The influence of occupant behaviour on indoor air quality and COVID-19 risk in refugee shelters and temporary houses

Abstract

There are growing concerns about poor indoor air quality in refugee shelters, particularly regarding chronic health conditions and the spread of airborne diseases like COVID-19. These issues are influenced by shelter design and occupants’ behaviours, such as cooking and window usage. However, behavioural aspects are often overlooked in shelter design due to the challenges of monitoring occupants, which can be costly and intrusive. To address this, we developed a cost-effective method for assessing shelters that combines self-assessed behavioural data, predicted ventilation rates, and a mathematical model for airborne disease transmission. This approach was tested in temporary housing following the 2020 floods in Kumamoto Prefecture, Japan. Results indicated that indoor CO₂ levels exceeded national thresholds over 70% of the time, suggesting inadequate ventilation to mitigate airborne disease transmission. We estimated a 60–80% risk of COVID-19 transmission under these conditions. Our findings highlight severe health inequalities in forcibly displaced populations and provide: (i) the first comprehensive guide to monitoring indoor conditions and behaviours in these settings; (ii) a new model for assessing airborne disease risk. While the study focuses on COVID-19, the results can be extended to other airborne respiratory diseases through our reproductive number (R₀) estimates.

Practical application

This study presents a novel, low-cost method for monitoring indoor air quality and ventilation in temporary shelters and refugee housing, which can be applied by built environment professionals and humanitarian workers without the need for advanced technical skills. By focusing on occupant behaviour and using minimal sensor data, this approach provides practical insights for improving shelter design, reducing airborne disease transmission risks like COVID-19, and enhancing overall indoor environmental quality. The method is particularly relevant for displaced populations, where ensuring healthy and sustainable living conditions is critical to occupant well-being.

Keywords

Indoor air quality carbon dioxide shelter design COVID-19

Introduction

The global forcibly displaced population has doubled over the last decade, reaching a historical high of 103 million in 2022, following the largest annual increase ever recorded by UNHCR.^1,2 While shelter provision is a critical component of disaster response and recovery,³ the creation of healthy indoor environments is often overlooked or inadequately addressed.^4,5 Displaced populations are particularly vulnerable to the health impacts of poor indoor air quality (IAQ) due to prolonged indoor exposure, reliance on unvented cooking stoves, and limited ventilation for reasons of privacy and security.^6–10 The elderly are especially at risk, spending up to 22 hours daily indoors,^7,11,12 where unhealthy levels of IAQ can exacerbate susceptibility to respiratory diseases, including COVID-19.^13–15

Proper ventilation is vital for mitigating IAQ risks and airborne disease transmission. For instance, a modelling study in Bangladeshi refugee camps predicted a severe COVID-19 epidemic, infecting 98% of the population without effective ventilation measures.¹⁶ Lower ventilation rates (<10 l·s⁻¹·p⁻¹) are strongly associated with increased transmission risks for diseases like COVID-19 and influenza, particularly among the elderly.¹⁷ For example, in the UK, flu can cause 14500 deaths per annum,¹⁸ predominately amongst the vulnerable. Increased ventilation in buildings can reduce the likelihood of COVID-19 spread, given that airborne transmission is a primary pathway.¹⁹

The classic Wells-Riley equation, first introduced by Rudnick & Milton,²⁰ serves as a framework for quantifying the transmission of airborne pathogens and assessing associated risks in enclosed spaces.^20,21 Despite the availability of newer, more complex models, the Wells-Riley model remains fit for purpose when model inputs are characterised adequately, and key assumptions are met.²² For instance, Zemouri et al. have incorporated the Wells-Riley equation into a standard susceptible-infectious-recovered (SIR) structure,²³ while Liao et al. have modified it to account for the outdoor air supply rate in terms of the proportion of indoor air that is exhaled.²⁴ This model has been instrumental in evaluating COVID-19 contagion risks under different ventilation rates. Buonomano et al.’s application of the model, which considers ventilation rate recommendations, suggests that the current ANSI/ASHRAE Standard 62.1 may underestimate the minimum ventilation rates required to mitigate COVID-19 transmission effectively.^25,26

Temporary housing, initially intended for short-term use, often remains in service far beyond its intended lifespan.²⁷ Studies of Japanese disaster recovery housing reveal IAQ challenges such as CO₂ levels exceeding 5000 ppm, inadequate air exchange rates, and elevated concentrations of pollutants like aldehydes and volatile organic compounds (VOCs).^28,29 These issues are compounded by occupant behaviours, including limited ventilation during winter to maintain thermal comfort.^30,31 The absence of comprehensive IAQ guidelines for temporary housing in Japan³² further highlights the need for studies linking ventilation, IAQ, and health outcomes in this context.

Occupant behaviour, including the operation of windows, cooking practices, and heating methods, significantly influences IAQ.^33,34 However, in the context of displacement, these behaviours are shaped by unique constraints, such as privacy concerns, socioeconomic limitations, and unfamiliar community dynamics.^35–37 Displaced populations may prioritise thermal comfort over ventilation, leading to poor IAQ.³⁸ Understanding these behaviours is crucial for designing effective ventilation strategies that align with occupant needs and preferences.³⁹

This study represents the first effort to systematically monitor indoor CO₂ concentrations, thermal conditions, and occupant behaviour in temporary shelters, including refugee camp-like settings. By combining environmental monitoring with survey-based behavioural analysis, the study provides a comprehensive understanding of the factors influencing IAQ and thermal comfort. Additionally, the integration of COVID-19 risk modelling using indoor CO₂ levels offers a novel approach to evaluating infection risks in displacement settings. This pioneering work bridges the fields of shelter design, public health, and occupant behaviour, contributing to the development of evidence-based guidelines for improving IAQ and mitigating health risks in temporary housing.

Additionally, the methodology is designed to be low-cost and scalable, using minimal sensors and simple surveys to enable implementation in resource-limited settings. This makes it particularly suitable for displaced populations, where logistical constraints and privacy concerns often hinder comprehensive data collection.⁴⁰ The incorporation of COVID-19 transmission risk modelling further distinguishes this approach, linking IAQ and behavioural data to health outcomes in a way that is directly actionable for shelter design and public health interventions.

Focusing on Japanese disaster recovery housing, the study aims to:

(1) Develop a survey and minimal sensor set for monitoring behaviours related to health outcomes influenced by ventilation;

(2) Quantify IAQ variability in shelters due to occupant behaviour;

(3) Evaluate a model linking COVID-19 transmission to shelter ventilation conditions;

(4) Estimate ventilation rates and associated transmission risks.

By addressing the intersection of ventilation, IAQ, and occupant behaviour, this study contributes to the development of evidence-based guidelines for healthier temporary housing, with implications for broader humanitarian contexts.

Materials and methods

Although this study introduces a novel general method for monitoring shelters and assessing the transmission risk of airborne respiratory diseases, it first presents the case study temporary houses. Information about the experimental in Japan are included in the Appendix. This helps demonstrate its application in a specific setting when introducing the method in subsequent sections.

The temporary houses

This study selects temporary housing in Japan as a representative best-in-case setting to appraise occupant behaviour, IAQ, and the risk of transmission of airborne respiratory diseases. Owing to its geographical characteristics, Japan experiences annual floods and earthquakes and has established processes to assist forcibly displaced populations. Compared to many countries hosting displacement camps, Japan’s affluence and infrastructure allow for temporary housing of relatively high construction quality and specifications. Therefore, if concerning performance related to ventilation is observed in this setting, it raises important questions about the likely challenges in countries with fewer resources and less robust shelter standards.

The temporary houses are located in Matsuba Temporary Housing Complex (see Appendix for more details about the experimental location), that hosts 24 people in 16 households (mean age of 78.7 years; 25% males, 75% females; 83% living alone, 17% in couples). All were invited to participate in the study and the 6 that volunteered (H1–H6) were representative of the key demographics in the complex (Table 1). Monitoring lasted for a week in December 2022 while they were fully occupied and it was designed to be unobtrusive, allowing occupants to use their homes as they usually do.

Table 1.

Characteristics of the households. Occasionally, in H1 there are two or three occupants. This has been considered in the calculations.

Household	Gender	Age
H1	F	82
H2	F	84
H2	M	87
H3	F	74
H4	F	75
H4	M	81
H5	F	74
H6	M	73

All houses are identical in terms of design and orientation (Figure 1 and Appendix– Table 9). They were built with a timber structure on a concrete foundation with 50 mm polystyrene insulation. The floor (structural plywood) included a 15-mm insulation layer filled with cellulose fibre and finished with cedar boards and tatami mats, and the inner walls were also filled with cellulose fibre. All houses use hybrid ventilation, with natural ventilation through windows and mechanic extraction through an electric fan in the kitchen

Figure 1.

Sagara temporary house. (a) Plan view (dimensions in mm; circles denote sensor location). (b) View of the front of one of the temporary houses.

Measurements of CO₂ concentration, indoor and outdoor temperature, and relative humidity

The measurement of CO₂, indoor and outdoor air temperature, and relative humidity took place from the 1^st to the 8^th of December 2022. This was based on ASTM standard D6245-98,⁴¹ which suggests monitoring CO₂ alongside building occupancy and ventilation system operation for at least 1 day. The indoor parameters were measured at 2-min intervals in all the rooms. Sensors were placed in every room away from windows and heat sources and at a height of 1.1 m (Figure 1). This is considered ideal as it provides good coverage in the absence of prior experience for this setting (CIBSE TM68,⁴² Albadra et al.⁶ and Shrestha et al.).⁴³ Outdoor CO₂ was recorded through spot measurements. Table 10 in the Appendix shows the characteristics of the sensors used.

The CO₂ sensors were calibrated at ambient conditions prior to monitoring. In addition, they have a calibration function to compensate for long-term sensor drift by gradually adjusting the lowest measured CO₂ concentration over a 1684-h period, to the global average concentration (atmospheric CO₂ level of around 400 ppm). This function did not apply here as monitoring lasted for 1 week.

Assessment of indoor CO₂

In this study the selected CO₂ threshold is 1000 ppm as per ANSI/ASHRAE Standard 62.2-2019²⁵ and the Japanese legislation.⁴⁴ Despite this CO₂ limit being removed from the ASHRAE Standard since the 2019 edition because of its questionable reliability as a proxy for overall IAQ acceptability, it continues to be a useful metric to contextualise findings.⁴⁵

Occupant behaviour survey

We employ a household questionnaire, an operational table, and a house survey (Table 2). The household questionnaire was used to enquire about occupants’ perception of their indoor environment. This questionnaire was also used to gather more general information regarding the use of ventilation (Table 3) and the source of income of the households. On the first day, the households were asked about their thermal sensation vote (TSV), overall comfort (OC) and thermal preference (TP), and the perceived IAQ. We used the Japanese scale⁴⁶ (Figure 2(a)) for the TSV, the OC, and the TP and translated them into Japanese. The clothes worn by the respondents during the field survey were recorded, and the corresponding clothing insulation value in the unit of ‘clo’ was determined in accordance with ISO 2003, also shown in Figure 2(b). Surveys about the use of the homes were also completed. The households were requested to complete a table (Figure 3) indicating the operation of windows, door, kitchen, kitchen fan, heating fan, and the Kotatsu (a small table with an electric heater underneath and covered with a heavy blanket) and how many people were in each room. Since the concentration of CO₂ and the ventilation rates depend on the activities performed in the zones, the households were requested to fill out the table every day for the duration of the data collection at least three times per day, in the morning, at midday and in the evening. The table was prepared according to previous literature^34,47–49 and adapted to the context of temporary houses. It was prepared in English and then translated into Japanese (Figure 3). On day 1 and day 7, the metered electricity and gas consumption were recorded.

Table 2.

Overview of the survey methods used in the study.

Survey	Questions/Notes	Study stage	Method	Reference
Household questionnaire	- TSV - OC - TP - Perceived IAQ - Clothing - Source of income	Day 1	Direct questions to the households	Supplemental material–S1
	- General question about use of ventilation and temporary house
Table	- Window status- Door status	Day 1 to day 7	Occupants fill the table on a daily basis	Figure 3
	- Use of kitchen
	- Use of cooking fan
	- Use of heating
	- Use of Kotatsu
	- Number of occupants
House survey	- House dimensions- House status	Day 1 and day 7^a	Notes recorded by authors	Supplemental material–S2
	- Electricity reading
	- Gas reading

^aOnly for electricity and gas reading.

Table 3.

Questions on perceived indoor air quality.

How satisfied are you with the air quality of your shelter?	How satisfied are you with your window dimension?	How satisfied are you with the daylight levels in your temporary house?	Explanation
1	1	1	Satisfied
0	0	0	Neither
−1	−1	−1	Unsatisfied

Figure 2.

Scales used for thermal comfort survey. (a) Thermal sensation vote scale adapted from.⁴⁶ (i) The Society of Heating, Air-Conditioning and Sanitary Engineers of Japan (SHASE) thermal sensation scale and (ii) The American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) thermal sensation scale. (b) Clothing insulation scale adapted from.⁴⁶

Figure 3.

Table used to collect data about the occupants’ behaviour and the indoor use of the temporary houses. (a) English version, (b) Japanese version. The occupants were requested to show with a line the time they open the window and/or the door when they use the kitchen fan, the kitchen, and the Kotatsu (the Kotatsu parameter was modified for houses using the electric stove or the AC). In the last field of the table, occupants indicated with a number the number of people in the temporary house and with a line the occupancy time.

Calculation of ventilation rate from indoor CO₂

We estimate the number of air changes (ACH) using the indoor CO₂ exhaled by occupants through the mass balance equation as suggested by Shrestha et al.⁴³

n_{p} G d t + C_{o} Q_{i n} d t = V d C_{r} + C_{r} Q_{o} d t

(1)

where n_p is the number of occupants, G is the CO₂ emission rate of the occupants [g (s⋅p)⁻¹], C_o outdoor CO₂ concentration [g⋅m⁻³], Q_in is the indoor airflow [m³⋅s⁻¹] assumed equal to the outdoor airflow Q_o [m³⋅s⁻¹], V is the shelter’s volume [m³], C_r is the indoor CO₂ concentration [g⋅m⁻³], dC_r is the infinitesimal change in CO₂ concentration [g⋅m⁻³] and dt is the infinitesimal time interval [s].

Calculated ACHs are then compared to EN 16798–1:2019⁵⁰ and ISO 17772-1:2017.⁵¹ The first provides guidelines for the assessment of indoor environmental parameters for buildings’ ventilation, considering both thermal comfort and indoor air quality. Similarly, the second outlines the parameters for assessing and achieving energy-efficient and comfortable indoor environments through heating, ventilation, and air conditioning (HVAC) systems. Both standards aim to promote occupants’ well-being by specifying criteria for the design and operation of building systems.

Linear regression models of COVID-19

Based on the estimated ACH, three parameters related to the COVID-19 infection risk are estimated: the probability (P) of infection for a susceptible population associated with an ACH, the basic reproductive number (R₀), and the probability of infection for a susceptible population associated with a CO₂ level (P_CO2). For calculating P, we adopt the Wells-Riley equation (2) proposed by Rudnick and Milton,²⁰ calibrated to COVID-19 with corrected quanta emission rates for an elderly population^52,53 as follows:

P = \frac{D}{S} = 1 - \exp {- \frac{i q b_{p} t}{Q} \times [1 - \frac{V}{Q t} (1 - \exp (- \frac{Q t}{V}))]}

(2)

where D is the number of disease cases, S is the number of susceptible, i is the number of infectors, q is the quantum generation rate by an infected person [quanta⋅S⁻¹], b_p is the breathing rate per person [m³⋅s⁻¹], t is the total exposure time [s], Q is the outdoor air supply rate [m³⋅s⁻¹], and V is the volume of the ventilated space [m³].

Then the rate of spread R₀ is determined using the Liao et al. model²⁴ which adopts a standard SIR structure to relate the Wells-Riley model with the transmission risk. R₀ is expressed as:

R_{0} = \frac{n β}{ν + μ}

(3)

where n is the number of people in the ventilated airspace, β is the transmission parameter indicating the relationship between the indoor CO₂ and the contact rate, μ is the birth rate of the infected population [d⁻¹], and ν the recovery rate [d⁻¹]. The input parameters for the estimation of R₀ are selected from Liao et al.²⁴ (Table 4)

Table 4.

Birth rate, recovery rate, and transmission parameters used in the SIR model for influenza and COVID-19 (adapted from Liao et al.²⁴).

Parameter	Estimate	95% CI
μ	3.43 × 10⁻²	3.20 × 10⁻³ – 6.54 × 10⁻²
ν	9.60 × 10⁻²	4.10 × 10⁻² – 1.51 × 10⁻¹
β	4.00 × 10⁻²	-

Finally, P_CO2 is estimated with the Zemouri et al. model.²³ This model associates the indoor concentration of CO₂ to a three-level risk scenario as shown in Table 5.

Table 5.

Three risk scenario model used in the SIR model for influenza and SARS (adapted from Zemouri et al.²³).

Description	Low risk	Medium risk	High risk
Indoor CO₂ (ppm)	<774	774 – 2375	2375<

The transmission parameters for COVID-19, including the quantum generation rate, breathing rate, and exposure time, are subject to significant variability in the literature. Rudnick and Milton²⁰ estimated a quantum generation rate of approximately 14 quanta per hour for normal breathing, whereas Buonanno et al.⁵⁴ reported higher rates for activities such as speaking or singing, which can exceed 100 quanta per hour depending on vocal intensity. For elderly populations, Liao et al.²⁴ and Zhang and Lin⁵³ suggested lower metabolic activity rates may reduce quantum generation, although this remains context-dependent.

The Wells-Riley model used in this study assumes uniform air mixing, a simplification that does not fully account for localised variations in ventilation, deposition, or filtration. While this approach provides a valuable first-order estimation of transmission risk, these limitations must be considered when interpreting the results. For example, specific ventilation strategies, such as cross-ventilation or filtration, could significantly reduce localised airborne concentrations, which the model does not capture.

Additionally, our assumptions include a fixed quantum generation rate specific to an elderly population and constant exposure times. These assumptions simplify the analysis but may not reflect individual differences in health, activity levels, or environmental conditions, all of which could influence transmission dynamics. Future research should incorporate more detailed modelling to address these complexities, particularly in settings with diverse population groups or ventilation strategies.

Linear regression was employed to examine the relationship between ventilation rates, CO₂ concentrations, and COVID-19 infection risks as estimated by the Wells-Riley and SIR models. This approach was chosen for its simplicity and interpretability, enabling the quantification of the influence of ventilation and IAQ parameters on transmission risks. By providing clear effect sizes for each variable, the regression analysis supported actionable insights for shelter design and ventilation strategies in temporary housing.

Methodology novelty

The novelty of this study lies in its integrative methodological approach, which combines environmental monitoring with behavioural self-assessment to examine IAQ and thermal comfort in temporary housing settings. While traditional IAQ studies often focus on physical parameters such as CO₂ concentrations and temperature, this approach captures the complex interplay between environmental conditions and occupant behaviours.

By applying this method to a vulnerable population group in a resource-limited context, the study provides insights into the behavioural drivers of IAQ and their implications for shelter design. Additionally, the incorporation of COVID-19 risk modelling, based on CO₂ levels and self-reported behaviours, represents a novel application in the context of displacement settings. This integrative approach enables the development of data-driven recommendations that are both technically effective and behaviourally informed.

Results

Indoor CO₂ concentration

The indoor CO₂ concentration in H2, H3, and H5 was in the range of 500-800 ppm while that of H1, H4, and H6 was approximately in the range of 500-1600 ppm for 95% of the time (Figure 4). The average CO₂ concentration in H2, H3, and H5 was 707 ppm, so below the Japan Building Regulation⁴⁴ threshold of 1000 ppm, while in H1, H4, and H6 the average (1387 ppm) was much higher.

Figure 4.

(a) Hourly indoor CO₂ concentration in the dining rooms of six temporary houses during the monitoring period. The red line indicates the 1000 ppm threshold established by the Japan Building Regulation.⁴⁴ (b) A closer view of CO₂ concentration values below 2500 ppm, showing fluctuations and comparisons between houses. (c) Temperature profiles of the same dining rooms, with outdoor temperature represented by the black dashed line. The shaded red area denotes the thermal comfort band as per Japanese government guidelines for residential buildings.⁵⁵ The data highlight significant differences in CO₂ levels and indoor temperatures between houses, particularly in relation to ventilation and heating practices.

The average CO₂ concentration at night was almost the same as during the day. This might be explained by the absence of ventilation at night. Indeed, indoor CO₂ levels exceeded 1000 ppm in H1, H4, and H6 at such times (Table 6). The maximum concentration of CO₂ for H1 is about 6500 ppm, being above the prescribed limit for almost 90% of the time. Furthermore, upon averaging the data into one-hour increments, it was observed that in H1, the CO₂ levels exceeded 1000 ppm for eight consecutive hours more than 40% of the time. While, CO₂ is not considered a harmful pollutant for human health, prolonged exposure to high concentrations of CO₂ has been associated with physiological changes in the human body, and more acute effects of headache and dizziness were found at indoor concentrations of 2000–5000 ppm for 2 hours of continuous exposure.

Table 6.

Percentage of time with CO₂ concentration above the Japanese Legislation safe threshold in the houses.

Temporary house	Percentage of time CO₂ > 1000 ppm (%)			Percentage of time CO₂ > 1000 ppm for 8 consecutive hours (%)
Temporary house	Kitchen	Dining room	Bedroom	Kitchen	Dining room	Bedroom
H1	88	89	91	30	42	42
H2	7	12	8	0	0	0
H3	3	5	5	0	0	0
H4	45	54	68	6	17	11
H5	4	6	11	0	0	0
H6	44	68	68	19	19	18

CO₂ concentrations are higher in the dining rooms and the bedrooms than in the kitchens (Figure 5). This is possibly due to the form of Japanese cooking (electric not gas) and the temporary house layout or by occupants spending most of their time in the dining room and bedroom area.

Figure 5.

Distribution of indoor CO₂ concentrations and air temperatures across the monitored temporary houses. The figure shows that dining rooms and bedrooms tend to have higher CO₂ concentrations compared to kitchens, likely due to occupant activity patterns and heating locations. Temperature distributions demonstrate variability between spaces, influenced by heating strategies and ventilation practices.

Indoor air temperature

Households were observed to approach space heating differently. H1 adopted an electric stove, H4 used air-conditioning despite the fact that they reported that they did not like the increased air movement, and H2, H3, H5, and H6 used a Kotatsu. The lowest average air temperature was recorded in H5 (kitchen) and the highest in H1 (bedroom) (Figure 5). In general, for the kitchens, the average air temperatures were lower than in the bedroom and the living room. This may be due to the heating being in the dining room. Indoor temperatures similar to each other were recorded in the dining room and the bedroom, except in H2 and H4.

Indoor CO₂ concentration and indoor temperature

H1 and H4 present both the highest average indoor temperature and the highest indoor concentration of CO₂ most of the time (Figure 6). This might be related to the reduced use of ventilation in these dwellings. The indoor concentration of CO₂ generated by occupants increases when a source of heating is on because the window is kept closed to maintain a comfortable thermal environment. However, H6 presents a high concentration of CO₂ but the lowest average indoor temperature. In fact, H4 and H6 at the same indoor concentration of CO₂ present an average difference in temperature of 6°C (Figure 6). This can be explained by observing that H4 ventilates less than H6 and by H4 using AC for heating.

Figure 6.

Comparison of indoor temperatures and CO₂ concentrations in the three temporary houses (H1, H4, and H6) with the highest monitored CO₂ levels. Highlighted in red are periods when CO₂ concentrations exceeded 1000 ppm, showing correlations with heating practices and ventilation patterns.

Indoor air quality and thermal comfort

Households were satisfied with the levels of indoor air quality when specifically asked through the questionnaire (Supplemental Material). None of them was perceiving any odour, despite different odours being present in the temporary houses. The results for thermal sensation vote (TSV), overall comfort (OC), and thermal preference (TP) are presented in Table 7. People preferring a slightly warmer thermal environment show an average clothing insulation level (I_cl) of 1.26 clo. Only H1 presents an hourly indoor temperature within the Japanese prescribed comfort range⁵⁵ for more than 50% of the time in all three monitored rooms, while H4 is only in the dining room, where the AC is used. Figure 4(c) also shows that the indoor temperature in H3 is out of the comfort range 72% of the time in the bedroom and 82% in the kitchen. Similar to H3, also indoor temperatures in H5 and H6 were out of the comfort band for more than 60% of the time. Indoor temperatures higher than the comfort band can be explained through the findings of Schellen et al.,⁵⁶ who showed how, due to the reduction in basal metabolic rate, the elderly population prefers a higher ambient temperature in comparison with the younger population, and they tend to reach it by reducing the ventilation rate or additional heating.

Table 7.

Thermal votes and clothing insulation for males and females and range value for the same region.

Household	Number of occupants	Sex	Clothing [clo]	TSV	OC	TP
H1	1	F	0.91	0	1	0
H2	1	F	0.81	0	1	0
H2	1	M	1.34	0	1	0
H3	1	F	1.60	0	1	0
H4	1	F	1.09	0	0	0
H4	1	M	1.50	0	0	0
H5	1	F	1.38	−1	−2	1
H6	1	M	1.52	−1	−2	1
Range value for the same region in winter⁵⁷			0.9–1.8	-	-	-

Occupant behaviour

The results described in this section are the first showing the use of windows, heating, and cooking in temporary houses collected in the table in Figure 3. The temporary houses are provided with single-side purged ventilation with windows in the dining room and the bedroom, while the entrance door and a small vent are located in the kitchen. According to the household survey, the window in the bedroom and the small vent in the kitchen were never operated in any of the six houses. Natural ventilation through the window in the dining room and the door in the kitchen was the main type of ventilation. Although the temporary houses are provided with air-conditioning, only H4 uses it as the resultant air movement is a cause of discomfort, as reported in the general question of the questionnaire (Supplemental Material). Figure 7 shows the frequency of window (a) and door (b) opening. As might be expected, the windows and doors were open more in temporary houses where the CO₂ levels were below the safe limit, and indeed H2, H3, and H5 with safe limits of CO₂ are more ventilated than H1, H4, and H6 with potentially unhealthy levels of CO₂.

Figure 7.

(a) Frequency of window openings in the dining rooms of six temporary houses during the monitoring period. (b) Frequency of door openings in kitchens across the same houses. The data show that temporary houses with lower CO₂ levels (H2, H3, and H5) had higher rates of window and door openings, while houses with higher CO₂ levels (H1, H4, and H6) demonstrated limited ventilation.

Because an announcement was made by the Japanese Government requesting residents to regularly ventilate to reduce the risk of COVID-19 spread, the household ventilated the dining room at least once per day (confirmed through the household questionnaire in the Supplemental Material). In five of the six temporary houses, the window in the dining room was open for at least 1 hour per day, as shown in Figure 7. However, as can be observed in Figure 8 both windows and doors were mainly closed during the observed period. This can be attributed to the fact that the monitoring campaign was conducted in winter and low outdoor temperatures.

Figure 8.

(a) Percentage of time windows remained open in the dining rooms of six temporary houses. (b) Percentage of time doors remained open in the kitchens. The results reveal limited use of natural ventilation during the winter monitoring period, particularly in houses with higher CO₂ levels (H1, H4, and H6). Low outdoor temperatures may have influenced occupants’ decisions to keep windows and doors closed.

In contrast to the findings of Hattori et al.,⁵⁸ in this study, the residents of the temporary houses were aware of the importance of properly naturally ventilating. In fact, in H2, H3, and H5, households preferred to ventilate despite the uncomfortable indoor temperatures. However, in H1 and H4, higher temperatures were recorded despite the low ventilation rates.

Figure 9 shows the frequency of use of heating and Kotatsu status. The heating system through AC is used in H4, while H1 uses an electric stove. H2, H3, H5 and H6 use a Kotatsu. The hourly use of heating is higher in temporary houses with higher hourly ventilation.

Figure 9.

Kotatsu (H2, H3, H5, and H6), electric stove (H1), and AC heating (H4) percentage of occurrence. The percentage of occurrence represents the total number of hours the Kotatsu or electric stove is used. The frequency is calculated over the total 168 hours of the monitoring campaign and is expressed as a percentage.

The survey conducted with the table in Figure 3 revealed that in all homes, the kitchen extractor fan was in use during cooking. However, the use of the fan did not prevent the increase in CO₂ levels whilst cooking (Figure 10). Even in temporary houses where the average CO₂ indoor concentration is below the safe limit, an increase in CO₂ above 1000 ppm in the kitchen is observed when the kitchen is in use. Interestingly, in H6, where there was low use of electric cooking (Japanese cuisine includes a high variety of raw food that does not need to be cooked), CO₂ in the kitchen reached high levels (Figure 5). This may be due to the lack of ventilation, as the window, in this case, was always closed. In H5, the Kotatsu was turned on for most of the day or whenever the room was occupied (Figure 9).

Figure 10.

(a) Frequency of cooking activities in the kitchens of six temporary houses during the monitoring period. (b) Frequency of kitchen fan use during cooking. The data indicate that kitchen extractor fans were used in all houses during cooking. However, CO₂ levels frequently exceeded the safe threshold, suggesting that ventilation was insufficient to mitigate the impact of cooking on indoor air quality.

Ventilation and COVID-19

Figure 11 shows the estimated ACH value calculated according to the time-variant mass balance equation (1) for each house based on the indoor CO₂ exhaled by occupants. On average, the airflow is 0.5 ACH (or 3.2 ls⁻¹p for an occupancy of 2 people) for H1 and H6 and 0.6 ACH (or 4.5 ls⁻¹p for an occupancy of 2 people) for H2, H3, and H4 (Table 8). These findings show a higher airflow if compared to previous studies on temporary houses in Japan,^29,30 where the estimated mean ACH was 0.3. A mean of 0.3 ACH was also recorded in temporary houses built in the aftermath of Hurricane Katrina.⁵⁹ However, recorded ACH in regular Japanese log houses was between 1.6 and 1.7 in summer and between 0.6 and 0.9 in winter.⁶⁰

Figure 11.

House airflow estimated according to Ref. 43. H6 does not ventilate regularly as the other temporary houses. H1 and H6 recorded average ACH values of 0.5. H2, H3, and H4 exhibited slightly higher average ACH values of 0.6. The airflow rates were lower than the 1.0 ACH recommended by the Japanese government for two occupants during the COVID-19 pandemic, highlighting inadequate ventilation.

Table 8.

Calculated airflow from indoor CO₂ in each temporary house. The airflow has been calculated according to Ref.43.

	Airflow (ACH)
Households	Max	Min	Avg
H1	0.7	0.5	0.5
H2	1.4	0.4	0.6
H3	1.6	0.2	0.6
H4	0.8	0.5	0.6
H5	1.3	0.4	0.6
H6	0.6	0.4	0.5

Our estimated airflow value per person for an occupancy of two persons is lower than the 8.3 ls⁻¹p prescribed by the Japanese government. During the Pandemic, the Japanese government required every building owner to keep a minimum ventilation rate of 30 m³h⁻¹p,⁶¹ which in this case implies 0.5 or 1 ACH depending on whether it is occupied by one or two people. The government also recommended reducing the number of people in the room and setting CO₂ meters to monitor ventilation if the ventilation rates were lower than the prescribed value. The estimated ventilation is below the 10 l·s⁻¹ per person suggested by standards such as EN 16798-1:2019 for elderly people^50,62 and ISO 17772-1:2017.⁵¹

Observed minimum values are an indicator of increased risk of aerosol transmission,⁶³ as shown in Figure 12. In this figure, the probability of infection risk is calculated for eight ventilation and assuming for all the scenarios: there is one infected person, two occupants in the temporary house, a 24-h exposure, and no use of masks. The probability of infection risk was estimated using the Wells-Riley equation (2) according to Rudnick and Milton.²⁰ Estimated percentages indicate a significant potential for COVID-19 transmission, which may be attributable to poor ventilation rate and long exposure time, due to the tendency of elderly people to spend around 22 hours indoors per day. The R₀ for this scenario is around 0.62. This value of R₀ suggests that, under these specific conditions, the transmission potential of COVID-19 is less than 1, indicating that each infected individual would, on average, transmit the virus to less than one other person. This implies a low potential for sustained transmission in this specific setting with the given parameters.

Figure 12.

(a) Indoor CO₂ concentration profiles in six temporary houses. (b) Estimated COVID-19 transmission risks based on CO₂ concentrations, occupancy, and exposure time, following the Wells-Riley model.¹⁹ The results demonstrate the relationship between poor ventilation and increased infection risk, emphasising the need for improved air exchange rates to mitigate health risks.

Figure 13 presents the CO₂ concentration profiles for the six temporary housing units alongside the associated COVID-19 infection risks modelled according to equation (3). Temporary houses H1 and H3 are included within the Low-Risk scenario. H2 intersects the intermediate threshold, whereas H4-H6 extends into the High-Risk band, suggesting increased transmission likelihood. The scenarios in Figure 13 agree with findings that associate high concentrations of CO₂ with an increased infection risk. However, these findings may necessitate more detailed analysis, including additional loss mechanisms like filtration or deposition. These results may also demonstrate how current standards EN 16798-1:2019⁵⁰ and ISO 17772-1:2017⁵¹ are based on the use of ventilation for diluting odours⁶³ and they should be reviewed to include airborne infection control⁶⁴ and household-related factors.⁶⁵

Figure 13.

COVID-19 transmission risk in each temporary house, based on the indoor concentration of CO₂ according to the Zemouri et al model.²³ Houses H1 and H3 fall within the low-risk category. H2 intersects the intermediate threshold. H4–H6 extend into the high-risk category, suggesting increased infection likelihood. The figure underscores the importance of maintaining adequate ventilation to reduce transmission risks in temporary housing settings.

Additional observations

This section presents the additional observations gathered during the monitoring campaign through observation and through the general question, “Do you have anything to add about the temporary house and the indoor environment?” asked at the end of the household questionnaire (Supplemental Material). The occupants reported preferring their previous traditional Japanese houses in summer to their temporary houses, as they said that Japanese houses are designed for summer. However, they prefer temporary houses for the winter season. They especially like the floor of the temporary house in winter. Japanese people traditionally sit on the floor, and in this case, as we are analysing an elderly population, they spend most of their day sitting on the floor, except for H1. The households, consisting mostly of retirees, rely on a pension as their sole source of income, except H3, who is not retired. Households explained that they wanted to remain in their temporary houses because they could not afford to build or buy a new house. They also complained about the need to pay for the land where their previous homes – now damaged or destroyed by flooding – stand. Households explained that paying the bills can be difficult in winter. In addition to this, when asked about their thermal sensation (TSV, OC, and TP, Table 2) in the household questionnaire, all the occupants answered a preference for layering clothing over using air-conditioning for heating to avoid the high cost associated with it. This habit of layering clothing rather than increasing the indoor temperature was also found by Wang et al.⁵⁷ in a study in regular houses in the same region. These practices suggest an adaptive strategy by occupants to maintain a comfortable indoor environment while managing financial limitations. This area of possible adaptive behaviour and fuel poverty warrants further investigation to fully understand its implications.

Discussion

Indoor CO₂ concentration and temperature

The indoor CO₂ concentration patterns in this study provided critical insights into ventilation practices and occupant behaviour in temporary housing. Houses H1, H4, and H6 consistently exceeded the 1000 ppm threshold, highlighting inadequate ventilation. In H1, CO₂ levels were above 1000 ppm for eight consecutive hours in more than 40% of the monitored period, which is a concerning finding. Although CO₂ itself is not considered toxic at these levels, exposure to concentrations exceeding 2000–5000 ppm has been associated with physiological changes, including headaches and dizziness. These results align with previous studies, such as Chen et al.,⁵² but differ in the observation that average CO₂ concentrations at night were similar to daytime levels, likely due to the absence of night-time ventilation.

Higher CO₂ levels in dining rooms and bedrooms compared to kitchens (Figure 5 and Table 6) could be attributed to the use of electric rather than gas cooking, as well as the spatial layout of the houses, which concentrates occupancy in dining and sleeping areas. This contrasts with Abdel-Salam’s findings in Egypt,⁶⁶ where kitchens exhibited the highest CO₂ levels.

Indoor temperature and thermal comfort

The observed thermal preferences of households underscored the challenges of balancing thermal comfort with adequate ventilation. H1 and H4 maintained temperatures within the Japanese prescribed comfort range⁵⁵ for over 50% of the monitoring period in the dining room. However, high temperatures in these houses were achieved by reducing ventilation rates, a practice consistent with findings by Schellen et al.,⁵⁶ who noted that elderly populations prefer higher ambient temperatures due to reduced basal metabolic rates.

Clothing insulation levels in this study (average 1.26 clo) were higher than those recorded in a previous winter study in Gifu, Japan (0.79 clo).⁶⁷ This suggests an adaptation to colder conditions, possibly exacerbated by limited ventilation.

Occupant behaviour and ventilation

Occupant behaviour played a significant role in influencing IAQ emerging as the primary determinant of poor IAQ rather than the design of the temporary houses themselves. Households generally ventilated their dining rooms at least once daily, following the Japanese government’s recommendations during the pandemic. However, windows and doors remained closed for the majority of the observed period, likely due to low outdoor temperatures.

A clear divergence in behaviour was observed among the households. H2, H3, and H5 prioritised ventilation over thermal comfort, reflecting a greater awareness of its health benefits. In contrast, H1 and H4 prioritised maintaining thermal comfort, leading to higher CO₂ concentrations. These behaviours were strongly influenced by age-specific factors, as elderly occupants exhibited a preference for higher ambient temperatures and lower ventilation rates, aligning with findings from Schellen et al.⁵⁶ regarding the reduced basal metabolic rates in older adults.

This study highlights the variability in behavioural responses, even within a small sample size, and underscores the importance of considering occupant behaviour when addressing IAQ. The integration of CO₂ monitoring with survey-based behavioural data provided a novel methodological approach, allowing for the identification of behavioural drivers of IAQ. This methodology, while initially applied to a specific population group, is scalable and can be adapted for broader studies.

The insights gained from this study lay the foundation for future research that incorporates behavioural variability into IAQ modelling. Stratified sampling across different age groups and housing contexts would help generalise findings and identify effective interventions. Furthermore, targeted campaigns to raise awareness of the health benefits of ventilation, coupled with behavioural nudges, could be instrumental in improving IAQ, particularly for vulnerable populations such as the elderly.

Ventilation rates and COVID-19 transmission

The average ACH in the monitored houses (0.5–0.6) was lower than recommended by the Japanese government and international standards (e.g., EN 16798-1:2019, ISO 17772-1:2017). Modelled infection risks indicated that houses H4–H6 were within the high-risk band for COVID-19 transmission (Figure 13). These findings emphasise the urgent need for ventilation strategies tailored to temporary housing, particularly for vulnerable populations such as the elderly, who spend up to 22 hours indoors daily.

Implications for design and policy

This study highlights the need for improved building designs that balance the use of natural ventilation, IAQ and thermal comfort. Houses with higher CO₂ concentrations (e.g., H1, H4, H6) highlighted the need for improved ventilation strategies, particularly during winter, when windows and doors remained closed to retain heat.

Although fabric use was not observed in the Sagara houses, the recommendation for better building materials reflects the broader applicability of this study’s findings to other temporary housing settings. Enhanced thermal insulation through improved materials could mitigate the trade-offs between IAQ and thermal comfort, benefiting occupants across diverse contexts. A low U-value building envelopes is suggested to enable higher ventilation rates without significant heat loss, providing a balance between IAQ and thermal comfort.

The behavioural patterns observed in the case study also emphasised the importance of occupant education and awareness. For instance, households prioritising thermal comfort over ventilation (H1, H4) faced higher IAQ risks, while others (H2, H3, H5) demonstrated better IAQ but often at the cost of thermal comfort. Educational campaigns could promote ventilation practices that optimise both parameters, ensuring healthier indoor environments.

Current ventilation standards should also be revisited to incorporate airborne infection control measures, particularly in settings housing vulnerable populations.

Methodology

The study used survey-based data to complement CO₂ monitoring, which provided insights into occupant behaviour. This combination allowed for a nuanced understanding of how specific actions, such as opening windows or using heating systems. However, survey data introduced potential biases, including recall inaccuracies, as participants may have struggled to accurately remember their actions, and social desirability effects, where respondents may have reported behaviours they perceived as desirable rather than actual practices.

Additionally, the Wells-Riley model, while validated in numerous studies, relied on simplifying assumptions such as uniform air mixing and fixed parameters. This limited its precision in complex environments like temporary shelters, where airflow patterns, localised ventilation effects, and deposition mechanisms could vary significantly. For instance, the model did not account for the potential impact of occupant clustering in specific zones, which may have increased localised transmission risks. Furthermore, the use of fixed quantum generation rates for elderly participants did not fully capture individual variability in health conditions, activity levels, or susceptibility to airborne diseases.

Despite these limitations, this study demonstrated the feasibility of integrating low-cost monitoring and survey methods to assess IAQ and occupant behaviour in displacement settings. However, limitations, including the short monitoring period and small sample size, indicate the need for future studies to extend these findings. We recommend conducting monitoring across multiple seasons to better capture seasonal variability and increasing the sample size to allow for statistical generalisation.

Limitations and future directions

This study acknowledges several limitations. First, the monitoring duration was limited to 1 week during the winter season. This timeframe may not fully capture the variability in indoor air quality and occupant behaviour across seasons, particularly during periods of higher natural ventilation use, such as summer. Future studies should aim to extend the monitoring period to include multiple seasons to gain a more comprehensive understanding of these dynamics.

Second, the small sample size of six households may be considered a limitation. However, this study is the first to monitor occupant behaviour alongside indoor CO₂ levels and calculate ventilation rates in shelters for displaced populations. Despite the small sample size, a high variability in occupant behaviour patterns was observed, which is both promising and insightful. This variability highlights the importance of understanding and monitoring occupant behaviour in this context, as it can have a significant impact on indoor air quality and ventilation. This study collected TSV, OC, TP, and Perceived IAQ data from this small sample of occupants. While the sample size was not sufficient for statistically significant numerical analysis, it provided valuable qualitative insights into occupant experiences and behavioural tendencies. These data, combined with environmental monitoring, offer the first perspective on the interaction between IAQ, thermal comfort, and occupant behaviour in a resource-limited setting.

Then, while the Wells-Riley model simplifies complex dynamics, its integration with occupant behaviour data allowed for a realistic appraisal of infection risks in temporary housing. These findings provide a foundation for exploring how ventilation practices and space design impact airborne disease transmission in resource-limited settings.

The findings of this study can serve as a model for future research in larger camps. While the current sample size limits statistical generalization, the results demonstrate the value of starting with pilot studies like this one. Pilot studies such as this are crucial for informing the development of more comprehensive research methodologies and determining appropriate sample sizes using statistical techniques. Given the lack of existing research on occupant behaviour in shelters for displaced populations, this study contributes a foundational understanding that can guide future investigations and ultimately improve indoor air quality through better-informed design strategies.

This study highlighted significant age-related behavioural patterns influencing IAQ in temporary shelters, such as the prioritisation of thermal comfort over ventilation. However, it did not attempt to isolate the causal effects of age on perceived IAQ outcomes within the elderly sample. The small sample size and specific context of temporary housing introduced confounding factors, including personal health, cultural practices, and shelter design, which interact with age to shape occupant behaviour and perceptions.

Future research should consider larger, stratified samples that include a broader range of age groups to better isolate the effects of age on IAQ perceptions. Experimental designs and advanced statistical methods, such as multivariate regression or structural equation modelling, could help disentangle these effects and provide a more comprehensive understanding of how age influences IAQ outcomes in diverse housing contexts.

Finally, while the findings are specific to the temporary housing setting in Sagara, Japan, the corresponding climatic conditions, and architectural designs, they highlight critical areas for consideration in other settings. Differences in climate, architectural design, and occupant behaviour in other regions may lead to different outcomes. Therefore, further studies across diverse climates and architectural typologies are essential to validate and extend these findings.

Conclusions

Natural ventilation and its relationship with indoor air quality and thermal comfort is an aspect largely overlooked in the shelter design process. However, the recent COVID-19 pandemic highlighted the importance of ventilation for healthy indoor spaces. Assessing IAQ in temporary houses has many implications for the health of the households, especially for an elderly displaced population, which is more at risk of developing airborne disease. The study aimed to develop a method for the monitoring of behaviour in shelters and the investigation of IAQ and aimed to define a COVID-19 transmission risk model based on ventilation rates and the indoor concentration of CO₂.

This paper is the first study investigating the role of natural ventilation as a determinant of IAQ and thermal comfort in temporary houses for elderly displaced people. Six temporary houses located in the south of Japan were monitored for 7 days. Indoor levels of CO₂ and comfort parameters such as indoor temperature and relative humidity were measured. For the first time in the field of temporary housing or shelter design, daily tables were used to investigate the occupant behaviour of the displaced population. A COVID-19 risk assessment was also used for the first time in a displacement setting.

From the results, the following conclusions can be drawn:

- The indoor concentration of CO₂ in half of the monitored temporary houses is above the 1000 ppm threshold for 70% of the time.

- Calculated ventilation rates are, on average 3.2 ls⁻¹p for an occupancy of 2 people, with a minimum of 1.5 ls⁻¹p. This is much lower than the 8.3 ls⁻¹p prescribed by the Japanese government to reduce the COVID-19 transmission risk.

- The COVID-19 transmission risk, associated with the lowest calculated ventilation, is 61% for an occupancy of two people and increases to 90% for an occupancy of five people (i.e. when visitors present).

- There may be a tension between indoor thermal comfort, the use of natural ventilation, and fuel poverty, as, more often than not, the occupants have to choose between a comfortable thermal environment or a well-ventilated environment.

This work clearly shows the benefit of assessing indoor air quality and disease transmission risk and their driving forces using an assessment method that includes occupant behaviour. It also shows that it is possible to create a practical, low-cost method that is suitable for application by humanitarian staff and that does not rely on controversial and expensive behavioural sensors. It is hoped that the use of the method in other camp settings will improve shelter design worldwide.

Supplemental Material

Supplemental Material - The influence of occupant behaviour on indoor air quality and COVID-19 risk in refugee shelters and temporary houses

Supplemental Material for The influence of occupant behaviour on indoor air quality and COVID-19 risk in refugee shelters and temporary houses by Anna Conzatti, Daniel Fosas, Tristan Kershaw, Takashi Nakaya, David Coley and Hom Bahadur Rijal in Building Services Engineering Research & Technology.

Footnotes

Acknowledgments

This research was conducted thanks to the collaboration of the Tokyo City University (TCU) in Japan. The authors particularly thank Miho Okuyama for facilitating the data collection. The authors also express their gratitude to all the families surveyed and all those involved, including, Dr Naja Aqilah, Dr Supriya Khadka and Dr Mishan Shrestha and Dr Rita Thapa.

Declaration of conflicting interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Sasakawa Foundation [grant number 6010]. Anna Conzatti appreciates the support of the McIntyre Scholarship in Healthy Housing.

ORCID iDs

Anna Conzatti

Daniel Fosas

Tristan Kershaw

Data Availability Statement

The data used to support the findings of this study have been deposited in the Bath University research data archive and can be downloaded from .

Supplemental Material

Supplemental material for this article is available online. Supplementary materials are available in a separate document. These materials contain the household questionnaire, the house survey, and the temporary houses working sheet.

Appendix

References

UNHCR . Global Trends report 2023. Accessible at https://www.unhcr.org/global-trends-report-2023.

UNHCR . UNCHR figure at glance. UNCHR Figure at Glance 2023. https://www.unhcr.org/uk/about-unhcr/who-we-are/figures-glance (accessed 15 July 2023).

Panosetti Pascal , Post disaster shelter ten designs IFRC - 2013, Shelter Cluster, 2013. https://sheltercluster.org/technical-toolkit/documents/post-disaster-shelter-ten-designs-ifrc-2013

International Rescue Committee . Poor Shelter Conditions: Threats to Health, Dignity and Safety. 2017; 3. https://www.rescue.org/sites/default/files/document/1664/ircsittweshelterbriefupdated.pdf.

Webb

Sheffield

Flinn

. Towards healthier homes in humanitarian settings 2020. Available online https://insights.careinternational.org.uk/publications/towards-healthier-homes-in-humanitarian-settings

Albadra

Kuchai

Acevedo-De-los-Ríos

, et al. Measurement and analysis of air quality in temporary shelters on three continents. Building and Environment. 2020; 185: 1–15. https://doi.org/10.1016/j.buildenv.2020.107259

Banerjee

. Investigating the age of air in rural Indian kitchens for sustainable built-environment design. Journal of Building Engineering. 2016; 7: 320–333. https://doi.org/10.1016/j.jobe.2016.07.011

Lam

Brant

, et al. Modeling indoor air pollution from cookstove emissions in developing countries using a Monte Carlo single-box model. Atmospheric Environment. 2011; 45: 3237–3243. https://doi.org/10.1016/j.atmosenv.2011.03.044

Thomas

SDM

Long

, et al. Populations at special health risk: displaced populations. In: Reference module in biomedical sciences. Amsterdam, Netherlands: Elsevier, 2023, p. B9780323999670001101.

10.

Zhou

Cheng

. Characterization of emissions from kerosene heaters in an unvented tent. Aerosol Sci Technol. 2000; 33: 510–514. https://doi.org/10.1080/02786820050195359

11.

Almeida-Silva

Almeida

Pegas

, et al. Exposure and dose assessment to particle components among an elderly population. Atmos Environ. 2015; 102: 156–166. https://doi.org/10.1016/j.atmosenv.2014.11.063

12.

Tanaka

Munaka

. Influence of human behavior on indoor air quality in a care facility for the elderly in Japan. Asian J Atmos Environ. 2021; 15: 2021089. https://doi.org/10.5572/ajae.2021.089

13.

Agarwal

Meena

Raj

, et al. Indoor air quality improvement in COVID-19 pandemic: review. Sustain Cities Soc. 2021; 70: 102942. https://doi.org/10.1016/j.scs.2021.102942

14.

Noorimotlagh

Jaafarzadeh

Martínez

, et al. A systematic review of possible airborne transmission of the COVID-19 virus (SARS-CoV-2) in the indoor air environment. Environ Res. 2021; 193: 110612. doi:10.1016/j.envres.2020.110612.

15.

WHO . Household air pollution attributable deaths. Geneva, Switzerland: World Health Organization, 2023. https://www.who.int/data/gho/data/indicators/indicator-details/GHO/household-air-pollution-attributable-deaths.

16.

Truelove

Abrahim

Altare

, et al. The potential impact of COVID-19 in refugee camps in Bangladesh and beyond: a modeling study. PLoS Med. 2020; 17: e1003144. https://doi.org/10.1371/journal.pmed.1003144

17.

Liu

Chen

Lin

, et al. Clinical features of COVID-19 in elderly patients: a comparison with young and middle-aged patients. J Infect. 2020; 80: e14-18. https://doi.org/10.1016/j.jinf.2020.03.005

18.

UKHSA . Surveillance of influenza and other seasonal respiratory viruses in the UK, winter 2022 to 2023. London: UK Health Security Agency, 2023. https://www.gov.uk/government/statistics/annual-flu-reports/surveillance-of-influenza-and-other-seasonal-respiratory-viruses-in-the-uk-winter-2022-to-2023.

19.

Motamedi

Shirzadi

Tominaga

, et al. CFD modeling of airborne pathogen transmission of COVID-19 in confined spaces under different ventilation strategies. Sustain Cities Soc. 2022; 76: 103397. https://doi.org/10.1016/j.scs.2021.103397

20.

Rudnick

Milton

. Risk of indoor airborne infection transmission estimated from carbon dioxide concentration. Indoor Air. 2003; 13(3): 237–245. https://doi.org/10.1034/j.1600-0668.2003.00189.x

21.

Sze To

Chao

CYH

. Review and comparison between the Wells–Riley and dose-response approaches to risk assessment of infectious respiratory diseases. Indoor Air. 2010; 20(1): 2–16. doi:10.1111/j.1600-0668.2009.00621.x.

22.

Thornton

Fleck

Kroeker

, et al. The impact of heating, ventilation, and air conditioning design features on the transmission of viruses, including the 2019 novel coronavirus: A systematic review of ventilation and coronavirus. Health Systems and Quality Improvement. 2021; 17(10): 1–23. doi:10.1371/journal.pone.0275654.

23.

Zemouri

Awad

Volgenant

CMC

, et al. Modeling of the transmission of coronaviruses, measles virus, influenza virus, Mycobacterium tuberculosis, and Legionella pneumophila in dental clinics. J Dent Res. 2020; 99(10): 1192–1198. doi:10.1177/0022034520940288.

24.

Liao

Chang

Liang

. A probabilistic transmission dynamic model to assess indoor airborne infection risks. Risk Anal. 2005; 25(5): 1097–1107. doi:10.1111/j.1539-6924.2005.00663.x.

25.

ASHRAE . ANSI/ASHRAE standard 62.2-2019 - ventilation and acceptable indoor air quality in residential. ASHRAE. 2019. https://www.ashrae.org/file002520library/technical002520resources/standards002520and002520guidelines/standards002520addenda/62_2_2019_h_20220429.pdf.

26.

Buonomano

Forzano

Giuzio

, et al. New ventilation design criteria for energy sustainability and indoor air quality in a post Covid-19 scenario. Renew Sustain Energy Rev. 2023; 182: 113378. https://doi.org/10.1016/j.rser.2023.113378

27.

Félix

Branco

Feio

. Temporary housing after disasters: a state of the art survey. Habitat Int. 2013; 40: 136–141. https://doi.org/10.1016/j.habitatint.2013.03.006

28.

Yanagi

Yoshino

Hasegawa

, et al. A study on indoor air environment of temporary houses after the Great East Japan Calamity. J Environ Eng (New York) 2013; 78: 917–921.

29.

Shinohara

Tokumura

Kazama

, et al. Indoor air quality, air exchange rates, and radioactivity in new built temporary houses following the Great East Japan earthquake in Minamisoma, Fukushima. Indoor Air. 2013; 23(4): 332–341. https://doi.org/10.1111/ina.12029

30.

Shinohara

Tokumura

Kazama

, et al. Indoor air quality and thermal comfort in temporary houses occupied after the Great East Japan Earthquake. Indoor Air. 2014; 24(4): 425–437. https://doi.org/10.1111/ina.12082

31.

Chen

Zhang

Yoshino

, et al. Winter indoor environment of elderly households: A case of rural regions in northeast and southeast China. Build Environ. 2019; 165. https://doi.org/10.1016/j.buildenv.2019.106388

32.

Kagawa

. Indoor air quality standardsand regulations in Japan. Indoor Built Environ. 1993; 2(4): 223–231. https://doi.org/10.1177/1420326X9300200407

33.

Thomas

Calderón

Senick

, et al. Investigation of indoor air quality determinants in a field study using three different data streams. Build Environ. 2019; 154: 281–295. https://doi.org/10.1016/j.buildenv.2019.03.022

34.

Liu

Wen

Huang Yau

, et al. Rethinking the limitations of research on occupants’ window-opening behavior: A review. Energy and Buildings. 2022; 277: 1–18. https://doi.org/10.1016/j.enbuild.2022.112552

35.

SAMSHA . Greater impact: how disasters affect people of low socioeconomic. Disaster Technical Assistance Center Supplemental Research Bulletin. 2017. https://www.samhsa.gov/sites/default/files/dtac/srb-low-ses_2.pdf.

36.

Fosas

Albadra

Natarajan

, et al. Refugee housing through cyclic design. Archit Sci Rev. 2018; 61(5): 327–337. https://doi.org/10.1080/00038628.2018.1502155

37.

Wang

Norbäck

. A systematic review of associations between energy use, fuel poverty, energy efficiency improvements and health. IJERPH. 2022; 19(12): 7393. https://doi.org/10.3390/ijerph19127393

38.

Bienvenido-Huertas

Sánchez-García

Rubio-Bellido

. Analysing natural ventilation to reduce the cooling energy consumption and the fuel poverty of social dwellings in coastal zones. Applied Energy. 2020; 279: 1–23. https://doi.org/10.1016/j.apenergy.2020.115845

39.

Burlinson

Giulietti

Battisti

. The elephant in the energy room: Establishing the nexus between housing poverty and fuel poverty. Energy Econ. 2018; 72: 135–144. https://doi.org/10.1016/j.eneco.2018.03.036

40.

Albadra

Elamin

Adeyeye

, et al. Participatory design in refugee camps: Comparison of different methods and visualization tools. Build Res Inf. 2020; 49(2): 248–264. https://doi.org/10.1080/09613218.2020.1740578

41.

ASHRAE . ASTM and ASHRAE standards for the assessment of indoor air quality. Handbook of indoor air quality. Singapore: Springer Singapore, 2021, pp. 1–36. https://cfpub.epa.gov/si/si_public_record_Report.cfm?dirEntryId=352887&Lab=CEMM.

42.

CIBSE.

TM68 Monitoring indoor environmental quality 2022. ISBN9781914543210

https://www.cibse.org/knowledge-research/knowledge-portal/tm68-indoor-environmental-quality-2022#:~:text=This%20Technical%20Memorandum%20aims%20to,properly%20manage%20the%20data%20flows

43.

Shrestha

Rijal

Kayo

, et al. An investigation on CO₂ concentration based on field survey and simulation in naturally ventilated Nepalese school buildings during summer. Build Environ. 2022; 207: 108405. https://doi.org/10.1016/j.buildenv.2021.108405

44.

Azuma

Uchiyama

Ikeda

. The regulations for indoor air pollution in Japan: A public health perspective. Journal of Risk Research. 2008; 11: 301–314. https://doi.org/10.1080/13669870801967119

45.

ASHRAE. ASHRAE position document on indoor carbon dioxide 2022. https://www.ashrae.org/file%20library/about/position%20documents/pd_indoorcarbondioxide_2022.pdf

46.

Rijal

Humphreys

Nicol

. Adaptive model and the adaptive mechanisms for thermal comfort in Japanese dwellings. Energy Build. 2019; 202: 1097371. https://doi.org/10.1016/j.enbuild.2019.109371

47.

Jia

Srinivasan

Raheem

. From occupancy to occupant behavior: An analytical survey of data acquisition technologies, modeling methodologies and simulation coupling mechanisms for building energy efficiency. Renew Sustain Energy Rev. 2017; 68: 525–540. https://doi.org/10.1016/j.rser.2016.10.011

48.

Hong

Yan

D’Oca

, et al. Ten questions concerning occupant behavior in buildings: The big picture. Build Environ. 2017; 114: 518–530. https://doi.org/10.1016/j.buildenv.2016.12.006

49.

Wagner

Brien William

O’

Dong

. Exploring occupant behaviotr in buildings. Salmon Tower Building, New York City: Springer, 2018.

50.

CEN En Standard. 16798–1. Energy performance of buildings—Ventilation for buildings—Part 1: Indoor environmental input parameters for design and assessment of energy performance of buildings addressing indoor air quality 2019. https://www.iso.org/standard/60498.html#:~:text=ISO%2017772%2D1%3A2017%20specifies,design%20and%20energy%20performance%20calculations

51.

ISO Standard. ISO Standard. 17772-1:2017 Energy performance of buildings - indoor environmental quality 2017. https://www.iso.org/standard/60498.html#:~:text=ISO17772%2D1%3A2017specifies,designandenergyperformancecalculations.

52.

Chen

Zhao

, et al. Makeshift hospitals for COVID-19 patients: where health-care workers and patients need sufficient ventilation for more protection. Journal of Hospital Infection. 2020; 105(1): 98–99. doi:10.1016/j.jhin.2020.03.008.

53.

Zhang

Lin

. Dilution-based evaluation of airborne infection risk - thorough expansion of Wells-Riley model. Build Environ. 2021; 194: 107674. https://doi.org/10.1016/j.buildenv.2021.107674

54.

Buonanno

Stabile

Morawska

. Estimation of airborne viral emission: quanta emission rate of SARS-CoV-2 for infection risk assessment. Environment International. 2020; 141: 105794. https://doi.org/10.1016/j.envint.2020.105794

55.

JMA (The Japan Meteorological Agency). Climate change monitoring report in 2021 2022. https://www.jma.go.jp/jma/en/NMHS/ccmr/ccmr2022.pdf

56.

Schellen

Van Marken Lichtenbelt

Loomans

MGLC

, et al. Differences between young adults and elderly in thermal comfort, productivity, and thermal physiology in response to a moderate temperature drift and a steady-state condition: Differences between young adults and elderly in thermal comfort, productivity, and thermal physiology. Indoor Air. 2010; 20(4): 273–283. https://doi.org/10.1111/j.1600-0668.2010.00657.x

57.

Wang

Hou

Yao

, et al. Research on indoor thermal environment evaluation and thermal adaptation in winter of Japanese wood-framed detached houses. Case Stud Therm Eng. 2024; 55: 104126. https://doi.org/10.1016/j.csite.2024.104126

58.

Hattori

Iwamatsu

Miura

, et al. Investigation of indoor air quality in residential buildings by measuring CO2 concentration and a questionnaire survey. Sensors. 2022; 22: 7331. https://doi.org/10.3390/s22197331

59.

Achutan

King

Adebayo

, et al. Niosh health hazard evaluation report. Hurricane Katrina response, 2007. https://www.cdc.gov/niosh/hhe/reports/pdfs/2005-0369-3034.pdf.

60.

Shinohara

Kataoka

Takamine

, et al. Distribution and variability of the 24-h average air exchange rates and interzonal flow rates in 26 Japanese residences in 5 seasons. Atmos Environ. 2011; 45: 3548–3552. https://doi.org/10.1016/j.atmosenv.2011.04.005

61.

Hayashi

Yanagi

Honma

, et al. Ventilation methods against indoor aerosol infection of COVID-19 in Japan. Atmosphere. 2023; 14: 150. doi:10.3390/atmos14010150.

62.

Dodd

Donatello

Cordella

. Level(s) indicator 4.1: Indoor air quality. Spain: European Commission, 2020. https://susproc.jrc.ec.europa.eu/product-bureau/sites/default/files/2020-10/20201013%20New%20Level(s)%20documentation_Indicator%204.1_Publication%20v1.0.pdf.

63.

Kurnitski

Kiil

Mikola

, et al. Post-COVID ventilation design: infection risk-based target ventilation rates and point source ventilation effectiveness. Energy Build. 2023; 296(74): 113386. doi:10.1016/j.enbuild.2023.113386.

64.

Morawska

Allen

Bahnfleth

, et al. A paradigm shift to combat indoor respiratory infection. Science. 2021; 372: 689–691. https://doi.org/10.1126/science.abg2025

65.

Taylor

Crabbe

Humphreys

, et al. Development and use of a directed acyclic graph (DAG) for conceptual framework and study protocol development exploring relationships between dwelling characteristics and household transmission of COVID-19 – England, 2020. Build Environ. 2024; 250: 111145. https://doi.org/10.1016/j.buildenv.2023.111145

66.

Abdel-Salam

MMM

. Indoor exposure of elderly to air pollutants in residential buildings in Alexandria, Egypt. Build Environ. 2022; 219: 109221. https://doi.org/10.1016/j.buildenv.2022.109221

67.

Watanabe

Rijal

Nakaya

. Investigation of clothing insulation and thermal comfort in Japanese houses. PLEA2013 - 29th conference, sustainable architecture for a renewable future. Munich, Germany, September 2013, pp. 10–12.

68.

NCAR

RDA. JRA-

55: Japanese 55-year Reanalysis, Monthly Means and Variances 2023. https://doi.org/10.5065/D60G3H5B

69.

Ministry of the Environment. Sagara . Environmental Statistics. 2023. https://www.env.go.jp/doc/toukei/tokeisyu.html.

70.

Fosas

Mitchell

Nikolaidou

Roberts

Allen

Walker

, et al. Novel super reduced, pedagogical model for scoping net zero buildings. Building and Environment. 2022; 208: 108570. https://doi.org/10.1016/j.buildenv.2021.108570

71.

Climate One Building 2023. climate.onebuilding.org.

72.

Evans

Shui

Takagi

. Country Report on Building Energy Codes in Japan. 2009. https://doi.org/10.2172/978545

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

5.12 MB

The influence of occupant behaviour on indoor air quality and COVID-19 risk in refugee shelters and temporary houses

Abstract

Practical application

Keywords

Introduction

Materials and methods

The temporary houses

Measurements of CO2 concentration, indoor and outdoor temperature, and relative humidity

Assessment of indoor CO2

Occupant behaviour survey

Calculation of ventilation rate from indoor CO2

Linear regression models of COVID-19

Methodology novelty

Results

Indoor CO2 concentration

Indoor air temperature

Indoor CO2 concentration and indoor temperature

Indoor air quality and thermal comfort

Occupant behaviour

Ventilation and COVID-19

Additional observations

Discussion

Indoor CO2 concentration and temperature

Indoor temperature and thermal comfort

Occupant behaviour and ventilation

Ventilation rates and COVID-19 transmission

Implications for design and policy

Methodology

Limitations and future directions

Conclusions

Supplemental Material

Supplemental Material - The influence of occupant behaviour on indoor air quality and COVID-19 risk in refugee shelters and temporary houses

Footnotes

Acknowledgments

Declaration of conflicting interest

Funding

ORCID iDs

Data Availability Statement

Supplemental Material

Appendix

References

Supplementary Material

Measurements of CO₂ concentration, indoor and outdoor temperature, and relative humidity

Assessment of indoor CO₂

Calculation of ventilation rate from indoor CO₂

Indoor CO₂ concentration

Indoor CO₂ concentration and indoor temperature

Indoor CO₂ concentration and temperature