305

The Wells-Riley method assumes a steady-state infectious particle concentration that 306 varies with the VR and well-mixed room air. In consequence, it supposes a limitation in 307 large rooms where the virus concentration is not necessarily well-mixed in the air. 308

Moreover, the quanta emission rates are currently being researched, are not definitive and 309 the uncertainty of these values is high [13, 44, 45] . 310

This section shows the results from the field measurement campaign carried out in the 313 selected classrooms at both locations. In this study, occupancy has been considered at a 314 normal scenario (100% occupancy) and COVID protocol scenario (reduced occupancy as 315 a measure taken during the reopening of the centers to minimize the transmission of 316 SARS-CoV-2 in educational centers in 2021). 2.5 CEW-MD 6.78±0.28 6.06±0.47 6.60±0.45 6.33±0.08 6.53±0.14 6.46±0.32 CAW 5.70±0.22 5.52±0.14 5.36±0.14 5.00±0.10 5.20±0.08 5.36±0.14 * These values have been calculated based on the Spanish national regulations. 330 ** Bold numbers indicate that the value is higher than the REHVA recommended ACH 331 value assuming the COVID-19 rule-based occupation. 332 In summary, in both locations, the cross-natural ventilation configuration (i.e. CAW-MD 374 and CAW-2D) provided more effective air renovation than the single-side ventilation 375 configuration (i.e. CAW). Moreover, it should be pointed out that the possible 376 configurations that can be implemented in each classroom depend on its characteristics 377 and, hence, the VR that is possible to achieve with these strategies is conditioned by the 378 classroom design. 379 380

The CO2 concentration inside the classrooms was estimated based on the results obtained 382 in the field measurements and the occupancy. Additionally, the probability of COVID 383 infection risk has also been estimated. A 2 hour duration was assumed to calculate the 384 probability, due to the fact that it is the average lecture time in both universities. The 385 results obtained are shown in Table 6 . 386 387 As can be seen, the CO2 concentration in indoor spaces is related to different factors such 404 as the volume of the space, the VR and the number of CO2 generation sources (i.e. 405 occupants). Fig. 4 shows the estimated CO2 concentrations based on these factors. Given 406 that CO2 concentration is a parameter that has been recommended for the assessment of 407 the effectiveness of indoor ventilation, this color map can be used to quickly identify how 408 to adapt these factors in order to ensure that the CO2 concentration limits are not 409 exceeded. From this figure and to accomplish the required limits, the following measures 410 could be adopted: (1) limitation of space occupancy, (2) increase of VR or (3) The first option requires limiting the number of occupants in the room to ensure that the 418 CO2 concentration remains below the limit. This measure has to be implemented in those 419 spaces where the maximum achievable VR is limited (e.g. classrooms with mechanical 420 ventilation systems sized below the required or recommended ACH value, ventilation 421 limitations arising from design characteristics, etc.) For example, as can be seen in Fig.  422 4, if the objective is to maintain the CO2 concentration at a level of 1.000 ppm in a 423 classroom whose volume and ACH is 300 m 3 and 4 h -1 respectively, the maximum 424 occupant/volume ratio in that scenario is 0.10 occupants•m -3 . i.e., 30 occupants. However, 425 many teaching spaces are designed for high volume occupancy with a low VR, so severely 426 limiting the number of occupants can lead to under-utilization. 427

Regarding the second option, the VR can be increased in those spaces where the CO2 428 concentration limit is exceeded while maintaining a 100% occupancy. This measure is 429 easily implemented in mechanically ventilated spaces where the size of the ventilation 430 system can be increased. However, ACH values above 10 are hardly achievable through 431 natural ventilation, which is the system mostly used in the analyzed classrooms. This is 432 similar in other European countries, where most schools (86%) use natural ventilation; 433 7% of schools use assisted ventilation and 7% of schools use mechanical ventilation [18] . 434

For example, classroom A2-2, (whose volume and occupancy is 341 m 3 and 120 seats, 435 respectively) requires 15 h -1 to maintain the CO2 concentration at a level around 1000 436 ppm (as shown in Fig. 4) . Moreover, in continuously naturally ventilated classrooms with 437 high ACH values, IEQ variables (such as temperature or pollutants) are closely related to 438 the outdoor environment. Therefore, keeping the indoor temperature in a comfortable 439 range will require higher energy consumption in the heating and cooling systems of 440 educational buildings. Consequently, in many cases, adapting spaces to limit the level of 441 As shown in Table 4 ventilation strategies also present problems due to their impact on the recommended IEQ 466 factors, especially those related to thermal and acoustic comfort [46] [47] [48] . The retrofitting of public spaces to make them more sustainable through the adoption of 483 strategies, measures and constructive solutions has been a concern at European level, with 484 a particular focus on educational buildings. Directive 2010/31/EU sets targets for 485 reducing energy consumption with the aim of "promoting the improvement of the energy 486 performance of buildings". Specifically, the directive states that all new buildings should 487 be Nearly Zero Energy Buildings by 2020 and public buildings by 2018. 488 In this context, the retrofitting and renovation process of existing educational buildings 489 offers an exceptional opportunity to not only take into account improvements in the 490 energy performance of buildings but also, to ensure retrofitting measures that will 491 guarantee an adequate IAQ after the renovation, at the design phase of the building. However, with the protocol for the reduction of room occupancy due to COVID-546 19 during the academic year 2020/2021, the CO2 concentration is significantly 547 reduced, ranging from 475 to 2201 ppm (a reduction between 11 and 44%). 548

-Regarding the infection risk, it is only higher than 1% in four of the scenarios 549 studied (A2-2, A3-1, A3-2 and A4-1). The analysis of the results has shown that, 550 although the estimated CO2 level is above the REHVA recommended level for 551 good ventilation, it does not necessarily imply that the risk of infection is higher. 552 -The characteristics and equipment in classrooms influence the possible ventilation 553 strategies that can be implemented. For this reason, and given the limitation to 554 achieve adequate indoor air quality, retrofitting interventions in teaching spaces 555 should not only prioritize energy efficiency, but should also ensure the IAQ is safe 556 for the occupants. In the case of buildings where retrofitting is not possible in the 557 short term, ventilation strategies should be analyzed and protocols (e.g. occupancy 558 limitation, ventilation strategies, etc.) should be established to ensure that they are 559 safe for use. 560

Finally, in light of the consequences of the recent COVID-19 pandemic, the ventilation 562 rate of buildings must be improved, either through adaptations of spaces or the adoption 563 of protocols, to ensure that they are safe for occupants to use. Additionally, it is 564 recommended that an action plan be developed that establishes protocols that not only 565 meet national indoor air quality standards, but also activate previously established 566 protocols, in the event of a new outbreak of an airborne virus. This rapid response will 567 enable the continued safe use of spaces without disrupting the learning of millions of 568 

comfort and indoor air quality in 627 nursery and elementary school buildings in the cold climatic zone of Greece, 628 Energy Build

Do indoor pollutants and thermal conditions in schools 631 influence student performance? A critical review of the literature

School air quality related to dry cough, 635 rhinitis and nasal patency in children

A long-term multi-parametric monitoring study: 639 Indoor air quality (IAQ) and the sources of the pollutants, prevalence of sick 640 building syndrome (SBS) symptoms, and respiratory health indicators

Particulate matter air 645 pollution and cardiovascular disease: an update to the scientific statement from 646 the American Heart Association

The effect of indoor air quality (IAQ) towards 649 occupants' psychological performance in office buildings

Indoor air 652 quality and sources in schools and related health effects

Indoor environmental quality of classrooms in 656

The role of classroom volume, occupancy, voice reduction 659 and ffp2 masks in transmission risk of SARS-CoV2 in schools

Medidas de prevención y 663 recomendaciones. Ministerio de Sanidad. Gobierno de España

Education: From disruption to recovery

American Society 670 of Heating, Ventilating, and Air-Conditioning Engineers

How to operate HVAC and other building service 674 systems to prevent the spread of the coronavirus (SARS-CoV-2) disease

Energy performance of buildings -Indoor environmental 680 quality -Part 1: Indoor environmental input parameters for the design and 681 assessment of energy performance of buildings

Energy performance of buildings -Ventilation for buildings -683

Part 1: Indoor environmental input parameters for design and assessment of 684 energy performance of buildings addressing indoor air quality, thermal 685 environment, lighting and acoustics -Module

Infection Prevention and Control during Health Care when COVID-19 Is 687 Suspected: Interim Guidance

State-of-the-art review of CO2 demand controlled 691 ventilation technology and application. National Institute of Standards and 692 Technology

health/air-quality/publications/2015/the-school-environment-policies-and-696 current-status (accessed on

Survey of ventilation rates in office 698 buildings

Diurnal and seasonal variation in air exchange rates and 702 interzonal airflows measured by active and passive tracer gas in homes

Tracer-gas techniques for measuring ventilation in a single zone

Air exchange rates from 709 atmospheric CO2 daily cycle

Review and extension of CO2-based methods to determine 712 ventilation rates with application to school classrooms

Natural ventilation strategy and related 715 issues to prevent coronavirus disease 2019 (COVID-19) airborne transmission in 716 a school building

Ventilation Assessment by Carbon Dioxide Levels in Dental Treatment Rooms, 720

A review of methods to 723 reduce the probability of the airborne spread of COVID-19 in ventilation systems 724 and enclosed spaces

Role of ventilation in airborne transmission of infectious agents in 728 the built environment-a multidisciplinary systematic review

Review and comparison of 731 HVAC operation guidelines in different countries during the COVID-19 732 pandemic

Re-thinking of engineering operation solutions to 735 HVAC systems under the emerging COVID-19 pandemic

Study on ventilation rates and assessment of infection risks of 738

COVID-19 in an outpatient building

Respiratory infection risk-based ventilation design method

Cedeno-Laurent 5-step guide to checking 744 ventilation rates in classrooms

Guía para 748 ventilación en aulas

Thermal performance of buildings and materials -751

Determination of specific airflow rate in buildings -Tracer gas dilution method

CO2 tracer gas concentration decay 754 method for measuring air change rate

Real Decreto 1027/2007, de 20 de julio, por el que se aprueba el Reglamento de 757

Energy performance of buildings -Ventilation for buildings -759

For non-residential buildings -Performance requirements for ventilation 760 and room-conditioning systems (Modules M5-1, M5-4)

Carbon dioxide generation rates for building occupants

Airborne spread of measles in a suburban 772 elementary school

Estimation of airborne viral emission

Quanta emission rate of SARS-CoV-2 for infection risk assessment

Quantitative assessment of the risk of 778 airborne transmission of SARS-CoV-2 infection: prospective and retrospective 779 applications

Thermodynamic equilibrium dose-response models for MERS-CoV 782 infection reveal a potential protective role of human lung mucus but not for SARS-783

Surfing the COVID-19 scientific wave

Monitoring 788 and Assessment of Indoor Environmental Conditions after the Implementation of 789

COVID-19-Based Ventilation Strategies in an Educational Building in Southern 790

Experimental assessment of the impact of natural 792 ventilation on indoor air quality and thermal comfort conditions of educational 793 buildings in the Eastern Mediterranean region during the heating period

An investigation of indoor thermal 796 environment in semi-cold region in Japan-Validity of thermal predictive indices 797

in Nagano during the summer season

Natural 800 ventilation for infection control in health-care settings. World Health 801 Organization

Natural ventilation in warm climates: The 805 challenges of thermal comfort, heatwave resilience and indoor air quality

Investigating natural ventilation potentials across 809 the globe: Regional and climatic variations

Acknowledgements: Antonio J. Aguilar Aguilera and María Luisa de la Hoz Torres wish to thank the support of the Ministerio de Ciencia, Innovación y Universidades of Spain under an FPU grant. This work has been supported by the "Junta de Andalucía" (Spain) under project B-TEP-362-UGR18, and the State Research Agency (SRA) of Spain and European Regional Development Funds (ERDF) under project PID2019-108761RB-I00.

-Best ventilation strategies are suggested from the analysis of in-situ measurements.-Appropriate combination of door and window openings provide sufficient air renewal.-Indoor CO2 concentration shows a wide dispersion in different ventilation configurations.-IAQ management should be also considered to improve risk management in future events.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.María Luisa de la Hoz Torres, on behalf of the authors University of Granada, Spain