- Email: [email protected]
The effect of outdoor air and indoor human activity on mass concentrations of PM10, PM2.5, and PM1 in a classroom
ARTICLE IN PRESS
Environmental Research 99 (2005) 143–149 www.elsevier.com/locate/envres
The effect of outdoor air and indoor human activity on mass concentrations of PM10, PM2.5, and PM1 in a classroom$ Martin Branisˇ , Pavla Rˇeza´cˇova´, Marke´ta Domasova´ Institute for Environmental Studies, Faculty of Science, Charles University in Prague, Albertov 6, 128 43 Prague 2, Czech Republic Received 1 July 2004; received in revised form 13 October 2004; accepted 1 December 2004 Available online 22 January 2005
Abstract The 12-h mass concentration of PM10, PM2.5, and PM1 was measured in a lecturing room by means of three co-located Harvard impactors. The filters were changed at 8 AM and at 8 PM to cover the periods of presence and absence of students. Concentrations were assessed by gravimetry. Ambient PM10 data were available for corresponding 12-h intervals from the nearest state air-qualitymonitoring network station. The data were pooled into four periods according to the presence and absence of students— Monday–Thursday day (workday daytime), Monday–Thursday night (workday night), Friday–Sunday day (weekend daytime), and Friday–Sunday night (weekend night). Average indoor workday daytime concentrations were 42.3, 21.9 and 13.7 mg m3, workday night were 20.9, 19.1 and 15.2 mg m3, weekend daytime were 21.9, 18.1 and 11.4 mg m3, and weekend night were 24.5, 21.3, and 15.6 mg m3 for PM10, PM2.5, and PM1, respectively. The highest 12-h mean, median, and maximum (42.3, 43.0, and 76.2 mg m3, respectively) indoor concentrations were recorded on workdays during the daytime for PM10. The statistically significant (r ¼ 0:68; Po0:0009) correlation between the number of students per hour per day and the indoor coarse fraction calculated as PM102.5 during daytime on workdays indicates that the presence of people is an important source of coarse particles indoor. On workdays, the daytime PM10 indoor/outdoor ratio was positively associated (r ¼ 0:93) with an increasing indoor coarse fraction (PM10–2.5), also indicating that an important portion of indoor PM10 had its source inside the classroom. With the exception of the calculated coarse fraction (PM10–2.5), all of the measured indoor particulate matter fractions were significantly highly correlated with outdoor PM10 and negatively correlated with wind velocity, showing that outdoor levels of particles influence their indoor concentrations. r 2004 Elsevier Inc. All rights reserved. Keywords: Indoor air pollution; Particulate matter; Coarse particles; PM10–2.5; Schools
1. Introduction People in developed countries are spending more and more of their time in various indoor environments (Farrow et al., 1997). For young individuals, schools represent the environment where they pass a substantial portion of the day (Silvers et al., 1994). A number of $ Human subjects were not directly involved in this study; however, full information about the measurements performed in the lecture room was provided to all students/teachers using the space. Corresponding author. Fax: +420 224 914 803. E-mail address: [email protected] (M. Branisˇ ).
0013-9351/$ - see front matter r 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2004.12.001
studies have revealed that school air may be a source of a wide spectrum of air pollutants, such as noise, NO2, CO, volatile organic compounds, aerosols, etc. The highest importance is currently attributed to aerosols because they represent a complex mixture of organic and inorganic substances with potential toxic, carcinogenic, inflammatory, allergenic, and other adverse properties (Maroni et al., 1995; Jones, 1999). It is generally accepted that indoor concentrations of particles derive from two sources: indoor and outdoor. However, the significance of both sources depends on a number of variables, e.g., air-exchange rate, outdoor air pollution, type of indoor activities, aerodynamic
ARTICLE IN PRESS 144
M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
diameter of particles emitted, etc. (Yocom, 1982; Owen et al., 1992; Wallace, 1996; Long et al., 2000; Monn, 2001). Recently, Monn et al. (1997) and Luoma and Batterman (2001) showed that in indoor environments, e.g., in homes and offices, where there is no specific source of pollution (like smoking and the combustion of fuel for heating and/or cooking), occupant-related activities may represent a principal source of dust (composed of cloth fibers, hair fragments, soil particles, skin cells, resuspended particles of various origin by walking, and emissions from materials handled, such as paper, fungi spores, and fibers, etc.). The aim of this study was to investigate the mass concentration of three fractions of particles in a classroom. We hypothesized that concentrations of coarse (4PM2.5) particles would be more associated with indoor activities while concentrations of smaller particles would be more correlated with outdoor particulate pollution. It was also expected that some association would exist between indoor particulate levels and indoor and/or outdoor microclimatic/meteorological factors such as temperature, relative humidity, and wind velocity.
2. Site and methods The study was conducted between 8 October and 11 November 2001 in Prague, Czech Republic. The period was selected as meteorologically stable without high pollution or pollen episodes. A university lecturing room with a capacity of about 80 persons was chosen for the measurements. The room on the second floor with dimensions of 13 m (length) 9 m (width) 4.5 m (height) was naturally ventilated through 10 large double-glazed windows (no forced ventilation or airconditioning system was in use in the building). The building was situated in a built-up area of central Prague and faced a street with an average traffic density of approximately 500 cars per hour during the day and several tens of cars per hour during the night; the building was backed by a university botanical garden. The classroom was occupied from Monday through Thursday, usually from 8 AM to 5 or 6 PM. With the exception of one Friday and one Saturday, no courses took place on Friday, Saturday, or Sunday. The week was therefore divided into two parts, one representing the high-occupancy period (Monday–Thursday) and the second the low/no-occupancy period (Friday–Sunday). The number of students present during a teaching hour (45 min) varied between a few individuals and the full capacity. The occupancy usually changed every second hour according to a traditional system of 2-h lectures. A short 5-min break during which the students moved in and out divided separate lectures. The presence of students was quantified by means of a personal-hour
indicator, which was calculated by multiplying the number of students by time in hours of their presence in the classroom. The average number of student-hours per the 12-h high-occupancy period was 269.3 (a maximum of 453 and a minimum of 126). The room was equipped with standard school furniture made of wooden chipboard and metal fixed to the floor. About 20 separate wood–metal chairs were available for excess visitors. There were one blackboard for chalk, one whiteboard for color-board markers, and one overhead, one slide, and one data projector in the classroom. The measuring equipment was placed on a table in the back of the room opposite the front wall, which accommodated the white- and blackboards. Three Harvard impactors (Air Diagnostic and Engineering Inc., Naples, ME, USA) with quiet pumps operating at 10 L min1 (for PM10 and PM2.5) and at 23 L min1 (for PM1) were used for sampling (Marple et al., 1987; Turner et al., 2000). The inlets of the impactors were situated in the breathing zone of the sitting students— 130 cm above the floor. Particles were collected on membrane Teflon filters (2-mm pore size, 37 mm in diameter). Filters were exposed for 12 h and changed twice a day, at 8 AM and 8 PM. Airflow was measured at the beginning and end of every filter exposure. The mass concentrations of PM were determined by gravimetry. The filters were equilibrated for constant humidity (50%) for at least 24 h before each weighing and weighed twice before and twice after each exposure on an electronic microbalance. The weighing accuracy calculated as the coefficient of variance was 2.4%. The mass difference of the filters before and after the exposure and the amount of sampled air volume were used for the PM concentration calculations. Indoor 12-h averages of temperature and relative humidity were calculated from continuous data measured by a sensor placed in the vicinity of the impactor inlets. Twenty-three blank filters (over 11% of the filters exposed) were used to determine the detection limit (DL). The blank filters were handled in the same manner (preweighed, mounted in filter holders, taken to the measuring room, placed in the impactors, removed without exposure, and reweighed) as the exposed filters. The mean difference (absolute number) of blanks was 3.9 mg. The DL, defined as 3s of field blanks divided by the nominal sample volume, was 6.5 mg m3 for PM10 and PM2.5 and 2.9 mg m3 for PM1. Fifteen (7%) of the total 204 exposed filters were below the DL. Six filters (3%) were discarded because of weighing inconsistencies. Outdoor PM10 concentrations (30-min integration time measured by means of b-radiometry) from the nearest monitoring station of the state air-qualitymonitoring network operated by the Czech HydroMeteorological Institute, situated about 1800 m from
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
the experimental locality, were available for the entire 30-day campaign. Twelve-hour averages of ambient PM10 concentrations covering the same intervals as the indoor Harvard impactor measurements were calculated from the continuous data. Ambient temperature, relative humidity, and wind velocity data from a nearby (300 m) state network measuring point were also collected from the Czech Hydro-Meteorological Institute database. Because of the lognormal distribution of data, Spearman correlation coefficients were used to indicate associations between the variables studied.
3. Results In order to ascertain the influence of the presence and absence of people in the classroom on indoor particulate Table 1 Twelve-hour dataa for indoor and outdoor PM mass concentrations (in mg m3) Workdays (Monday–Thursday) Daytime (8 AM– 8 PM) 39.0/38.3 PM10 out PM10 in 42.3/43.0 21.9/18.7 PM2.5 in PM1 in 13.7/10.9
Weekends (Friday–Sunday)
(23.7–61.0) (12.9–76.2) (7.6–44.0) (3.5–34.4)
44.3/40.3 24.2/23.5 19.1/21.4 13.5/13.0
(28.7–69.3) (4.6–54.4) (4.2–33.3) (3.4–26.6)
Nights (8 PM– 8 AM) 34.4/32.3 (23.7–51.8) PM10 out PM10 in 22.5/19.4 (3.8–52.4) PM2.5 in 18.1/13.8 (2.3–48.5) 11.4/9.9 (1.2–28.8) PM1 in
43.3/42.5 24.5/22.2 21.3/23.6 15.6/16.1
(23.6–73.6) (6.8–48.0) (5.5–34.4) (0.3–27.6)
a
Arithmetic average/median (range).
145
matter concentrations, the data obtained during the course of the 30-day measuring campaign were divided into four separate periods: (i) workday daytime, (ii) workday night, (iii) weekend daytime, and (iv) weekend night. The highest 12-h mean, median, and maximum indoor concentrations were recorded on workdays during the day for PM10. Both PM2.5 and PM1 fractions reached similar average and median values within all of the periods studied. No substantial differences were found for the two finer fractions (PM2.5 and PM1) between nights and daytime and between workdays and weekends. Basic statistical information about the mass concentration of the ambient PM10 and the three PM fractions measured are presented in Table 1. Time series for the three PM indoor fractions and PM10 outdoor are presented in Fig. 1. The ratio between the indoor and ambient PM10 exceeded unity (1.07) during the day on workdays when the classroom was occupied by students. The absence of people resulted in average ratios of 0.60, 0.56, and 0.50 for workday nights, weekend nights, and weekend daytime, respectively. A comparison of only indoor fractions showed higher ratios in periods without the presence of people, i.e., during nights and during weekend daytime (Table 2). The correlation between indoor and outdoor PM10 was lowest on weekends at night and highest during workdays at night. A comparison of finer indoor fractions with outdoor PM10 showed a higher correlation during workdays. Correlations between all three indoor fractions revealed that the highest correlation (close to unity) was found during weekends, when there were no students in the classroom. All correlations were highly significant (Table 3).
Fig. 1. Time series of indoor (IN) and outdoor (OUT) particulate matter (PM) concentrations in the studied classroom.
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
146
Table 2 Ratio of indoor and outdoor PM mass concentrationsa Day
PM10 in/PM10 out PM2.5 in/PM10 in PM1 in/PM2.5 in PM1 in/PM10 in a
Night
Workday
Weekend
Workday
Weekend
1.07 0.51 0.62 0.32
0.50 0.81 0.77 0.63
0.60 0.74 0.66 0.48
0.56 0.87 0.71 0.62
(70.32) (70.12) (70.18) (70.12)
(70.19) (70.17) (70.15) (70.09)
(70.22) (70.21) (70.18) (70.18)
(70.21) (70.12) (70.20) (70.18)
Arithmetic average (standard deviation).
Table 3 Correlation coefficients between indoor PM fractions and outdoor PM10 and between all indoor PM fractions
PM10 in/PM10 out PM2.5 in/PM10 out PM1 in/PM10 out PM2.5 in/PM10 in PM1 in/PM10 in PM1 in/PM2.5 in
Workday—day
Workday—night
Weekend—day
Weekend—night
0.800*** 0.943*** 0.851*** 0.872*** 0.762*** 0.862***
0.884*** 0.955*** 0.889*** 0.912*** 0.897*** 0.944***
0.800** 0.839*** 0.749*** 0.918*** 0.956*** 0.938***
0.707** 0.749** 0.743** 0.991*** 0.939*** 0.943***
**Po0.01; ***Po0.001.
A comparison of all indoor fractions and the presence of persons—the student-hours per day (SH)— revealed that on workdays PM2.5 and PM1 were poorly correlated with SH (PM2.5, r ¼ 0:135; PM1, r ¼ 0:069). A better correlation (but not significant, Po0:09) was found between SH and PM10 (r ¼ 0:391). The best fit was observed for the coarse fraction expressed as PM102.5 (r ¼ 0:683; Po0:001). All indoor PM fractions were compared with indoor microclimatic and ambient meteorological factors, which may have an effect on aerosol formation and dispersion. Indoor temperature, indoor relative humidity and ambient temperature, relative humidity, and wind velocity were selected as the most important variables. The strongest and statistically highly significant negative correlation was found between wind velocity and all measured indoor fractions. A high positive association was recorded between PM concentrations and indoor or ambient relative humidity. With one exception (PM10 versus ambient temperature on weekdays at night), no statistically significant correlation was found between PM and temperature. No statistically significant association was registered between the calculated coarse fraction and either of the indoor and/or ambient climatic variables (Table 4).
4. Discussion A healthy indoor school environment is one of the basic preconditions of education. However, a number of
scientific papers have revealed that classrooms and other spaces of schools sometimes represent hazardous environments. In their review, Daisey et al. (2003) reported a wide spectrum of air pollutants that may negatively affect the health of students, teachers, and other regular visitors such as participants in extracurricular programs. Other studies confirm that indoor air quality in schools is far from what may be characterized as a ‘‘healthy microenvironment’’ (Siskos et al., 2001; Smedje et al., 1997). Lee and Chang (1999) studied indoor air quality, including PM10 levels, in five classrooms in Hong Kong. The concentrations they report (in the range of 21–617 mg m3) seem very high in comparison to those of our study, and the authors suggested the use of air cleaners to mitigate the problem. However, it has been pointed out several times that the readings of photometers, such as those used in their study (TSI DustTrak), may exaggerate the real mass concentrations measured by reference gravimetrical methods (Gorner et al., 1995; Ramachandran et al., 2000). The results may be biased when the air contains a large proportion of fine (oPM2.5) particles, which presumably was the case in the Lee and Chang study. The penetration of such particle fractions was also demonstrated in our study. According to our unpublished results, the readings of the DustTrak photometer exceeded under various conditions the reference gravimetrical method (i.e., the Harvard impactor) by 2.2–6.3 times depending on aerosol characteristics and microclimatic factors such as the proportion of fine particles in the PM10 fraction and changes in temperature or relative humidity.
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149 Table 4 Correlation between temperature (T), relative humidity (RH), wind velocity (WV), and all indoor PM fractions PM10
PM2.5
PM1
PM10–2.5
Daytime weekday T in RH in T out RH out WV
0.179 0.597** 0.224 0.489* 0.748***
0.009 0.723*** 0.229 0.600** 0.797***
0.083 0.629** 0.212 0.579** 0.839***
0.293 0.322 0.158 0.268 0.432
Nighttime weekday T in RH in T out RH out WV
0.093 0.689*** 0.468* 0.649** 0.849***
0.044 0.595** 0.301 0.756*** 0.845***
0.032 0.605** 0.340 0.712*** 0.849***
0.335 0.016 0.230 0.279 0.304
Daytime weekend T in RH in T out RH out WV
0.227 0.209 0.155 0.364 0.673*
0.021 0.343 0.279 0.293 0.689**
0.213 0.314 0.099 0.534** 0.560**
0.582 0.164 0.118 0.000 0.073
Nighttime weekend T in RH in T out RH out WV
0.307 0.082 0.021 0.796*** 0.721**
0.424 0.095 0.037 0.807*** 0.750**
0.229 0.007 0.086 0.789*** 0.780***
0.511 0.200 0.068 0.596* 0.311
*Po0.05; **Po0.01; ***Po0.001.
Wheeler et al. (2000), Roorda-Knape et al. (1998), and Janssen et al. (1997) demonstrated that mass concentrations of PM10 were much higher in classrooms than in ambient air. This was also shown for most of the elements analyzed (Janssen et al., 1999). The results of the present study also show some difference of this type; however, it is not so pronounced. The average and median PM10 values in our lecturing room were slightly higher than the outdoor values only during the workday, 12-h teaching period, but the difference was not statistically significant. Seeming differences occurred when the classroom was empty, at night or during weekends, when ambient PM10 concentrations were approximately double than the indoor concentrations. A high association between the presence of students (the student-hour indicator) and the concentration of the PM10–2.5 fraction confirmed that human presence and related activities represent a significant source of coarse fraction indoors. Finer particulate fractions did not behave the same way as PM10. Both PM2.5 and PM1 measured indoors did not reflect the presence of people, and absolute values of both fractions remained stable throughout the entire period of study. In accord with our findings, Luoma and Batterman (2001) concluded that in offices occupant activities such as the movement
147
of people and paperwork increase the concentrations of coarse particles. Elevated coarse fraction (PM10–2.5) levels were similarly reported for homes with no apparent combustion sources (Abt et al., 2000; Jones et al., 2000; Monn et al., 1997). Also, the high correlation (r ¼ 0:933) between the PM10 indoor/outdoor (I/O) ratio and the coarse fraction (PM10–2.5) during the 12-h teaching period (workdays daytime) indicated that an important portion of indoor PM10 had its source inside the classroom. Since the classroom did not contain any other specific source of particles (smoking is prohibited in the entire building), human activity was seemingly the most important factor to account for the higher indoor levels of coarse particles. Without this source, during nights and on weekends the indoor PM10 was composed predominantly of PM2.5 (over 80%). The influence of human activities on particulate matter levels also can be demonstrated by a very high correlation between all of the three indoor fractions during weekends. All correlation coefficients exceeded 0.9, suggesting that there is only one source of particles. Lower correlations between PM10, two finer fractions, and the second highest PM10 I/O ratio during workday nights can be explained by activities taking place in the building in the late evening and early morning. Some students and researchers who worked in neighboring rooms and labs and/or early morning cleaning activities in corridors could have influenced, to some extent, the quality of air penetrating into the closed lecturing room through door gaps. Without the presence of people, the PM10 I/O ratio fit well values given by Monn et al. (1997) for naturally ventilated homes without smokers (0.7) and by Wallace (1996), who estimated that if there are no indoor sources of particles, indoor concentrations are about 0.43 the outdoor concentrations for coarse particles. The I/O ratio of PM10 given by Janssen et al. (1997) for Dutch (Wageningen and Amsterdam) schools varied between 0.46 and 0.75. These figures are in accord with those of our study but only for periods without the presence of students in the classroom. Contrary to the results of Patterson and Eatough (2000), we could not demonstrate a good correlation between PM2.5 mass and the presence of students. Neither could we confirm our previous pilot results of an association between the number of students and the PM1 concentration in the same lecturing room (Branisˇ et al., 2002). This controversy presumably arose from a limited set of data available for the previous analysis. The stronger association of indoor PM2.5 and PM1 with outdoor PM10 found in our case suggests that outdoor air may be an important source of fine indoor particles. These results are in accord with those of previous studies (Morawska et al., 2001; Wallace, 2000; Chao et al., 1998; Janssen et al., 1997), which showed that under natural ventilation outdoor particle concentrations
ARTICLE IN PRESS 148
M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
could be used to predict indoor levels of smaller particles, although not so well for PM10. Lower correlation between outdoor PM10 and all of the three indoor fractions during weekends cannot be explained simply. It is possible that while during the workdays the doors and sometimes the windows could be partially open, allowing for better air exchange, on weekends the air exchange might be limited. Faster deposition of coarse fraction and/or better penetration of fine fractions through window gaps could also contribute to a lower accord between indoor and outdoor PM10 and a lower association between indoor PM10 and the finer fraction concentrations. As was already demonstrated, smaller particles below approximately PM2.5 penetrate more easily through narrow gaps than do larger ones (Mosley et al., 2001; Thornburg et al., 2001; Liu and Nazaroff, 2003). Nevertheless, the correlation coefficients between all indoor and outdoor particulate matter variables were not poor even during this time of slow or limited air exchange (r40:7 in all cases). A strong effect of ambient wind velocity on indoor PM fractions offers indirect evidence that without important sources the indoor air may be strongly influenced by ambient particle concentrations; the higher the wind speed the lower the ambient and also the indoor particulate concentrations. Adverse health effects of particulate matter are mostly attributed to finer particulate matter fractions (Schwartz and Neas, 2000; Kaiser, 2000). However, it has been demonstrated that ambient coarse particles may, under specific conditions, also have negative effects on human health (Castillejos et al., 2000; Cifuentes et al., 2000; Smith et al., 2000; Zhang et al., 2002). Coarse particles known to be made up of soil material brought in on shoes, of skin flakes, of cloth and furniture fragments, of viable molds and bacteria, of insect and mite debris, of pollen granules, and of other materials may be significant carriers of allergenic properties. As already shown in the literature, carpeted and smooth-floor classrooms contain crude dust with allergenic properties of significantly higher amounts per unit area than homes (Dybendal and Elsayed, 1994). It has also been noted that coarse particles in schools have a high allergenic and proinflammatory potential (Gravesen et al., 1986; Einarsson et al., 1995; Wheeler et al., 2000; Nordback et al., 2000; Tortolero et al., 2002). According to Fox et al. (2003), occupied schoolrooms show a much larger content of bacterial markers than ambient air of the same unoccupied rooms during weekends. Our results confirm that without other sources of particles (such as smoking) human activity is an important factor accounting for elevated indoor levels of coarse particles and that finer indoor fractions could be significantly influenced by concentrations of ambient particles, which originate in traffic-related combustion processes. According to published data coarse particles
in schools and other public spaces have a potential to negatively influence human health.
Acknowledgments The work was supported by the Czech Ministry of Education Youth and Sports under the State research scheme (Grant No. MSM 113100007). The valuable comments of Prof. Robert Holub from the Colorado School of Mines (Golden, CO, USA) are greatly appreciated. References Abt, E., Suh, H.H., Allen, G., Koutrakis, P., 2000. Characterization of indoor particle sources: a study conducted in the metropolitan Boston area. Environ. Health Perspect. 108, 35–44. Branisˇ , M., Rˇeza´cˇova´, P., Guignon, N., 2002. Fine particles in four different indoor environments. Indoor Built Environ. 11, 184–190. Castillejos, M., Borja-Aburto, V.H., Dockery, D.W., Gold, D.R., Loomis, D., 2000. Airborne coarse particles and mortality. Inhalation Toxicol. 12 (Suppl. 1), 61–72. Chao, C.Y.H., Tung, T.C.W., Burnett, J., 1998. Influence of different indoor activities on the indoor particulate levels in residential buildings. Indoor Built Environ. 7, 110–121. Cifuentes, L.A., Vega, J., Kopfer, K., 2000. Effect of fine fraction of particulate matter versus the coarse mass and other pollutants on daily mortality in Santiago, Chile. J. Air Waste Manage. Assoc. 50, 1287–1298. Daisey, J.M., Angell, W.J., Apte, M.G., 2003. Indoor air quality, ventilation and health symptoms in schools: analysis of existing information. Indoor Air 13, 53–64. Dybendal, T., Elsayed, S., 1994. Dust from carpeted and smooth floors. 6. Allergen in homes compared with those in schools in Norway. Allergy 49, 210–216. Einarsson, R., Munir, A.K.M., Dreborg, S.K.G., 1995. Allergens in school dust. 2. Major mite (DER-P-1, DER-F-1) allergens in dust from Swedish schools. J. Allergy Clin. Immunol. 95, 1049–1053. Farrow, A., Taylor, H., Golding, J., 1997. Time spent in the home by different family members. Environ. Technol. 18, 605–613. Fox, A., Harley, W., Feigley, C., Salzberg, D., Sebastian, A., Larson, L., 2003. Increased levels of bacterial markers and CO2 in occupied school rooms. J. Environ. Monit. 5, 246–252. Gorner, P., Bemer, D., Fabries, J.F., 1995. Photometer measurement of polydisperse aerosol. J. Aerosol Sci. 26, 1281–1302. Gravesen, S., Larsen, L., Gyntelberg, F., Skov, P., 1986. Demonstration of microorganisms and dust in schools and offices—an observational study of nonindustrial buildings. Allergy 41, 520–525. Janssen, N.A.H., Hoek, G., Harssema, H., Brunekreef, B., 1997. Childhood exposure to PM10: relation between personal, classroom and outdoor concentrations. Occup. Environ. Med. 54, 888–894. Janssen, N.A.H., Hoek, G., Brunekreef, B., Harssema, H., 1999. Mass concentration and elemental composition of PM10 in classrooms. Occup. Environ. Med. 56, 482–487. Jones, A.P., 1999. Indoor air quality and health. Atmos. Environ. 33, 4535–4564. Jones, N.C., Thornton, C.A., Mark, D., Harrison, R.M., 2000. Indoor/outdoor relationships of particulate matter in domestic homes with roadside, urban and rural locations. Atmos. Environ. 34, 2603–2612.
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149 Kaiser, J., 2000. Air pollution—evidence mounts that tiny particles can kill. Science 289, 22–23. Lee, S.C., Chang, M., 1999. Indoor air quality investigation at five classrooms. Indoor Air Int. J. Indoor Air Qual. Clim. 9, 134–138. Liu, D.-L., Nazaroff, W.W., 2003. Particle penetration through building cracks. Aerosol Sci. Technol. 37, 565–573. Long, C.M., Suh, H.H., Koutrakis, P., 2000. Characterization of indoor particle sources using continuous mass and size monitors. J. Air Waste Manage. Assoc. 50, 1236–1250. Luoma, M., Batterman, A., 2001. Characterization of particulate emissions from occupant activities in offices. Indoor Air 11, 35–48. Maroni, M., Seifert, B., Lindvall, T. (Eds.), 1995. Indoor Air Quality, A Comprehensive Reference Book. Air Quality Monographs, vol. 3. Elsevier, Amsterdam. Marple, V.A., Rubow, K.L., Turner, W., Spengler, J.D., 1987. Lowflow rate sharp cut impactors for indoor air sampling—design and calibration. JAPCA Int. J. Air Pollut. Control Hazard. Waste Manage. 37, 1303–1307. Monn, C., 2001. Exposure assessment of air pollutants: a review on spatial heterogeneity and indoor/outdoor/personal exposure to suspended particulate matter, nitrogen dioxide and ozone. Atmos. Environ. 35, 1–32. Monn, C., Fuchs, A., Hogger, D., Junker, M., Kogelschatz, D., Roth, N., Wanner, H.-U., 1997. Particulate matter less than 10 mm (PM10) and fine particles less than 2.5 mm (PM2.5): relationships between indoor, outdoor and personal concentrations. Sci. Total Environ. 208, 15–21. Morawska, L., He, C.R., Hitchins, J., Gilbert, D., Parappukkaran, S., 2001. The relationships between indoor and outdoor airborne particles in the residential environment. Atmos. Environ. 35, 3463–3473. Mosley, R.B., Greenwell, D.J., Sparks, L.E., Guo, Z., Tucker, W.G., Fortman, R., Whitfield, C., 2001. Penetration of ambient fine particles into the indoor environmnent. Aerosol Sci. Technol. 34, 127–136. Nordback, D., Walinder, R., Wieslander, G., Smedje, G., Erwall, C., Venge, P., 2000. Indoor air pollutants in schools: nasal patency and biomarkers in nasal lavage. Allergy 55, 163–170. Owen, M.K., Ensor, D.S., Sparks, L.E., 1992. Airborne particle sizes and sources found in indoor air. Atmos. Environ. 26, 2149–2162. Patterson, E., Eatough, D.J., 2000. Indoor and outdoor relationships for ambient PM2.5 and associated pollutants: epidemiological implications in Lindon, Utah. J. Air Waste Manage. Assoc. 50, 103–110. Ramachandran, G., Adgate, J.L., Hill, N., Sexton, K., 2000. Comparison of short term variation (15-minute averages) in outdoor and indoor PM2.5 concentrations. J. Air Waste Manage. Assoc. 50, 1157–1166.
149
Roorda-Knape, M.C., Janssen, N.A.H., De Hartog, J.J., Van Vliet, P.H.N., Harssema, H., Brunekreef, B., 1998. Air pollution from traffic in city districts near major motorways. Atmos. Environ. 32 (11), 1921–1930. Schwartz, J., Neas, L.M., 2000. Fine particles are more strongly associated than coarse particles with acute respiratory health effects in schoolchildren. Epidemiology 11, 6–10. Silvers, A., Florence, B.T., Rourke, D.L., Lorimor, R.J., 1994. How children spend their time—a sample survey for use in exposure and risk assessment. Risk Anal. 14, 931–944. Siskos, P.A., Bouba, K.E., Stroubou, A.P., 2001. Determination of selected pollutants and measurement of physical parameters for the evaluation of indoor air quality in school buildings in Athens, Greece. Indoor Built Environ. 10 (3–4), 185–192. Smedje, G., Norback, D., Edling, C., 1997. Subjective indoor air quality in schools in relation to exposure. Indoor Air 7 (2), 143–150. Smith, R.L., Spitzner, D., Kim, Y., Fuentes, M., 2000. Threshold dependence of mortality effects for fine and coarse particles in Phoenix, Arizona. J. Air Waste Manage. Assoc. 50, 1367–1379. Thornburg, J., Ensor, D.S., Rodes, C.E., Lawless, P.A., Sparks, L.E., Mosley, R.B., 2001. Penetration of particles into buildings and associated physical factors. Part I: model development and computer simulations. Aerosol Sci. Technol. 34, 284–296. Tortolero, S.R., Bartholomew, L.K., Tyrrell, S., Abramson, S.L., Sockrider, M.M., Markham, C.M., Whitehead, L.W., Parcel, G.S., 2002. Environmental allergens and irritants in schools: a focus on asthma. J. School Health 72, 33–38. Turner, W.A., Olson, B.A., Allen, G.A., 2000. Calibration of sharp cut impactors for indoor and outdoor particle sampling. J. Air Waste Manage. Assoc. 50, 484–487. Wallace, L., 1996. Indoor particles: a review. J. Air Waste Manage. Assoc. 46, 98–126. Wallace, L., 2000. Correlations of personal exposure to particles with outdoor air measurements, a review of recent studies. Aerosol Sci. Technol. 32, 15–25. Wheeler, A.J., Williams, I., Beaumont, R.A., Hamilton, R.S., 2000. Characterisation of particulate matter sampled during a study of children’s personal exposure to airborne particulate matter in a UK urban environment. Environ. Monit. Assess. 65, 69–77. Yocom, J.E., 1982. Indoor–outdoor air quality relationships: a critical review. J. Air Pollut. Control Assoc. 32, 500–520. Zhang, J., Hu, W., Wei, F., Wu, G., Korn, L.R., Chapman, R., 2002. Children’s respiratory morbidity prevalence in relation to air pollution in four Chinese cities. Environ. Health Perspect. 110, 961–967.
Environmental Research 99 (2005) 143–149 www.elsevier.com/locate/envres
The effect of outdoor air and indoor human activity on mass concentrations of PM10, PM2.5, and PM1 in a classroom$ Martin Branisˇ , Pavla Rˇeza´cˇova´, Marke´ta Domasova´ Institute for Environmental Studies, Faculty of Science, Charles University in Prague, Albertov 6, 128 43 Prague 2, Czech Republic Received 1 July 2004; received in revised form 13 October 2004; accepted 1 December 2004 Available online 22 January 2005
Abstract The 12-h mass concentration of PM10, PM2.5, and PM1 was measured in a lecturing room by means of three co-located Harvard impactors. The filters were changed at 8 AM and at 8 PM to cover the periods of presence and absence of students. Concentrations were assessed by gravimetry. Ambient PM10 data were available for corresponding 12-h intervals from the nearest state air-qualitymonitoring network station. The data were pooled into four periods according to the presence and absence of students— Monday–Thursday day (workday daytime), Monday–Thursday night (workday night), Friday–Sunday day (weekend daytime), and Friday–Sunday night (weekend night). Average indoor workday daytime concentrations were 42.3, 21.9 and 13.7 mg m3, workday night were 20.9, 19.1 and 15.2 mg m3, weekend daytime were 21.9, 18.1 and 11.4 mg m3, and weekend night were 24.5, 21.3, and 15.6 mg m3 for PM10, PM2.5, and PM1, respectively. The highest 12-h mean, median, and maximum (42.3, 43.0, and 76.2 mg m3, respectively) indoor concentrations were recorded on workdays during the daytime for PM10. The statistically significant (r ¼ 0:68; Po0:0009) correlation between the number of students per hour per day and the indoor coarse fraction calculated as PM102.5 during daytime on workdays indicates that the presence of people is an important source of coarse particles indoor. On workdays, the daytime PM10 indoor/outdoor ratio was positively associated (r ¼ 0:93) with an increasing indoor coarse fraction (PM10–2.5), also indicating that an important portion of indoor PM10 had its source inside the classroom. With the exception of the calculated coarse fraction (PM10–2.5), all of the measured indoor particulate matter fractions were significantly highly correlated with outdoor PM10 and negatively correlated with wind velocity, showing that outdoor levels of particles influence their indoor concentrations. r 2004 Elsevier Inc. All rights reserved. Keywords: Indoor air pollution; Particulate matter; Coarse particles; PM10–2.5; Schools
1. Introduction People in developed countries are spending more and more of their time in various indoor environments (Farrow et al., 1997). For young individuals, schools represent the environment where they pass a substantial portion of the day (Silvers et al., 1994). A number of $ Human subjects were not directly involved in this study; however, full information about the measurements performed in the lecture room was provided to all students/teachers using the space. Corresponding author. Fax: +420 224 914 803. E-mail address: [email protected] (M. Branisˇ ).
0013-9351/$ - see front matter r 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.envres.2004.12.001
studies have revealed that school air may be a source of a wide spectrum of air pollutants, such as noise, NO2, CO, volatile organic compounds, aerosols, etc. The highest importance is currently attributed to aerosols because they represent a complex mixture of organic and inorganic substances with potential toxic, carcinogenic, inflammatory, allergenic, and other adverse properties (Maroni et al., 1995; Jones, 1999). It is generally accepted that indoor concentrations of particles derive from two sources: indoor and outdoor. However, the significance of both sources depends on a number of variables, e.g., air-exchange rate, outdoor air pollution, type of indoor activities, aerodynamic
ARTICLE IN PRESS 144
M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
diameter of particles emitted, etc. (Yocom, 1982; Owen et al., 1992; Wallace, 1996; Long et al., 2000; Monn, 2001). Recently, Monn et al. (1997) and Luoma and Batterman (2001) showed that in indoor environments, e.g., in homes and offices, where there is no specific source of pollution (like smoking and the combustion of fuel for heating and/or cooking), occupant-related activities may represent a principal source of dust (composed of cloth fibers, hair fragments, soil particles, skin cells, resuspended particles of various origin by walking, and emissions from materials handled, such as paper, fungi spores, and fibers, etc.). The aim of this study was to investigate the mass concentration of three fractions of particles in a classroom. We hypothesized that concentrations of coarse (4PM2.5) particles would be more associated with indoor activities while concentrations of smaller particles would be more correlated with outdoor particulate pollution. It was also expected that some association would exist between indoor particulate levels and indoor and/or outdoor microclimatic/meteorological factors such as temperature, relative humidity, and wind velocity.
2. Site and methods The study was conducted between 8 October and 11 November 2001 in Prague, Czech Republic. The period was selected as meteorologically stable without high pollution or pollen episodes. A university lecturing room with a capacity of about 80 persons was chosen for the measurements. The room on the second floor with dimensions of 13 m (length) 9 m (width) 4.5 m (height) was naturally ventilated through 10 large double-glazed windows (no forced ventilation or airconditioning system was in use in the building). The building was situated in a built-up area of central Prague and faced a street with an average traffic density of approximately 500 cars per hour during the day and several tens of cars per hour during the night; the building was backed by a university botanical garden. The classroom was occupied from Monday through Thursday, usually from 8 AM to 5 or 6 PM. With the exception of one Friday and one Saturday, no courses took place on Friday, Saturday, or Sunday. The week was therefore divided into two parts, one representing the high-occupancy period (Monday–Thursday) and the second the low/no-occupancy period (Friday–Sunday). The number of students present during a teaching hour (45 min) varied between a few individuals and the full capacity. The occupancy usually changed every second hour according to a traditional system of 2-h lectures. A short 5-min break during which the students moved in and out divided separate lectures. The presence of students was quantified by means of a personal-hour
indicator, which was calculated by multiplying the number of students by time in hours of their presence in the classroom. The average number of student-hours per the 12-h high-occupancy period was 269.3 (a maximum of 453 and a minimum of 126). The room was equipped with standard school furniture made of wooden chipboard and metal fixed to the floor. About 20 separate wood–metal chairs were available for excess visitors. There were one blackboard for chalk, one whiteboard for color-board markers, and one overhead, one slide, and one data projector in the classroom. The measuring equipment was placed on a table in the back of the room opposite the front wall, which accommodated the white- and blackboards. Three Harvard impactors (Air Diagnostic and Engineering Inc., Naples, ME, USA) with quiet pumps operating at 10 L min1 (for PM10 and PM2.5) and at 23 L min1 (for PM1) were used for sampling (Marple et al., 1987; Turner et al., 2000). The inlets of the impactors were situated in the breathing zone of the sitting students— 130 cm above the floor. Particles were collected on membrane Teflon filters (2-mm pore size, 37 mm in diameter). Filters were exposed for 12 h and changed twice a day, at 8 AM and 8 PM. Airflow was measured at the beginning and end of every filter exposure. The mass concentrations of PM were determined by gravimetry. The filters were equilibrated for constant humidity (50%) for at least 24 h before each weighing and weighed twice before and twice after each exposure on an electronic microbalance. The weighing accuracy calculated as the coefficient of variance was 2.4%. The mass difference of the filters before and after the exposure and the amount of sampled air volume were used for the PM concentration calculations. Indoor 12-h averages of temperature and relative humidity were calculated from continuous data measured by a sensor placed in the vicinity of the impactor inlets. Twenty-three blank filters (over 11% of the filters exposed) were used to determine the detection limit (DL). The blank filters were handled in the same manner (preweighed, mounted in filter holders, taken to the measuring room, placed in the impactors, removed without exposure, and reweighed) as the exposed filters. The mean difference (absolute number) of blanks was 3.9 mg. The DL, defined as 3s of field blanks divided by the nominal sample volume, was 6.5 mg m3 for PM10 and PM2.5 and 2.9 mg m3 for PM1. Fifteen (7%) of the total 204 exposed filters were below the DL. Six filters (3%) were discarded because of weighing inconsistencies. Outdoor PM10 concentrations (30-min integration time measured by means of b-radiometry) from the nearest monitoring station of the state air-qualitymonitoring network operated by the Czech HydroMeteorological Institute, situated about 1800 m from
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
the experimental locality, were available for the entire 30-day campaign. Twelve-hour averages of ambient PM10 concentrations covering the same intervals as the indoor Harvard impactor measurements were calculated from the continuous data. Ambient temperature, relative humidity, and wind velocity data from a nearby (300 m) state network measuring point were also collected from the Czech Hydro-Meteorological Institute database. Because of the lognormal distribution of data, Spearman correlation coefficients were used to indicate associations between the variables studied.
3. Results In order to ascertain the influence of the presence and absence of people in the classroom on indoor particulate Table 1 Twelve-hour dataa for indoor and outdoor PM mass concentrations (in mg m3) Workdays (Monday–Thursday) Daytime (8 AM– 8 PM) 39.0/38.3 PM10 out PM10 in 42.3/43.0 21.9/18.7 PM2.5 in PM1 in 13.7/10.9
Weekends (Friday–Sunday)
(23.7–61.0) (12.9–76.2) (7.6–44.0) (3.5–34.4)
44.3/40.3 24.2/23.5 19.1/21.4 13.5/13.0
(28.7–69.3) (4.6–54.4) (4.2–33.3) (3.4–26.6)
Nights (8 PM– 8 AM) 34.4/32.3 (23.7–51.8) PM10 out PM10 in 22.5/19.4 (3.8–52.4) PM2.5 in 18.1/13.8 (2.3–48.5) 11.4/9.9 (1.2–28.8) PM1 in
43.3/42.5 24.5/22.2 21.3/23.6 15.6/16.1
(23.6–73.6) (6.8–48.0) (5.5–34.4) (0.3–27.6)
a
Arithmetic average/median (range).
145
matter concentrations, the data obtained during the course of the 30-day measuring campaign were divided into four separate periods: (i) workday daytime, (ii) workday night, (iii) weekend daytime, and (iv) weekend night. The highest 12-h mean, median, and maximum indoor concentrations were recorded on workdays during the day for PM10. Both PM2.5 and PM1 fractions reached similar average and median values within all of the periods studied. No substantial differences were found for the two finer fractions (PM2.5 and PM1) between nights and daytime and between workdays and weekends. Basic statistical information about the mass concentration of the ambient PM10 and the three PM fractions measured are presented in Table 1. Time series for the three PM indoor fractions and PM10 outdoor are presented in Fig. 1. The ratio between the indoor and ambient PM10 exceeded unity (1.07) during the day on workdays when the classroom was occupied by students. The absence of people resulted in average ratios of 0.60, 0.56, and 0.50 for workday nights, weekend nights, and weekend daytime, respectively. A comparison of only indoor fractions showed higher ratios in periods without the presence of people, i.e., during nights and during weekend daytime (Table 2). The correlation between indoor and outdoor PM10 was lowest on weekends at night and highest during workdays at night. A comparison of finer indoor fractions with outdoor PM10 showed a higher correlation during workdays. Correlations between all three indoor fractions revealed that the highest correlation (close to unity) was found during weekends, when there were no students in the classroom. All correlations were highly significant (Table 3).
Fig. 1. Time series of indoor (IN) and outdoor (OUT) particulate matter (PM) concentrations in the studied classroom.
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
146
Table 2 Ratio of indoor and outdoor PM mass concentrationsa Day
PM10 in/PM10 out PM2.5 in/PM10 in PM1 in/PM2.5 in PM1 in/PM10 in a
Night
Workday
Weekend
Workday
Weekend
1.07 0.51 0.62 0.32
0.50 0.81 0.77 0.63
0.60 0.74 0.66 0.48
0.56 0.87 0.71 0.62
(70.32) (70.12) (70.18) (70.12)
(70.19) (70.17) (70.15) (70.09)
(70.22) (70.21) (70.18) (70.18)
(70.21) (70.12) (70.20) (70.18)
Arithmetic average (standard deviation).
Table 3 Correlation coefficients between indoor PM fractions and outdoor PM10 and between all indoor PM fractions
PM10 in/PM10 out PM2.5 in/PM10 out PM1 in/PM10 out PM2.5 in/PM10 in PM1 in/PM10 in PM1 in/PM2.5 in
Workday—day
Workday—night
Weekend—day
Weekend—night
0.800*** 0.943*** 0.851*** 0.872*** 0.762*** 0.862***
0.884*** 0.955*** 0.889*** 0.912*** 0.897*** 0.944***
0.800** 0.839*** 0.749*** 0.918*** 0.956*** 0.938***
0.707** 0.749** 0.743** 0.991*** 0.939*** 0.943***
**Po0.01; ***Po0.001.
A comparison of all indoor fractions and the presence of persons—the student-hours per day (SH)— revealed that on workdays PM2.5 and PM1 were poorly correlated with SH (PM2.5, r ¼ 0:135; PM1, r ¼ 0:069). A better correlation (but not significant, Po0:09) was found between SH and PM10 (r ¼ 0:391). The best fit was observed for the coarse fraction expressed as PM102.5 (r ¼ 0:683; Po0:001). All indoor PM fractions were compared with indoor microclimatic and ambient meteorological factors, which may have an effect on aerosol formation and dispersion. Indoor temperature, indoor relative humidity and ambient temperature, relative humidity, and wind velocity were selected as the most important variables. The strongest and statistically highly significant negative correlation was found between wind velocity and all measured indoor fractions. A high positive association was recorded between PM concentrations and indoor or ambient relative humidity. With one exception (PM10 versus ambient temperature on weekdays at night), no statistically significant correlation was found between PM and temperature. No statistically significant association was registered between the calculated coarse fraction and either of the indoor and/or ambient climatic variables (Table 4).
4. Discussion A healthy indoor school environment is one of the basic preconditions of education. However, a number of
scientific papers have revealed that classrooms and other spaces of schools sometimes represent hazardous environments. In their review, Daisey et al. (2003) reported a wide spectrum of air pollutants that may negatively affect the health of students, teachers, and other regular visitors such as participants in extracurricular programs. Other studies confirm that indoor air quality in schools is far from what may be characterized as a ‘‘healthy microenvironment’’ (Siskos et al., 2001; Smedje et al., 1997). Lee and Chang (1999) studied indoor air quality, including PM10 levels, in five classrooms in Hong Kong. The concentrations they report (in the range of 21–617 mg m3) seem very high in comparison to those of our study, and the authors suggested the use of air cleaners to mitigate the problem. However, it has been pointed out several times that the readings of photometers, such as those used in their study (TSI DustTrak), may exaggerate the real mass concentrations measured by reference gravimetrical methods (Gorner et al., 1995; Ramachandran et al., 2000). The results may be biased when the air contains a large proportion of fine (oPM2.5) particles, which presumably was the case in the Lee and Chang study. The penetration of such particle fractions was also demonstrated in our study. According to our unpublished results, the readings of the DustTrak photometer exceeded under various conditions the reference gravimetrical method (i.e., the Harvard impactor) by 2.2–6.3 times depending on aerosol characteristics and microclimatic factors such as the proportion of fine particles in the PM10 fraction and changes in temperature or relative humidity.
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149 Table 4 Correlation between temperature (T), relative humidity (RH), wind velocity (WV), and all indoor PM fractions PM10
PM2.5
PM1
PM10–2.5
Daytime weekday T in RH in T out RH out WV
0.179 0.597** 0.224 0.489* 0.748***
0.009 0.723*** 0.229 0.600** 0.797***
0.083 0.629** 0.212 0.579** 0.839***
0.293 0.322 0.158 0.268 0.432
Nighttime weekday T in RH in T out RH out WV
0.093 0.689*** 0.468* 0.649** 0.849***
0.044 0.595** 0.301 0.756*** 0.845***
0.032 0.605** 0.340 0.712*** 0.849***
0.335 0.016 0.230 0.279 0.304
Daytime weekend T in RH in T out RH out WV
0.227 0.209 0.155 0.364 0.673*
0.021 0.343 0.279 0.293 0.689**
0.213 0.314 0.099 0.534** 0.560**
0.582 0.164 0.118 0.000 0.073
Nighttime weekend T in RH in T out RH out WV
0.307 0.082 0.021 0.796*** 0.721**
0.424 0.095 0.037 0.807*** 0.750**
0.229 0.007 0.086 0.789*** 0.780***
0.511 0.200 0.068 0.596* 0.311
*Po0.05; **Po0.01; ***Po0.001.
Wheeler et al. (2000), Roorda-Knape et al. (1998), and Janssen et al. (1997) demonstrated that mass concentrations of PM10 were much higher in classrooms than in ambient air. This was also shown for most of the elements analyzed (Janssen et al., 1999). The results of the present study also show some difference of this type; however, it is not so pronounced. The average and median PM10 values in our lecturing room were slightly higher than the outdoor values only during the workday, 12-h teaching period, but the difference was not statistically significant. Seeming differences occurred when the classroom was empty, at night or during weekends, when ambient PM10 concentrations were approximately double than the indoor concentrations. A high association between the presence of students (the student-hour indicator) and the concentration of the PM10–2.5 fraction confirmed that human presence and related activities represent a significant source of coarse fraction indoors. Finer particulate fractions did not behave the same way as PM10. Both PM2.5 and PM1 measured indoors did not reflect the presence of people, and absolute values of both fractions remained stable throughout the entire period of study. In accord with our findings, Luoma and Batterman (2001) concluded that in offices occupant activities such as the movement
147
of people and paperwork increase the concentrations of coarse particles. Elevated coarse fraction (PM10–2.5) levels were similarly reported for homes with no apparent combustion sources (Abt et al., 2000; Jones et al., 2000; Monn et al., 1997). Also, the high correlation (r ¼ 0:933) between the PM10 indoor/outdoor (I/O) ratio and the coarse fraction (PM10–2.5) during the 12-h teaching period (workdays daytime) indicated that an important portion of indoor PM10 had its source inside the classroom. Since the classroom did not contain any other specific source of particles (smoking is prohibited in the entire building), human activity was seemingly the most important factor to account for the higher indoor levels of coarse particles. Without this source, during nights and on weekends the indoor PM10 was composed predominantly of PM2.5 (over 80%). The influence of human activities on particulate matter levels also can be demonstrated by a very high correlation between all of the three indoor fractions during weekends. All correlation coefficients exceeded 0.9, suggesting that there is only one source of particles. Lower correlations between PM10, two finer fractions, and the second highest PM10 I/O ratio during workday nights can be explained by activities taking place in the building in the late evening and early morning. Some students and researchers who worked in neighboring rooms and labs and/or early morning cleaning activities in corridors could have influenced, to some extent, the quality of air penetrating into the closed lecturing room through door gaps. Without the presence of people, the PM10 I/O ratio fit well values given by Monn et al. (1997) for naturally ventilated homes without smokers (0.7) and by Wallace (1996), who estimated that if there are no indoor sources of particles, indoor concentrations are about 0.43 the outdoor concentrations for coarse particles. The I/O ratio of PM10 given by Janssen et al. (1997) for Dutch (Wageningen and Amsterdam) schools varied between 0.46 and 0.75. These figures are in accord with those of our study but only for periods without the presence of students in the classroom. Contrary to the results of Patterson and Eatough (2000), we could not demonstrate a good correlation between PM2.5 mass and the presence of students. Neither could we confirm our previous pilot results of an association between the number of students and the PM1 concentration in the same lecturing room (Branisˇ et al., 2002). This controversy presumably arose from a limited set of data available for the previous analysis. The stronger association of indoor PM2.5 and PM1 with outdoor PM10 found in our case suggests that outdoor air may be an important source of fine indoor particles. These results are in accord with those of previous studies (Morawska et al., 2001; Wallace, 2000; Chao et al., 1998; Janssen et al., 1997), which showed that under natural ventilation outdoor particle concentrations
ARTICLE IN PRESS 148
M. Branisˇ et al. / Environmental Research 99 (2005) 143–149
could be used to predict indoor levels of smaller particles, although not so well for PM10. Lower correlation between outdoor PM10 and all of the three indoor fractions during weekends cannot be explained simply. It is possible that while during the workdays the doors and sometimes the windows could be partially open, allowing for better air exchange, on weekends the air exchange might be limited. Faster deposition of coarse fraction and/or better penetration of fine fractions through window gaps could also contribute to a lower accord between indoor and outdoor PM10 and a lower association between indoor PM10 and the finer fraction concentrations. As was already demonstrated, smaller particles below approximately PM2.5 penetrate more easily through narrow gaps than do larger ones (Mosley et al., 2001; Thornburg et al., 2001; Liu and Nazaroff, 2003). Nevertheless, the correlation coefficients between all indoor and outdoor particulate matter variables were not poor even during this time of slow or limited air exchange (r40:7 in all cases). A strong effect of ambient wind velocity on indoor PM fractions offers indirect evidence that without important sources the indoor air may be strongly influenced by ambient particle concentrations; the higher the wind speed the lower the ambient and also the indoor particulate concentrations. Adverse health effects of particulate matter are mostly attributed to finer particulate matter fractions (Schwartz and Neas, 2000; Kaiser, 2000). However, it has been demonstrated that ambient coarse particles may, under specific conditions, also have negative effects on human health (Castillejos et al., 2000; Cifuentes et al., 2000; Smith et al., 2000; Zhang et al., 2002). Coarse particles known to be made up of soil material brought in on shoes, of skin flakes, of cloth and furniture fragments, of viable molds and bacteria, of insect and mite debris, of pollen granules, and of other materials may be significant carriers of allergenic properties. As already shown in the literature, carpeted and smooth-floor classrooms contain crude dust with allergenic properties of significantly higher amounts per unit area than homes (Dybendal and Elsayed, 1994). It has also been noted that coarse particles in schools have a high allergenic and proinflammatory potential (Gravesen et al., 1986; Einarsson et al., 1995; Wheeler et al., 2000; Nordback et al., 2000; Tortolero et al., 2002). According to Fox et al. (2003), occupied schoolrooms show a much larger content of bacterial markers than ambient air of the same unoccupied rooms during weekends. Our results confirm that without other sources of particles (such as smoking) human activity is an important factor accounting for elevated indoor levels of coarse particles and that finer indoor fractions could be significantly influenced by concentrations of ambient particles, which originate in traffic-related combustion processes. According to published data coarse particles
in schools and other public spaces have a potential to negatively influence human health.
Acknowledgments The work was supported by the Czech Ministry of Education Youth and Sports under the State research scheme (Grant No. MSM 113100007). The valuable comments of Prof. Robert Holub from the Colorado School of Mines (Golden, CO, USA) are greatly appreciated. References Abt, E., Suh, H.H., Allen, G., Koutrakis, P., 2000. Characterization of indoor particle sources: a study conducted in the metropolitan Boston area. Environ. Health Perspect. 108, 35–44. Branisˇ , M., Rˇeza´cˇova´, P., Guignon, N., 2002. Fine particles in four different indoor environments. Indoor Built Environ. 11, 184–190. Castillejos, M., Borja-Aburto, V.H., Dockery, D.W., Gold, D.R., Loomis, D., 2000. Airborne coarse particles and mortality. Inhalation Toxicol. 12 (Suppl. 1), 61–72. Chao, C.Y.H., Tung, T.C.W., Burnett, J., 1998. Influence of different indoor activities on the indoor particulate levels in residential buildings. Indoor Built Environ. 7, 110–121. Cifuentes, L.A., Vega, J., Kopfer, K., 2000. Effect of fine fraction of particulate matter versus the coarse mass and other pollutants on daily mortality in Santiago, Chile. J. Air Waste Manage. Assoc. 50, 1287–1298. Daisey, J.M., Angell, W.J., Apte, M.G., 2003. Indoor air quality, ventilation and health symptoms in schools: analysis of existing information. Indoor Air 13, 53–64. Dybendal, T., Elsayed, S., 1994. Dust from carpeted and smooth floors. 6. Allergen in homes compared with those in schools in Norway. Allergy 49, 210–216. Einarsson, R., Munir, A.K.M., Dreborg, S.K.G., 1995. Allergens in school dust. 2. Major mite (DER-P-1, DER-F-1) allergens in dust from Swedish schools. J. Allergy Clin. Immunol. 95, 1049–1053. Farrow, A., Taylor, H., Golding, J., 1997. Time spent in the home by different family members. Environ. Technol. 18, 605–613. Fox, A., Harley, W., Feigley, C., Salzberg, D., Sebastian, A., Larson, L., 2003. Increased levels of bacterial markers and CO2 in occupied school rooms. J. Environ. Monit. 5, 246–252. Gorner, P., Bemer, D., Fabries, J.F., 1995. Photometer measurement of polydisperse aerosol. J. Aerosol Sci. 26, 1281–1302. Gravesen, S., Larsen, L., Gyntelberg, F., Skov, P., 1986. Demonstration of microorganisms and dust in schools and offices—an observational study of nonindustrial buildings. Allergy 41, 520–525. Janssen, N.A.H., Hoek, G., Harssema, H., Brunekreef, B., 1997. Childhood exposure to PM10: relation between personal, classroom and outdoor concentrations. Occup. Environ. Med. 54, 888–894. Janssen, N.A.H., Hoek, G., Brunekreef, B., Harssema, H., 1999. Mass concentration and elemental composition of PM10 in classrooms. Occup. Environ. Med. 56, 482–487. Jones, A.P., 1999. Indoor air quality and health. Atmos. Environ. 33, 4535–4564. Jones, N.C., Thornton, C.A., Mark, D., Harrison, R.M., 2000. Indoor/outdoor relationships of particulate matter in domestic homes with roadside, urban and rural locations. Atmos. Environ. 34, 2603–2612.
ARTICLE IN PRESS M. Branisˇ et al. / Environmental Research 99 (2005) 143–149 Kaiser, J., 2000. Air pollution—evidence mounts that tiny particles can kill. Science 289, 22–23. Lee, S.C., Chang, M., 1999. Indoor air quality investigation at five classrooms. Indoor Air Int. J. Indoor Air Qual. Clim. 9, 134–138. Liu, D.-L., Nazaroff, W.W., 2003. Particle penetration through building cracks. Aerosol Sci. Technol. 37, 565–573. Long, C.M., Suh, H.H., Koutrakis, P., 2000. Characterization of indoor particle sources using continuous mass and size monitors. J. Air Waste Manage. Assoc. 50, 1236–1250. Luoma, M., Batterman, A., 2001. Characterization of particulate emissions from occupant activities in offices. Indoor Air 11, 35–48. Maroni, M., Seifert, B., Lindvall, T. (Eds.), 1995. Indoor Air Quality, A Comprehensive Reference Book. Air Quality Monographs, vol. 3. Elsevier, Amsterdam. Marple, V.A., Rubow, K.L., Turner, W., Spengler, J.D., 1987. Lowflow rate sharp cut impactors for indoor air sampling—design and calibration. JAPCA Int. J. Air Pollut. Control Hazard. Waste Manage. 37, 1303–1307. Monn, C., 2001. Exposure assessment of air pollutants: a review on spatial heterogeneity and indoor/outdoor/personal exposure to suspended particulate matter, nitrogen dioxide and ozone. Atmos. Environ. 35, 1–32. Monn, C., Fuchs, A., Hogger, D., Junker, M., Kogelschatz, D., Roth, N., Wanner, H.-U., 1997. Particulate matter less than 10 mm (PM10) and fine particles less than 2.5 mm (PM2.5): relationships between indoor, outdoor and personal concentrations. Sci. Total Environ. 208, 15–21. Morawska, L., He, C.R., Hitchins, J., Gilbert, D., Parappukkaran, S., 2001. The relationships between indoor and outdoor airborne particles in the residential environment. Atmos. Environ. 35, 3463–3473. Mosley, R.B., Greenwell, D.J., Sparks, L.E., Guo, Z., Tucker, W.G., Fortman, R., Whitfield, C., 2001. Penetration of ambient fine particles into the indoor environmnent. Aerosol Sci. Technol. 34, 127–136. Nordback, D., Walinder, R., Wieslander, G., Smedje, G., Erwall, C., Venge, P., 2000. Indoor air pollutants in schools: nasal patency and biomarkers in nasal lavage. Allergy 55, 163–170. Owen, M.K., Ensor, D.S., Sparks, L.E., 1992. Airborne particle sizes and sources found in indoor air. Atmos. Environ. 26, 2149–2162. Patterson, E., Eatough, D.J., 2000. Indoor and outdoor relationships for ambient PM2.5 and associated pollutants: epidemiological implications in Lindon, Utah. J. Air Waste Manage. Assoc. 50, 103–110. Ramachandran, G., Adgate, J.L., Hill, N., Sexton, K., 2000. Comparison of short term variation (15-minute averages) in outdoor and indoor PM2.5 concentrations. J. Air Waste Manage. Assoc. 50, 1157–1166.
149
Roorda-Knape, M.C., Janssen, N.A.H., De Hartog, J.J., Van Vliet, P.H.N., Harssema, H., Brunekreef, B., 1998. Air pollution from traffic in city districts near major motorways. Atmos. Environ. 32 (11), 1921–1930. Schwartz, J., Neas, L.M., 2000. Fine particles are more strongly associated than coarse particles with acute respiratory health effects in schoolchildren. Epidemiology 11, 6–10. Silvers, A., Florence, B.T., Rourke, D.L., Lorimor, R.J., 1994. How children spend their time—a sample survey for use in exposure and risk assessment. Risk Anal. 14, 931–944. Siskos, P.A., Bouba, K.E., Stroubou, A.P., 2001. Determination of selected pollutants and measurement of physical parameters for the evaluation of indoor air quality in school buildings in Athens, Greece. Indoor Built Environ. 10 (3–4), 185–192. Smedje, G., Norback, D., Edling, C., 1997. Subjective indoor air quality in schools in relation to exposure. Indoor Air 7 (2), 143–150. Smith, R.L., Spitzner, D., Kim, Y., Fuentes, M., 2000. Threshold dependence of mortality effects for fine and coarse particles in Phoenix, Arizona. J. Air Waste Manage. Assoc. 50, 1367–1379. Thornburg, J., Ensor, D.S., Rodes, C.E., Lawless, P.A., Sparks, L.E., Mosley, R.B., 2001. Penetration of particles into buildings and associated physical factors. Part I: model development and computer simulations. Aerosol Sci. Technol. 34, 284–296. Tortolero, S.R., Bartholomew, L.K., Tyrrell, S., Abramson, S.L., Sockrider, M.M., Markham, C.M., Whitehead, L.W., Parcel, G.S., 2002. Environmental allergens and irritants in schools: a focus on asthma. J. School Health 72, 33–38. Turner, W.A., Olson, B.A., Allen, G.A., 2000. Calibration of sharp cut impactors for indoor and outdoor particle sampling. J. Air Waste Manage. Assoc. 50, 484–487. Wallace, L., 1996. Indoor particles: a review. J. Air Waste Manage. Assoc. 46, 98–126. Wallace, L., 2000. Correlations of personal exposure to particles with outdoor air measurements, a review of recent studies. Aerosol Sci. Technol. 32, 15–25. Wheeler, A.J., Williams, I., Beaumont, R.A., Hamilton, R.S., 2000. Characterisation of particulate matter sampled during a study of children’s personal exposure to airborne particulate matter in a UK urban environment. Environ. Monit. Assess. 65, 69–77. Yocom, J.E., 1982. Indoor–outdoor air quality relationships: a critical review. J. Air Pollut. Control Assoc. 32, 500–520. Zhang, J., Hu, W., Wei, F., Wu, G., Korn, L.R., Chapman, R., 2002. Children’s respiratory morbidity prevalence in relation to air pollution in four Chinese cities. Environ. Health Perspect. 110, 961–967.