Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration

Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration Vasileios Salamalikis, Amirhossein Hassani, Paweł Zawadzki, Sebastian Bykuć, and 1 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4618450/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 11 You are reading this latest preprint version Abstract Indoor air pollution poses a significant environmental concern, leading to adverse health effects. Fine particulates (PM 2.5 ) observed indoors exhibit high variability, influenced by both indoor emission sources and the infiltration of outdoor particles through open spaces and the incomplete building insulation. This study examines the relationship between indoor and outdoor PM 2.5 levels using data from a network of citizen-operated low-cost air quality sensors, deployed in Legionowo, Poland. Our results showed that generally, indoor PM 2.5 was lower than outdoor levels, with occasional peaks attributed to potential indoor emission sources. Statistical analysis identified emission events, particularly during cooking and household-heating periods, occurring more frequently from October to April. In the absence of indoor sources, outdoor particles accounted for 29–75% of indoor particle concentration, highlighting the significance of infiltration. This study shows how citizen-generated data using low-cost sensors, after post-processing, can provide decision-ready information as for example outdoor particulate matter infiltration factors for each building. This information can help decision-makers in devising effective interventions such as prioritizing indoor ventilation or addressing outdoor pollution sources. Earth and environmental sciences/Climate sciences/Atmospheric science Earth and environmental sciences/Environmental sciences/Environmental impact Indoor Air Quality Particulate Matter Indoor-to-Outdoor Ratio Infiltration Factor Low-Cost Sensor Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Household air pollution accounts for three million deaths worldwide according to the World Health Organization (WHO) 1 . Fine particulate matter (PM 2.5 ) — particulates with an aerodynamic diameter < 2.5 µm — is a major atmospheric pollutant attributed to both outdoor and indoor sources. Due to their small sizes, fine particles suspended to the air can easily penetrate into the respiratory system through inhalation, causing adverse health effects 2 . Therefore, assessing the indoor air pollutant concentrations and human exposure to indoor air pollution is crucial, considering that people typically spend from 65% to more than 90% of their time in indoor environments 3,4 . Indoor particulate matter (PM) is regarded as a mixture of: particles generated by indoor sources and outdoor generated particles that have infiltrated into the indoor environment. Due to different origins, such particles exhibit diverse chemical compositions, toxicities, and sizes. When indoor sources are absent or not significant, particles with outdoor origin potentially affect the indoor air quality. Outdoor particles can enter a building through open doors and windows, but also from porous media and cracks of the building shell 5–10 . The presence of potential indoor emission sources coupled with the absent or inadequate ventilation could lead to higher indoor PM 2.5 concentrations compared to outdoor levels. Therefore, for a better understanding of indoor air quality, it’s essential to understand the ratio at which outdoor PM affects indoor conditions. Cooking, smoking, candles and wood burning appliances are mainly responsible for indoor fine PM, while coarser particles (PM 2.5 – PM 10 ) are typically the result of non-combustion activities like cleaning and particle resuspension due to residents’ movements 11,12 . The initial step to investigate the connection between indoor and outdoor concentrations involves the calculation of Indoor-to-Outdoor concentration ratio (I/O). The I/O ratio varies dynamically throughout the day and across seasons 10 . A ratio greater than unity shows the dominance of indoor emission sources. In general, since I/O depends on various factors and processes, it cannot solely offer reliable information about the relationship between indoor and outdoor air quality 8 unless additional information on ventilation and the frequency and duration of window openings is available 13 . A detailed overview of I/O is provided in 8 and 14 , showing a large range of values as explained by the different types of buildings, emission sources, monitoring periods, type of ventilation, etc. 8 . The impact of outdoor-generated particles on indoor air quality is mainly represented by the infiltration factor (f inf ) which quantifies the proportion of ambient PM entering indoor environments and remaining suspended under steady state conditions. The f inf depends on building characteristics and operation, indoor and outdoor meteorological conditions, season and time of the day, and particle size 4,10,12,15–24 , while socioeconomic and demographic factors may also have indirect influences 18,23 . Various methodologies have been evaluated to estimate f inf 13 . The simplest approach when both information about indoor and outdoor air pollutants is available involves regressing indoor concentration on outdoor concentrations. The gain and the offset of the linear relationship act as the infiltration factor and the average concentration of the indoor-generated particles, respectively 25–27 . However, this approach estimates a “global“ f inf and indoor-generated concentration neglecting the variability of both variables in time and their dependencies on auxiliary factors. The relationship between indoor and outdoor concentration levels was also statistically addressed using distributed-lag regression models 22,23,28 . These models use indoor concentrations as an independent parameter and outdoor PM levels in several lags combined with meteorological parameters or other auxiliary factors as independent information. The f inf was derived by summing the regression coefficients of the lagged outdoor PM levels. Using this approach, Krebs et al. 22 estimated f inf from 40–70% in the state of California, USA. A large extent of f inf in single-family homes across USA was also estimated by Burke et al. 23 , where f inf relationship with the household income and the outdoor concentration levels was also investigated. Some other studies have calculated f inf using the average I/O ratio over long-time intervals (e.g., day and week) excluding the extreme indoor emission sessions 15,17,18,24 . In those studies, it is assumed that the indoor concentrations closely mirror the corresponding outdoor concentrations when indoor emissions are minimal. Indoor emission-related PM concentrations typically exhibit abrupt, high-magnitude changes, and can be identified using rule-based approaches. Comparisons between indoor and outdoor concentrations, along with their time derivatives (consecutive concentration changes in time) help the detection of indoor emission events. Usually, higher values of f inf are observed in summer compared to winter because of increased window opening frequency, with values varying depending on the building type. This study explores the relationship between indoor and outdoor PM 2.5 concentrations in typical buildings of Legionowo, Poland. This is accomplished by integrating meteorological and PM 2.5 data from low-cost air quality sensors (LCSs) operated by citizens and installed in various locations across the city. In recent years, the use of LCSs for indoor air quality monitoring has risen. They not only provide information about air pollutant concentrations but also help identify pollution sources 29–32 . Despite challenges, particularly concerning data reliability compared to regulatory-grade instruments 33 , LCSs provide insights into both indoor and outdoor air quality levels 34 . This paper shows how data collected by citizens can be converted to decision-ready information. The ultimate goals are to explore the linkage between indoor and outdoor PM 2.5 , to estimate the proportion of indoor PM 2.5 that originates from infiltration from the outdoor environment and to understand the factors affecting indoor PM 2.5 variability, such as indoor emissions and f inf . The identified f inf values for each building provide insights for policymakers. For instance, low f inf indicates that dominant sources of pollution are indoor, suggesting that more interventions are required to improve indoor air quality, such as enhancing ventilation or regulating indoor heating sources. Conversely, high infiltration values suggest that the building has better insulation, but there may be a need to address outdoor pollution sources. 2. Data 2.1. Study area Legionowo is a city in central-eastern Poland (Longitude: 20.9369°E, Latitude: 52.4012°N), approximately 23 km north of Warsaw (Figure S1 a). It covers a surface of 13.5 km 2 featuring a relatively flat terrain, predominantly surrounded by forests. The highest elevation in the area is around 80 km with no notable hills or mountainous ranges nearby. The population in Legionowo area is ~ 54170 inhabitants with a population density of 4000 people km –2 ( https://pl.wikipedia.org/wiki/Legionowo , assessed on 06.05.2022). The climate is classified as temperate oceanic of type Cfb based on the Köppen-Geiger climate classification system. Figure S1 b shows the climograph for Legionowo area. Meteorological data from 1951 to 2013 were obtained from the Polish Institute of Meteorology and Water Management—National Research Institute (IMGW-PIB) via https://dane.impw.pl/ using the “climate” R package 35 (red star in Figure S1 a). The average temperature is 8.3 ° C, with January being the coldest month at − 2.3 ° C and July the warmest at 19.2 ° C. The yearly average precipitation is 537.2 mm per year, with the majority occurring during the summer period (341.1 mm). July stands out as the wettest month with 74.5 mm while the lowest precipitation amounts are recorded in February (27.9 mm) and March (27.1 mm). In Legionowo, based on local experts, multi-family buildings are typically heated using central heating provided by a district heating system. For structures not connected to the central heating network, individual systems are used. Natural gas is the predominant choice for heating, followed by less common options such as hard coal, wood, electricity, and heat pumps. However, wood or coal is used for heating purposes 36 . 2.2. Low-cost sensor network PM 2.5 and basic meteorological parameters are measured using low-cost sensors in various buildings situated in the Legionowo area. Figure 2 shows the spatial distribution of the air quality sensors. Indoor air quality was measured at nine locations while seven outdoor sensors were also installed in the same or nearby buildings. In six out of nine locations, both indoor and outdoor sensors were installed in the same building (yellow triangles in Fig. 1 ). For sensors 5485 and 5736 (red diamonds in Fig. 1 ), information about the outdoor air quality conditions was obtained through sensors installed in nearby buildings (green stars in Fig. 2 ). For 6030, outdoor air quality was determined through the nearest outdoor sensor (6033) which is positioned at a distance of 445 m. Outdoor air quality was assessed using Airly sensors ( https://airly.org/en ). The Airly sensor system incorporates a Plantower PMS5003 sensor ( https://www.plantower.com/en/products_33/74.html ), which uses laser-based light scattering for measuring airborne particulate matter with aerodynamic diameters ranging from 0.3 to 10 µm. PM mass concentrations are provided in hourly resolution in the fractions of PM 1 , PM 2.5 and PM 10 . According to the manufacturer, the measurement range is 0–1000 µg m –3 with an accuracy of ± 1 µg m –3 for PM 2.5 . In addition to PM, the Airly system also records air temperature ( ° C), humidity (%), and air pressure (hPa). For indoor air quality monitoring, AIRTHINGS Cloudberry systems ( https://airthings.com ) were deployed. The sensor system is equipped with an air quality sensor that utilizes laser light scattering technology to measure air particulates. PM 1 and PM 2.5 are recorded within the measurement range of 0 to 1000 µg m –3 . The claimed accuracies include ± 10 µg m –3 for concentrations lower than 100 µg m –3 and ± 10% for concentrations above 100 µg m –3 . The AIRTHINGS Cloudberry kit also includes sensors for measuring air temperature ( ° C), relative humidity (%), air pressure (hPa), and volatile organic compounds (µg m –3 ). It provides data at irregular sub-hourly time intervals. Indoor observations are hourly aggregated for being temporally consistent with outdoor data. Indoor and outdoor air quality data were obtained through the AIRTHINGS dashboard and the Airly Data Platform from April 2022 to December 2023 based on the availability of indoor air quality measurements. 3. Methodology In general, the indoor concentration of an atmospheric pollutant, such as PM 2.5 , can be modeled through a single mass-balance equation (Eq. 1 ) considering that indoor concentration changes mainly depend on a) the infiltration of outdoor particles, b) indoor particles that escape outdoors, c) deposition of indoor particles and d) indoor emissions. Eq. 1 assumes uniform mixing of the air pollutants and negates concentration changes due to gas-phase interactions or variations in the environmental parameters between the indoor and outdoor environments 13 . $$\frac{\text{d}{\text{C}}_{\text{i}\text{n}}\left(\text{t}\right)}{\text{d}\text{t}}=\text{p}{\alpha }{\text{C}}_{\text{o}\text{u}\text{t}}\left(\text{t}\right)-{\alpha }{\text{C}}_{\text{i}\text{n}}\left(\text{t}\right)-\text{k}{\text{C}}_{\text{i}\text{n}}\left(\text{t}\right)+\frac{\text{S}\left(\text{t}\right)}{\text{V}}$$ 1 where dC in is the change in indoor PM 2.5 concentration (µg m –3 ) during the time interval dt, C in and C out are the indoor and outdoor particle’s concentrations (µg m –3 ), t is the time (h), p is the penetration efficiency of particles (dimensionless), α is the air exchange rate (h –1 ), k is the deposition rate of particles (h –1 ), S is the indoor emission rate (µg h –1 ) and V is the volume of the building or the room (m 3 ). Under the assumption that the instantaneous temporal change of indoor concentration is substantially lower than the average indoor concentration over a considerable time interval, the steady-state form of Eq. 1 can be re-arranged as follows: $${\text{C}}_{\text{i}\text{n}}=\frac{\text{p}{\alpha }}{{\alpha }+\text{k}}{\text{C}}_{\text{o}\text{u}\text{t}}+\frac{\text{S}}{\text{V}\left({\alpha }+\text{k}\right)}={\text{f}}_{\text{i}\text{n}\text{f}}{\text{C}}_{\text{o}\text{u}\text{t}}+{\text{C}}_{\text{i}\text{n}, \text{g}\text{e}\text{n}}$$ 2 $$\frac{{\text{C}}_{\text{i}\text{n}}}{{\text{C}}_{\text{o}\text{u}\text{t}}}={\text{f}}_{\text{i}\text{n}\text{f}}+\frac{{\text{C}}_{\text{i}\text{n}, \text{g}\text{e}\text{n}}}{{\text{C}}_{\text{o}\text{u}\text{t}}}$$ 3 with \({\text{f}}_{\text{i}\text{n}\text{f}}=\frac{\text{p}{\alpha }}{{\alpha }+\text{k}}\) the infiltration factor and \({\text{C}}_{\text{i}\text{n}, \text{g}\text{e}\text{n}}=\frac{\text{S}}{\text{V}\left({\alpha }+\text{k}\right)}\) the indoor concentration (µg m –3 ) generated solely by indoor activities. The indoor-to-outdoor concentration ratio converges to the f inf when the outdoor concentration is significantly high or during periods with minimal indoor contribution (Eq. 3 ). The latter requires the detection and the removal of the indoor-generated peaks before the calculation of the infiltration factor. The optimal approach to detect such specific instances or/and periods is to record in detail the indoor activities. In this case, the peaks can also be assigned to the major activities that occurred in the indoor environment. If this kind of information is missing, machine learning and rule-based approaches can be evaluated for peak detection 15,17,18,24,28,37 . Here, the indoor emission cases were identified through analysis of the indoor PM 2.5 series using the Robust Extraction of Baseline Signal (REBS) methodology 38,39 . REBS has been applied for detecting local sources in various air pollutants time series 39–41 . According to Ruckstuhl et al. 39 , the indoor time series can be decomposed as: $${\text{C}}_{\text{i}\text{n}}\left(\text{t}\right)={\text{C}}_{\text{B}}\left(\text{t}\right)+{\text{C}}_{\text{R}}\left(\text{t}\right)+{\epsilon }$$ 4 where C B (t) is the background concentration levels, C R (t) is the concentration due to indoor emissions and other contributions (e.g., outdoor concentration) and ε is the normally distributed and independent errors. The local emissions that are responsible for spikes in time series then can be identified using the REBS method through a two-stage approach. Initially, the background concentration is determined using local linear regression over a moving window of a specific duration. Then, any data points greater than a designated threshold relative to the background concentrations are classified as emissions: $${\text{C}}_{\text{i}\text{n}}\left(\text{t}\right)>\widehat{{\text{C}}_{\text{B}}}\left(\text{t}\right)+{\beta }\times {\sigma }$$ 5 where \(\widehat{{\text{C}}_{\text{B}}}\left(\text{t}\right)\) is the estimated background curve, σ is the standard deviation of the data falling below the background curve. β is a user-defined parameter. β controls the width of the threshold curve with higher values attributing to wider threshold concentrations. Here β is set equal to 3 as initially proposed in 39 . In this study, the REBS method was implemented using the rfbaseline function of the “IDPmisc” R package 42 . Once the peak emission events were detected, the f inf at each timestamp was determined, assuming steady-state conditions after T hours (Section S1 of the supplementary material): $${\text{f}}_{\text{i}\text{n}\text{f}}\left(\text{t}\right)=\frac{{⟨{\text{C}}_{\text{i}\text{n}}\left(\text{t}\right)⟩}_{\text{T}}}{{⟨{\text{C}}_{\text{o}\text{u}\text{t}}\left(\text{t}\right)⟩}_{\text{T}}}$$ 6 with T and T the T-hour rolling averages for the indoor and the outdoor PM 2.5 concentrations, considering only the timestamps that were not identified as indoor emission events and at least 30% data coverage within the T-hour window. A steady state is reached when the instantaneous temporal change of indoor concentration is significantly lower than the average indoor concentration over a considerable time span (Eq. 1 ). Following the methodology outlined in section S1, the hourly concentration change is calculated by the concentration difference of the current and previous timestamp, ΔC in (t) = C in (t) – C in (t–1). Then, running averages at various temporal intervals T are computed for the indoor concentration, T , and the respective changes T . It is expected that – T almost equals to T at increasing temporal intervals (Fig. S6 and S7). Section S1 and Fig. S6 and S7 showed steady state conditions are achieved after 48 hours. Based on this result, f inf is derived through Eq. 6 using 48-hour running averages of indoor and outdoor PM 2.5 . 4. Results 4.1. Indoor and outdoor PM 2.5 concentrations PM 2.5 was simultaneously measured in both indoor and outdoor environments using LCSs from February 2022 to August 2023. The data availability for each sensor is presented in Fig. S2. Regarding the outdoor PM 2.5 time series, most sensors provided continuous measurements with minimal data gaps (Fig. S2b). However, substantial data gaps were present in the indoor PM 2.5 measurements, primarily due to sensor malfunctions (Fig. S2a). Sensors 5189 and 5165 yielded almost complete time series for over five months, while the other sensors exhibited significant data gaps (Fig. S2a). Fig. S2c presents the periods when both indoor and outdoor PM 2.5 recordings were simultaneously available. Notably, sensors 5165 and 6030 were active for less than one month and provided the lowest data availability over the whole deployment period. Since this study focuses on the relationship between indoor and outdoor PM 2.5 levels, only the periods where both sensors were operational were included in the subsequent analysis. Summary of the indoor and outdoor hourly PM 2.5 LCS measurements for the buildings is shown in Fig. 2a. Only timestamps with both indoor and outdoor measurements are used to generate the boxplots. Indoor PM 2.5 concentrations generally appear lower than the outdoor readings. The median ranges for indoor and outdoor PM 2.5 are 1.9–17.3 µg m –3 and 6.7–27.9 µg m –3 , respectively (Table A1). Although significant peaks are present in indoor PM 2.5 time series (outlier points in Fig. 2a), mainly attributed to dominant indoor emission sources, 75% of the data were lower than 28 µg m − 3 . Figure 2b and 2c illustrate the diurnal distribution of average indoor and outdoor PM 2.5 based on the available measurements of the entire period for each sensor. It is important to mention that the diurnal distribution for some sensors was drawn using a limited amount of data. For example, sensor 6030 measures indoor and outdoor PM 2.5 only for a two-month period implying that the diurnal pattern is not representative for the general temporal pattern at this specific location. Analyzing outdoor levels, all sensors exhibited a U-shaped pattern throughout the entire period, with lower concentrations between 09:00 and 18:00 local time (LT). However, analysis of the diurnal profiles on a seasonal basis revealed a different pattern in winter compared to other seasons, primarily displaying a bimodal distribution with morning and evening peaks (Fig. S3). This pattern is commonly observed in urban settings, where the morning peak is typically attributed to traffic, and the elevated concentrations during late afternoon and nighttime hours are due to heating activities and traffic 34,43 . The outdoor concentration remains consistently at high levels throughout the evening period. In winters, the low temperature conditions increase the heating activities while the low boundary layer height also traps the air pollutants close to the ground surface, resulting in relatively high PM 2.5 concentrations 44 . Conversely, the diurnal distribution of indoor PM 2.5 is primarily influenced by indoor activities. Most locations exhibit early morning and early afternoon and nighttime peaks. The sensor 5189 shows similar diurnal patterns for both indoor and outdoor PM 2.5 . This behavior may be because indoor concentration is influenced by outdoor concentrations. It is worth noting that three sensors (5165, 5377, and 6030) were installed in kindergartens. Indoor PM 2.5 shows minimal variation throughout the day mainly because of the absence of significant indoor activities. In kindergartens, indoor emissions primarily stem from particle resuspension due to children's activities, as combustion activities are generally absent 45,46 . The type of ventilation used in indoor environments also affects the observed PM 2.5 levels. Sensor 6033 was installed in a kindergarten with a mechanical ventilation system, whereas other buildings rely on natural ventilation. In this particular location (sensor 6033), intra-day variations of the average indoor PM 2.5 are low ranging between 7.3 and 10.5 µg m –3 . Figure 3a presents the Cumulative Distribution Functions (CDF) for the calculated hourly I/O ratios. On an hourly timescale, I/O exhibited significant variability both within a specific location and between different locations. In most places, 90% of the hourly I/O ratios were below unity implying that indoor PM 2.5 levels mainly reflect the contribution from outdoor particles. However, sensors 5351 and 5736 showed I/O > 1 in over 40% of hourly measurements, suggesting that in these two locations, the contribution and the strength of the indoor emission activities are greater compared to the other sensor locations. CDFs are also drawn based on the building type (kindergarten vs. other). Due to the absence of significant indoor sources in kindergartens, it is expected the diurnal distribution of I/O ratio to be lower than 1. Statistical analysis showed significant differences in the I/O distributions between the two building types (two-sample Kolmogorov-Smirnov test at 95% confidence level: KS statistic = 0.08 and p–value < 1 for kindergartens, while the for the other buildings, timestamps with I/O ratios > 1 are near 20% of the total calculated I/O ratios. Figure 3b shows the diurnal distribution of the I/O ratio. The findings of Fig. 3b align with those discussed in X 10 concerning the diurnal distribution of I/O. Generally, I/O exhibits higher values from 09:00 to 18:00 LT, primarily due to the decrease in outdoor PM 2.5 during this period and increased indoor activities emitting PM 2.5 (Fig. 2c). Locations where the diurnal patterns of indoor and outdoor PM 2.5 closely resemble each other tend to display less variability in I/O throughout the day, as observed in sensors 5189 and 5485. A nearly stable I/O variation is presented in sensor 5189 which might be attributed to the impact of mechanical ventilation within the building and the relatively stable intra-day indoor PM 2.5 levels. Sensor 5736 showed I/O > 1 after 06:00 LT and the discrepancies between the diurnal indoor and outdoor PM 2.5 extended from − 0.1 µg m –3 (I/O = 0.99) to 15 µg m –3 (I/O = 2.9). 4.2. Detection of indoor emission events The detection of elevated indoor PM 2.5 due to intense indoor activities was conducted using the REBS method with an illustrative example provided in Fig. 4 for sensor 5351. As illustrated in Fig. 4a, the REBS method efficiently captures the spikes present in the time series. Overall, 895 out of 8706 hourly data points (~ 10%) are attributed to significant indoor emission events. Figure 4b displays the statistical distributions of the indoor PM 2.5 levels for cases with PM 2.5 > threshold (“Emission”) and PM 2.5 < threshold (“Non-Emission”). The distributions are statistically different (two-sample t-test at 95% confidence level: p–value < < 0.05), with average (± 1 standard deviation) concentrations are 17.2 ± 11.1 µg m –3 and 54.0 ± 27.7 µg m –3 for the “Non-Emission” and “Emission” classes, respectively. The methodology for identifying indoor emission events was applied to all indoor sensors with the results summarized in the boxplots of Fig. 5a. The average (± 1 standard deviation) PM 2.5 for the emission events ranged from 7.5 ± 2.1 µg m –3 (sensor 5165) to 62.4 ± 68.1 µg m –3 (sensor 5736). The ‘Emission’ and ‘Non-Emission’ classes were statistically different among the various sensors (based on the pairwise t-test under the 95% confidence level). Cooking and household-heating activities potentially influence indoor PM 2.5 patterns. Consequently, an increase in the frequency of emission events is anticipated during the periods with high levels of indoor activities. As already mentioned in the data session, the use of coal or wood for residential heating purposes is very common in Poland. In general, Polish households have the highest coal consumption rate across the European Union 47 while strict measures of solid-fuel burning for heating purposes have not widely implemented. As pointed out in 48 , fine PM concentrations shows a strong positive correlation with the number of heating systems in private-homes in Poland . Figure 5c illustrates the diurnal distribution of detected emission events, expressed as a percentage of the total number of indoor LCSs measurements in each hour of day, from October to April (“household-heating” period) and from May to September (“non-household-heating” period). The percentages were calculated using detected emission events from all sensors. During the “non-household-heating” period, the percentage of emission events fluctuates between 4% and 10% of total measured events, with higher values observed in the early morning hours and after 16:00 LT. Assuming that indoor activities like cooking remain consistent during a day/year, the difference between the diurnal curves for the two periods could provide insights about household-heating activity contribution to the number of detected indoor emission events. In the “heating-household” months, people stay more at home, engage in more indoor activities, and the demand for heating increases. During “household-heating” period, nearly 25–27% of the indoor PM 2.5 measurements were identified to have a dominant indoor source of emission after 18:00 LT, indicating a 15% increase compared to the “non-household-heating” period. Similar analyses were conducted at each sensor location separately (Fig. S6). Only three out of nine sensors have available data for both “household-heating” and “non-household-heating” periods. 4.3. Infiltration factor, f inf The box plots in Fig. 6a and the summary statistics in Table A1 present the results of the f inf calculation for various locations in Legionowo. The estimated f inf values ranged from 0 and 1, with average values extending between 0.29 (sensors 5165 and 5189) and 0.75 (sensor 5751). This suggests 29%-75% of indoor measured concentrations are of outdoor origin. The standard deviation of f inf is nearly similar across all locations (0.13–0.16). f inf values are consistent with those reported in the literature 8 . Seasonal analysis of f inf is presented in Fig. 6b. Overall, during the “household-heating” period, f inf was 0.45 ± 0.20, and 0.53 ± 0.22 during the “non-household-heating” period. Significant seasonal differences in the average f inf were also observed for individual locations (based on t-tests at the 95% confidence level), ranging from 0.03 (for sensor 6033) to 0.19 (for sensor 5485). During the “non-household-heating” period, windows are opened more frequently, allowing the infiltration of outdoor particles into the indoor environment. The f inf depends on the penetration efficiency, the deposition rate of the indoor air pollutants, and the air-exchange rate (Eq. 5). In addition to building characteristics and residents’ activities, meteorological conditions may also influence f inf 8,24,49 . In order to assess the relation between meteorology and f inf , the differences between outdoor-indoor temperatures during the “household-heating” and “non-household-heating” periods are plotted against f inf in Fig. S8a an Fig. S8b, respectively. Similar analysis was conducted in Lunderberg et al. 24 . Based on these results, the contribution of outdoor/indoor temperature difference to the f inf is rather marginal, suggesting that this parameter cannot be considered as significant proxy for describing the f inf variations. 5. Conclusions This study investigated the link between indoor and outdoor PM 2.5 levels using a network of citizen-operated LCSs in Legionowo, Poland. Sensors were deployed both indoors and outdoors at nine locations throughout the city, providing data from April 2022 to December 2023. One limitation of this study is the absence of information regarding the daily residents’ activities contributing to indoor PM 2.5 . Nonetheless, simultaneous measurements of indoor and outdoor PM 2.5 helped to detect indoor emission events and estimating proportion of the outdoor particles infiltrating into the buildings. The key findings are outlined as follows: Indoor PM 2.5 levels were significantly lower than the outdoor levels, with median concentration ranging from 1.9 to 17.3 µg m –3 indoors and 6.7 to 27.9 µg m –3 outdoors. Occasional spikes in indoor PM 2.5 were attributed to potential indoor emission sources. The diurnal distribution of outdoor PM 2.5 concentrations exhibited a U-shape pattern throughout the measurement period peaking in the morning and late afternoon to evening hours. Indoor PM 2.5 varied during the day driven by the indoor activities, with evening peaks likely associated with dinner cooking and household heating activities. Sensors installed in kindergartens (5165, 5377 and 6030) showed minimal diurnal PM 2.5 variation, mainly reflecting the influence of children’s activities on indoor particle levels. The I/O concentration ratio was analyzed to understand the relationship between indoor and outdoor PM 2.5 . I/O varied significantly both within a specific location and between different locations, mostly falling below unity. Sensors 5351 and 5736 exhibited significant indoor emissions, with over 40% of hourly measurements showing I/O > 1. Higher I/O ratios were observed from 09:00 to 18:00 LT on a diurnal basis, with less variability in intra-day I/O ratios observed in locations with similar indoor and outdoor PM 2.5 patterns. A statistical approach was used to detect the potential indoor emissions events. The number of emission events was higher during early afternoon and nighttime hours, where activities like dinner cooking and house heating were more prevalent. The emission events were more distinguishable from October to April due to the contribution of household heating activities. The influence of outdoor PM2.5 on indoor air pollution was also quantified in the f inf , showing that 29–75% of indoor measured concentrations originated from outdoor sources. The f inf was higher during warmer months (May – September) with little to no household-heating activities, attributable to more-frequent door and window openings. The contribution of outdoor-indoor temperature difference to the f inf was found minimal. Declarations Data availability: Data supporting the findings of this study will be made available upon reasonable request. Funding: This research leading to these results has been received from the EEA Grants 2014–2021 via the National Center for Research and Development. The grant is provided under “Applied Research” Program for the “GREEN HEAT – towards collaborative local decarbonization” project (acronym GREEN HEAT); grant number PL-Applied Research-0043. Also, this research has received partial funding from the European Union’s Horizon 2020 Research and Innovation programme under grant agreement No 952433, VIDIS project. We also acknowledge funding for SOCIO-BEE project from the European Union’s Horizon Europe research and innovation programme under grant agreement No. 101037648. Acknowledgments: The contribution of all citizens who participated in the GREEN HEAT project is acknowledged. Conflicts of Interest: 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. References WHO. Household air pollution. https://www.who.int/news-room/fact-sheets/detail/household-air-pollution-and-health (2022). Pope, C. A., Coleman, N., Pond, Z. A. & Burnett, R. T. Fine particulate air pollution and human mortality: 25 + years of cohort studies. Environmental Research 183, 108924 (2020). Schweizer, C. et al. Indoor time–microenvironment–activity patterns in seven regions of Europe. J Expo Sci Environ Epidemiol 17, 170–181 (2007). Liu, Y., Ma, H., Zhang, N. & Li, Q. A systematic literature review on indoor PM2.5 concentrations and personal exposure in urban residential buildings. Heliyon 8, e10174 (2022). Liu, D.-L. & Nazaroff, W. W. Modeling pollutant penetration across building envelopes. Atmospheric Environment 35, 4451–4462 (2001). Liu, D.-L. & Nazaroff, W. W. Particle Penetration Through Building Cracks. Aerosol Science and Technology 37, 565–573 (2003). Bennett, D. H. & Koutrakis, P. Determining the infiltration of outdoor particles in the indoor environment using a dynamic model. Journal of Aerosol Science 37, 766–785 (2006). Chen, C. & Zhao, B. Review of relationship between indoor and outdoor particles: I/O ratio, infiltration factor and penetration factor. Atmospheric Environment 45, 275–288 (2011). Liu, C. et al. Influence of natural ventilation rate on indoor PM2.5 deposition. Building and Environment 144, 357–364 (2018). Stamp, S. et al. A critical evaluation of the dynamic nature of indoor-outdoor air quality ratios. Atmospheric Environment 273, 118955 (2022). Isaxon, C. et al. Contribution of indoor-generated particles to residential exposure. Atmospheric Environment 106, 458–466 (2015). Li, J., Fan, G., Ou, Y. & Deng, Q. Characteristics and control strategies of indoor particles: An updated review. Energy and Buildings 294, 113232 (2023). Diapouli, E., Chaloulakou, A. & Koutrakis, P. Estimating the concentration of indoor particles of outdoor origin: A review. Journal of the Air & Waste Management Association 63, 1113–1129 (2013). Ilacqua, V., Scharko, N., Zambrana, J. & Malashock, D. Survey of residential indoor particulate matter measurements 1990–2019. Indoor Air 32, (2022). Allen, R., Larson, T., Sheppard, L., Wallace, L. & Liu, L.-J. S. Use of Real-Time Light Scattering Data To Estimate the Contribution of Infiltrated and Indoor-Generated Particles to Indoor Air. Environ. Sci. Technol. 37, 3484–3492 (2003). Rim, D., Wallace, L. & Persily, A. Infiltration of Outdoor Ultrafine Particles into a Test House. Environ. Sci. Technol. 44, 5908–5913 (2010). Kearney, J. et al. Residential indoor and outdoor ultrafine particles in Windsor, Ontario. Atmospheric Environment 45, 7583–7593 (2011). MacNeill, M. et al. Factors influencing variability in the infiltration of PM2.5 mass and its components. Atmospheric Environment 61, 518–532 (2012). Tang, C. H. et al. Development of a modeling approach to estimate indoor-to-outdoor sulfur ratios and predict indoor PM2.5 and black carbon concentrations for Eastern Massachusetts households. J Expo Sci Environ Epidemiol 28, 125–130 (2018). Vu, T. V. et al. Assessing the contributions of outdoor and indoor sources to air quality in London homes of the SCAMP cohort. Building and Environment 222, 109359 (2022). Liang, Y. et al. Wildfire smoke impacts on indoor air quality assessed using crowdsourced data in California. Proc. Natl. Acad. Sci. U.S.A. 118, e2106478118 (2021). Krebs, B., Burney, J., Zivin, J. G. & Neidell, M. Using Crowd-Sourced Data to Assess the Temporal and Spatial Relationship between Indoor and Outdoor Particulate Matter. Environ. Sci. Technol. 55, 6107–6115 (2021). Burke, M. et al. Exposures and behavioural responses to wildfire smoke. Nat Hum Behav 6, 1351–1361 (2022). Lunderberg, D. M. et al. Assessing residential PM 2.5 concentrations and infiltration factors with high spatiotemporal resolution using crowdsourced sensors. Proc. Natl. Acad. Sci. U.S.A. 120, e2308832120 (2023). Ott, W., Wallace, L. & Mage, D. Predicting Particulate (PM 10 ) Personal Exposure Distributions Using a Random Component Superposition Statistical Model. Journal of the Air & Waste Management Association 50, 1390–1406 (2000). Wallace, L. & Ott, W. Long-Term Indoor-Outdoor PM2.5 Measurements Using PurpleAir Sensors: An Improved Method of Calculating Indoor-Generated and Outdoor-Infiltrated Contributions to Potential Indoor Exposure. Sensors 23, 1160 (2023). Mousavi, A. & Wu, J. Indoor-Generated PM 2.5 During COVID-19 Shutdowns Across California: Application of the PurpleAir Indoor–Outdoor Low-Cost Sensor Network. Environ. Sci. Technol. 55, 5648–5656 (2021). Nguyen, P. D. M. et al. Using Low-Cost Sensors to Assess Fine Particulate Matter Infiltration (PM2.5) during a Wildfire Smoke Episode at a Large Inpatient Healthcare Facility. IJERPH 18, 9811 (2021). Morawska, L. et al. Applications of low-cost sensing technologies for air quality monitoring and exposure assessment: How far have they gone? Environment International 116, 286–299 (2018). Ródenas García, M. et al. Review of low-cost sensors for indoor air quality: Features and applications. Applied Spectroscopy Reviews 57, 747–779 (2022). Bousiotis, D., Alconcel, L.-N. S., Beddows, D. C. S., Harrison, R. M. & Pope, F. D. Monitoring and apportioning sources of indoor air quality using low-cost particulate matter sensors. Environment International 174, 107907 (2023). Higgins, C., Kumar, P. & Morawska, L. Indoor air quality monitoring and source apportionment using low-cost sensors. Environ. Res. Commun. 6, 012001 (2024). Castell, N. et al. Can commercial low-cost sensor platforms contribute to air quality monitoring and exposure estimates? Environment International 99, 293–302 (2017). Hassani, A. et al. Low-cost sensors and Machine Learning aid in identifying environmental factors affecting particulate matter emitted by household heating. Atmospheric Environment 314, 120108 (2023). Czernecki, B., Głogowski, A. & Nowosad, J. Climate: An R Package to Access Free In-Situ Meteorological and Hydrological Datasets For Environmental Assessment. Sustainability 12, 394 (2020). Karpinska, L., Śmiech, S., Gouveia, J. P. & Palma, P. Mapping Regional Vulnerability to Energy Poverty in Poland. Sustainability 13, 10694 (2021). Chan, W. R. et al. Quantifying fine particle emission events from time-resolved measurements: Method description and application to 18 California low‐income apartments. Indoor Air 28, 89–101 (2018). Ruckstuhl, A. F., Jacobson, M. P., Field, R. W. & Dodd, J. A. Baseline subtraction using robust local regression estimation. Journal of Quantitative Spectroscopy and Radiative Transfer 68, 179–193 (2001). Ruckstuhl, A. F. et al. Robust extraction of baseline signal of atmospheric trace species using local regression. Atmos. Meas. Tech. 5, 2613–2624 (2012). et al. Identification of spikes associated with local sources in continuous time series of atmospheric CO, CO 2 and CH 4 Atmos. Meas. Tech. 11, 1599–1614 (2018). Cristofanelli, P. et al. Identification of spikes in continuous ground-based in situ time series of CO 2 , CH 4 and CO: an extended experiment within the European ICOS Atmosphere network. Atmos. Meas. Tech. 16, 5977–5994 (2023). Locher, R. IDPmisc: ‘Utilities of Institute of Data Analyses and Process Design ( . (2024). Hassani, A., Schneider, P., Vogt, M. & Castell, N. Low-Cost Particulate Matter Sensors for Monitoring Residential Wood Burning. Environ. Sci. Technol. 57, 15162–15172 (2023). Barlow, J. F. Progress in observing and modelling the urban boundary layer. Urban Climate 10, 216–240 (2014). Morawska, L. et al. Airborne particles in indoor environment of homes, schools, offices and aged care facilities: The main routes of exposure. Environment International 108, 75–83 (2017). Kalimeri, K. K., Bartzis, J. G., Sakellaris, I. A. & De Oliveira Fernandes, E. Investigation of the PM2.5, NO2 and O3 I/O ratios for office and school microenvironments. Environmental Research 179, 108791 (2019). Sokołowski, J. & Bouzarovski, S. Decarbonisation of the Polish residential sector between the 1990s and 2021: A case study of policy failures. Energy Policy 163, 112848 (2022). Zgłobicki, W. & Baran-Zgłobicka, B. Air pollution in major Polish cities in the period 2005–2021: Intensity, effects and attempts to reduce it. Environmental Research 240, 117497 (2024). Liang, D. et al. Estimating climate change-related impacts on outdoor air pollution infiltration. Environmental Research 196, 110923 (2021). Additional Declarations (Not answered) Supplementary Files AppendixA.docx SupplementaryMaterialSalamalikisetal.docx Supplementary Material: Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: revise 29 Jul, 2024 Review # 2 received at journal 26 Jul, 2024 Review # 3 received at journal 23 Jul, 2024 Reviewer # 3 agreed at journal 05 Jul, 2024 Reviewer # 2 agreed at journal 04 Jul, 2024 Reviewer # 1 agreed at journal 04 Jul, 2024 Reviewers invited by journal 02 Jul, 2024 Editor assigned by journal 01 Jul, 2024 Submission checks completed at journal 01 Jul, 2024 First submitted to journal 27 Jun, 2024 Unknown event 25 Jun, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4618450","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":321696320,"identity":"aebdc8b5-764a-4696-9aea-0df37d8ec28a","order_by":0,"name":"Vasileios Salamalikis","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAkUlEQVRIiWNgGAWjYHACxgMfGCRkSNNzcAaDBA9pWg4D1ZOgxeD84QeHbXdY8Bgc4D34gDgtN9IMDueekQBq4Us2IEqL5AwGoJY2kBYeMwnitPQf/3DYkiQt/Aw5BocZSdMikVNwsBeoRfIwjzFxfmHjP77xwc+2Ojm+4z2GD4jSggDMJKofBaNgFIyCUYAHAACoLCjDDJQCPwAAAABJRU5ErkJggg==","orcid":"","institution":"NILU","correspondingAuthor":true,"prefix":"","firstName":"Vasileios","middleName":"","lastName":"Salamalikis","suffix":""},{"id":321696321,"identity":"e071c84e-d028-40d7-800c-8fb4ff49bca7","order_by":1,"name":"Amirhossein Hassani","email":"","orcid":"","institution":"NILU","correspondingAuthor":false,"prefix":"","firstName":"Amirhossein","middleName":"","lastName":"Hassani","suffix":""},{"id":321696322,"identity":"176c759c-1e22-4401-b8c1-c0fff7d70a21","order_by":2,"name":"Paweł Zawadzki","email":"","orcid":"","institution":"Institute of Fluid-Flow Machinery Polish Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Paweł","middleName":"","lastName":"Zawadzki","suffix":""},{"id":321696323,"identity":"1254605d-a73a-4d7c-90e9-b34e37f443b2","order_by":3,"name":"Sebastian Bykuć","email":"","orcid":"","institution":"Institute of Fluid-Flow Machinery Polish Academy of Sciences","correspondingAuthor":false,"prefix":"","firstName":"Sebastian","middleName":"","lastName":"Bykuć","suffix":""},{"id":321696324,"identity":"6c73feac-f903-4136-b6ea-f28a270da4ec","order_by":4,"name":"Núria Castell","email":"","orcid":"","institution":"NILU","correspondingAuthor":false,"prefix":"","firstName":"Núria","middleName":"","lastName":"Castell","suffix":""}],"badges":[],"createdAt":"2024-06-21 16:55:53","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4618450/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4618450/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":60981198,"identity":"f8e6fe49-24cf-4316-a8fb-05eecc46034e","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":48727,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/5744e9d68c9e1b5a09a9e92e.jpg"},{"id":60981199,"identity":"96a6be11-786f-4f70-8006-012c5ef7212d","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"jpg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":78595,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eBoxplots of hourly indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations. The y-axis is in logarithmic scale, and it is truncated to 0.05 μg m\u003csup\u003e–3\u003c/sup\u003e. The vertical line inside the boxplots corresponds to the median and the boxplot edges are the 25% and 75% quantiles. Observations falling outside the interval [Q\u003csub\u003e25%\u003c/sub\u003e – 1.5×IQR, Q\u003csub\u003e75%\u003c/sub\u003e + 1.5×IQR] were marked as outliers (grey circles), with Q\u003csub\u003e25%\u003c/sub\u003e and Q\u003csub\u003e75%\u003c/sub\u003e representing the 25% and 75% quantiles, and IQR denoting the interquartile range (IQR = Q\u003csub\u003e75%\u003c/sub\u003e – Q\u003csub\u003e25%\u003c/sub\u003e). \u003cstrong\u003e(b)\u003c/strong\u003e average indoor PM\u003csub\u003e2.5\u003c/sub\u003e, and \u003cstrong\u003e(c)\u003c/strong\u003e average outdoor PM\u003csub\u003e2.5\u003c/sub\u003e for each hour of day. The values are smoothed using a loess fit. The shaded colored areas show the standard errors of the loess fits. Different colors correspond to the various indoor sites (IDs). PM\u003csub\u003e2.5\u003c/sub\u003e concentration levels are in μg m\u003csup\u003e–3\u003c/sup\u003e.\u003c/p\u003e","description":"","filename":"2.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/94d6207a51ddfa90af1a7b8c.jpg"},{"id":60981205,"identity":"c1491e24-e2ee-4500-af0d-93d627d0e247","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"jpg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":39855,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eCumulative Distribution Function (CDF) for the I/O ratio. The black lines represent the cumulative distributions for different building types (kindergarten: solid and other buildings: dashed). The sensors 5165, 5377, and 6030, were installed in kindergartens. \u003cstrong\u003e(b)\u003c/strong\u003e Average I/O ratio for each hour of day. The ratio for each hour was calculated by dividing the average measured indoor PM\u003csub\u003e2.5\u003c/sub\u003e by the average measured outdoor PM\u003csub\u003e2.5\u003c/sub\u003e. Thick grey lines denote I/O = 1. The solid and dashed black lines correspond to the diurnal distribution of I/O ratio for kindergartens and other buildings, respectively.\u003c/p\u003e","description":"","filename":"3.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/dc68d7045c68df6d1d19bd5b.jpg"},{"id":60981203,"identity":"1b100bb3-f2a2-4b70-ad4c-dccb064cfab1","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"jpg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":63235,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"4.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/b46db216be91a14274acd9d8.jpg"},{"id":60982041,"identity":"08ab08da-c065-47eb-bc00-032465f879ef","added_by":"auto","created_at":"2024-07-24 09:20:06","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":75147,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eBoxplots of indoor PM\u003csub\u003e2.5\u003c/sub\u003e for the “Emission” and “Non-Emission” classes after applying the REBS method. The y-axis is in logarithmic scale, and it is truncated to 0.05 μg m\u003csup\u003e–3\u003c/sup\u003e. The vertical line inside the boxplots corresponds to the median and the boxplot edges are the 25% and 75% quantiles. Observations falling outside the interval [Q\u003csub\u003e25%\u003c/sub\u003e – 1.5×IQR, Q\u003csub\u003e75%\u003c/sub\u003e + 1.5×IQR] were marked as outliers (grey circles), with Q\u003csub\u003e25%\u003c/sub\u003e and Q\u003csub\u003e75%\u003c/sub\u003e representing the 25% and 75% quantiles, and IQR denoting the interquartile range (IQR = Q\u003csub\u003e75%\u003c/sub\u003e – Q\u003csub\u003e25%\u003c/sub\u003e). \u003cstrong\u003e(b)\u003c/strong\u003e Diurnal distribution of the occurrence of the indoor emission events detected in the nine locations using the REBS method\u003cstrong\u003e (c)\u003c/strong\u003e Diurnal distribution of the number of detected indoor emission events relative to the total number of indoor measurements per hour, for October-April, period with high household-heating activities and April-September, period with little to no household-heating activities.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/f8ebab4d2f1ec1632e8fdeb2.jpg"},{"id":60981200,"identity":"ef59672b-d499-4574-9d50-ab44ddf1fcac","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"jpg","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":51630,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003e(a) \u003c/strong\u003eBoxplots of the infiltration factor, f\u003csub\u003einf\u003c/sub\u003e. The red diamonds are the average values. \u003cstrong\u003e(b)\u003c/strong\u003e Boxplots of the f\u003csub\u003einf\u003c/sub\u003e, for October-April (period with high household-heating activities) and May-September (period with low to little household-heating activities). The vertical line inside the boxplots corresponds to the median and the boxplot edges are the 25% and 75% quantiles. f\u003csub\u003einf\u003c/sub\u003e values outside the interval [Q\u003csub\u003e25%\u003c/sub\u003e – 1.5×IQR, Q\u003csub\u003e75%\u003c/sub\u003e + 1.5×IQR] were marked as outliers (grey circles), with Q\u003csub\u003e25%\u003c/sub\u003e and Q\u003csub\u003e75%\u003c/sub\u003e representing the 25% and 75% quantiles and IQR denoting the interquartile range (IQR = Q\u003csub\u003e75%\u003c/sub\u003e – Q\u003csub\u003e25%\u003c/sub\u003e).\u003c/p\u003e","description":"","filename":"6.jpg","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/c9970d36de6932c2a21d6f6a.jpg"},{"id":60982943,"identity":"7891974f-eb7d-4fc6-949d-32d25543de0b","added_by":"auto","created_at":"2024-07-24 09:28:07","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":884567,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/35176889-f13a-4bb5-a72f-b963fc5d9baf.pdf"},{"id":60981202,"identity":"8e7b286e-f3cc-4f44-955b-91509e190810","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":20445,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cbr\u003e\u003c/p\u003e","description":"","filename":"AppendixA.docx","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/d32ac9e4ed41f46389a716cd.docx"},{"id":60981204,"identity":"e6c2c4d5-b06e-4c94-9846-80b8812cbd50","added_by":"auto","created_at":"2024-07-24 09:12:06","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":13512546,"visible":true,"origin":"","legend":"\u003cp\u003eSupplementary Material: Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration\u003c/p\u003e","description":"","filename":"SupplementaryMaterialSalamalikisetal.docx","url":"https://assets-eu.researchsquare.com/files/rs-4618450/v1/2e02b78bebfe00b0841e3139.docx"}],"financialInterests":"(Not answered)","formattedTitle":"Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eHousehold air pollution accounts for three million deaths worldwide according to the World Health Organization (WHO) \u003csup\u003e1\u003c/sup\u003e. Fine particulate matter (PM\u003csub\u003e2.5\u003c/sub\u003e) \u0026mdash; particulates with an aerodynamic diameter\u0026thinsp;\u0026lt;\u0026thinsp;2.5 \u0026micro;m \u0026mdash; is a major atmospheric pollutant attributed to both outdoor and indoor sources. Due to their small sizes, fine particles suspended to the air can easily penetrate into the respiratory system through inhalation, causing adverse health effects \u003csup\u003e2\u003c/sup\u003e. Therefore, assessing the indoor air pollutant concentrations and human exposure to indoor air pollution is crucial, considering that people typically spend from 65% to more than 90% of their time in indoor environments \u003csup\u003e3,4\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eIndoor particulate matter (PM) is regarded as a mixture of: particles generated by indoor sources and outdoor generated particles that have infiltrated into the indoor environment. Due to different origins, such particles exhibit diverse chemical compositions, toxicities, and sizes. When indoor sources are absent or not significant, particles with outdoor origin potentially affect the indoor air quality. Outdoor particles can enter a building through open doors and windows, but also from porous media and cracks of the building shell \u003csup\u003e5\u0026ndash;10\u003c/sup\u003e. The presence of potential indoor emission sources coupled with the absent or inadequate ventilation could lead to higher indoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations compared to outdoor levels. Therefore, for a better understanding of indoor air quality, it\u0026rsquo;s essential to understand the ratio at which outdoor PM affects indoor conditions. Cooking, smoking, candles and wood burning appliances are mainly responsible for indoor fine PM, while coarser particles (PM\u003csub\u003e2.5\u003c/sub\u003e \u0026ndash; PM\u003csub\u003e10\u003c/sub\u003e) are typically the result of non-combustion activities like cleaning and particle resuspension due to residents\u0026rsquo; movements \u003csup\u003e11,12\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe initial step to investigate the connection between indoor and outdoor concentrations involves the calculation of Indoor-to-Outdoor concentration ratio (I/O). The I/O ratio varies dynamically throughout the day and across seasons \u003csup\u003e10\u003c/sup\u003e. A ratio greater than unity shows the dominance of indoor emission sources. In general, since I/O depends on various factors and processes, it cannot solely offer reliable information about the relationship between indoor and outdoor air quality \u003csup\u003e8\u003c/sup\u003e unless additional information on ventilation and the frequency and duration of window openings is available \u003csup\u003e13\u003c/sup\u003e. A detailed overview of I/O is provided in \u003csup\u003e8\u003c/sup\u003e and \u003csup\u003e14\u003c/sup\u003e, showing a large range of values as explained by the different types of buildings, emission sources, monitoring periods, type of ventilation, etc. \u003csup\u003e8\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eThe impact of outdoor-generated particles on indoor air quality is mainly represented by the infiltration factor (f\u003csub\u003einf\u003c/sub\u003e) which quantifies the proportion of ambient PM entering indoor environments and remaining suspended under steady state conditions. The f\u003csub\u003einf\u003c/sub\u003e depends on building characteristics and operation, indoor and outdoor meteorological conditions, season and time of the day, and particle size \u003csup\u003e4,10,12,15\u0026ndash;24\u003c/sup\u003e, while socioeconomic and demographic factors may also have indirect influences \u003csup\u003e18,23\u003c/sup\u003e. Various methodologies have been evaluated to estimate f\u003csub\u003einf\u003c/sub\u003e \u003csup\u003e13\u003c/sup\u003e. The simplest approach when both information about indoor and outdoor air pollutants is available involves regressing indoor concentration on outdoor concentrations. The gain and the offset of the linear relationship act as the infiltration factor and the average concentration of the indoor-generated particles, respectively \u003csup\u003e25\u0026ndash;27\u003c/sup\u003e. However, this approach estimates a \u0026ldquo;global\u0026ldquo; f\u003csub\u003einf\u003c/sub\u003e and indoor-generated concentration neglecting the variability of both variables in time and their dependencies on auxiliary factors. The relationship between indoor and outdoor concentration levels was also statistically addressed using distributed-lag regression models \u003csup\u003e22,23,28\u003c/sup\u003e. These models use indoor concentrations as an independent parameter and outdoor PM levels in several lags combined with meteorological parameters or other auxiliary factors as independent information. The f\u003csub\u003einf\u003c/sub\u003e was derived by summing the regression coefficients of the lagged outdoor PM levels. Using this approach, Krebs et al. \u003csup\u003e22\u003c/sup\u003e estimated f\u003csub\u003einf\u003c/sub\u003e from 40\u0026ndash;70% in the state of California, USA. A large extent of f\u003csub\u003einf\u003c/sub\u003e in single-family homes across USA was also estimated by Burke et al. \u003csup\u003e23\u003c/sup\u003e, where f\u003csub\u003einf\u003c/sub\u003e relationship with the household income and the outdoor concentration levels was also investigated.\u003c/p\u003e \u003cp\u003eSome other studies have calculated f\u003csub\u003einf\u003c/sub\u003e using the average I/O ratio over long-time intervals (e.g., day and week) excluding the extreme indoor emission sessions \u003csup\u003e15,17,18,24\u003c/sup\u003e. In those studies, it is assumed that the indoor concentrations closely mirror the corresponding outdoor concentrations when indoor emissions are minimal. Indoor emission-related PM concentrations typically exhibit abrupt, high-magnitude changes, and can be identified using rule-based approaches. Comparisons between indoor and outdoor concentrations, along with their time derivatives (consecutive concentration changes in time) help the detection of indoor emission events. Usually, higher values of f\u003csub\u003einf\u003c/sub\u003e are observed in summer compared to winter because of increased window opening frequency, with values varying depending on the building type.\u003c/p\u003e \u003cp\u003eThis study explores the relationship between indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations in typical buildings of Legionowo, Poland. This is accomplished by integrating meteorological and PM\u003csub\u003e2.5\u003c/sub\u003e data from low-cost air quality sensors (LCSs) operated by citizens and installed in various locations across the city. In recent years, the use of LCSs for indoor air quality monitoring has risen. They not only provide information about air pollutant concentrations but also help identify pollution sources \u003csup\u003e29\u0026ndash;32\u003c/sup\u003e. Despite challenges, particularly concerning data reliability compared to regulatory-grade instruments \u003csup\u003e33\u003c/sup\u003e, LCSs provide insights into both indoor and outdoor air quality levels \u003csup\u003e34\u003c/sup\u003e. This paper shows how data collected by citizens can be converted to decision-ready information. The ultimate goals are to explore the linkage between indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e, to estimate the proportion of indoor PM\u003csub\u003e2.5\u003c/sub\u003e that originates from infiltration from the outdoor environment and to understand the factors affecting indoor PM\u003csub\u003e2.5\u003c/sub\u003e variability, such as indoor emissions and f\u003csub\u003einf\u003c/sub\u003e. The identified f\u003csub\u003einf\u003c/sub\u003e values for each building provide insights for policymakers. For instance, low f\u003csub\u003einf\u003c/sub\u003e indicates that dominant sources of pollution are indoor, suggesting that more interventions are required to improve indoor air quality, such as enhancing ventilation or regulating indoor heating sources. Conversely, high infiltration values suggest that the building has better insulation, but there may be a need to address outdoor pollution sources.\u003c/p\u003e"},{"header":"2. Data","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1. Study area\u003c/h2\u003e \u003cp\u003eLegionowo is a city in central-eastern Poland (Longitude: 20.9369\u0026deg;E, Latitude: 52.4012\u0026deg;N), approximately 23 km north of Warsaw (Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea). It covers a surface of 13.5 km\u003csup\u003e2\u003c/sup\u003e featuring a relatively flat terrain, predominantly surrounded by forests. The highest elevation in the area is around 80 km with no notable hills or mountainous ranges nearby. The population in Legionowo area is ~\u0026thinsp;54170 inhabitants with a population density of 4000 people km\u003csup\u003e\u0026ndash;2\u003c/sup\u003e (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://pl.wikipedia.org/wiki/Legionowo\u003c/span\u003e\u003cspan address=\"https://pl.wikipedia.org/wiki/Legionowo\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e, assessed on 06.05.2022).\u003c/p\u003e \u003cp\u003eThe climate is classified as temperate oceanic of type Cfb based on the K\u0026ouml;ppen-Geiger climate classification system. Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eb shows the climograph for Legionowo area. Meteorological data from 1951 to 2013 were obtained from the Polish Institute of Meteorology and Water Management\u0026mdash;National Research Institute (IMGW-PIB) via \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://dane.impw.pl/\u003c/span\u003e\u003cspan address=\"https://dane.impw.pl/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e using the \u0026ldquo;climate\u0026rdquo; R package \u003csup\u003e35\u003c/sup\u003e (red star in Figure \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea). The average temperature is 8.3\u003csup\u003e\u0026deg;\u003c/sup\u003eC, with January being the coldest month at \u0026minus;\u0026thinsp;2.3\u003csup\u003e\u0026deg;\u003c/sup\u003eC and July the warmest at 19.2\u003csup\u003e\u0026deg;\u003c/sup\u003eC. The yearly average precipitation is 537.2 mm per year, with the majority occurring during the summer period (341.1 mm). July stands out as the wettest month with 74.5 mm while the lowest precipitation amounts are recorded in February (27.9 mm) and March (27.1 mm).\u003c/p\u003e \u003cp\u003eIn Legionowo, based on local experts, multi-family buildings are typically heated using central heating provided by a district heating system. For structures not connected to the central heating network, individual systems are used. Natural gas is the predominant choice for heating, followed by less common options such as hard coal, wood, electricity, and heat pumps. However, wood or coal is used for heating purposes \u003csup\u003e36\u003c/sup\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2. Low-cost sensor network\u003c/h2\u003e \u003cp\u003ePM\u003csub\u003e2.5\u003c/sub\u003e and basic meteorological parameters are measured using low-cost sensors in various buildings situated in the Legionowo area. Figure\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e shows the spatial distribution of the air quality sensors. Indoor air quality was measured at nine locations while seven outdoor sensors were also installed in the same or nearby buildings. In six out of nine locations, both indoor and outdoor sensors were installed in the same building (yellow triangles in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). For sensors 5485 and 5736 (red diamonds in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e), information about the outdoor air quality conditions was obtained through sensors installed in nearby buildings (green stars in Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e). For 6030, outdoor air quality was determined through the nearest outdoor sensor (6033) which is positioned at a distance of 445 m.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOutdoor air quality was assessed using Airly sensors (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://airly.org/en\u003c/span\u003e\u003cspan address=\"https://airly.org/en\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). The Airly sensor system incorporates a Plantower PMS5003 sensor (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.plantower.com/en/products_33/74.html\u003c/span\u003e\u003cspan address=\"https://www.plantower.com/en/products_33/74.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), which uses laser-based light scattering for measuring airborne particulate matter with aerodynamic diameters ranging from 0.3 to 10 \u0026micro;m. PM mass concentrations are provided in hourly resolution in the fractions of PM\u003csub\u003e1\u003c/sub\u003e, PM\u003csub\u003e2.5\u003c/sub\u003e and PM\u003csub\u003e10\u003c/sub\u003e. According to the manufacturer, the measurement range is 0\u0026ndash;1000 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e with an accuracy of \u0026plusmn;\u0026thinsp;1 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e for PM\u003csub\u003e2.5\u003c/sub\u003e. In addition to PM, the Airly system also records air temperature (\u003csup\u003e\u0026deg;\u003c/sup\u003eC), humidity (%), and air pressure (hPa).\u003c/p\u003e \u003cp\u003eFor indoor air quality monitoring, AIRTHINGS Cloudberry systems (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://airthings.com\u003c/span\u003e\u003cspan address=\"https://airthings.com\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) were deployed. The sensor system is equipped with an air quality sensor that utilizes laser light scattering technology to measure air particulates. PM\u003csub\u003e1\u003c/sub\u003e and PM\u003csub\u003e2.5\u003c/sub\u003e are recorded within the measurement range of 0 to 1000 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e. The claimed accuracies include\u0026thinsp;\u0026plusmn;\u0026thinsp;10 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e for concentrations lower than 100 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e and \u0026plusmn;\u0026thinsp;10% for concentrations above 100 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e. The AIRTHINGS Cloudberry kit also includes sensors for measuring air temperature (\u003csup\u003e\u0026deg;\u003c/sup\u003eC), relative humidity (%), air pressure (hPa), and volatile organic compounds (\u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e). It provides data at irregular sub-hourly time intervals. Indoor observations are hourly aggregated for being temporally consistent with outdoor data.\u003c/p\u003e \u003cp\u003eIndoor and outdoor air quality data were obtained through the AIRTHINGS dashboard and the Airly Data Platform from April 2022 to December 2023 based on the availability of indoor air quality measurements.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Methodology","content":"\u003cp\u003eIn general, the indoor concentration of an atmospheric pollutant, such as PM\u003csub\u003e2.5\u003c/sub\u003e, can be modeled through a single mass-balance equation (Eq.\u0026nbsp;\u003cspan refid=\"Equ1\" class=\"InternalRef\"\u003e1\u003c/span\u003e) considering that indoor concentration changes mainly depend on a) the infiltration of outdoor particles, b) indoor particles that escape outdoors, c) deposition of indoor particles and d) indoor emissions. Eq.\u0026nbsp;\u003cspan refid=\"Equ1\" class=\"InternalRef\"\u003e1\u003c/span\u003e assumes uniform mixing of the air pollutants and negates concentration changes due to gas-phase interactions or variations in the environmental parameters between the indoor and outdoor environments \u003csup\u003e13\u003c/sup\u003e.\u003cdiv id=\"Equ1\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ1\" name=\"EquationSource\"\u003e\n$$\\frac{\\text{d}{\\text{C}}_{\\text{i}\\text{n}}\\left(\\text{t}\\right)}{\\text{d}\\text{t}}=\\text{p}{\\alpha }{\\text{C}}_{\\text{o}\\text{u}\\text{t}}\\left(\\text{t}\\right)-{\\alpha }{\\text{C}}_{\\text{i}\\text{n}}\\left(\\text{t}\\right)-\\text{k}{\\text{C}}_{\\text{i}\\text{n}}\\left(\\text{t}\\right)+\\frac{\\text{S}\\left(\\text{t}\\right)}{\\text{V}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e1\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere dC\u003csub\u003ein\u003c/sub\u003e is the change in indoor PM\u003csub\u003e2.5\u003c/sub\u003e concentration (\u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e) during the time interval dt, C\u003csub\u003ein\u003c/sub\u003e and C\u003csub\u003eout\u003c/sub\u003e are the indoor and outdoor particle\u0026rsquo;s concentrations (\u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e), t is the time (h), p is the penetration efficiency of particles (dimensionless), α is the air exchange rate (h\u003csup\u003e\u0026ndash;1\u003c/sup\u003e), k is the deposition rate of particles (h\u003csup\u003e\u0026ndash;1\u003c/sup\u003e), S is the indoor emission rate (\u0026micro;g h\u003csup\u003e\u0026ndash;1\u003c/sup\u003e) and V is the volume of the building or the room (m\u003csup\u003e3\u003c/sup\u003e). Under the assumption that the instantaneous temporal change of indoor concentration is substantially lower than the average indoor concentration over a considerable time interval, the steady-state form of Eq.\u0026nbsp;\u003cspan refid=\"Equ1\" class=\"InternalRef\"\u003e1\u003c/span\u003e can be re-arranged as follows:\u003cdiv id=\"Equ2\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ2\" name=\"EquationSource\"\u003e\n$${\\text{C}}_{\\text{i}\\text{n}}=\\frac{\\text{p}{\\alpha }}{{\\alpha }+\\text{k}}{\\text{C}}_{\\text{o}\\text{u}\\text{t}}+\\frac{\\text{S}}{\\text{V}\\left({\\alpha }+\\text{k}\\right)}={\\text{f}}_{\\text{i}\\text{n}\\text{f}}{\\text{C}}_{\\text{o}\\text{u}\\text{t}}+{\\text{C}}_{\\text{i}\\text{n}, \\text{g}\\text{e}\\text{n}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e2\u003c/div\u003e\u003c/div\u003e\u003cdiv id=\"Equ3\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ3\" name=\"EquationSource\"\u003e\n$$\\frac{{\\text{C}}_{\\text{i}\\text{n}}}{{\\text{C}}_{\\text{o}\\text{u}\\text{t}}}={\\text{f}}_{\\text{i}\\text{n}\\text{f}}+\\frac{{\\text{C}}_{\\text{i}\\text{n}, \\text{g}\\text{e}\\text{n}}}{{\\text{C}}_{\\text{o}\\text{u}\\text{t}}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e3\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewith \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{f}}_{\\text{i}\\text{n}\\text{f}}=\\frac{\\text{p}{\\alpha }}{{\\alpha }+\\text{k}}\\)\u003c/span\u003e\u003c/span\u003e the infiltration factor and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\({\\text{C}}_{\\text{i}\\text{n}, \\text{g}\\text{e}\\text{n}}=\\frac{\\text{S}}{\\text{V}\\left({\\alpha }+\\text{k}\\right)}\\)\u003c/span\u003e\u003c/span\u003e the indoor concentration (\u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e) generated solely by indoor activities.\u003c/p\u003e \u003cp\u003eThe indoor-to-outdoor concentration ratio converges to the f\u003csub\u003einf\u003c/sub\u003e when the outdoor concentration is significantly high or during periods with minimal indoor contribution (Eq.\u0026nbsp;\u003cspan refid=\"Equ3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). The latter requires the detection and the removal of the indoor-generated peaks before the calculation of the infiltration factor. The optimal approach to detect such specific instances or/and periods is to record in detail the indoor activities. In this case, the peaks can also be assigned to the major activities that occurred in the indoor environment. If this kind of information is missing, machine learning and rule-based approaches can be evaluated for peak detection \u003csup\u003e15,17,18,24,28,37\u003c/sup\u003e. Here, the indoor emission cases were identified through analysis of the indoor PM\u003csub\u003e2.5\u003c/sub\u003e series using the Robust Extraction of Baseline Signal (REBS) methodology \u003csup\u003e38,39\u003c/sup\u003e. REBS has been applied for detecting local sources in various air pollutants time series \u003csup\u003e39\u0026ndash;41\u003c/sup\u003e. According to Ruckstuhl et al. \u003csup\u003e39\u003c/sup\u003e, the indoor time series can be decomposed as:\u003cdiv id=\"Equ4\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ4\" name=\"EquationSource\"\u003e\n$${\\text{C}}_{\\text{i}\\text{n}}\\left(\\text{t}\\right)={\\text{C}}_{\\text{B}}\\left(\\text{t}\\right)+{\\text{C}}_{\\text{R}}\\left(\\text{t}\\right)+{\\epsilon }$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e4\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere C\u003csub\u003eB\u003c/sub\u003e(t) is the background concentration levels, C\u003csub\u003eR\u003c/sub\u003e(t) is the concentration due to indoor emissions and other contributions (e.g., outdoor concentration) and ε is the normally distributed and independent errors. The local emissions that are responsible for spikes in time series then can be identified using the REBS method through a two-stage approach. Initially, the background concentration is determined using local linear regression over a moving window of a specific duration. Then, any data points greater than a designated threshold relative to the background concentrations are classified as emissions:\u003cdiv id=\"Equ5\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ5\" name=\"EquationSource\"\u003e\n$${\\text{C}}_{\\text{i}\\text{n}}\\left(\\text{t}\\right)\u0026gt;\\widehat{{\\text{C}}_{\\text{B}}}\\left(\\text{t}\\right)+{\\beta }\\times {\\sigma }$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e5\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewhere \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\widehat{{\\text{C}}_{\\text{B}}}\\left(\\text{t}\\right)\\)\u003c/span\u003e\u003c/span\u003e is the estimated background curve, σ is the standard deviation of the data falling below the background curve. β is a user-defined parameter. β controls the width of the threshold curve with higher values attributing to wider threshold concentrations. Here β is set equal to 3 as initially proposed in \u003csup\u003e39\u003c/sup\u003e. In this study, the REBS method was implemented using the \u003cem\u003erfbaseline\u003c/em\u003e function of the \u0026ldquo;IDPmisc\u0026rdquo; R package \u003csup\u003e42\u003c/sup\u003e.\u003c/p\u003e \u003cp\u003eOnce the peak emission events were detected, the f\u003csub\u003einf\u003c/sub\u003e at each timestamp was determined, assuming steady-state conditions after T hours (Section S1 of the supplementary material):\u003cdiv id=\"Equ6\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equ6\" name=\"EquationSource\"\u003e\n$${\\text{f}}_{\\text{i}\\text{n}\\text{f}}\\left(\\text{t}\\right)=\\frac{{\u0026lang;{\\text{C}}_{\\text{i}\\text{n}}\\left(\\text{t}\\right)\u0026rang;}_{\\text{T}}}{{\u0026lang;{\\text{C}}_{\\text{o}\\text{u}\\text{t}}\\left(\\text{t}\\right)\u0026rang;}_{\\text{T}}}$$\u003c/div\u003e\u003cdiv class=\"EquationNumber\"\u003e6\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003ewith \u0026lt;\u0026thinsp;C\u003csub\u003ein\u003c/sub\u003e(t)\u0026gt;\u003csub\u003eT\u003c/sub\u003e and \u0026lt;\u0026thinsp;C\u003csub\u003eout\u003c/sub\u003e(t)\u0026gt;\u003csub\u003eT\u003c/sub\u003e the T-hour rolling averages for the indoor and the outdoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations, considering only the timestamps that were not identified as indoor emission events and at least 30% data coverage within the T-hour window. A steady state is reached when the instantaneous temporal change of indoor concentration is significantly lower than the average indoor concentration over a considerable time span (Eq.\u0026nbsp;\u003cspan refid=\"Equ1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Following the methodology outlined in section S1, the hourly concentration change is calculated by the concentration difference of the current and previous timestamp, ΔC\u003csub\u003ein\u003c/sub\u003e(t)\u0026thinsp;=\u0026thinsp;C\u003csub\u003ein\u003c/sub\u003e(t) \u0026ndash; C\u003csub\u003ein\u003c/sub\u003e(t\u0026ndash;1). Then, running averages at various temporal intervals T are computed for the indoor concentration, \u0026lt;C\u003csub\u003ein\u003c/sub\u003e(t)\u0026gt;\u003csub\u003eT\u003c/sub\u003e, and the respective changes\u0026thinsp;\u0026lt;\u0026thinsp;ΔC\u003csub\u003ein\u003c/sub\u003e(t)\u0026gt;\u003csub\u003eT\u003c/sub\u003e. It is expected that\u0026thinsp;\u0026lt;\u0026thinsp;ΔC\u003csub\u003ein\u003c/sub\u003e(t)\u0026gt; \u0026ndash; \u0026lt;C\u003csub\u003ein\u003c/sub\u003e(t)\u0026gt;\u003csub\u003eT\u003c/sub\u003e almost equals to \u0026lt;\u0026thinsp;C\u003csub\u003ein\u003c/sub\u003e(t)\u0026gt;\u003csub\u003eT\u003c/sub\u003e at increasing temporal intervals (Fig. S6 and S7). Section S1 and Fig. S6 and S7 showed steady state conditions are achieved after 48 hours. Based on this result, f\u003csub\u003einf\u003c/sub\u003e is derived through Eq.\u0026nbsp;\u003cspan refid=\"Equ6\" class=\"InternalRef\"\u003e6\u003c/span\u003e using 48-hour running averages of indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e.\u003c/p\u003e"},{"header":"4. Results","content":"\u003cdiv id=\"Sec7\"\u003e\n \u003ch2\u003e4.1. Indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations\u003c/h2\u003e\n \u003cp\u003ePM\u003csub\u003e2.5\u003c/sub\u003e was simultaneously measured in both indoor and outdoor environments using LCSs from February 2022 to August 2023. The data availability for each sensor is presented in Fig. S2. Regarding the outdoor PM\u003csub\u003e2.5\u003c/sub\u003e time series, most sensors provided continuous measurements with minimal data gaps (Fig. S2b). However, substantial data gaps were present in the indoor PM\u003csub\u003e2.5\u003c/sub\u003e measurements, primarily due to sensor malfunctions (Fig. S2a). Sensors 5189 and 5165 yielded almost complete time series for over five months, while the other sensors exhibited significant data gaps (Fig. S2a). Fig. S2c presents the periods when both indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e recordings were simultaneously available. Notably, sensors 5165 and 6030 were active for less than one month and provided the lowest data availability over the whole deployment period. Since this study focuses on the relationship between indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e levels, only the periods where both sensors were operational were included in the subsequent analysis.\u003c/p\u003e\n \u003cp\u003eSummary of the indoor and outdoor hourly PM\u003csub\u003e2.5\u003c/sub\u003e LCS measurements for the buildings is shown in Fig. 2a. Only timestamps with both indoor and outdoor measurements are used to generate the boxplots. Indoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations generally appear lower than the outdoor readings. The median ranges for indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e are 1.9\u0026ndash;17.3 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e and 6.7\u0026ndash;27.9 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e, respectively (Table A1). Although significant peaks are present in indoor PM\u003csub\u003e2.5\u003c/sub\u003e time series (outlier points in Fig. 2a), mainly attributed to dominant indoor emission sources, 75% of the data were lower than 28 \u0026micro;g m\u003csup\u003e\u0026minus;\u0026thinsp;3\u003c/sup\u003e.\u003c/p\u003e\n \u003cp\u003eFigure\u0026nbsp;2b and 2c illustrate the diurnal distribution of average indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e based on the available measurements of the entire period for each sensor. It is important to mention that the diurnal distribution for some sensors was drawn using a limited amount of data. For example, sensor 6030 measures indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e only for a two-month period implying that the diurnal pattern is not representative for the general temporal pattern at this specific location. Analyzing outdoor levels, all sensors exhibited a U-shaped pattern throughout the entire period, with lower concentrations between 09:00 and 18:00 local time (LT). However, analysis of the diurnal profiles on a seasonal basis revealed a different pattern in winter compared to other seasons, primarily displaying a bimodal distribution with morning and evening peaks (Fig. S3). This pattern is commonly observed in urban settings, where the morning peak is typically attributed to traffic, and the elevated concentrations during late afternoon and nighttime hours are due to heating activities and traffic \u003csup\u003e34,43\u003c/sup\u003e. The outdoor concentration remains consistently at high levels throughout the evening period. In winters, the low temperature conditions increase the heating activities while the low boundary layer height also traps the air pollutants close to the ground surface, resulting in relatively high PM\u003csub\u003e2.5\u003c/sub\u003e concentrations \u003csup\u003e44\u003c/sup\u003e. Conversely, the diurnal distribution of indoor PM\u003csub\u003e2.5\u003c/sub\u003e is primarily influenced by indoor activities. Most locations exhibit early morning and early afternoon and nighttime peaks. The sensor 5189 shows similar diurnal patterns for both indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e. This behavior may be because indoor concentration is influenced by outdoor concentrations. It is worth noting that three sensors (5165, 5377, and 6030) were installed in kindergartens. Indoor PM\u003csub\u003e2.5\u003c/sub\u003e shows minimal variation throughout the day mainly because of the absence of significant indoor activities. In kindergartens, indoor emissions primarily stem from particle resuspension due to children\u0026apos;s activities, as combustion activities are generally absent \u003csup\u003e45,46\u003c/sup\u003e. The type of ventilation used in indoor environments also affects the observed PM\u003csub\u003e2.5\u003c/sub\u003e levels. Sensor 6033 was installed in a kindergarten with a mechanical ventilation system, whereas other buildings rely on natural ventilation. In this particular location (sensor 6033), intra-day variations of the average indoor PM\u003csub\u003e2.5\u003c/sub\u003e are low ranging between 7.3 and 10.5 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e.\u003c/p\u003e\n \u003cp\u003eFigure\u0026nbsp;3a presents the Cumulative Distribution Functions (CDF) for the calculated hourly I/O ratios. On an hourly timescale, I/O exhibited significant variability both within a specific location and between different locations. In most places, 90% of the hourly I/O ratios were below unity implying that indoor PM\u003csub\u003e2.5\u003c/sub\u003e levels mainly reflect the contribution from outdoor particles. However, sensors 5351 and 5736 showed I/O\u0026thinsp;\u0026gt;\u0026thinsp;1 in over 40% of hourly measurements, suggesting that in these two locations, the contribution and the strength of the indoor emission activities are greater compared to the other sensor locations. CDFs are also drawn based on the building type (kindergarten vs. other). Due to the absence of significant indoor sources in kindergartens, it is expected the diurnal distribution of I/O ratio to be lower than 1. Statistical analysis showed significant differences in the I/O distributions between the two building types (two-sample Kolmogorov-Smirnov test at 95% confidence level: KS statistic\u0026thinsp;=\u0026thinsp;0.08 and p\u0026ndash;value\u0026thinsp;\u0026lt;\u0026thinsp;\u0026lt;\u0026thinsp;0.05). Less than 10% of the I/O ratios is \u0026gt;\u0026thinsp;1 for kindergartens, while the for the other buildings, timestamps with I/O ratios\u0026thinsp;\u0026gt;\u0026thinsp;1 are near 20% of the total calculated I/O ratios.\u003c/p\u003e\n \u003cp\u003eFigure 3b shows the diurnal distribution of the I/O ratio. The findings of Fig. 3b align with those discussed in X \u003csup\u003e10\u003c/sup\u003e concerning the diurnal distribution of I/O. Generally, I/O exhibits higher values from 09:00 to 18:00 LT, primarily due to the decrease in outdoor PM\u003csub\u003e2.5\u003c/sub\u003e during this period and increased indoor activities emitting PM\u003csub\u003e2.5\u003c/sub\u003e (Fig. 2c). Locations where the diurnal patterns of indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e closely resemble each other tend to display less variability in I/O throughout the day, as observed in sensors 5189 and 5485. A nearly stable I/O variation is presented in sensor 5189 which might be attributed to the impact of mechanical ventilation within the building and the relatively stable intra-day indoor PM\u003csub\u003e2.5\u003c/sub\u003e levels. Sensor 5736 showed I/O\u0026thinsp;\u0026gt;\u0026thinsp;1 after 06:00 LT and the discrepancies between the diurnal indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e extended from \u0026minus;\u0026thinsp;0.1 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e (I/O\u0026thinsp;=\u0026thinsp;0.99) to 15 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e (I/O\u0026thinsp;=\u0026thinsp;2.9).\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec8\"\u003e\n \u003ch2\u003e4.2. Detection of indoor emission events\u003c/h2\u003e\n \u003cp\u003eThe detection of elevated indoor PM\u003csub\u003e2.5\u003c/sub\u003e due to intense indoor activities was conducted using the REBS method with an illustrative example provided in Fig. 4 for sensor 5351. As illustrated in Fig. 4a, the REBS method efficiently captures the spikes present in the time series. Overall, 895 out of 8706 hourly data points (~\u0026thinsp;10%) are attributed to significant indoor emission events. Figure 4b displays the statistical distributions of the indoor PM\u003csub\u003e2.5\u003c/sub\u003e levels for cases with PM\u003csub\u003e2.5\u003c/sub\u003e \u0026gt; threshold (\u0026ldquo;Emission\u0026rdquo;) and PM\u003csub\u003e2.5\u003c/sub\u003e \u0026lt; threshold (\u0026ldquo;Non-Emission\u0026rdquo;). The distributions are statistically different (two-sample t-test at 95% confidence level: p\u0026ndash;value\u0026thinsp;\u0026lt;\u0026thinsp;\u0026lt;\u0026thinsp;0.05), with average (\u0026plusmn;\u0026thinsp;1 standard deviation) concentrations are 17.2\u0026thinsp;\u0026plusmn;\u0026thinsp;11.1 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e and 54.0\u0026thinsp;\u0026plusmn;\u0026thinsp;27.7 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e for the \u0026ldquo;Non-Emission\u0026rdquo; and \u0026ldquo;Emission\u0026rdquo; classes, respectively.\u003c/p\u003e\n \u003cp\u003eThe methodology for identifying indoor emission events was applied to all indoor sensors with the results summarized in the boxplots of Fig.\u0026nbsp;5a. The average (\u0026plusmn;\u0026thinsp;1 standard deviation) PM\u003csub\u003e2.5\u003c/sub\u003e for the emission events ranged from 7.5\u0026thinsp;\u0026plusmn;\u0026thinsp;2.1 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e (sensor 5165) to 62.4\u0026thinsp;\u0026plusmn;\u0026thinsp;68.1 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e (sensor 5736). The \u0026lsquo;Emission\u0026rsquo; and \u0026lsquo;Non-Emission\u0026rsquo; classes were statistically different among the various sensors (based on the pairwise t-test under the 95% confidence level).\u003c/p\u003e\n \u003cp\u003eCooking and household-heating activities potentially influence indoor PM\u003csub\u003e2.5\u003c/sub\u003e patterns. Consequently, an increase in the frequency of emission events is anticipated during the periods with high levels of indoor activities. As already mentioned in the data session, the use of coal or wood for residential heating purposes is very common in Poland. In general, Polish households have the highest coal consumption rate across the European Union \u003csup\u003e47\u003c/sup\u003e while strict measures of solid-fuel burning for heating purposes have not widely implemented. As pointed out in \u003csup\u003e48\u003c/sup\u003e, fine PM concentrations shows a strong positive correlation with the number of heating systems in private-homes in Poland .\u003c/p\u003e\n \u003cp\u003eFigure\u0026nbsp;5c illustrates the diurnal distribution of detected emission events, expressed as a percentage of the total number of indoor LCSs measurements in each hour of day, from October to April (\u0026ldquo;household-heating\u0026rdquo; period) and from May to September (\u0026ldquo;non-household-heating\u0026rdquo; period). The percentages were calculated using detected emission events from all sensors. During the \u0026ldquo;non-household-heating\u0026rdquo; period, the percentage of emission events fluctuates between 4% and 10% of total measured events, with higher values observed in the early morning hours and after 16:00 LT. Assuming that indoor activities like cooking remain consistent during a day/year, the difference between the diurnal curves for the two periods could provide insights about household-heating activity contribution to the number of detected indoor emission events. In the \u0026ldquo;heating-household\u0026rdquo; months, people stay more at home, engage in more indoor activities, and the demand for heating increases. During \u0026ldquo;household-heating\u0026rdquo; period, nearly 25\u0026ndash;27% of the indoor PM\u003csub\u003e2.5\u003c/sub\u003e measurements were identified to have a dominant indoor source of emission after 18:00 LT, indicating a 15% increase compared to the \u0026ldquo;non-household-heating\u0026rdquo; period. Similar analyses were conducted at each sensor location separately (Fig. S6). Only three out of nine sensors have available data for both \u0026ldquo;household-heating\u0026rdquo; and \u0026ldquo;non-household-heating\u0026rdquo; periods.\u003c/p\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec9\"\u003e\n \u003ch2\u003e4.3. Infiltration factor, f\u003csub\u003einf\u003c/sub\u003e\u003c/h2\u003e\n \u003cp\u003eThe box plots in Fig.\u0026nbsp;6a and the summary statistics in Table A1 present the results of the f\u003csub\u003einf\u003c/sub\u003e calculation for various locations in Legionowo. The estimated f\u003csub\u003einf\u003c/sub\u003e values ranged from 0 and 1, with average values extending between 0.29 (sensors 5165 and 5189) and 0.75 (sensor 5751). This suggests 29%-75% of indoor measured concentrations are of outdoor origin. The standard deviation of f\u003csub\u003einf\u003c/sub\u003e is nearly similar across all locations (0.13\u0026ndash;0.16). f\u003csub\u003einf\u003c/sub\u003e values are consistent with those reported in the literature \u003csup\u003e8\u003c/sup\u003e. Seasonal analysis of f\u003csub\u003einf\u003c/sub\u003e is presented in Fig. 6b. Overall, during the \u0026ldquo;household-heating\u0026rdquo; period, f\u003csub\u003einf\u003c/sub\u003e was 0.45\u0026thinsp;\u0026plusmn;\u0026thinsp;0.20, and 0.53\u0026thinsp;\u0026plusmn;\u0026thinsp;0.22 during the \u0026ldquo;non-household-heating\u0026rdquo; period. Significant seasonal differences in the average f\u003csub\u003einf\u003c/sub\u003e were also observed for individual locations (based on t-tests at the 95% confidence level), ranging from 0.03 (for sensor 6033) to 0.19 (for sensor 5485). During the \u0026ldquo;non-household-heating\u0026rdquo; period, windows are opened more frequently, allowing the infiltration of outdoor particles into the indoor environment.\u003c/p\u003e\n \u003cp\u003eThe f\u003csub\u003einf\u003c/sub\u003e depends on the penetration efficiency, the deposition rate of the indoor air pollutants, and the air-exchange rate (Eq. 5). In addition to building characteristics and residents\u0026rsquo; activities, meteorological conditions may also influence f\u003csub\u003einf\u003c/sub\u003e \u003csup\u003e8,24,49\u003c/sup\u003e. In order to assess the relation between meteorology and f\u003csub\u003einf\u003c/sub\u003e, the differences between outdoor-indoor temperatures during the \u0026ldquo;household-heating\u0026rdquo; and \u0026ldquo;non-household-heating\u0026rdquo; periods are plotted against f\u003csub\u003einf\u003c/sub\u003e in Fig. S8a an Fig. S8b, respectively. Similar analysis was conducted in Lunderberg et al. \u003csup\u003e24\u003c/sup\u003e. Based on these results, the contribution of outdoor/indoor temperature difference to the f\u003csub\u003einf\u003c/sub\u003e is rather marginal, suggesting that this parameter cannot be considered as significant proxy for describing the f\u003csub\u003einf\u003c/sub\u003e variations.\u003c/p\u003e\n\u003c/div\u003e"},{"header":"5. Conclusions","content":"\u003cp\u003eThis study investigated the link between indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e levels using a network of citizen-operated LCSs in Legionowo, Poland. Sensors were deployed both indoors and outdoors at nine locations throughout the city, providing data from April 2022 to December 2023. One limitation of this study is the absence of information regarding the daily residents\u0026rsquo; activities contributing to indoor PM\u003csub\u003e2.5\u003c/sub\u003e. Nonetheless, simultaneous measurements of indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e helped to detect indoor emission events and estimating proportion of the outdoor particles infiltrating into the buildings. The key findings are outlined as follows:\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eIndoor PM\u003csub\u003e2.5\u003c/sub\u003e levels were significantly lower than the outdoor levels, with median concentration ranging from 1.9 to 17.3 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e indoors and 6.7 to 27.9 \u0026micro;g m\u003csup\u003e\u0026ndash;3\u003c/sup\u003e outdoors. Occasional spikes in indoor PM\u003csub\u003e2.5\u003c/sub\u003e were attributed to potential indoor emission sources.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe diurnal distribution of outdoor PM\u003csub\u003e2.5\u003c/sub\u003e concentrations exhibited a U-shape pattern throughout the measurement period peaking in the morning and late afternoon to evening hours. Indoor PM\u003csub\u003e2.5\u003c/sub\u003e varied during the day driven by the indoor activities, with evening peaks likely associated with dinner cooking and household heating activities. Sensors installed in kindergartens (5165, 5377 and 6030) showed minimal diurnal PM\u003csub\u003e2.5\u003c/sub\u003e variation, mainly reflecting the influence of children\u0026rsquo;s activities on indoor particle levels.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe I/O concentration ratio was analyzed to understand the relationship between indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e. I/O varied significantly both within a specific location and between different locations, mostly falling below unity. Sensors 5351 and 5736 exhibited significant indoor emissions, with over 40% of hourly measurements showing I/O\u0026thinsp;\u0026gt;\u0026thinsp;1. Higher I/O ratios were observed from 09:00 to 18:00 LT on a diurnal basis, with less variability in intra-day I/O ratios observed in locations with similar indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e patterns.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eA statistical approach was used to detect the potential indoor emissions events. The number of emission events was higher during early afternoon and nighttime hours, where activities like dinner cooking and house heating were more prevalent. The emission events were more distinguishable from October to April due to the contribution of household heating activities.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eThe influence of outdoor PM2.5 on indoor air pollution was also quantified in the f\u003csub\u003einf\u003c/sub\u003e, showing that 29\u0026ndash;75% of indoor measured concentrations originated from outdoor sources. The f\u003csub\u003einf\u003c/sub\u003e was higher during warmer months (May \u0026ndash; September) with little to no household-heating activities, attributable to more-frequent door and window openings. The contribution of outdoor-indoor temperature difference to the f\u003csub\u003einf\u003c/sub\u003e was found minimal.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability:\u0026nbsp;\u003c/strong\u003eData supporting the findings of this study will be made available upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding:\u003c/strong\u003e This research leading to these results has been received from the EEA Grants 2014–2021 via the National Center for Research and Development. The grant is provided under “Applied Research” Program for the “GREEN HEAT – towards collaborative local decarbonization” project (acronym GREEN HEAT); grant number PL-Applied Research-0043. Also, this research has received partial funding from the European Union’s Horizon 2020 Research and Innovation programme under grant agreement No 952433, VIDIS project. We also acknowledge funding for SOCIO-BEE project from the European Union’s Horizon Europe research and innovation programme under grant agreement No. 101037648.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments:\u003c/strong\u003e The contribution of all citizens who participated in the GREEN HEAT project is acknowledged.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflicts of Interest:\u003c/strong\u003e 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.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eWHO. Household air pollution. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.who.int/news-room/fact-sheets/detail/household-air-pollution-and-health\u003c/span\u003e\u003cspan address=\"https://www.who.int/news-room/fact-sheets/detail/household-air-pollution-and-health\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePope, C. A., Coleman, N., Pond, Z. A. \u0026amp; Burnett, R. T. Fine particulate air pollution and human mortality: 25\u0026thinsp;+\u0026thinsp;years of cohort studies. Environmental Research 183, 108924 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchweizer, C. \u003cem\u003eet al.\u003c/em\u003e Indoor time\u0026ndash;microenvironment\u0026ndash;activity patterns in seven regions of Europe. J Expo Sci Environ Epidemiol 17, 170\u0026ndash;181 (2007).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, Y., Ma, H., Zhang, N. \u0026amp; Li, Q. A systematic literature review on indoor PM2.5 concentrations and personal exposure in urban residential buildings. Heliyon 8, e10174 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, D.-L. \u0026amp; Nazaroff, W. W. Modeling pollutant penetration across building envelopes. Atmospheric Environment 35, 4451\u0026ndash;4462 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, D.-L. \u0026amp; Nazaroff, W. W. Particle Penetration Through Building Cracks. Aerosol Science and Technology 37, 565\u0026ndash;573 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBennett, D. H. \u0026amp; Koutrakis, P. Determining the infiltration of outdoor particles in the indoor environment using a dynamic model. Journal of Aerosol Science 37, 766\u0026ndash;785 (2006).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, C. \u0026amp; Zhao, B. Review of relationship between indoor and outdoor particles: I/O ratio, infiltration factor and penetration factor. Atmospheric Environment 45, 275\u0026ndash;288 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiu, C. \u003cem\u003eet al.\u003c/em\u003e Influence of natural ventilation rate on indoor PM2.5 deposition. Building and Environment 144, 357\u0026ndash;364 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStamp, S. \u003cem\u003eet al.\u003c/em\u003e A critical evaluation of the dynamic nature of indoor-outdoor air quality ratios. Atmospheric Environment 273, 118955 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIsaxon, C. \u003cem\u003eet al.\u003c/em\u003e Contribution of indoor-generated particles to residential exposure. Atmospheric Environment 106, 458\u0026ndash;466 (2015).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, J., Fan, G., Ou, Y. \u0026amp; Deng, Q. Characteristics and control strategies of indoor particles: An updated review. Energy and Buildings 294, 113232 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDiapouli, E., Chaloulakou, A. \u0026amp; Koutrakis, P. Estimating the concentration of indoor particles of outdoor origin: A review. Journal of the Air \u0026amp; Waste Management Association 63, 1113\u0026ndash;1129 (2013).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIlacqua, V., Scharko, N., Zambrana, J. \u0026amp; Malashock, D. Survey of residential indoor particulate matter measurements 1990\u0026ndash;2019. Indoor Air 32, (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAllen, R., Larson, T., Sheppard, L., Wallace, L. \u0026amp; Liu, L.-J. S. Use of Real-Time Light Scattering Data To Estimate the Contribution of Infiltrated and Indoor-Generated Particles to Indoor Air. Environ. Sci. Technol. 37, 3484\u0026ndash;3492 (2003).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRim, D., Wallace, L. \u0026amp; Persily, A. Infiltration of Outdoor Ultrafine Particles into a Test House. Environ. Sci. Technol. 44, 5908\u0026ndash;5913 (2010).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKearney, J. \u003cem\u003eet al.\u003c/em\u003e Residential indoor and outdoor ultrafine particles in Windsor, Ontario. Atmospheric Environment 45, 7583\u0026ndash;7593 (2011).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMacNeill, M. \u003cem\u003eet al.\u003c/em\u003e Factors influencing variability in the infiltration of PM2.5 mass and its components. Atmospheric Environment 61, 518\u0026ndash;532 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTang, C. H. \u003cem\u003eet al.\u003c/em\u003e Development of a modeling approach to estimate indoor-to-outdoor sulfur ratios and predict indoor PM2.5 and black carbon concentrations for Eastern Massachusetts households. J Expo Sci Environ Epidemiol 28, 125\u0026ndash;130 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVu, T. V. \u003cem\u003eet al.\u003c/em\u003e Assessing the contributions of outdoor and indoor sources to air quality in London homes of the SCAMP cohort. Building and Environment 222, 109359 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang, Y. \u003cem\u003eet al.\u003c/em\u003e Wildfire smoke impacts on indoor air quality assessed using crowdsourced data in California. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 118, e2106478118 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKrebs, B., Burney, J., Zivin, J. G. \u0026amp; Neidell, M. Using Crowd-Sourced Data to Assess the Temporal and Spatial Relationship between Indoor and Outdoor Particulate Matter. Environ. Sci. Technol. 55, 6107\u0026ndash;6115 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBurke, M. \u003cem\u003eet al.\u003c/em\u003e Exposures and behavioural responses to wildfire smoke. Nat Hum Behav 6, 1351\u0026ndash;1361 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLunderberg, D. M. \u003cem\u003eet al.\u003c/em\u003e Assessing residential PM \u003csub\u003e2.5\u003c/sub\u003e concentrations and infiltration factors with high spatiotemporal resolution using crowdsourced sensors. \u003cem\u003eProc. Natl. Acad. Sci. U.S.A.\u003c/em\u003e 120, e2308832120 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eOtt, W., Wallace, L. \u0026amp; Mage, D. Predicting Particulate (PM \u003csub\u003e10\u003c/sub\u003e) Personal Exposure Distributions Using a Random Component Superposition Statistical Model. Journal of the Air \u0026amp; Waste Management Association 50, 1390\u0026ndash;1406 (2000).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWallace, L. \u0026amp; Ott, W. Long-Term Indoor-Outdoor PM2.5 Measurements Using PurpleAir Sensors: An Improved Method of Calculating Indoor-Generated and Outdoor-Infiltrated Contributions to Potential Indoor Exposure. Sensors 23, 1160 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMousavi, A. \u0026amp; Wu, J. Indoor-Generated PM \u003csub\u003e2.5\u003c/sub\u003e During COVID-19 Shutdowns Across California: Application of the PurpleAir Indoor\u0026ndash;Outdoor Low-Cost Sensor Network. Environ. Sci. Technol. 55, 5648\u0026ndash;5656 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNguyen, P. D. M. \u003cem\u003eet al.\u003c/em\u003e Using Low-Cost Sensors to Assess Fine Particulate Matter Infiltration (PM2.5) during a Wildfire Smoke Episode at a Large Inpatient Healthcare Facility. \u003cem\u003eIJERPH\u003c/em\u003e 18, 9811 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorawska, L. \u003cem\u003eet al.\u003c/em\u003e Applications of low-cost sensing technologies for air quality monitoring and exposure assessment: How far have they gone? Environment International 116, 286\u0026ndash;299 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eR\u0026oacute;denas Garc\u0026iacute;a, M. \u003cem\u003eet al.\u003c/em\u003e Review of low-cost sensors for indoor air quality: Features and applications. Applied Spectroscopy Reviews 57, 747\u0026ndash;779 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBousiotis, D., Alconcel, L.-N. S., Beddows, D. C. S., Harrison, R. M. \u0026amp; Pope, F. D. Monitoring and apportioning sources of indoor air quality using low-cost particulate matter sensors. Environment International 174, 107907 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHiggins, C., Kumar, P. \u0026amp; Morawska, L. Indoor air quality monitoring and source apportionment using low-cost sensors. Environ. Res. Commun. 6, 012001 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCastell, N. \u003cem\u003eet al.\u003c/em\u003e Can commercial low-cost sensor platforms contribute to air quality monitoring and exposure estimates? Environment International 99, 293\u0026ndash;302 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHassani, A. \u003cem\u003eet al.\u003c/em\u003e Low-cost sensors and Machine Learning aid in identifying environmental factors affecting particulate matter emitted by household heating. Atmospheric Environment 314, 120108 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCzernecki, B., Głogowski, A. \u0026amp; Nowosad, J. Climate: An R Package to Access Free In-Situ Meteorological and Hydrological Datasets For Environmental Assessment. \u003cem\u003eSustainability\u003c/em\u003e 12, 394 (2020).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKarpinska, L., Śmiech, S., Gouveia, J. P. \u0026amp; Palma, P. Mapping Regional Vulnerability to Energy Poverty in Poland. Sustainability 13, 10694 (2021).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChan, W. R. \u003cem\u003eet al.\u003c/em\u003e Quantifying fine particle emission events from time-resolved measurements: Method description and application to 18 California low‐income apartments. Indoor Air 28, 89\u0026ndash;101 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuckstuhl, A. F., Jacobson, M. P., Field, R. W. \u0026amp; Dodd, J. A. Baseline subtraction using robust local regression estimation. Journal of Quantitative Spectroscopy and Radiative Transfer 68, 179\u0026ndash;193 (2001).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRuckstuhl, A. F. \u003cem\u003eet al.\u003c/em\u003e Robust extraction of baseline signal of atmospheric trace species using local regression. Atmos. Meas. Tech. 5, 2613\u0026ndash;2624 (2012).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003eet al.\u003c/em\u003e Identification of spikes associated with local sources in continuous time series of atmospheric CO, CO\u003csub\u003e2\u003c/sub\u003e and CH\u003csub\u003e4\u003c/sub\u003e \u003cem\u003eAtmos. Meas. Tech.\u003c/em\u003e 11, 1599\u0026ndash;1614 (2018).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCristofanelli, P. \u003cem\u003eet al.\u003c/em\u003e Identification of spikes in continuous ground-based in situ time series of CO \u003csub\u003e2\u003c/sub\u003e, CH \u003csub\u003e4\u003c/sub\u003e and CO: an extended experiment within the European ICOS Atmosphere network. Atmos. Meas. Tech. 16, 5977\u0026ndash;5994 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLocher, R. IDPmisc: \u0026lsquo;Utilities of Institute of Data Analyses and Process Design (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003e\u003c/span\u003e\u003cspan address=\"http://www.zhaw.ch/idp)\u0026rsquo;\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHassani, A., Schneider, P., Vogt, M. \u0026amp; Castell, N. Low-Cost Particulate Matter Sensors for Monitoring Residential Wood Burning. Environ. Sci. Technol. 57, 15162\u0026ndash;15172 (2023).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBarlow, J. F. Progress in observing and modelling the urban boundary layer. Urban Climate 10, 216\u0026ndash;240 (2014).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorawska, L. \u003cem\u003eet al.\u003c/em\u003e Airborne particles in indoor environment of homes, schools, offices and aged care facilities: The main routes of exposure. Environment International 108, 75\u0026ndash;83 (2017).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKalimeri, K. K., Bartzis, J. G., Sakellaris, I. A. \u0026amp; De Oliveira Fernandes, E. Investigation of the PM2.5, NO2 and O3 I/O ratios for office and school microenvironments. Environmental Research 179, 108791 (2019).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSokołowski, J. \u0026amp; Bouzarovski, S. Decarbonisation of the Polish residential sector between the 1990s and 2021: A case study of policy failures. Energy Policy 163, 112848 (2022).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZgłobicki, W. \u0026amp; Baran-Zgłobicka, B. Air pollution in major Polish cities in the period 2005\u0026ndash;2021: Intensity, effects and attempts to reduce it. Environmental Research 240, 117497 (2024).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLiang, D. \u003cem\u003eet al.\u003c/em\u003e Estimating climate change-related impacts on outdoor air pollution infiltration. Environmental Research 196, 110923 (2021).\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"npj-climate-and-atmospheric-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjclimatsci","sideBox":"Learn more about [npj Climate and Atmospheric Science](http://www.nature.com/npjclimatsci/)","snPcode":"41612","submissionUrl":"https://submission.springernature.com/new-submission/41612/3","title":"npj Climate and Atmospheric Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Indoor Air Quality, Particulate Matter, Indoor-to-Outdoor Ratio, Infiltration Factor, Low-Cost Sensor","lastPublishedDoi":"10.21203/rs.3.rs-4618450/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4618450/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eIndoor air pollution poses a significant environmental concern, leading to adverse health effects. Fine particulates (PM\u003csub\u003e2.5\u003c/sub\u003e) observed indoors exhibit high variability, influenced by both indoor emission sources and the infiltration of outdoor particles through open spaces and the incomplete building insulation. This study examines the relationship between indoor and outdoor PM\u003csub\u003e2.5\u003c/sub\u003e levels using data from a network of citizen-operated low-cost air quality sensors, deployed in Legionowo, Poland. Our results showed that generally, indoor PM\u003csub\u003e2.5\u003c/sub\u003e was lower than outdoor levels, with occasional peaks attributed to potential indoor emission sources. Statistical analysis identified emission events, particularly during cooking and household-heating periods, occurring more frequently from October to April. In the absence of indoor sources, outdoor particles accounted for 29\u0026ndash;75% of indoor particle concentration, highlighting the significance of infiltration. This study shows how citizen-generated data using low-cost sensors, after post-processing, can provide decision-ready information as for example outdoor particulate matter infiltration factors for each building. This information can help decision-makers in devising effective interventions such as prioritizing indoor ventilation or addressing outdoor pollution sources.\u003c/p\u003e","manuscriptTitle":"Citizen-Operated Low-Cost Sensors for Estimating Outdoor Particulate Matter Infiltration","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-07-24 09:12:01","doi":"10.21203/rs.3.rs-4618450/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"revise","date":"2024-07-29T10:54:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-07-26T10:09:04+00:00","index":2,"fulltext":"This content is not available."},{"type":"editorInvitedReview","content":"This content is not available.","date":"2024-07-23T22:26:10+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-07-05T15:25:15+00:00","index":3,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-07-04T08:55:19+00:00","index":2,"fulltext":"This content is not available."},{"type":"reviewerAgreed","content":"This content is not available.","date":"2024-07-04T06:57:27+00:00","index":1,"fulltext":"This content is not available."},{"type":"reviewersInvited","content":"","date":"2024-07-02T11:46:42+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-07-01T20:46:03+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-07-01T13:40:26+00:00","index":"","fulltext":""},{"type":"submitted","content":"npj Climate and Atmospheric Science","date":"2024-06-27T07:19:15+00:00","index":"","fulltext":""},{"type":"checksFailed","content":"","date":"2024-06-25T13:25:17+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"npj-climate-and-atmospheric-science","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"npjclimatsci","sideBox":"Learn more about [npj Climate and Atmospheric Science](http://www.nature.com/npjclimatsci/)","snPcode":"41612","submissionUrl":"https://submission.springernature.com/new-submission/41612/3","title":"npj Climate and Atmospheric Science","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"NPJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"61d10662-09fd-483d-b72d-29d43da90f49","owner":[],"postedDate":"July 24th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[{"id":34010444,"name":"Earth and environmental sciences/Climate sciences/Atmospheric science"},{"id":34010445,"name":"Earth and environmental sciences/Environmental sciences/Environmental impact"}],"tags":[],"updatedAt":"2024-12-06T08:55:15+00:00","versionOfRecord":[],"versionCreatedAt":"2024-07-24 09:12:01","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4618450","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4618450","identity":"rs-4618450","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}