- Email: [email protected]
Air pollution and young children's inhalation exposure to organophosphorus pesticide in an agricultural community in Japan
Environment International 31 (2005) 1123 – 1132 www.elsevier.com/locate/envint
Air pollution and young children’s inhalation exposure to organophosphorus pesticide in an agricultural community in Japan Junko Kawaharaa,*, Ryoko Horikoshib, Takashi Yamaguchic, Kazukiyo Kumagaia, Yukio Yanagisawaa a
Department of Environmental System Institute of Environmental Studies, Graduate School of Frontier Sciences, The University of Tokyo, Japan b The Graduate School of Home Economics, Japan Women’s University, Japan c Gunma Prefectural Institute of Public Health and Environmental Sciences, Japan Received 4 November 2004 Available online 24 June 2005
Abstract Assessment of airborne organophosphorus pesticides in houses of young children (1 – 6 years old) and childcare facilities was conducted following pesticide applications in an agricultural community in Japan. Trichlorfon and fenitrothion, applied in two separate periods, were frequently detected from outdoor and indoor air. Dichlorvos, the primary degradation product of trichlorfon, was also detected after the application of trichlorfon. Both the outdoors and indoor concentration of applied pesticide were shown to increase with decreasing distance from the pesticide-applied farm. Indoor concentration of these pesticides significantly correlated with outdoor concentration ( p = 0.001 for trichlorfon and p = 0.001 for fenitrothion), indicating infiltration of applied pesticide inside. Ratio of indoor to outdoor concentration (I/O ratio) of fenitrothion was higher for houses with windows open during the application than those with closed windows (median value: 0.74 vs. 0.16, p = 0.003). However, a similar trend was not observed for trichlorfon as well as dichlorvos in the first period. Dichlorvos was found to have a higher I/O ratio than trichlorfon during the period, and clear correlation between indoor concentrations of dichlorvos and those of trichlorfon suggested increased decomposition of trichlorfon in the indoor environment. Daily inhalation exposure estimated by using the fixed measurement data and time – activity questionnaire ranged from 0 to 35 ng/kg/day for trichlorfon, from 0 to 26 ng/kg/day for dichlorvos, and from 0 to 44 ng/kg/day for fenitrothion. Median inhalation exposure from indoor air accounted for 74%, 86.3%, and 45% of the daily inhalation exposure, respectively. For kindergarteners or nursery school children, inhalation exposure at childcare facilities was comparable with or more than that at home, indicating that pollution level at childcare facilities had potential of high impact on children’s exposure. Estimated daily inhalation exposures were inversely correlated to the proximity of their activity location to the pesticide-applied farm. D 2005 Elsevier Ltd. All rights reserved. Keywords: Organophosphorus pesticides; Air; Young children; Agricultural area
1. Introduction Young children’s exposure to organophosphorus (OP) pesticide and the associated health effects have been a great public health concern (Pope and Liu, 1994; Eskenazi et al.,
* Corresponding author. 5th Building of Faculty of Engineering, 7-3-1 Hongo Bunkyo-ku Tokyo, 113-8656, Japan. Tel.: +81 3 5841 7335; fax: +81 3 5841 8583. E-mail address: [email protected] (J. Kawahara). 0160-4120/$ - see front matter D 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.envint.2005.04.001
1999; Faustman et al., 2000; Sheets, 2000). OP pesticide is one of the main compounds used worldwide for agriculture. Several studies have reported pollution of ambient air by agricultural pesticide (Beard et al., 1995; Baker et al., 1996; Bradman et al., 1997; Foreman et al., 2000), as well as pollution of in indoor air (Camann et al., 1993; Mukerjee et al., 1997), where people spent most of their time. Furthermore, higher levels of OP pesticide metabolites were observed in urine of young children living in agricultural communities (Loewenherz et al., 1997; Lu et al., 2000; Fenske et al., 2002). These results indicate that young
1124
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
children living in agricultural areas have a greater potential to be exposed to OP pesticides. Use of agricultural pesticide near residential areas has been increasing in Japan due to the urbanization of agricultural areas. There are concerns that young children will suffer health effects from exposure to applied pesticide, however, available data are limited to those obtained from ambient monitoring at fixed sites near farms. Also there are few assessments of pesticide exposure for young children based on the monitoring of residential areas or public places where young children frequently spend time, while taking into consideration their unique environment and behavioral pattern (Bearer, 1995; Goldman, 1995; Cohen Hubal et al., 2000). Study of OP pesticide exposure for young children was conducted following pesticide application in an agricultural community in Japan. In this paper we present the results of a measurement of airborne OP pesticide in young children’s environment and the estimation of their inhalation exposure using the data obtained by fixed measurement and time – activity questionnaire.
2. Materials and methods 2.1. Study community and recruitment of subject The study took place in an agricultural community in the suburbs of Tokyo, Japan. The total area of the community is approximately 2600 ha. Every summer, OP pesticides are applied to 240 ha of paddy fields using remote controlled helicopters. For the other 220 ha of paddy fields, farmers using a spray pump individually applied OP pesticides. Families with young children aged below 6 years old were recruited for this survey. Briefing sessions were held in health care centers, three nursery schools, and one kindergarten in the community to ask for participation in the study. After the sessions, parents who showed positive response to participating in the study were contacted individually by phone. Details of the sampling method and schedules were explained. Only the families that gave oral consent were involved in this study. The number of families who participated in the study was 31 during the first survey period from 23rd to 26th July, and 24 during the second period from 19th to 21st August 2003. The ages of subject children ranged from 1 to 6 years old in both periods. 2.2. Pesticide application in the study community Pesticide applications to the rice paddy fields were conducted between 23rd and 26th of July and between 19th and 21st of August 2003. The active ingredient in the pesticide solution was trichlorfon [Dimethyl 2,2,2-trichloro1-hydroxyethyl phosphonate (Chemical Abstracts Service (CAS) no. 52-68-6)] in the first application period and fenitrothion [O,O-Dimethyl O-(3-methyl-4-nitrophenyl)
phosphorothioate (CAS no. 122-14-5)] in the second period. They were diluted with water prior to application and applied from approximately 3 m above the leaves of the paddies by remote-controlled helicopters. Weight of applied pesticide was approximately 1.3 kg/ha for trichlorfon and 0.65 kg/ha for fenitrothion. Pesticide applications were performed in the morning (from 5 am to 11 am) when low drift diffusion was expected due to low updraft and wind. The applications were conducted for several consecutive days (4 days in July and 3 days in August). Diagram of pesticide application with wind rose in respective days is presented in Fig. 1. The meteorological data for the survey periods were obtained from those the neighboring monitoring station (located at 5 km north east of the community), since the data were not recorded in the study community due to the malfunction of aerovane. 2.3. Target pesticide Trichlorfon and fenitrothion were targeted in this study. In addition to them, five other OP pesticides were included in the analysis: dichlorvos [2,2-dichlorovinyl dimethyl phosphate (CAS no. 62-73-7)], chlorpyrifos [O,O-Diethyl O-(3,5,6-trichloro-2-pyridinyl)phosphorothioate (CAS no. 2921-88-2)], diazinon [O,O-Diethyl O-(2-isopropyl-4methyl-6-pyrimidinyl) phosphorothioate (CAS no. 333-415)], malathion [S-[1,2-Bis(ethoxycarbonyl)ethyl] O,Odimethyl phosphorodithioate (CAS no. 121-75-5)], and fenthion [O,O-Dimethyl O-(3-methyl-4-methylmercaptophenyl) phosphorothioate (CAS no. 55-38-9)]. These were identified as commonly used pesticides in the study area by using agricultural shipment data for Japan. Dichlorvos is also the primary degradation product of trichlorfon (Samuelsen, 1987; Murphy et al., 1996). It has higher toxicity than the parent does. 2.4. Air sampling strategy Air sampling was conducted for the first 24 h following application when the air pollution would be significant, based on our previous pilot study (Kawahara and Yanagisawa, 2003a,b). Sampling at each subject’s house and childcare facilities was scheduled depending on the day of application in the area. Children on holiday as well as young children who do not attend daycare centers were expected to have spent most of their time at home. For children who attend childcare facilities, their own house and the facilities were expected to be their main environments. For the former case, air sampling was performed at outdoors and indoors of their own houses only. Additional sampling was performed at outdoors and indoors of childcare facilities of the children for the latter case. In the houses, air sample was collected for 24 h at outdoors and indoor where the subject spent most of their time, such as the child’s room or the living room. In the childcare facilities, air was sampled at the classrooms and the playground.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
1125
N
a) 1st survey period (23rd to 26th July, 2003) 25 20 15 10 5 0
Area a-4 (26th Jul.)
CALM: 12.5% Average wind speed: 1.6m/s rd
Area a-2 (24th Jul.)
Area a-1 (23 Jul.)
330° 300° 270°
25 20 15 10 5 0
25 20 15 10 5 0
0°
25 20 15 10 5 0
30° 60° 90° 120°
240° 210°
180°
150°
frequency(%)
Area a-3 (25th Jul.)
CALM: 20.8% Average wind speed: 1.2m/s
CALM: 12.5% Average wind speed: 1.5m/s
25 20 15 10 5 0
1000m
CALM: 20.8% Average wind speed: 0.9m/s
b) 2nd. Survey period (19th to 21st August, 2003) 25 20 15 10 5 0
Area b-3 (20th Aug.)
CALM:41.7% Average wind speed:0.9m/s Area b-2 (21st Aug.) 25 20 15 10 5 0
25 20 15 10 5 0
Area b-1, 19th Aug CALM: 16.7% Average wind speed: 1.1m/s
CALM: 8.6% Average wind speed: 1.6m/s
1000m
:Town border :Area border of pesticide application by day :House (Families participated in both survey periods) :House (Families participated in either 1st or 2nd survey period) :Childcare facilities (participated in both survey period) :Childcare facilities (participated in either 1st or 2nd survey period) Fig. 1. Diagram of pesticide application with wind rose in respective days; a) 23rd to 26th July and b) 19th to 21st August; the date wrote in each area means the date of pesticide application within the area.
2.5. Air sampling procedure A Chromosorb102 sampling tube (8 mm OD 110 mm length, adsorbent 50/100 mg, SKC Inc.) was used for air sampling. Our previous study obtained good collection
efficiencies of the adsorbent for the target OP pesticides (Kawahara and Yanagisawa, 2003b), and earlier works had shown the efficiencies of the adsorbent for pesticides (e.g. dichlorvos or chlorpyrifos) being comparable to PUF (polyurethane foam), Tenax GC, or ORBO42 (Roper and
1126
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Wright, 1984; Thomas and Nishioka, 1985; Leidy and Wright, 1991). Chromosorb102 tube was preceded by a 37 mm diameter quarts fiber filters (QM-A, Whatman Inc.), and air was drawn through them at a flow rate of 2.0 l/min by using portable air pump (Aircheck2000, SKC Inc.). Inlet of the samplers was set at 70 cm to 100 cm above the floor or ground to ensure air collection at the children’s breathing zone. Flow rates were determined before and after the sampling period using a primary flow meter (Dry Cal DCLite, BIOS Inc.). To reduce the noise of the pump, the main body was put in a box filled with noise adsorbing material, or a quiet-type air pump (IAQ-3, SKC. Inc.) was used instead. After sampling, the sampling tubes and filter cassettes were capped, sealed in a zip-lock bag, stored in an icebox at 0 -C, and sent back to the laboratory. The samples were kept at 10 -C in a freezer until the day of analysis (within 14 days after sampling). The field reproducibility of this method for analytes was evaluated through duplicate sampling. 2.6. Sample analysis Filters and adsorbents were extracted twice with 2 ml of acetone in an ultrasonic bath for 40 min (20 min 2). An aliquot (10 Al) of acetic anhydride was added as the derivertizing agent for trichlorfon to prevent its thermal decomposition to dichlorvos in the following gas chromatographic analysis (Bowman and Dame, 1974; Slahck, 1988). To examine the breakthrough of target compounds in the glass tube sampler, adsorbents in the primary and back-up section were extracted and analyzed separately. Breakthrough was defined as the point at which 5% of the total collected on the entire adsorbents was found in the back-up section. For samples where breakthrough was found in back-up section, the data was removed from quantitative evaluation. One analytical blank sample was prepared in each analytical batch. The target compounds were analyzed with a gas chromatograph (HP6890 Hewlett Packard Inc.) equipped with a flame photometric detector (FPD). Three micro liters of the extract was injected using the pulsed splitless mode. A column, 30 m length 0.32 mm ID HP-1 with 0.25 Am film thickness (DB-1MS, J&W Inc.) was used for the gas chromatographic analysis. Super purified helium gas was used as the carrier gas and set at a flow rate of 5.3 ml/min. Injector temperature was set at 200 -C. Initial oven temperature was set at 60 -C and held for 2 min, then ramped to 250 -C at 8 -C/min and then held for 3 min. Detector temperature was set at 240 -C. Calibration curve was prepared by plotting the area of detected peak against concentration of five different standard solutions (pesticide standard mixture no. 15, 16, and 17; Wako Inc.). Linearity was determined by curve fitting. Detail of analytical procedure and result is reported elsewhere (Kawahara and Yanagisawa, 2003b). Table 1 shows the spike recovery of target pesticides from Chromosorb102 tubes that had sampled 40 m 3 of
Table 1 Spike recovery of target analytes from ambient air matrix Compounds
Trichlorfon Dichlorvos Fenitrothion Chlorpyrifos Diazinon Malathion Fenthion
LODa MLOQb (ng/ml) (ng/m3)
Extraction efficiencyc (%)
4 3 3 4 3 4 3
88 T 5.8 92 T 12 90 T 3.5 91 T 3.8 82 T 1.9 91 T 3.0 81 T1.8
10 6 8 6 6 8 7
Gas Particle (Chromosorb102) (glass fiber filter) mean T S.D. (n = 3) mean T S.D. (n = 4) 61 T 5.2 119 T 2.3 100 T 3.2 99 T 2.3 102 T 2.2 87 T 0.9 125 T 1.3
a Instrumental limit of quantitation (LOD) were determined with analytical standards in solvent (no matrix effect). b Method limit of quantitation (MLOQ). Extraction efficiency was based on laboratory spike recovery tests from air matrix. c Method extraction efficiency was based on spike recovery tests from Chromosorb102 tubes containing ambient air matrix.
ambient air, the instrument limits of detection (LOD), and the method limit of quantitation (MLOQ). Recoveries of pesticides from ambient air matrix fell within the range 80 to 100% and from 60 to 125% for filters, with relative standard deviation of the triplicate sample below 12%. Pesticide concentrations in the air samples were adjusted by these extraction efficiencies. Instrumental LOD was defined as the concentration at which the height of the signal peak is three times the baseline noise. The method had MLOQ of 6 to 10 ng/m3 by 24-h air sampling (sample volume: 2.8 m3). Whitmore et al. (1994) had used PUF plug in the non-occupational pesticide exposure study. In their report, the detection limits for target pesticides in air sampled at a flow rate of 3.8 l/min for 24 h were shown to be 1.5 – 79.0 ng/m3 for dichlorvos and 0.5– 4.5 ng/m3 for chlorpyrifos analyzed with GC/ECD, and 11– 48 ng/m3 for diazinon, and 10 –60 ng/m3 for malathion with GC/MS. These detection limits are comparable to those by our method. The detail of analytical procedure and result are reported elsewhere (Kawahara and Yanagisawa, 2003b). In the statistical analysis, sample with no recognizable signal (signal to noise ratio <3:1) was assigned values of zero, and those below the respective MLOQ were assigned one half the value of the MLOQ. 2.7. Time – activity log and questionnaire Parents were asked to record the location and the activity level of their child on the day of air sampling, using the time– location/activity log sheet. The log sheet consists of two categories, one for the location and one for the level of their child’s activity for a period of 24 h. Parents checked the appropriate box to report the location of their child categorized by home indoor, home outdoor, childcare facility, other place indoor, or other place outdoor, every 30 min. Activity levels were recorded by checking box of Fheavy_, Fmoderate_, Flight_, and Fresting_. Time – location
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
data for when the subjects were at childcare facilities was obtained from childcare workers in the facilities. Parents were also asked to answer a self-administrated questionnaire about the physical characteristics of their children, the family members’ activities in the houses on the day of the air sampling, and the history of pesticide usage in or around their houses. 2.8. Estimation of inhalation exposure In this study, fixed measurements in each subject’s microenvironment and time – location/activity questionnaires completed by parents were used for estimating their inhalation exposure to the applied pesticides. The subject’s daily inhalation exposure following pesticide application was estimated using the following equation (US EPA, 1997; WHO, 1999): Daily inhalation exposure ðAg=kg=dayÞ ¼ RðCi IRinh EDi Þ=ðBW ATÞ: Here i is the given location where the child was at a given period of time (indoor/outdoor at home, childcare facility, and other place were given in this study), C i is the OP pesticide concentration in the air at the given locations (ng/ m3). IRinh is the inhalation rate per hour (m3/h). The inhalation rate for each subject was estimated using
1127
reference values for normal Japanese children as a function of age, reported by International Atomic Energy Agency (1998). In the report, activity levels were limited to that of resting, light, and heavy. Therefore, inhalation rate for Fmoderate_ activity obtained from the time –activity log was substituted by Flight_ activity. EDi is the exposure duration (h) to air at each given location. BW is body weight of subject (kg), as reported by the parents. AT is the average time of exposure (24 h). For a few children who lacked time –activity data, mean values of data for the same age group was substituted.
3. Results 3.1. Sampling performance Air was sampled at 31 houses and 2 childcare facilities where the 14 subjects were attending in the first survey period. In the second period, 24 houses and 4 childcare facilities of 16 children were sampled. All sampling started within 5 h after the pesticide application except for that of two houses and one childcare facility, which started within 12 h. This was due to a change of the application schedule. One outdoor air sampling in each season failed due to malfunction or accidental power loss of the sampling pump. Breakthrough was found in 21% of the samples for
Table 2 Average concentrations of the target pesticides in air sample of homes and childcare facilities for 24 h following pesticide application (ng/m3)a Compound
Home (ng/m3)
Childcare facilities (ng/m3)
Outdoor
Indoor
Outdoor
Indoor
Frequency Median Mean Range (%)b
Frequency Median Mean Range Frequency Median Mean Range (%) (%)
Frequency Median Mean Range (%)
First application period (23rd – 26th July) (n = 26) (n = 31) Trichlorfon 24 (92) 43 63 0 – 367c 20 (65) Dichlorvos 25 (96) 13 19 0 – 55c 29 (94) d Fenitrothion 5 (19) 0 3 0 – 33 9 (29) Chlorpyrifos 1 (4) 0 2 0 – 51 7 (23) Diazinon 0 (0) 0 0 0 0 (0) Malathion 1 (4) 0 0 0 – 12 1 (3) Fenthion 0 (0) 0 0 0 1 (3)
9 9 0 0 0 0 0
13 13 2 2 0 0 0
0 – 50 0 – 32c 0 – 13 0 – 12 0 0–5 0–2
(n = 2) 2 (100) 2 (100) 1 (50) 0 (0) 0 (0) 1 (50) 0 (0)
40 23 8 0 8 5 0
40 23 8 0 8 5 0
26 – 53 9 – 38 0 – 16 0 0 0 – 11 0
(n = 2) 2 (100) 2 (100) 1 (50) 0 (0) 0 (0) 0 (0) 0 (0)
28 13 0 0 0 0 0
28 13 0 0 0 0 0
15 – 42 8 – 18 0–5 0 0 0 0
Second application period (19th – 21st August) (n = 24) (n = 24) Trichlorfon 13 (56) 9 12 0 – 61 7 (30) Dichlorvos 23 (100) 7 8 4 – 11c 22 (95) Fenitrothion 23 (100) 51 136 2 – 567c 19 (82) Chlorpyrifos 0 (0) 0 0 0 6 (26) Diazinon 9 (39) 0 1 0–5 6 (25) Malathion 9 (39) 0 1 0 – 11c 7 (30) Fenthion 0 (0) 0 0 0 1 (4)
0 7 18 0 0 0 0
9 7 23 4 1 0 0
0 – 117 0 – 16c 0 – 56c 0 – 51c 0–9 0c 0–2
(n = 4) 3 (75) 4 (100) 4 (100) 0 (0) 0 (0) 0 (0) 0 (0)
9 6 108 0 0 0 0
8 6 122 0 0 0 0
0 – 13 5 – 6c 41 – 232 0 0 0 0
(n = 4) 1 (25) 4 (100) 4 (100) 1 (25) 1 (25) 0 (0) 0 (0)
0 5 11 0 0 0 0
2 5 12 0 0 0 0
0–7 4 – 5c 2 – 25 0–2 0–2 0 0
a Method limits of quantitation (MLOQ) in air (ng/m3) for 24 h air sampling (sample volume: 2.8 m3); trichlorfon, 10; dichlorvos, 6; fenitrothion, 8, chlorpyrifos, 6; malathion, 6; fenthion 8. Data LOD); percentages are in parentheses. c Samples of which breakthrough were found in Chromosorb102, was removed from statistical analysis. d Samples that were not detectable (signal:noise ratio <3:1) was assigned values of zero.
1128
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
dichlorvos and malathion, 18% for fenitrothion, and 2% for chlorpyrifos. Breakthroughs for malathion and fenitrothion were mainly observed in samples collected after the rain. Duplicate samplers showed the percent deference to be below 20%.
further away (348 vs. 70 ng/m3, Mann – Whitney U – Wilcoxon’s rank sum W, p = 0.017).
3.2. Pesticide use in houses
Applied pesticide was detected from indoor air in each period. Inverse correlations were also observed between indoor air concentrations of applied pesticides and distances from pesticide-applied farm (Spearman’s r = 0.507, p = 0.003, n = 33 for trichlorfon in the first period, and r = 0.570, p = 0.011 n = 19 for fenitrothion in the second). However, the trend was less significant for dichlorvos in the first survey period (Spearman’s r = 0.329, p = 0.081, n = 29).
Table 2 provides the frequency of detection, median, mean, and range of each target compound from air samples by sampled location. In the first survey period, trichlorfon was frequently detected from outdoor air at concentrations from below the detection limit to 367 ng/m3. Dichlorvos was also detected from outdoor in this season, at concentrations ranged from below the detection limit to 55 ng/m3. In the second survey period, outdoor concentration of fenitrothion ranged from 2 to 567 ng/m3. Dichlorvos was also frequently detected outdoors but the concentration was lower than the first period. Besides applied pesticides, diazinon and malathion were also detected outdoors. Outdoor concentrations of applied pesticides showed significant inverse correlation with the distances from pesticide-applied farm to the house or childcare facilities (Spearman’s r = 0.608, p = 0.001, n = 28 for trichlorfon in the first application period, and r = 0.801, p < 0.001. n = 18 for fenitrothion in the second period). However, a similar trend was not seen for dichlorvos in the first period (Spearman’s r = 0.155, p = 0.46, n = 25). Houses and childcare facilities were categorized to < 30 m and > 30 m from the pesticide-applied field. 30 m is the distance that finer particles (< 100 Am) are expected to drift when under the 1 m/s wind (Association of Agriculture, Forestry and Fisheries Aviation, Japan, 1996). Median outdoor concentrations of trichlorfon in the first survey period was 4.7 times higher for houses and facilities within 30 m than those further away (median value: 137 vs. 29 ng/m3, Mann – Whitney U – Wilcoxon’s rank sum W, p = 0.007). In the second survey period, median outdoors concentrations of fenitrothion within 30 m was 5.0 times higher than those
Trichlorfon 50 Indoor concentration. (ng/m3)
3.3. Outdoor concentration and relationship between distance from applied farm
60
Dichlorvos
40
30
20
Trichlorfon: Spearman's r =0.565 (p =0.001)
10
Dichlorvos Spearman's r =0.310 (p =0.116)
0 0
100 200 300 400 Outdoor concentration (ng/m3)
500
100 90
MEP
80 Indoor concentration. (ng/m3)
In the first survey period, 9 families reported the use of pest control measures around their house within a month of air sampling, and one family within a week. Five families had used products including OP pesticides within a month (2 for trichlorfon, 1 for dichlorvos, and 1 for acephate). One family had used products including glyhosate within a week. In the second period, 7 families reported pest control use around their house within a month, of which 3 families reported the use of OP pesticides (1 for fenitrothion and 2 for trichlorfon). Pest control inside of the house was reported from one family (within a month) in the first period but no use of OP pesticide.
3.4. Indoor pesticide and factor that influenced to the concentration
70 60 50 40 30
Fenitrothion: Spearman's r =0.686 (p =0.001)
20 10 0 0
100 200 300 400 Outdoor concentration (ng/m3)
500
Fig. 2. Relationships between indoor and outdoor concentrations for applied pesticides. a) Trichlorfon and dichlorvos in the first survey period, b) fenitrothion in the second survey period.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Significant relation was seen between indoor and outdoor concentrations of applied pesticides in both periods (Spearman’s r = 0.565, p = 0.001, n = 32, for trichlorfon, and r = 0.686, p = 0.001, n = 20, for fenitrothion), indicating evident impact of outside pesticide application on indoor air pollution (Fig. 2a, b). But for dichlorvos in the first period, the relationship was not significant (Spearman’s r = 0.310, p = 0.116, n = 27). Median of the indoor to outdoor air concentration ratio (I/O ratio) was 0.29 for trichlorfon, and 0.74 for dichlorvos in the first period. The I/O ratio was 0.18 for fenitrothion in the second period. Chlorpyrifos was detected in indoor air sample of 13 houses in the two survey periods. None of them, except one house, were detected chlorpyrifos in outdoor air. However, seven had a history of termite control within 5 to 10 years (range of indoor concentration: 1 to 51 ng/ m3). Chlorpyrifos was found only in the outdoor air of one house, with its concentration of 51 ng/m3, had 4 ng/ m3 of chlorpyrifos in indoor air. Since this house also had a history of termite control within 7 years, it was not clear how outdoor pollution by the pesticide influenced indoor air. 3.5. Window opening and indoor air pollution The reported resident’s activity on the day of pesticide application showed that 20 out of 31 families in the first survey period and 16 out of 23 families in the second period opened the windows of their houses during the morning (5 am to 11 pm). I/O ratio was higher for the houses with window open in the morning than those with windows closed (median value: 0.74 vs. 0.16, Mann – Whitney U – Wilcoxon’s rank sum W, p = 0.004). In first survey period, however, I/O ratio was not significantly affected by window opening (median value: 0.29 vs. 0.25, Mann – Whitney U –
1129
Wilcoxon’s rank sum W, p = 0.802 for trichlorfon, and 0.99 vs. 0.56, p = 0.354 for dichlorvos). 3.6. Source of dichlorvos in indoor Using non-parametric analysis, significant difference in the I/O ratio was found between trichlorfon and dichlorvos for the first period (Wilcoxon’s matched pairs signed-rank test: p = 0.002). This suggests that there is a different mechanism of indoor air pollution of these pesticides. As already presented, indoor concentrations of dichlorvos in the first period did not show associations with outdoor concentration, distance from pesticide-applied farm, or window opening by the resident. We found clear correlation between indoor concentrations of dichlorvos and those of trichlorfon (Spearman’s r = 0.544, p = 0.002, n = 29). This finding suggested that the indoor air pollution by dichlorvos was not only caused by the infiltration from outdoors, but also secondary generation from trichlorfon in the indoor environment. 3.7. Time –activity data of subjects Time – location data of subject children was obtained from families and the childcare facilities in both survey periods, although two families failed to fill the time – location log sheet. Table 3 provides the statistical descriptions of the time subject children spent at categorized locations by day category and survey period. Children who were on holiday as well as children who did not attend childcare facilities spent 60 –80% of day at home indoors and 0– 8% of day at home outdoors. Children who attended childcare facilities spent about 60% of their day at home and 30% of their day at childcare facilities. These children had more time to spend outdoors at the facilities than at their own houses in the second survey period (median value: 2.0 and 0.3 h/day,
Table 3 Time spent in the categorized microenvironment during survey perioda Survey period
Location Category
23rd – 26th July
Indoor
Without attendance at childcare facilitiesa n
Outdoor
19th – 21st August Indoor
Outdoor
Own house 14 Childcare facility All other indoor location Total indoors Own house Childcare facility All other outdoor location Total outdoors Own house 6 Childcare facility All other indoor location Total indoors Own house Childcare facility All other outdoor location Total outdoors
With attendance at childcare facilities
Median (h)
Mean (h)
Range (h)
n
Median (h)
Mean (h)
Range (h)
21.0 – 0.3 22.0 0.5 – 0.5 2.0 19.0 – 1.5 21.5 1.0 – 1.5 2.5
20.6 – 1.4 21.9 0.6 – 1.5 2.1 18.3 – 2.9 21.1 0.9 – 1.9 2.9
18.0 – 23.5 – 0.0 – 4.5 17.5 – 24.0 0.0 – 2.0 – 0.0 – 4.5 0.0 – 6.5 10.0 – 22.5 – 0.0 – 11.0 13.5 – 23.0 0.0 – 2.5 – 1.0 – 3.0 1.0 – 5.5
14
14.5 5.6 0.3 21.3 0.0 1.6 0.5 2.8 15.0 5.6 0.0 21.7 0.3 2.0 0.0 2.3
14.3 7.5 0.9 21.1 0.5 1.2 0.8 2.9 15.4 5.7 0.4 21.6 0.6 1.5 0.4 2.4
11.5 – 16.8 2.0 – 7.5 0.0 – 3.0 19.0 – 23.5 0.0 – 2.0 0.0 – 2.5 0.0 – 2.3 0.0 – 5.0 12.5 – 170 5.0 – 8.4 0.0 – 2.0 0.0 – 4.0 0.0 – 3.5 0.0 – 3.5 0.0 – 1.5 1.0 – 4.0
16
a Children without attendance at childcare facilities include children who did not attend childcare facilities and children who were on holiday on the day of monitoring.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Trichlorfon in the 1st survey period Home Indoor Childcare facilities Indoor Home Outdoor Childcare facilities Outdoor
0
10
20
30
25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 1 1
40
Daily inhalation exposure (ng/kg/day)
11 10 9 8 7
Rank
30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 1 1
Rank
Rank
1130
6 5
Dichlorvos in the 1st survey period
Fenitrothion in the 2nd survey period
4
Home Indoor Childcare facilities Indoor Home Outdoor Childcare facilities Outdoor
Home Indoor Childcare facilities Indoor Home Outdoor Childcare facilities Outdoor
3 2 1
0
10
20
30
40
0
Daily inhalation exposure (ng/kg/day)
10
20
30
40
50
Daily inhalation exposure (ng/kg/day)
Fig. 3. Estimated OP pesticide exposure from air (ng/kg/day) for individual subject ranked in ascending order, and the contribution of each microenvironment.
3.8. Daily inhalation exposure for subject children following pesticide application Estimation of daily inhalation exposure to trichlorfon and dichlorvos following the first application was performed for 30 and 25 subjects, respectively, and fenitrothion following the second application for 11 subjects, whose microenvironment monitoring was completed with acceptable sampling performance. Estimated inhalation exposure ranged from 0 to 35 ng/kg/day for trichlorfon, from 0 to 26 ng/kg/day for dichlorvos, and from 5 to 44 ng/kg/day for fenitrothion (Fig. 3 ). These estimated exposure levels were below the level of the acceptable daily intake for food intake set by the Ministry of Health, Labor and Welfare in Japan (trichlorfon, 10 Ag/kg/day; dichlorvos, 3.3 Ag/kg/day; fenitrothion, 5.0 Ag/kg/day). The hazard quotients (HQs), the ratio of the estimated dose to the reference dose, was from 0 to 0.0035 for trichlorfon, from 0 to 0.0078 for dichlorvos, and from 0.0011 to 0.0088 for fenitrothion.
based on the median value, and 86% for dichlorvos in the first application period. Whereas for fenitrothion in the second period, outdoor exposure accounted for a relatively high portion of the daily inhalation exposure at 55%. This higher amount of outdoor exposure is thought to be due to the low infiltration of fenitrothion indoors (Fig. 3). For children who attended childcare facilities, inhalation exposure during their stay at the facilities was comparable with that at home (median value: 78% for trichlorfon, 55% for fenitrothion), although the total time they spent at the facilities was less than that at home. This result demonstrated the great impact of the environment for their exposure. 50
Daily inhalation exposure (ng/kg/day)
respectively, Wilcoxon’s matched pairs signed-rank test: p = 0.04). The trend was less significant in the first period (median value: 1.6 and 0.0 h/day, respectively, Wilcoxon’s matched pairs signed-rank test: p = 0.06). As for the activity log, only a few parents recorded the subject’s activity level other than resting. This is supposedly due to the limitation of the ability to follow-up on the child’s activity. Therefore, in the calculation of their inhalation exposure, inhalation rate of a light activity was used for activity levels other than resting.
Trichlorfon/ 1st. period Dichlorvos/ 1st. period Fenitrothion/ 2nd. period
40
30
20
10
0 0
3.9. Inhalation exposure and activity location Inhalation exposure to trichlorfon from indoor air accounted for 74% of the daily inhalation exposure
100
200
300
400
500
Distance from pesticide applied farm to subject's activity location (m)
Fig. 4. Relationship between estimated daily inhalation exposure and average distance from applied farm to subject’s activity location.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
The relationship between estimated inhalation exposure and the distance from pesticide-applied farm to their activity location was evaluated. For children with attendance at childcare facilities, time-weighted-average distances from applied farm to their own houses or facilities were substituted, because pesticide exposure at the facilities was considered not to be negligible. The distance was calculated by weighting the distance from farm to home or facilities by the time the subject spent at each location. Results showed that subjects had higher inhalation exposure to trichlorfon and dichlorvos if they stayed closer to the pesticide applied-farm. As for fenitrothion in the second survey period, the trend was not so significant (Fig. 4).
4. Conclusion OP pesticides were frequently detected from the air of residential environments and childcare facilities following pesticide application in an agricultural community. Correlation between indoor and outdoor concentration indicated evident impact of outside pesticide application on indoor air pollution. Houses and childcare facilities located near the applied farms had higher air pollution levels than those farther away. Window opening during pesticide application was found to increases the infiltration of the pesticide indoors. The study indicated that distance from pesticide farm to children’s activity location is an important factor that affects their inhalation exposure. Moreover, depending on nature of the applied pesticide, exposure from indoor air constitutes a significant fraction of daily inhalation exposure to the pesticide, although the application was conducted outside of the buildings. The pesticide pollution level was found to have potentially high impact on the children’s exposure. This suggested the need for reducing the pollution in such an environment where young children spend time frequently. More detailed air monitoring of children’s microenvironment, time – activity data, and identification of other exposure pathways are needed for more accurate estimation and development of effective measures to reduce exposure of children. Acknowledgements We appreciate all the participants in this study and the office staffs of the agriculture public corporation in the studied community for their understanding and cooperation in our study. This study was funded by Sumitomo Foundation 2002. References Association of Agriculture, Forestry and Fisheries Aviation of Japan. Guideline for agriculture and forestry aviation safety measure; 1996.
1131
Baker LW, Fitzell DL, Seiber JN, Parker TR, Shibamoto T, Poore MW, et al. Ambient air concentration of pesticides in California. Environ Sci Technol 1996;30:1365 – 8. Beard J, Wise WV, Sullivan G. Exposure to pesticides in ambient air. Aust J Public Health 1995;19:357 – 62. Bearer CF. How are children different from adults? Environ Health Perspect 1995;103(Suppl. 6):7 – 12. Bowman PB, Dame PW. Gas – liquid chromatography of trichlorfon in soluble powder formulations. J Assoc Off Anal Chem 1974;57: 1128 – 31. Bradman MA, Harnly ME, Draper W, Seidel S, Teran S, Wakeham D, et al. Pesticide exposure to children from California’s central valley: results of a pilot study. J Expo Anal Environ Epidemiol 1997;7:217 – 34. Camann DE, Geno PW, Harding HJ, Giardino NJ, Bond AE, Lewis RG, et al. A pilot study of pesticides in indoor and in relation to agricultural application. Proceedings of the 6th International Conference on Indoor Air Quality and Climate, Indoor Air 1993, Helsinki, Finland, vol. 2; 1993. p. 207 – 12. Cohen Hubal EA, Sheldon LS, Burke JM, McCurdy TR, Berry MR, Rigas ML, et al. Children’s exposure assessment: a review of factors influencing children’s exposure, and the data available to characterize and assess that exposure. Environ Health Perspect 2000;108:475 – 86. Eskenazi B, Bradman A, Castorina R. Exposure of children to organophosphate pesticides and their potential adverse health effects. Environ Health Perspect 1999;107(Suppl. 3):409 – 19. Faustman EM, Silbernagel SM, Fenske RA, Burbacher TM, Ponce RA. Mechanism underlying children’s susceptibility to environmental toxicants. Environ Health Perspect 2000;108(Suppl. 1):13 – 21. Fenske RA, Lu C, Barr D, Needham L. Children’s exposure to chlorpyrifos and parathion in an agricultural community in central Washington State. Environ Health Perspect 2002;110:549 – 53. Foreman WT, Majewski MS, Goolsby DA, Wiebe FW, Coupe RH. Pesticides in the atmosphere of the Mississippi River Valley: Part IIR Air. Sci Total Environ 2000;248:213 – 6. Goldman LR. Children—unique and vulnerableR Environmental risks facing children and recommendations for response. Environ Health Perspect 1995;103(Suppl. 6):13 – 8. International Atomic Energy Agency. Compilation of anatomical, physiological and metabolic characteristics for a reference Asian man, vol. 2. Country reports. H. Physical and anatomical data, and part of physiological and metabolic data for normal Japanese with special reference to establishing reference Asian man model for the anatomical characteristics; 1998. Kawahara J, Yanagisawa Y. Indoor and outdoor organophosphorus pesticides in an agricultural area in Japan. Proceedings ofR 7th International Conference on Healthy Buildings, Singapore, vol. 1; 2003a. p. 452 – 5. Kawahara J, Yanagisawa Y. Simplified analytical method for organophosphorus pesticide in air using adsorbent filled tube and solvent extraction by ultrasonic wave. Bunseki Kagaku 2003b;52:519 – 25. Leidy RB, Wright CG. Trapping efficiency of selected adsorbents for airborne pesticides. Environ Sci Health B 1991;26:367 – 82. Loewenherz C, Fenske RA, Simcox NJ, Bellamy G, Kalman D. Biological monitoring of organophosphorus pesticide exposure among children of agricultural workers in central Washington State. Environ Health Perspect 1997;105:1344 – 653. Lu C, Fenske RA, Simcox NJ, Kalman D. Pesticide exposure of children in an agricultural community: evidence of household proximity to farmland and take home exposure pathways. Environ Res A 2000;84: 290 – 302. Mukerjee WD, Ellenson WD, Lewis RG, Stevens RK, Somerville MC, Shadwick D, et al. An environmental scoping study in the Lower Rio grand Grande Valley of Texas: IIIR Residential microenvironmental monitoring for air, house dust, and soil. Environ Int 1997;23:657 – 73. Murphy KC, Cooper RJ, Clark JM. Volatile and dislodgeable residues following trichlorfon and isazofos application to turfgrass and implications for human exposure. Crop Sci 1996;36:1446 – 54.
1132
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Pope CN, Liu J. Age related differences in sensitivity to organophosphorus pesticides. Environ Toxicol Pharmacol 1994;4:309 – 14. Roper EM, Wright CG. Sampling efficiency of five solid sorbents for trapping airborne pesticides. Bull Environ Contam Toxicol 1984;33: 476 – 783. Samuelsen OB. Aeration rate, pH, and temperature effects on the degradation of trichlorfon to DDVP and the half-lives of trichlorfon and DDVP in seawater. Aquaculture 1987;66:373 – 80. Sheets LP. A consideration of age-dependent differences in susceptibility to organophosphorus and pyrethroid insecticides. Neuro Toxicol 2000;21: 57 – 64. Slahck SC. Direct gas chlomatographic analysis of trichlorfon for its decomposition products. J Assoc Off Anal Chem 1988;71:440 – 2.
Thomas C, Nishioka YA. Sampling of airborne pesticide using Chromosorb102. Bull Environ Contam Toxicol 1985;35:460 – 5. US EPA (United States Environmental Protection Agency). Exposure Factors Handbook. EPA/600/P-95/002/Fa; 1997. Whitmore RW, Immerman FW, Camann DE, Bond AE, Lewis RG, Schaum JL. Non-occupational exposures to pesticide for residents of two US cities. Arch Environ Contam Toxicol 1994;26:47 – 59. WHO (World Health Organization). Principle for the assessment of risks to human health from exposure to chemicals. Environ Health Criteria 1999;210.
Air pollution and young children’s inhalation exposure to organophosphorus pesticide in an agricultural community in Japan Junko Kawaharaa,*, Ryoko Horikoshib, Takashi Yamaguchic, Kazukiyo Kumagaia, Yukio Yanagisawaa a
Department of Environmental System Institute of Environmental Studies, Graduate School of Frontier Sciences, The University of Tokyo, Japan b The Graduate School of Home Economics, Japan Women’s University, Japan c Gunma Prefectural Institute of Public Health and Environmental Sciences, Japan Received 4 November 2004 Available online 24 June 2005
Abstract Assessment of airborne organophosphorus pesticides in houses of young children (1 – 6 years old) and childcare facilities was conducted following pesticide applications in an agricultural community in Japan. Trichlorfon and fenitrothion, applied in two separate periods, were frequently detected from outdoor and indoor air. Dichlorvos, the primary degradation product of trichlorfon, was also detected after the application of trichlorfon. Both the outdoors and indoor concentration of applied pesticide were shown to increase with decreasing distance from the pesticide-applied farm. Indoor concentration of these pesticides significantly correlated with outdoor concentration ( p = 0.001 for trichlorfon and p = 0.001 for fenitrothion), indicating infiltration of applied pesticide inside. Ratio of indoor to outdoor concentration (I/O ratio) of fenitrothion was higher for houses with windows open during the application than those with closed windows (median value: 0.74 vs. 0.16, p = 0.003). However, a similar trend was not observed for trichlorfon as well as dichlorvos in the first period. Dichlorvos was found to have a higher I/O ratio than trichlorfon during the period, and clear correlation between indoor concentrations of dichlorvos and those of trichlorfon suggested increased decomposition of trichlorfon in the indoor environment. Daily inhalation exposure estimated by using the fixed measurement data and time – activity questionnaire ranged from 0 to 35 ng/kg/day for trichlorfon, from 0 to 26 ng/kg/day for dichlorvos, and from 0 to 44 ng/kg/day for fenitrothion. Median inhalation exposure from indoor air accounted for 74%, 86.3%, and 45% of the daily inhalation exposure, respectively. For kindergarteners or nursery school children, inhalation exposure at childcare facilities was comparable with or more than that at home, indicating that pollution level at childcare facilities had potential of high impact on children’s exposure. Estimated daily inhalation exposures were inversely correlated to the proximity of their activity location to the pesticide-applied farm. D 2005 Elsevier Ltd. All rights reserved. Keywords: Organophosphorus pesticides; Air; Young children; Agricultural area
1. Introduction Young children’s exposure to organophosphorus (OP) pesticide and the associated health effects have been a great public health concern (Pope and Liu, 1994; Eskenazi et al.,
* Corresponding author. 5th Building of Faculty of Engineering, 7-3-1 Hongo Bunkyo-ku Tokyo, 113-8656, Japan. Tel.: +81 3 5841 7335; fax: +81 3 5841 8583. E-mail address: [email protected] (J. Kawahara). 0160-4120/$ - see front matter D 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.envint.2005.04.001
1999; Faustman et al., 2000; Sheets, 2000). OP pesticide is one of the main compounds used worldwide for agriculture. Several studies have reported pollution of ambient air by agricultural pesticide (Beard et al., 1995; Baker et al., 1996; Bradman et al., 1997; Foreman et al., 2000), as well as pollution of in indoor air (Camann et al., 1993; Mukerjee et al., 1997), where people spent most of their time. Furthermore, higher levels of OP pesticide metabolites were observed in urine of young children living in agricultural communities (Loewenherz et al., 1997; Lu et al., 2000; Fenske et al., 2002). These results indicate that young
1124
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
children living in agricultural areas have a greater potential to be exposed to OP pesticides. Use of agricultural pesticide near residential areas has been increasing in Japan due to the urbanization of agricultural areas. There are concerns that young children will suffer health effects from exposure to applied pesticide, however, available data are limited to those obtained from ambient monitoring at fixed sites near farms. Also there are few assessments of pesticide exposure for young children based on the monitoring of residential areas or public places where young children frequently spend time, while taking into consideration their unique environment and behavioral pattern (Bearer, 1995; Goldman, 1995; Cohen Hubal et al., 2000). Study of OP pesticide exposure for young children was conducted following pesticide application in an agricultural community in Japan. In this paper we present the results of a measurement of airborne OP pesticide in young children’s environment and the estimation of their inhalation exposure using the data obtained by fixed measurement and time – activity questionnaire.
2. Materials and methods 2.1. Study community and recruitment of subject The study took place in an agricultural community in the suburbs of Tokyo, Japan. The total area of the community is approximately 2600 ha. Every summer, OP pesticides are applied to 240 ha of paddy fields using remote controlled helicopters. For the other 220 ha of paddy fields, farmers using a spray pump individually applied OP pesticides. Families with young children aged below 6 years old were recruited for this survey. Briefing sessions were held in health care centers, three nursery schools, and one kindergarten in the community to ask for participation in the study. After the sessions, parents who showed positive response to participating in the study were contacted individually by phone. Details of the sampling method and schedules were explained. Only the families that gave oral consent were involved in this study. The number of families who participated in the study was 31 during the first survey period from 23rd to 26th July, and 24 during the second period from 19th to 21st August 2003. The ages of subject children ranged from 1 to 6 years old in both periods. 2.2. Pesticide application in the study community Pesticide applications to the rice paddy fields were conducted between 23rd and 26th of July and between 19th and 21st of August 2003. The active ingredient in the pesticide solution was trichlorfon [Dimethyl 2,2,2-trichloro1-hydroxyethyl phosphonate (Chemical Abstracts Service (CAS) no. 52-68-6)] in the first application period and fenitrothion [O,O-Dimethyl O-(3-methyl-4-nitrophenyl)
phosphorothioate (CAS no. 122-14-5)] in the second period. They were diluted with water prior to application and applied from approximately 3 m above the leaves of the paddies by remote-controlled helicopters. Weight of applied pesticide was approximately 1.3 kg/ha for trichlorfon and 0.65 kg/ha for fenitrothion. Pesticide applications were performed in the morning (from 5 am to 11 am) when low drift diffusion was expected due to low updraft and wind. The applications were conducted for several consecutive days (4 days in July and 3 days in August). Diagram of pesticide application with wind rose in respective days is presented in Fig. 1. The meteorological data for the survey periods were obtained from those the neighboring monitoring station (located at 5 km north east of the community), since the data were not recorded in the study community due to the malfunction of aerovane. 2.3. Target pesticide Trichlorfon and fenitrothion were targeted in this study. In addition to them, five other OP pesticides were included in the analysis: dichlorvos [2,2-dichlorovinyl dimethyl phosphate (CAS no. 62-73-7)], chlorpyrifos [O,O-Diethyl O-(3,5,6-trichloro-2-pyridinyl)phosphorothioate (CAS no. 2921-88-2)], diazinon [O,O-Diethyl O-(2-isopropyl-4methyl-6-pyrimidinyl) phosphorothioate (CAS no. 333-415)], malathion [S-[1,2-Bis(ethoxycarbonyl)ethyl] O,Odimethyl phosphorodithioate (CAS no. 121-75-5)], and fenthion [O,O-Dimethyl O-(3-methyl-4-methylmercaptophenyl) phosphorothioate (CAS no. 55-38-9)]. These were identified as commonly used pesticides in the study area by using agricultural shipment data for Japan. Dichlorvos is also the primary degradation product of trichlorfon (Samuelsen, 1987; Murphy et al., 1996). It has higher toxicity than the parent does. 2.4. Air sampling strategy Air sampling was conducted for the first 24 h following application when the air pollution would be significant, based on our previous pilot study (Kawahara and Yanagisawa, 2003a,b). Sampling at each subject’s house and childcare facilities was scheduled depending on the day of application in the area. Children on holiday as well as young children who do not attend daycare centers were expected to have spent most of their time at home. For children who attend childcare facilities, their own house and the facilities were expected to be their main environments. For the former case, air sampling was performed at outdoors and indoors of their own houses only. Additional sampling was performed at outdoors and indoors of childcare facilities of the children for the latter case. In the houses, air sample was collected for 24 h at outdoors and indoor where the subject spent most of their time, such as the child’s room or the living room. In the childcare facilities, air was sampled at the classrooms and the playground.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
1125
N
a) 1st survey period (23rd to 26th July, 2003) 25 20 15 10 5 0
Area a-4 (26th Jul.)
CALM: 12.5% Average wind speed: 1.6m/s rd
Area a-2 (24th Jul.)
Area a-1 (23 Jul.)
330° 300° 270°
25 20 15 10 5 0
25 20 15 10 5 0
0°
25 20 15 10 5 0
30° 60° 90° 120°
240° 210°
180°
150°
frequency(%)
Area a-3 (25th Jul.)
CALM: 20.8% Average wind speed: 1.2m/s
CALM: 12.5% Average wind speed: 1.5m/s
25 20 15 10 5 0
1000m
CALM: 20.8% Average wind speed: 0.9m/s
b) 2nd. Survey period (19th to 21st August, 2003) 25 20 15 10 5 0
Area b-3 (20th Aug.)
CALM:41.7% Average wind speed:0.9m/s Area b-2 (21st Aug.) 25 20 15 10 5 0
25 20 15 10 5 0
Area b-1, 19th Aug CALM: 16.7% Average wind speed: 1.1m/s
CALM: 8.6% Average wind speed: 1.6m/s
1000m
:Town border :Area border of pesticide application by day :House (Families participated in both survey periods) :House (Families participated in either 1st or 2nd survey period) :Childcare facilities (participated in both survey period) :Childcare facilities (participated in either 1st or 2nd survey period) Fig. 1. Diagram of pesticide application with wind rose in respective days; a) 23rd to 26th July and b) 19th to 21st August; the date wrote in each area means the date of pesticide application within the area.
2.5. Air sampling procedure A Chromosorb102 sampling tube (8 mm OD 110 mm length, adsorbent 50/100 mg, SKC Inc.) was used for air sampling. Our previous study obtained good collection
efficiencies of the adsorbent for the target OP pesticides (Kawahara and Yanagisawa, 2003b), and earlier works had shown the efficiencies of the adsorbent for pesticides (e.g. dichlorvos or chlorpyrifos) being comparable to PUF (polyurethane foam), Tenax GC, or ORBO42 (Roper and
1126
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Wright, 1984; Thomas and Nishioka, 1985; Leidy and Wright, 1991). Chromosorb102 tube was preceded by a 37 mm diameter quarts fiber filters (QM-A, Whatman Inc.), and air was drawn through them at a flow rate of 2.0 l/min by using portable air pump (Aircheck2000, SKC Inc.). Inlet of the samplers was set at 70 cm to 100 cm above the floor or ground to ensure air collection at the children’s breathing zone. Flow rates were determined before and after the sampling period using a primary flow meter (Dry Cal DCLite, BIOS Inc.). To reduce the noise of the pump, the main body was put in a box filled with noise adsorbing material, or a quiet-type air pump (IAQ-3, SKC. Inc.) was used instead. After sampling, the sampling tubes and filter cassettes were capped, sealed in a zip-lock bag, stored in an icebox at 0 -C, and sent back to the laboratory. The samples were kept at 10 -C in a freezer until the day of analysis (within 14 days after sampling). The field reproducibility of this method for analytes was evaluated through duplicate sampling. 2.6. Sample analysis Filters and adsorbents were extracted twice with 2 ml of acetone in an ultrasonic bath for 40 min (20 min 2). An aliquot (10 Al) of acetic anhydride was added as the derivertizing agent for trichlorfon to prevent its thermal decomposition to dichlorvos in the following gas chromatographic analysis (Bowman and Dame, 1974; Slahck, 1988). To examine the breakthrough of target compounds in the glass tube sampler, adsorbents in the primary and back-up section were extracted and analyzed separately. Breakthrough was defined as the point at which 5% of the total collected on the entire adsorbents was found in the back-up section. For samples where breakthrough was found in back-up section, the data was removed from quantitative evaluation. One analytical blank sample was prepared in each analytical batch. The target compounds were analyzed with a gas chromatograph (HP6890 Hewlett Packard Inc.) equipped with a flame photometric detector (FPD). Three micro liters of the extract was injected using the pulsed splitless mode. A column, 30 m length 0.32 mm ID HP-1 with 0.25 Am film thickness (DB-1MS, J&W Inc.) was used for the gas chromatographic analysis. Super purified helium gas was used as the carrier gas and set at a flow rate of 5.3 ml/min. Injector temperature was set at 200 -C. Initial oven temperature was set at 60 -C and held for 2 min, then ramped to 250 -C at 8 -C/min and then held for 3 min. Detector temperature was set at 240 -C. Calibration curve was prepared by plotting the area of detected peak against concentration of five different standard solutions (pesticide standard mixture no. 15, 16, and 17; Wako Inc.). Linearity was determined by curve fitting. Detail of analytical procedure and result is reported elsewhere (Kawahara and Yanagisawa, 2003b). Table 1 shows the spike recovery of target pesticides from Chromosorb102 tubes that had sampled 40 m 3 of
Table 1 Spike recovery of target analytes from ambient air matrix Compounds
Trichlorfon Dichlorvos Fenitrothion Chlorpyrifos Diazinon Malathion Fenthion
LODa MLOQb (ng/ml) (ng/m3)
Extraction efficiencyc (%)
4 3 3 4 3 4 3
88 T 5.8 92 T 12 90 T 3.5 91 T 3.8 82 T 1.9 91 T 3.0 81 T1.8
10 6 8 6 6 8 7
Gas Particle (Chromosorb102) (glass fiber filter) mean T S.D. (n = 3) mean T S.D. (n = 4) 61 T 5.2 119 T 2.3 100 T 3.2 99 T 2.3 102 T 2.2 87 T 0.9 125 T 1.3
a Instrumental limit of quantitation (LOD) were determined with analytical standards in solvent (no matrix effect). b Method limit of quantitation (MLOQ). Extraction efficiency was based on laboratory spike recovery tests from air matrix. c Method extraction efficiency was based on spike recovery tests from Chromosorb102 tubes containing ambient air matrix.
ambient air, the instrument limits of detection (LOD), and the method limit of quantitation (MLOQ). Recoveries of pesticides from ambient air matrix fell within the range 80 to 100% and from 60 to 125% for filters, with relative standard deviation of the triplicate sample below 12%. Pesticide concentrations in the air samples were adjusted by these extraction efficiencies. Instrumental LOD was defined as the concentration at which the height of the signal peak is three times the baseline noise. The method had MLOQ of 6 to 10 ng/m3 by 24-h air sampling (sample volume: 2.8 m3). Whitmore et al. (1994) had used PUF plug in the non-occupational pesticide exposure study. In their report, the detection limits for target pesticides in air sampled at a flow rate of 3.8 l/min for 24 h were shown to be 1.5 – 79.0 ng/m3 for dichlorvos and 0.5– 4.5 ng/m3 for chlorpyrifos analyzed with GC/ECD, and 11– 48 ng/m3 for diazinon, and 10 –60 ng/m3 for malathion with GC/MS. These detection limits are comparable to those by our method. The detail of analytical procedure and result are reported elsewhere (Kawahara and Yanagisawa, 2003b). In the statistical analysis, sample with no recognizable signal (signal to noise ratio <3:1) was assigned values of zero, and those below the respective MLOQ were assigned one half the value of the MLOQ. 2.7. Time – activity log and questionnaire Parents were asked to record the location and the activity level of their child on the day of air sampling, using the time– location/activity log sheet. The log sheet consists of two categories, one for the location and one for the level of their child’s activity for a period of 24 h. Parents checked the appropriate box to report the location of their child categorized by home indoor, home outdoor, childcare facility, other place indoor, or other place outdoor, every 30 min. Activity levels were recorded by checking box of Fheavy_, Fmoderate_, Flight_, and Fresting_. Time – location
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
data for when the subjects were at childcare facilities was obtained from childcare workers in the facilities. Parents were also asked to answer a self-administrated questionnaire about the physical characteristics of their children, the family members’ activities in the houses on the day of the air sampling, and the history of pesticide usage in or around their houses. 2.8. Estimation of inhalation exposure In this study, fixed measurements in each subject’s microenvironment and time – location/activity questionnaires completed by parents were used for estimating their inhalation exposure to the applied pesticides. The subject’s daily inhalation exposure following pesticide application was estimated using the following equation (US EPA, 1997; WHO, 1999): Daily inhalation exposure ðAg=kg=dayÞ ¼ RðCi IRinh EDi Þ=ðBW ATÞ: Here i is the given location where the child was at a given period of time (indoor/outdoor at home, childcare facility, and other place were given in this study), C i is the OP pesticide concentration in the air at the given locations (ng/ m3). IRinh is the inhalation rate per hour (m3/h). The inhalation rate for each subject was estimated using
1127
reference values for normal Japanese children as a function of age, reported by International Atomic Energy Agency (1998). In the report, activity levels were limited to that of resting, light, and heavy. Therefore, inhalation rate for Fmoderate_ activity obtained from the time –activity log was substituted by Flight_ activity. EDi is the exposure duration (h) to air at each given location. BW is body weight of subject (kg), as reported by the parents. AT is the average time of exposure (24 h). For a few children who lacked time –activity data, mean values of data for the same age group was substituted.
3. Results 3.1. Sampling performance Air was sampled at 31 houses and 2 childcare facilities where the 14 subjects were attending in the first survey period. In the second period, 24 houses and 4 childcare facilities of 16 children were sampled. All sampling started within 5 h after the pesticide application except for that of two houses and one childcare facility, which started within 12 h. This was due to a change of the application schedule. One outdoor air sampling in each season failed due to malfunction or accidental power loss of the sampling pump. Breakthrough was found in 21% of the samples for
Table 2 Average concentrations of the target pesticides in air sample of homes and childcare facilities for 24 h following pesticide application (ng/m3)a Compound
Home (ng/m3)
Childcare facilities (ng/m3)
Outdoor
Indoor
Outdoor
Indoor
Frequency Median Mean Range (%)b
Frequency Median Mean Range Frequency Median Mean Range (%) (%)
Frequency Median Mean Range (%)
First application period (23rd – 26th July) (n = 26) (n = 31) Trichlorfon 24 (92) 43 63 0 – 367c 20 (65) Dichlorvos 25 (96) 13 19 0 – 55c 29 (94) d Fenitrothion 5 (19) 0 3 0 – 33 9 (29) Chlorpyrifos 1 (4) 0 2 0 – 51 7 (23) Diazinon 0 (0) 0 0 0 0 (0) Malathion 1 (4) 0 0 0 – 12 1 (3) Fenthion 0 (0) 0 0 0 1 (3)
9 9 0 0 0 0 0
13 13 2 2 0 0 0
0 – 50 0 – 32c 0 – 13 0 – 12 0 0–5 0–2
(n = 2) 2 (100) 2 (100) 1 (50) 0 (0) 0 (0) 1 (50) 0 (0)
40 23 8 0 8 5 0
40 23 8 0 8 5 0
26 – 53 9 – 38 0 – 16 0 0 0 – 11 0
(n = 2) 2 (100) 2 (100) 1 (50) 0 (0) 0 (0) 0 (0) 0 (0)
28 13 0 0 0 0 0
28 13 0 0 0 0 0
15 – 42 8 – 18 0–5 0 0 0 0
Second application period (19th – 21st August) (n = 24) (n = 24) Trichlorfon 13 (56) 9 12 0 – 61 7 (30) Dichlorvos 23 (100) 7 8 4 – 11c 22 (95) Fenitrothion 23 (100) 51 136 2 – 567c 19 (82) Chlorpyrifos 0 (0) 0 0 0 6 (26) Diazinon 9 (39) 0 1 0–5 6 (25) Malathion 9 (39) 0 1 0 – 11c 7 (30) Fenthion 0 (0) 0 0 0 1 (4)
0 7 18 0 0 0 0
9 7 23 4 1 0 0
0 – 117 0 – 16c 0 – 56c 0 – 51c 0–9 0c 0–2
(n = 4) 3 (75) 4 (100) 4 (100) 0 (0) 0 (0) 0 (0) 0 (0)
9 6 108 0 0 0 0
8 6 122 0 0 0 0
0 – 13 5 – 6c 41 – 232 0 0 0 0
(n = 4) 1 (25) 4 (100) 4 (100) 1 (25) 1 (25) 0 (0) 0 (0)
0 5 11 0 0 0 0
2 5 12 0 0 0 0
0–7 4 – 5c 2 – 25 0–2 0–2 0 0
a Method limits of quantitation (MLOQ) in air (ng/m3) for 24 h air sampling (sample volume: 2.8 m3); trichlorfon, 10; dichlorvos, 6; fenitrothion, 8, chlorpyrifos, 6; malathion, 6; fenthion 8. Data
1128
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
dichlorvos and malathion, 18% for fenitrothion, and 2% for chlorpyrifos. Breakthroughs for malathion and fenitrothion were mainly observed in samples collected after the rain. Duplicate samplers showed the percent deference to be below 20%.
further away (348 vs. 70 ng/m3, Mann – Whitney U – Wilcoxon’s rank sum W, p = 0.017).
3.2. Pesticide use in houses
Applied pesticide was detected from indoor air in each period. Inverse correlations were also observed between indoor air concentrations of applied pesticides and distances from pesticide-applied farm (Spearman’s r = 0.507, p = 0.003, n = 33 for trichlorfon in the first period, and r = 0.570, p = 0.011 n = 19 for fenitrothion in the second). However, the trend was less significant for dichlorvos in the first survey period (Spearman’s r = 0.329, p = 0.081, n = 29).
Table 2 provides the frequency of detection, median, mean, and range of each target compound from air samples by sampled location. In the first survey period, trichlorfon was frequently detected from outdoor air at concentrations from below the detection limit to 367 ng/m3. Dichlorvos was also detected from outdoor in this season, at concentrations ranged from below the detection limit to 55 ng/m3. In the second survey period, outdoor concentration of fenitrothion ranged from 2 to 567 ng/m3. Dichlorvos was also frequently detected outdoors but the concentration was lower than the first period. Besides applied pesticides, diazinon and malathion were also detected outdoors. Outdoor concentrations of applied pesticides showed significant inverse correlation with the distances from pesticide-applied farm to the house or childcare facilities (Spearman’s r = 0.608, p = 0.001, n = 28 for trichlorfon in the first application period, and r = 0.801, p < 0.001. n = 18 for fenitrothion in the second period). However, a similar trend was not seen for dichlorvos in the first period (Spearman’s r = 0.155, p = 0.46, n = 25). Houses and childcare facilities were categorized to < 30 m and > 30 m from the pesticide-applied field. 30 m is the distance that finer particles (< 100 Am) are expected to drift when under the 1 m/s wind (Association of Agriculture, Forestry and Fisheries Aviation, Japan, 1996). Median outdoor concentrations of trichlorfon in the first survey period was 4.7 times higher for houses and facilities within 30 m than those further away (median value: 137 vs. 29 ng/m3, Mann – Whitney U – Wilcoxon’s rank sum W, p = 0.007). In the second survey period, median outdoors concentrations of fenitrothion within 30 m was 5.0 times higher than those
Trichlorfon 50 Indoor concentration. (ng/m3)
3.3. Outdoor concentration and relationship between distance from applied farm
60
Dichlorvos
40
30
20
Trichlorfon: Spearman's r =0.565 (p =0.001)
10
Dichlorvos Spearman's r =0.310 (p =0.116)
0 0
100 200 300 400 Outdoor concentration (ng/m3)
500
100 90
MEP
80 Indoor concentration. (ng/m3)
In the first survey period, 9 families reported the use of pest control measures around their house within a month of air sampling, and one family within a week. Five families had used products including OP pesticides within a month (2 for trichlorfon, 1 for dichlorvos, and 1 for acephate). One family had used products including glyhosate within a week. In the second period, 7 families reported pest control use around their house within a month, of which 3 families reported the use of OP pesticides (1 for fenitrothion and 2 for trichlorfon). Pest control inside of the house was reported from one family (within a month) in the first period but no use of OP pesticide.
3.4. Indoor pesticide and factor that influenced to the concentration
70 60 50 40 30
Fenitrothion: Spearman's r =0.686 (p =0.001)
20 10 0 0
100 200 300 400 Outdoor concentration (ng/m3)
500
Fig. 2. Relationships between indoor and outdoor concentrations for applied pesticides. a) Trichlorfon and dichlorvos in the first survey period, b) fenitrothion in the second survey period.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Significant relation was seen between indoor and outdoor concentrations of applied pesticides in both periods (Spearman’s r = 0.565, p = 0.001, n = 32, for trichlorfon, and r = 0.686, p = 0.001, n = 20, for fenitrothion), indicating evident impact of outside pesticide application on indoor air pollution (Fig. 2a, b). But for dichlorvos in the first period, the relationship was not significant (Spearman’s r = 0.310, p = 0.116, n = 27). Median of the indoor to outdoor air concentration ratio (I/O ratio) was 0.29 for trichlorfon, and 0.74 for dichlorvos in the first period. The I/O ratio was 0.18 for fenitrothion in the second period. Chlorpyrifos was detected in indoor air sample of 13 houses in the two survey periods. None of them, except one house, were detected chlorpyrifos in outdoor air. However, seven had a history of termite control within 5 to 10 years (range of indoor concentration: 1 to 51 ng/ m3). Chlorpyrifos was found only in the outdoor air of one house, with its concentration of 51 ng/m3, had 4 ng/ m3 of chlorpyrifos in indoor air. Since this house also had a history of termite control within 7 years, it was not clear how outdoor pollution by the pesticide influenced indoor air. 3.5. Window opening and indoor air pollution The reported resident’s activity on the day of pesticide application showed that 20 out of 31 families in the first survey period and 16 out of 23 families in the second period opened the windows of their houses during the morning (5 am to 11 pm). I/O ratio was higher for the houses with window open in the morning than those with windows closed (median value: 0.74 vs. 0.16, Mann – Whitney U – Wilcoxon’s rank sum W, p = 0.004). In first survey period, however, I/O ratio was not significantly affected by window opening (median value: 0.29 vs. 0.25, Mann – Whitney U –
1129
Wilcoxon’s rank sum W, p = 0.802 for trichlorfon, and 0.99 vs. 0.56, p = 0.354 for dichlorvos). 3.6. Source of dichlorvos in indoor Using non-parametric analysis, significant difference in the I/O ratio was found between trichlorfon and dichlorvos for the first period (Wilcoxon’s matched pairs signed-rank test: p = 0.002). This suggests that there is a different mechanism of indoor air pollution of these pesticides. As already presented, indoor concentrations of dichlorvos in the first period did not show associations with outdoor concentration, distance from pesticide-applied farm, or window opening by the resident. We found clear correlation between indoor concentrations of dichlorvos and those of trichlorfon (Spearman’s r = 0.544, p = 0.002, n = 29). This finding suggested that the indoor air pollution by dichlorvos was not only caused by the infiltration from outdoors, but also secondary generation from trichlorfon in the indoor environment. 3.7. Time –activity data of subjects Time – location data of subject children was obtained from families and the childcare facilities in both survey periods, although two families failed to fill the time – location log sheet. Table 3 provides the statistical descriptions of the time subject children spent at categorized locations by day category and survey period. Children who were on holiday as well as children who did not attend childcare facilities spent 60 –80% of day at home indoors and 0– 8% of day at home outdoors. Children who attended childcare facilities spent about 60% of their day at home and 30% of their day at childcare facilities. These children had more time to spend outdoors at the facilities than at their own houses in the second survey period (median value: 2.0 and 0.3 h/day,
Table 3 Time spent in the categorized microenvironment during survey perioda Survey period
Location Category
23rd – 26th July
Indoor
Without attendance at childcare facilitiesa n
Outdoor
19th – 21st August Indoor
Outdoor
Own house 14 Childcare facility All other indoor location Total indoors Own house Childcare facility All other outdoor location Total outdoors Own house 6 Childcare facility All other indoor location Total indoors Own house Childcare facility All other outdoor location Total outdoors
With attendance at childcare facilities
Median (h)
Mean (h)
Range (h)
n
Median (h)
Mean (h)
Range (h)
21.0 – 0.3 22.0 0.5 – 0.5 2.0 19.0 – 1.5 21.5 1.0 – 1.5 2.5
20.6 – 1.4 21.9 0.6 – 1.5 2.1 18.3 – 2.9 21.1 0.9 – 1.9 2.9
18.0 – 23.5 – 0.0 – 4.5 17.5 – 24.0 0.0 – 2.0 – 0.0 – 4.5 0.0 – 6.5 10.0 – 22.5 – 0.0 – 11.0 13.5 – 23.0 0.0 – 2.5 – 1.0 – 3.0 1.0 – 5.5
14
14.5 5.6 0.3 21.3 0.0 1.6 0.5 2.8 15.0 5.6 0.0 21.7 0.3 2.0 0.0 2.3
14.3 7.5 0.9 21.1 0.5 1.2 0.8 2.9 15.4 5.7 0.4 21.6 0.6 1.5 0.4 2.4
11.5 – 16.8 2.0 – 7.5 0.0 – 3.0 19.0 – 23.5 0.0 – 2.0 0.0 – 2.5 0.0 – 2.3 0.0 – 5.0 12.5 – 170 5.0 – 8.4 0.0 – 2.0 0.0 – 4.0 0.0 – 3.5 0.0 – 3.5 0.0 – 1.5 1.0 – 4.0
16
a Children without attendance at childcare facilities include children who did not attend childcare facilities and children who were on holiday on the day of monitoring.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Trichlorfon in the 1st survey period Home Indoor Childcare facilities Indoor Home Outdoor Childcare facilities Outdoor
0
10
20
30
25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 1 1
40
Daily inhalation exposure (ng/kg/day)
11 10 9 8 7
Rank
30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 8 7 6 5 4 3 1 1
Rank
Rank
1130
6 5
Dichlorvos in the 1st survey period
Fenitrothion in the 2nd survey period
4
Home Indoor Childcare facilities Indoor Home Outdoor Childcare facilities Outdoor
Home Indoor Childcare facilities Indoor Home Outdoor Childcare facilities Outdoor
3 2 1
0
10
20
30
40
0
Daily inhalation exposure (ng/kg/day)
10
20
30
40
50
Daily inhalation exposure (ng/kg/day)
Fig. 3. Estimated OP pesticide exposure from air (ng/kg/day) for individual subject ranked in ascending order, and the contribution of each microenvironment.
3.8. Daily inhalation exposure for subject children following pesticide application Estimation of daily inhalation exposure to trichlorfon and dichlorvos following the first application was performed for 30 and 25 subjects, respectively, and fenitrothion following the second application for 11 subjects, whose microenvironment monitoring was completed with acceptable sampling performance. Estimated inhalation exposure ranged from 0 to 35 ng/kg/day for trichlorfon, from 0 to 26 ng/kg/day for dichlorvos, and from 5 to 44 ng/kg/day for fenitrothion (Fig. 3 ). These estimated exposure levels were below the level of the acceptable daily intake for food intake set by the Ministry of Health, Labor and Welfare in Japan (trichlorfon, 10 Ag/kg/day; dichlorvos, 3.3 Ag/kg/day; fenitrothion, 5.0 Ag/kg/day). The hazard quotients (HQs), the ratio of the estimated dose to the reference dose, was from 0 to 0.0035 for trichlorfon, from 0 to 0.0078 for dichlorvos, and from 0.0011 to 0.0088 for fenitrothion.
based on the median value, and 86% for dichlorvos in the first application period. Whereas for fenitrothion in the second period, outdoor exposure accounted for a relatively high portion of the daily inhalation exposure at 55%. This higher amount of outdoor exposure is thought to be due to the low infiltration of fenitrothion indoors (Fig. 3). For children who attended childcare facilities, inhalation exposure during their stay at the facilities was comparable with that at home (median value: 78% for trichlorfon, 55% for fenitrothion), although the total time they spent at the facilities was less than that at home. This result demonstrated the great impact of the environment for their exposure. 50
Daily inhalation exposure (ng/kg/day)
respectively, Wilcoxon’s matched pairs signed-rank test: p = 0.04). The trend was less significant in the first period (median value: 1.6 and 0.0 h/day, respectively, Wilcoxon’s matched pairs signed-rank test: p = 0.06). As for the activity log, only a few parents recorded the subject’s activity level other than resting. This is supposedly due to the limitation of the ability to follow-up on the child’s activity. Therefore, in the calculation of their inhalation exposure, inhalation rate of a light activity was used for activity levels other than resting.
Trichlorfon/ 1st. period Dichlorvos/ 1st. period Fenitrothion/ 2nd. period
40
30
20
10
0 0
3.9. Inhalation exposure and activity location Inhalation exposure to trichlorfon from indoor air accounted for 74% of the daily inhalation exposure
100
200
300
400
500
Distance from pesticide applied farm to subject's activity location (m)
Fig. 4. Relationship between estimated daily inhalation exposure and average distance from applied farm to subject’s activity location.
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
The relationship between estimated inhalation exposure and the distance from pesticide-applied farm to their activity location was evaluated. For children with attendance at childcare facilities, time-weighted-average distances from applied farm to their own houses or facilities were substituted, because pesticide exposure at the facilities was considered not to be negligible. The distance was calculated by weighting the distance from farm to home or facilities by the time the subject spent at each location. Results showed that subjects had higher inhalation exposure to trichlorfon and dichlorvos if they stayed closer to the pesticide applied-farm. As for fenitrothion in the second survey period, the trend was not so significant (Fig. 4).
4. Conclusion OP pesticides were frequently detected from the air of residential environments and childcare facilities following pesticide application in an agricultural community. Correlation between indoor and outdoor concentration indicated evident impact of outside pesticide application on indoor air pollution. Houses and childcare facilities located near the applied farms had higher air pollution levels than those farther away. Window opening during pesticide application was found to increases the infiltration of the pesticide indoors. The study indicated that distance from pesticide farm to children’s activity location is an important factor that affects their inhalation exposure. Moreover, depending on nature of the applied pesticide, exposure from indoor air constitutes a significant fraction of daily inhalation exposure to the pesticide, although the application was conducted outside of the buildings. The pesticide pollution level was found to have potentially high impact on the children’s exposure. This suggested the need for reducing the pollution in such an environment where young children spend time frequently. More detailed air monitoring of children’s microenvironment, time – activity data, and identification of other exposure pathways are needed for more accurate estimation and development of effective measures to reduce exposure of children. Acknowledgements We appreciate all the participants in this study and the office staffs of the agriculture public corporation in the studied community for their understanding and cooperation in our study. This study was funded by Sumitomo Foundation 2002. References Association of Agriculture, Forestry and Fisheries Aviation of Japan. Guideline for agriculture and forestry aviation safety measure; 1996.
1131
Baker LW, Fitzell DL, Seiber JN, Parker TR, Shibamoto T, Poore MW, et al. Ambient air concentration of pesticides in California. Environ Sci Technol 1996;30:1365 – 8. Beard J, Wise WV, Sullivan G. Exposure to pesticides in ambient air. Aust J Public Health 1995;19:357 – 62. Bearer CF. How are children different from adults? Environ Health Perspect 1995;103(Suppl. 6):7 – 12. Bowman PB, Dame PW. Gas – liquid chromatography of trichlorfon in soluble powder formulations. J Assoc Off Anal Chem 1974;57: 1128 – 31. Bradman MA, Harnly ME, Draper W, Seidel S, Teran S, Wakeham D, et al. Pesticide exposure to children from California’s central valley: results of a pilot study. J Expo Anal Environ Epidemiol 1997;7:217 – 34. Camann DE, Geno PW, Harding HJ, Giardino NJ, Bond AE, Lewis RG, et al. A pilot study of pesticides in indoor and in relation to agricultural application. Proceedings of the 6th International Conference on Indoor Air Quality and Climate, Indoor Air 1993, Helsinki, Finland, vol. 2; 1993. p. 207 – 12. Cohen Hubal EA, Sheldon LS, Burke JM, McCurdy TR, Berry MR, Rigas ML, et al. Children’s exposure assessment: a review of factors influencing children’s exposure, and the data available to characterize and assess that exposure. Environ Health Perspect 2000;108:475 – 86. Eskenazi B, Bradman A, Castorina R. Exposure of children to organophosphate pesticides and their potential adverse health effects. Environ Health Perspect 1999;107(Suppl. 3):409 – 19. Faustman EM, Silbernagel SM, Fenske RA, Burbacher TM, Ponce RA. Mechanism underlying children’s susceptibility to environmental toxicants. Environ Health Perspect 2000;108(Suppl. 1):13 – 21. Fenske RA, Lu C, Barr D, Needham L. Children’s exposure to chlorpyrifos and parathion in an agricultural community in central Washington State. Environ Health Perspect 2002;110:549 – 53. Foreman WT, Majewski MS, Goolsby DA, Wiebe FW, Coupe RH. Pesticides in the atmosphere of the Mississippi River Valley: Part IIR Air. Sci Total Environ 2000;248:213 – 6. Goldman LR. Children—unique and vulnerableR Environmental risks facing children and recommendations for response. Environ Health Perspect 1995;103(Suppl. 6):13 – 8. International Atomic Energy Agency. Compilation of anatomical, physiological and metabolic characteristics for a reference Asian man, vol. 2. Country reports. H. Physical and anatomical data, and part of physiological and metabolic data for normal Japanese with special reference to establishing reference Asian man model for the anatomical characteristics; 1998. Kawahara J, Yanagisawa Y. Indoor and outdoor organophosphorus pesticides in an agricultural area in Japan. Proceedings ofR 7th International Conference on Healthy Buildings, Singapore, vol. 1; 2003a. p. 452 – 5. Kawahara J, Yanagisawa Y. Simplified analytical method for organophosphorus pesticide in air using adsorbent filled tube and solvent extraction by ultrasonic wave. Bunseki Kagaku 2003b;52:519 – 25. Leidy RB, Wright CG. Trapping efficiency of selected adsorbents for airborne pesticides. Environ Sci Health B 1991;26:367 – 82. Loewenherz C, Fenske RA, Simcox NJ, Bellamy G, Kalman D. Biological monitoring of organophosphorus pesticide exposure among children of agricultural workers in central Washington State. Environ Health Perspect 1997;105:1344 – 653. Lu C, Fenske RA, Simcox NJ, Kalman D. Pesticide exposure of children in an agricultural community: evidence of household proximity to farmland and take home exposure pathways. Environ Res A 2000;84: 290 – 302. Mukerjee WD, Ellenson WD, Lewis RG, Stevens RK, Somerville MC, Shadwick D, et al. An environmental scoping study in the Lower Rio grand Grande Valley of Texas: IIIR Residential microenvironmental monitoring for air, house dust, and soil. Environ Int 1997;23:657 – 73. Murphy KC, Cooper RJ, Clark JM. Volatile and dislodgeable residues following trichlorfon and isazofos application to turfgrass and implications for human exposure. Crop Sci 1996;36:1446 – 54.
1132
J. Kawahara et al. / Environment International 31 (2005) 1123 – 1132
Pope CN, Liu J. Age related differences in sensitivity to organophosphorus pesticides. Environ Toxicol Pharmacol 1994;4:309 – 14. Roper EM, Wright CG. Sampling efficiency of five solid sorbents for trapping airborne pesticides. Bull Environ Contam Toxicol 1984;33: 476 – 783. Samuelsen OB. Aeration rate, pH, and temperature effects on the degradation of trichlorfon to DDVP and the half-lives of trichlorfon and DDVP in seawater. Aquaculture 1987;66:373 – 80. Sheets LP. A consideration of age-dependent differences in susceptibility to organophosphorus and pyrethroid insecticides. Neuro Toxicol 2000;21: 57 – 64. Slahck SC. Direct gas chlomatographic analysis of trichlorfon for its decomposition products. J Assoc Off Anal Chem 1988;71:440 – 2.
Thomas C, Nishioka YA. Sampling of airborne pesticide using Chromosorb102. Bull Environ Contam Toxicol 1985;35:460 – 5. US EPA (United States Environmental Protection Agency). Exposure Factors Handbook. EPA/600/P-95/002/Fa; 1997. Whitmore RW, Immerman FW, Camann DE, Bond AE, Lewis RG, Schaum JL. Non-occupational exposures to pesticide for residents of two US cities. Arch Environ Contam Toxicol 1994;26:47 – 59. WHO (World Health Organization). Principle for the assessment of risks to human health from exposure to chemicals. Environ Health Criteria 1999;210.