- Email: [email protected]
Health risk assessment of emissions of dioxins and furans from a municipal waste incinerator: comparison with other emission sources
Environment International 30 (2004) 481 – 489 www.elsevier.com/locate/envint
Health risk assessment of emissions of dioxins and furans from a municipal waste incinerator: comparison with other emission sources Montse Meneses a, Marta Schuhmacher a,b, Jose´ L. Domingo b,* a
b
Environmental Engineering Laboratory, Department of Chemical Engineering, ‘‘Rovira i Virgili’’ University, 43007 Tarragona, Spain Laboratory of Toxicology and Environmental Health, School of Medicine, ‘‘Rovira i Virgili’’ University, San Lorenzo 21, 43201 Reus, Spain Received 20 June 2003; accepted 26 September 2003
Abstract The aim of this study was to calculate the incremental lifetime-risk to dioxins and furans (PCDD/Fs) for the population living in the surroundings of a municipal solid waste incinerator (MSWI), as well as to establish the potential reduction on human health risks as a consequence of the adaptation to the EU legislation on pollutant emissions from the MSWI stack. Analytical and modelled results were obtained. PCDD/F concentrations in environmental media were determined by means of a simple-compartment-multimedia model (air – soil – vegetation model). Predicted and measured PCDD/F concentrations in soils and vegetation were compared, and the effects of MSWI emissions in the environmental media were determined. Human health risks due to PCDD/F emissions from the MSWI were also estimated based on I-TEQ measured and modelled in various environmental media. Cancer risks due to PCDD/F emissions of the plant were 1.07E 07 and 3.08E 09, before and after installation of the clean air system, respectively. On the other hand, cancer risks due to other PCDD/F emission sources in the area were 5.54E 06 and 1.86E 06. Total PCDD/F cancer risks (including those from diet) for the population living in the vicinity of the MSWI were 1.3E 04 and 4.25E 05, respectively (67.6% of reduction). Hazard ratio for total PCDD/F exposure (including diet) decreased during the last 5 years from 1.16 to 0.38. The above data show that other emission sources of PCDD/Fs also have a notable environmental impact on the area under direct influence of the MSWI. D 2003 Elsevier Ltd. All rights reserved. Keywords: Health risks; Municipal solid waste incinerator; Dioxins and furans; Emissions; Air dispersion; Soils; Herbage
1. Introduction Although incineration is an effective way of treating municipal solid waste, the potential public health effects associated with stack emissions has become a major public concern. Some of the chemicals emitted are constituents of the waste, which travel through the combustion chamber and are not captured by pollution control devices. Chemicals emitted into the atmosphere as air emissions are directly transmitted to humans through inhalation. However, these chemicals can also cross environmental media boundaries becoming distributed in different media: soil, vegetation, water, biota, etc. As a result, human health can be indirectly affected through different pathways such as drinking water or groundwater, skin absorption of the chemicals present in water (i.e., bath, river, lake), eating contaminated foodstuffs, and * Corresponding author. Tel.: +34-977-759380; fax: +34-977-759322. E-mail address: [email protected] (J.L. Domingo). 0160-4120/$ - see front matter D 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.envint.2003.10.001
through ingestion and skin absorption of those chemicals adsorbed to soil. Hence, for an accurate health risk estimate, the pollutant concentrations in each environmental media must be determined (Nessel et al., 1991; Zemba et al., 1996). There is clear evidence that in recent years, atmospheric emission reductions of polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs) have been occurring, with PCDD/F concentrations in ‘‘atmospherically impacted’’ media (e.g., vegetation, cow’s milk), human dietary intake and body burdens showing significant declines. It is interesting to note that data from sediment cores and archived samples suggest that peak inputs of PCDD/Fs to the environment occurred probably in the late 1960s/early 1970s (Alcock and Jones, 1996; Duarte-Davidson et al., 1997; Alcock et al., 1998). However, the large-scale efforts at primary source reduction were not initiated until later in Spain, with efforts to tackle emissions from municipal solid waste incinerators (MSWI) as well as other important primary sources.
482
M. Meneses et al. / Environment International 30 (2004) 481–489
Since 1975, a MSWI has been operating in Montcada (Barcelona, Spain). Until recently, an electrostatic precipitator was used as emission control device. In March 1999, a modernization of the flue gas cleaning system was carried out. An acid gas (HCl/SO2) and metal emission limit reduction equipment was installed and an active-carbon adsorption filter was added to the fabric filter. As a consequence of this, PCDD/F emissions were notably reduced. However, the MSWI is located in an active industrial zone with a heavy traffic. Therefore, other PCDD/F emissions additional to the MSWI were expected in the area under direct influence of the facility. These emissions could mask and/or distort the environmental benefits of the pronounced reduction in the emissions of PCDD/Fs from the MSWI. Taking this into account, the main goals of the present study were the following: – to assess the human exposure to PCDD/Fs due to the plant before and after the installation of the new gas cleaning system – to compare the human exposure to PCDD/Fs due to the plant with the exposure due to others emissions sources in the area – to evaluate the total health risks (carcinogenic and noncarcinogenic) for the population living in the vicinity of the MSWI To achieve these objectives, the concentrations of PCDD/Fs in air, soil and herbage due to emissions from the MSWI were estimated by application of multimedia (air – soil – vegetation) models. These levels were determined before and after installation of the new technical equipment. Subsequently, predicted and measured PCDD/F concentrations were compared and PCDD/F exposure and health risks were assessed.
2. Materials and methods
are some small mountains at the south and southeast of the facility. A motorway with a heavy traffic is near to the incinerator. To establish the environmental PCDD/F levels in the area under potential influence of the MSWI, soil and vegetation monitoring was conducted in 1996 (before the installation of the air-cleaning device). Twenty-four soils and 24 herbage samples (Pipatherum paradoxum L.) were collected in a radius of 3 km at the main wind directions in the area under direct influence of the facility. Samples were taken at 100, 250, 500, 750, 1000, 1500, 2000 and 3000 m from the stack in each of the three main historically wind rose directions in the area: S, NW and NE. Details about sample collection and analytical procedures were previously reported (Schuhmacher et al., 1997, 1998). Only a brief summary of these details is here given. About 50 g (dry matter) of soil sample were used for the analysis. The samples were dried and then ground before to be transferred to a Soxhlet thimble. Every 10 samples, a blank was introduced. Herbage samples were dried and cut to pieces less than 5 mm using an electrical cutter. About 100 g of sample were used for the analysis. The sample material was pretreated by immersion in a 1 mol l 1 hydrochloric acid solution in water to release PCDD/Fs enclosed in fly ash particles deposited on the vegetation. PCDD/F analyses were performed by HRGC/HRMS. In March 2000, 1 year after environmental improvements in the plant were implemented, a new further soil and vegetation monitoring was carried out. Twenty-four soil and 24 vegetation samples were again collected at the same sampling points. The results showed a small reduction in the median PCDD/F concentrations in herbage but not in soil samples (Domingo et al., 2001). However, this reduction was not as great as it could be expected according to the very pronounced decreases in PCDD/F emissions from the stack. It was a clear indication that other emission sources of PCDD/Fs (traffic, industrial activities, local fires, etc.) had also a notable impact on the area under direct influence of the MSWI.
2.1. Description 2.2. Air dispersion modeling The current case study is a MSWI that has been operating since 1975 and processing an average of 49,000 tons of MSW per year. Until March 1999, PCDD/F concentrations emitted by the plant were, on average, 111.39 ng I-TEQ N m 3, indicating that the annual emission rate of total PCDD/F equivalents (I-TEQs) was approximately 32.3 mg. In March 1999, technical improvements were carried out in the facility. Measures were taken to reduce PCDD/F emissions and a new gas equipment (semi-dry scrubber and active carbon addition) was installed. As a consequence, PCDD/F emissions dropped to 0.086 ng I-TEQ N m 3 (Abad et al., 2003). The rate of emission from the new situation was 0.032 mg I-TEQ/year, indicating a 99.9% reduction in PCDD/F emissions. The plant is placed in a residential area with an important industrial activity. There
The fate and transport of PCDD/Fs are often described as a function of the number of chlorine atoms. Thus, lower chlorinated compounds are partitioned in more proportion into the vapor phase than the higher chlorinated compounds. If the air concentrations and deposition fluxes of the pollutants cannot be measured, these can be estimated using an air dispersion model, which simulate the atmospheric dispersion using meteorological and topographic information of the area under evaluation. In the current study, the air dispersion of the emitted 17 toxic PCDD/F congeners was simulated for the surroundings of the MSWI using the commercial software ‘‘BEEST for Windows 95’’. This simulation program provides a graphical interface for the ISCST3 (Industrial Source Complex-Short
M. Meneses et al. / Environment International 30 (2004) 481–489
Term, Version 3) model. This model estimates not only the concentration of the different congeners in the particle phase, but also those in the gas phase. ISCST3 is based on a Gaussian plume model. It can compute ambient air concentrations and surface deposition fluxes at specific receptors near a steady-state emission source. Results of the air dispersion model rely on three basic data sets: meteorological conditions (wind speed and flow vector, ambient air temperature, stability class and rural and urban mixing height), facility characteristics and cartographic (length, latitude and height) dates. The Meteorological Service of Catalonia provided hourly meteorological dates from the study area. These data belonged to a meteorological station located inside the municipality of Montcada. The data showed a high percentage of calm hours. About 27% and 40% of all hourly wind speed values did not exceed 0.1 m/s in 1996 and 2000, respectively. The average wind speed was 2.75 and 3.08 m/s in 1996 and 2000, respectively. The characteristics of the gas source (before and after the installation of the new cleaning gas system) needed to run the model are shown in Table 1. The dispersion of PCDD/F emissions in the atmosphere is primarily governed by their gas-particle partitioning. This process has a decisive influence on transport, deposition and degradation processes of PCDD/Fs. Only PCDD/Fs in the gaseous phase are believed to be depleted due to degradation reactions, while the particle properties determine the transport of particle-bound PCDD/Fs away from sources. Although in different proportion, all 17 toxic congeners of PCDD/Fs are adhered to the particulate emissions. Thus, each congener is distributed between the particulate phase and the gas phase (Kaupp and McLachlan, 1999; Lee and Jones, 1999; Lorber and Pinsky, 2000). Therefore, it was necessary to model the air dispersion of particles and gases. Data about emissions from the stack before and after installation of the new cleaning gas system are shown in Table 2. Pollutants are removed from the atmosphere via deposition processes and deposit on terrestrial and water surfaces. Accordingly, the atmospheric concentration decreases. The ISCST3 model includes algorithms that consider deposition and depletion, which accounts for dry and wet plume depletion depending on the topography of
483
Table 2 Air emission concentrations of PCDD/Fs before (1996) and after (2000) the installation of the air cleaning devicea
2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
1996
2000
1.29 15.41 21.10 48.88 55.28 227.66 251.24 94.10 10.81 81.04 121.76 69.21 10.47 135.83 227.06 68.01 198.91 111.39
0.007 0.026 0.032 0.085 0.074 0.533 0.665 0.029 0.010 0.028 0.052 0.050 0.003 0.102 0.231 0.040 0.345 0.086
a Values are expressed in (ng N m 3) and correspond to one sample in 1996 and the mean of four samples in 2000.
the region. A special terrain grid file with the required topographic information was introduced in the model. Thus, plume depletion could be regarded in the air dispersion modeling. Dry deposition of particles is a result of gravitational settling, turbulent and molecular diffusion. For the calculation of the dry deposition, a distribution of the particle sizes and the related mass density of the particle emissions were required. These values were obtained from the literature (Meneses et al., 2002). The mass density was set to 1 g/cm3 for all sizes. Wet deposition is the removal of particles from the plume through precipitation. To obtain these results, specific scavenging coefficients were needed. The scavenging coefficients of particles depend only on the particle size. Corresponding values were taken from the ISC3 User’s Guide. As these specific values were not available, the PCDD/F vapor form was assumed to have a behavior corresponding to extremely small particles. The ISCST-3 model was run in plume depletion mode, meaning that PCDD/Fs were depleted from the plume moving away from the MSWI by an amount equal to the PCDD/Fs depositing by dry and wet particle deposition. 2.3. Vegetation modeling
Table 1 Characteristics of the municipal solid waste incinerator Height (m) Elevation (m) Diameter (m)
Temperature (jC) Volumetric flow rate (N m3/h) Exit velocity (m/s)
48 50 1.77 Without cleaning device
With cleaning device
236.4 33,332 8.0
175.1 42,951 8.9
There are two possible pathways for the accumulation of PCDD/Fs in vegetation: (a) uptake from soil and (b) direct deposition. Apparently, accumulation from soil is a minor pathway (Wagrowski and Hites, 1998). In addition, the lipid content of the plant, the roughness of the leaves, and the orientation of the vegetation to the atmosphere (i.e., horizontal or vertical) may influence the ability of vegetation to collect and retain PCDD/Fs from both the gas and particle phase (Wagrowski and Hites, 1998).
484
M. Meneses et al. / Environment International 30 (2004) 481–489
There are three general forms of atmospheric deposition: dry gaseous deposition, dry particle-bound deposition and wet deposition of dissolved and/or particle-bound substances as rain, snow or fog. Initially, it was thought that dry particle-bound deposition was the dominant atmospheric pathway to plants (Welsch-Pausch et al., 1995). However, further studies revealed that substantial fractions of airborne PCDD/Fs are present in the gas phase during the growing season (Welsch-Pausch et al., 1995). Taking into account the above considerations, vegetation concentrations can be calculated as the sum of four different pathway contributions (EPA, 1990, 1998). The vegetation concentration Cabv is: Cabv ¼ Cva þ Cddp þ Cwdp þ Cur where Cabv: concentration in above-ground vegetation (ppm), Cva: concentration due to vapor-phase absorption (ppm), Cddp: concentration due to dry particulate deposition onto plant matter (ppm), Cwdp: concentration due to wet particulate deposition onto plant matter (ppm), Cur: concentration due to uptake by root onto plant matter (ppm). Particle-borne contaminants released from the MSWI will deposit on vegetation in the surroundings. The potential of local vegetation to intercept and incorporate settling particles depend on a variety of factors, including rate of contaminant deposition, fraction of deposition intercepted (which depends mainly on the cross-sectional area covered by the vegetation), the rate at which degradation and weathering processes remove contaminants, the length of the growing season and the yield of the particular crop. Particle bound contaminants are assumed to deposit into surface soils (Zemba et al., 1996). The vapor-phase absorption assumes equilibrium partitioning of airborne and sorbed pollutants. 2.4. Soil modeling The accumulation of PCDD/Fs in soil is a result of a particle dry and wet deposition, uptake by roots, background concentration, as well as loss of contaminants through processes such as leaching or volatilization. Distribution and mobility of contaminants are affected by soil conditions such as pH, soil structure and characteristics, and water content (EPA, 1990, 1998; Lorber et al., 1994). As for vegetation, soil concentrations were calculated taking into account the contribution of four pathways: Cs ¼ Cbg þ Csddp þ Cswdp Cur where Cs: concentration in above-ground soil (ppm), Cbg: background soil concentration (ppm), Csddp: concentration due to dry particulate deposition in soils (ppm), Cswdp: concentration due to wet particulate deposition in soils (ppm), Cur: concentration due to the uptake by roots onto plant matter (ppm).
The contaminant mass fraction in surface soils was estimated using a simple accumulation model in which pollutants were assumed to deposit, mix and remain within a fixed soil depth. The processes responsible of distributing contaminants throughout the soil include transport by infiltrating rainwater, mechanical mixing by a variety of fauna (earthworms, ants, termites and burrowing animals) and biological decay and tree uprooting. Contaminants may be removed from surface soils as a result of leaching. In addition, biodegradation, hydrolysis and photolysis serve to destroy some chemicals (Zemba et al., 1996). To take into account these loss mechanisms, a first order removal model was used. Details about the modeling algorithms here used for both vegetation and soil were previously reported (Meneses et al., 2002). The soil and vegetation models presented above were previously assessed. Four different pathways of contribution to the vegetation concentrations were taken into account: vapor-phase absorption, dry particle deposition, wet particle deposition and root uptake. The most important pathway was vapor-phase absorption, while the less was root uptake. In the soil model, four pathways were considered: background soil concentration, dry particle deposition, wet particle deposition and root uptake After background concentration, the most important pathway was wet deposition (Meneses et al., 2002). For I-TEQs, a quite good agreement with the measured levels in soil and vegetation was obtained. Although the models predicted higher concentrations for the higher chlorinated PCDD/Fs, the predictions on total I-TEQ concentrations fit very accurately (Meneses et al., 2002). 2.5. Human health risk assessment Human health risk assessment requires identification of the pathways through which people can be potentially exposed to the chemicals of concern (PCDD/Fs in this case). The quantitative estimation of health risk due to a PCDD/F exposure was considered as a combination of five ways: (1) Intake of contaminated soil. Humans ingest small amounts of soil indirectly (hand-to-mouth transfer), which depend on the potential extent of exposure to soil. (2) Ingestion of vegetables grown in the area under evaluation. Although the population may consume vegetables grown in the vicinity of a MSWI, usually most vegetables in an urban diet does not come from the area of potential influence of the plant. Therefore, only a fraction between 1 and 10% of the total ingestion of vegetables from the home-grown was here considered. (3) Inhalation of re-suspended soil particles. Subjects were exposed to air contaminants, assuming that indoor air exposure was equal to outdoor exposure.
M. Meneses et al. / Environment International 30 (2004) 481–489
(4) Inhalation of air. It depends on the contaminant’s concentration in the atmosphere and the inhalation rate. Both vapour and particle air concentration can be inhaled. (5) Dermal absorption. It was assumed to occur only in case of direct contact of contaminated soil to the skin. It depends on the human activity, seasonal differences and skin surface exposed to soil. Exposure was calculated as the average daily intake of I-TEQ equivalents per unit body weight. Details on the parameters and equations used in human exposure calculations were previously reported (Schuhmacher et al., 2001). Because ingestion of PCDD/Fs through the diet is known to be the major pathway of exposure (Schuhmacher et al., 2001), in the present study PCDD/F intakes from foods (1996 and 2000) were calculated for an adult population (Domingo et al., 1999; Llobet et al., 2003). Food samples were randomly obtained from local markets, supermarkets and grocery stores from various towns in Catalonia. From all groups of analyzed samples, the quantity of each food was estimated according to the dietary habits of the population living in the area under assessment population (Domingo et al., 1999; Llobet et al., 2003). Risks for adverse human health effects were estimated depending on they were carcinogenic or non-carcinogenic. For the non-carcinogenic evaluation, in order to determine if the contaminant can possess a risk to human health, the daily intake was compared with the reference dose for chronic exposure (RfD). Reference dose for chronic exposure (RfD) of PCDD/Fs is 1 – 4 pg/kg/day (van Leeuwen and Younes, 1998). The hazard ratio was calculated by dividing the exposure dose by the RfD. In turn, the carcinogenic risk was calculated by multiplying the estimated dose by the cancer potency factor for PCDD/Fs. The carcinogenic potency factor was considered to be 34,000– 56,000 (mg/ kg/day) 1 (Katsumata and Kastenberg, 1997). Details about the modeling algorithms here used were previously given (Schuhmacher et al., 2001).
3. Results and discussion Since the environmental behavior of individual PCDD/F congeners varies substantially, a serious error might be introduced by the 2,3,7,8-TCDD approach. This is especially important for MSWI emissions, for which PCDD/F distributions are frequently dominated by congeners that have transports properties, which differ widely from those of 2,3,7,8-TCDD. EPA guidance emphasizes the need to consider each PCDD/F congener separately, and data to perform congener-specific fate and transport assessment are rapidly becoming available. However, such analysis requires a large number of parameters. Considerable re-
485
search is still needed to characterize all the media-to-media transfer of interest. Another important issue in PCDD/F fate and transport modeling involves the distinction between vapor and particle-bound phase in air. Recent studies have suggested that vegetation readily assimilates vapor phase PCDD/Fs. Although for some congeners a small vapor fraction may contribute, the majority of the pollutant becomes incorporated in leafy tissues (Zemba et al., 1996). In the present study, air concentrations of the 17 toxic PCDD/Fs congeners were simulated at the 24 monitoring points, before and after the new gas cleaning system was installed in the MSWI. Tables 3 and 4 show the simulated mean air concentrations due to PCDD/F emissions for the 24 sampling points, before and after installation of the new equipment. These tables also show the measured air concentrations in the study area (data provided by the Generalitat of Catalonia) in 1996 and 2000. A notable decrease of the I-TEQ values due to reductions in MSWI emissions from 10.33 to 0.014 fg ITEQ m 3 was observed. A comparison of the total (measured) PCDD/F concentration in the air of the area under evaluation, with the concentration due to MSWI emissions, without and with the new cleaning system gas (Tables 3 and 4), shows that PCDD/F emissions from the MSWI have a low contribution to the total air concentration. PCDD/F emissions from the plant before installation of the new equipment contributed with a 4.25% to the total I-TEQ (air), while after the new cleaning system was installed, only a 0.024% of total I-TEQ (air) would correspond to PCDD/F emissions from the MSWI.
Table 3 Emission concentrations of PCDD/Fs (fg m 3) before installation of the new cleaning gas system (1996) in the MSWI of Montcada Cmeasureda
Cestimated 2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
Cgas
Cparticle
Ctotal
0.04 0.52 0.72 1.67 0.58 2.31 2.06 6.85 0.52 3.96 4.93 2.49 0.10 2.74 3.57 0.53 0.65 4.38
0.08 0.90 1.23 2.85 4.44 18.36 20.71 2.09 0.48 3.61 6.37 3.91 0.85 9.68 17.12 5.63 17.31 5.95
0.12 1.42 1.95 4.51 5.02 20.66 22.77 8.94 1.01 7.57 11.30 6.40 0.95 12.42 20.70 6.16 17.96 10.33
40.8 24.9 40.8 89.6 65.63 814.34 300.0 330.34 59.70 127.88 244.0 103.84 8.11 180.03 481.79 74.5 415.27 243.01
a Values correspond to one sample. Data provided by the Regional Government of Catalonia.
486
M. Meneses et al. / Environment International 30 (2004) 481–489
Table 4 Emission concentrations of PCDD/Fs (fg m 3) after installation of the new cleaning gas system (2000) in the MSWI of Montcada Cmeasureda
Cestimated 2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
Cgas
Cparticle
Ctotal
3.81E 04 1.48E 03 1.88E 03 4.89E 03 1.31E 03 9.15E 03 9.23E 03 3.58E 03 7.91E 04 2.35E 03 3.55E 03 3.04E 03 4.94E 05 3.47E 03 6.16E 03 5.27E 04 1.92E 03 0.005
7.49E 04 2.74E 03 3.41E 03 8.92E 03 1.07E 02 7.76E 02 9.90E 02 1.16E 03 7.79E 04 2.28E 03 4.91E 03 5.09E 03 4.28E 04 1.31E 02 3.15E 02 5.95E 03 5.43E 02 0.009
1.13E 03 4.22E 03 5.29E 03 1.38E 02 1.20E 02 8.68E 02 1.08E 01 4.75E 03 1.57E 03 4.63E 03 8.46E 03 8.12E 03 4.78E 04 1.65E 02 3.77E 02 6.48E 03 5.62E 02 0.014
4.1 7.2 7.2 13.6 27.9 70.7 185.0 91.2 10.2 35.8 78.7 29.9 1.8 44.8 131.0 10.5 101.0 58.0
a Values correspond to the mean of two samples. Data provided by the Regional Government of Catalonia.
Figs. 1 and 2 show the I-TEQ congener profiles for air PCDD/F concentrations due to emissions from the MSWI, as well as the measured concentrations before and after installation of the new technical equipment. It can be noted that the congeners showing the greatest contribution to ITEQ were 2,3,4,7,8-PeCDF, TCDF, TCDD and 2,3,4,6,7,8-HxCDF. A source of certain uncertainty of the results can be due to the dechlorination process, which was not here considered (Lorber et al., 2000). This process may occur between
the emission point and the ambient measuring station. Generally, polychlorinated organic compounds easily experience photochemical loss of chlorine atoms. If the higher chlorinated PCDD/Fs are dechlorinated to form lower chlorinated PCDD/Fs in the atmosphere, more lower chlorinated PCDD/Fs would have arrived at the ambient air-monitoring point causing distinct air profile. An evaluation of the ISCST3 model showed the great importance of using onsite data sets in the modeling of air dispersion (Eschenroeder and Lorber, 1999). However, no onsite meteorological data of the MSWI could be obtained. Therefore, information from the meteorological station of Montcada was used. The assumption of equal meteorological conditions in Montcada and the plant leads to an uncertainty in the computed air concentrations. Table 5 summarizes the levels of the 17 toxic congeners of PCDD/Fs in herbage, which were calculated based on the simulated air concentration due to MSWI emissions and measured air concentrations. The calculated herbage concentrations correspond to PCDD/Fs accumulated during a period of 6 months (the lifetime of herbage). It can be seen that emissions from the MSWI contributed with 0.89% and 0.20% to the total vegetation concentrations before and after the installation of the new air cleaning device, respectively. It can be concluded that emissions from the MSWI mean a minimal contribution to the herbage concentrations, less than 1%. The soil concentration was calculated considering a continuous deposition during the period of time in which the MSWI has been operating. Taking into account that the soil is a media where the pollutants accumulate during long periods of time, soil concentrations before installation of the new equipment were calculated estimating a period of accumulation of 22 years. In turn, soil concentrations after
Fig. 1. Profile of air PCDD/F concentrations without the new cleaning gas system in the MSWI.
M. Meneses et al. / Environment International 30 (2004) 481–489
487
Fig. 2. Profile of air PCDD/F concentrations with the new cleaning gas system in the MSWI.
installation of the new equipment were calculated considering the 25 years that the plant was operating without the new cleaning gas system, plus the accumulation during the last year in which the MSWI was operating with the new equipment. The levels of the 17 toxic congeners of PCDD/ Fs in soils are shown in Table 6. Emissions from the MSWI contributed with a 1.59% and 2.62%, respectively, to the ITEQ soil concentration before and after the new cleaning gas system was installed. It can be concluded that emissions
Table 5 PCDD/Fs levels (ng/kg dry weight) in herbage samples before (1996) and after (2000) installation of the air gas cleaning device in the MSWI of Montcada 1996 Csimulated 2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ a
2.1E 04 2.2E 03 7.4E 03 1.7E 02 9.3E 03 5.0E 02 3.9E 01 1.4E 02 1.7E 03 8.2E 03 2.1E 02 1.1E 02 1.1E 03 1.8E 02 4.4E 02 8.4E 03 2.3E 02 1.7E 02
2000 Cmeasureda 1.0E 01 4.4E 01 3.0E 01 2.4E + 00 1.1E + 00 5.2E + 00 1.3E + 01 1.3E + 00 9.8E 01 9.3E 01 1.0E + 00 9.6E 01 3.0E 02 1.3E + 00 4.0E + 00 3.5E 01 1.3E + 00 1.9E + 00
Csimulated 1.1E 04 5.1E 04 1.5E 04 3.1E 04 4.3E 04 3.5E 03 7.6E 03 9.8E 04 7.2E 04 9.7E 04 9.0E 04 6.3E 04 4.7E 05 1.1E 03 2.4E 03 1.8E 04 7.4E 04 1.4E 03
from the MSWI mean also a minimal contribution to total PCDD/F concentrations in soil. Table 7 shows PCDD/F exposure for the population living in the surroundings of the MSWI before (1996) and after (2000) pronounced reductions in the emissions of PCDD/Fs from the facility. Dietary exposure for the two years of study is included. Human carcinogenic and noncarcinogenic risks (hazard ratio), based on the total dose, are also given. It can be observed that PCDD/F exposure due to emissions from the MSWI decreased from 2.37E 06 to Table 6 PCDD/Fs levels (ng/kg dry weight) in soil samples before (1996) and after (2000) installation of the air gas cleaning device in the MSWI of Montcada 1996
Cmeasureda 5.5E 02 2.0E 01 1.0E 01 3.5E 01 2.5E 01 3.1E + 00 7.5E + 00 7.0E 01 4.5E 01 5.0E 01 4.0E 01 4.0E 01 2.5E 02 4.0E 01 1.6E + 00 1.5E 01 1.3E + 00 7.0E 01
Values correspond to the mean of PCDD/F concentrations in 24 herbage samples collected in the vicinity of the MSWI in 1996 and again in 2000.
2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
2000 a
Csimulated
Cmeasured
4.70E 04 4.97E 03 1.11E 02 2.58E 02 3.95E 02 3.19E 01 5.12E 01 2.20E 02 4.87E 03 3.63E 02 6.13E 02 3.75E 02 7.96E 03 9.18E 02 1.16E 01 3.79E 02 6.59E 02 5.60E 02
1.45E 01 5.90E 01 7.20E 01 1.15E + 00 1.15E + 00 3.00E + 01 1.60E + 02 1.50E + 00 1.20E + 00 1.55E + 00 2.20E + 00 1.60E + 00 1.75E 01 3.25E + 00 1.30E + 01 1.10E + 00 1.45E + 01 3.52E + 00
Csimulated
Cmeasureda
4.72E 04 4.98E 03 1.11E 02 2.58E 02 3.95E 02 3.19E 01 5.13E 01 2.20E 02 4.87E 03 3.63E 02 6.13E 02 3.75E 02 7.98E 03 9.18E 02 1.16E 01 3.79E 02 6.59E 02 5.64E 02
9.00E 02 5.00E 01 5.50E 01 1.00E + 00 9.00E 01 1.35E + 01 6.40E + 01 1.10E + 00 9.00E 01 1.15E + 00 1.45E + 00 1.35E + 00 1.00E 01 1.85E + 00 9.25E + 00 8.00E 01 8.75E + 00 2.15E + 00
a Values correspond to the mean of PCDD/F concentrations in 24 soil samples collected in the vicinity of the MSWI in 1996 and again in 2000.
488
M. Meneses et al. / Environment International 30 (2004) 481–489
Table 7 PCDD/F exposure (ng I-TEQ/kg/day) and health risks for the population living in the vicinity of the plant before (1996) and after (2000) the installation of the air-cleaning device 1996
2000
MSWI emissions Environmental exposure due to MSWI Hazard ratio Cancer risk
2.37E 06 9.48E 04 1.07E 07
6.85E 08 2.74E 05 3.08E 09
Other emission sourcesb Environmental exposure Hazard ratio Cancer risk
1.23E 04 4.92E 02 5.54E 06
4.14E 05 1.66E 02 1.86E 06
Diet Dietary exposure Hazard ratioc Cancer riskd Total exposure Hazard ratio Cancer risk
2.77E 03 1.11E + 00 1.25E 04 2.90E 03 1.16E + 00 1.31E 04
9.03E 04 3.61E 01 4.06E 05 9.45E 04 3.78E 01 4.25E 05
a
a
Values obtained from simulation data. Values obtained from measured data minus simulation data. c RfD = 2.5 pg/kg/day. d Carcinogenic Potency Factor: 25,000. b
6.85E 08 ng/kg/day (97.1% of reduction) throughout the period 1996 – 2000. These values mean 1.93% and 0.17%, respectively, of the contribution to total PCDD/F exposure (excluding diet). Hazard ratio for total PCDD/F exposure (including diet) was reduced during the last 5 years from 1.16 to 0.38 when a RfD of 2.5 pg/kg/day was considered. Cancer risks due to PCDD/F emissions from the plant was 1.07E 07 and 3.08E 09 before and after installation of the clean air system, respectively. On the other hand, cancer risks due to other emission sources in the area were 5.54E 06 and 1.86E 06, respectively. It means that cancer risk contribution of the MSWI with respect to other PCDD/F emissions in the area decreased from 1.93% to 0.17% after the installation of the clean system device. The total cancer risks due to PCDD/F exposure (including diet) for the population living in the vicinity of the MSWI were 1.3E 04 and 4.25E 05 for 1996 and 2000, respectively. Contribution of PCDD/F emissions from the MSWI to total health risks of PCDD/F (including diet) for the population living in the vicinity of the facility were 0.08% and 0.007% before and after the installation of the new cleaning gas system, respectively. Although the reduction of PCDD/F emissions from the MSWI after the new equipment was installed has been very notable (99.9%), the important exposure reduction during the period 1996 – 2000 (67.6%) has been especially due to the decrease in PCDD/F intake throughout the diet (Llobet et al., 2003). Recent emphasis in the risk analysis field has been focused on uncertainty in the risk assessment process.
Some weakness owing the uncertainties in the data and model parameters must be taken into account. The small number of air stack emissions before installation of the new equipment, as well as the lack of consideration of all possible plume depletion mechanisms in the dispersion and deposition modeling (atmospheric degradation), can be considered as a weakness of the model. Congener-specific atmospheric degradation rates, photolytic dechlorination and/or congener-specific soil dissipation rates have not been considered. The number of PCDD/F concentrations measured in air was only one and two for 1996 and 2000, respectively (mean values are here presented). It made the comparison of predicted and measured air concentrations a tenuous exercise. A similar problem was also found with the emissions from the stack: one and four samples, respectively, before and after installation of the new technical equipment. In summary, the present results show that although in general terms, PCDD/F health risks decreased during the examined period (before and after the installation of the new cleaning gas system), this reduction has not been as great as it could be expected according to the very pronounced decreases in PCDD/Fs emissions from the stack. It should be taken into account that these emissions diminished, on average, from 111.39 ng I-TEQN m3 (before) to 0.036 ng I-TEQ/N m3 (after). It clearly indicates that other sources of PCDD/Fs (traffic, industrial activities, local fires, etc.) also have a notable impact on the area under direct influence of the MSWI. Similar conclusions were also reached following a comparison of the congener profile of PCDD/Fs in soil and herbage samples collected near the MSWI with those from samples collected in a close area, which is outside the direct influence of the plant. According to the above, it can be concluded that the health risks due to the current PCDD/F emissions from the MSWI would be of a relatively small significance for the population living in the neighborhood of the facility. References Abad E, Caixach J, Rivera J. Improvements in dioxin abatement strategies at a municipal waste management plant in Barcelona. Chemosphere 2003;50:1175 – 82. Alcock RE, Jones KC. Dioxins in the environment: a review of trend data. Environ Sci Technol 1996;33:3133 – 43. Alcock RE, Gemmill R, Jones KC. Improvements to the UK PCDD/F and PCB atmospheric emission inventory following an emissions measurements programme. Chemosphere 1998;38:759 – 70. Domingo JL, Schuhmacher M, Granero S, Llobet JM. PCDDs and PCDFs in food samples from Catalonia, Spain. An assessment of dietary intake. Chemosphere 1999;38:3517 – 28. Domingo JL, Schuhmacher M, Granero S, de Kok HAM. Temporal variation of PCDD/F levels in environmental samples collected near an old municipal waste incinerator. Environ Monit Assess 2001;69: 179 – 93. Duarte-Davidson R, Alcock RE, Jones KC. Exploring the balance between sources, deposition and the environmental burden of PCDD/Fs in the
M. Meneses et al. / Environment International 30 (2004) 481–489 UK: an aid to identifying uncertainties and research needs. Environ Sci Technol 1997;31:1 – 11. EPA (US Environmental Protection Agency). Methodology for assessing health risks associated with indirect exposure to combustor emissions. Office of Health and Environmental Assessment. Cincinnati, OH, EPA/ 600/6-90/003; 1990. EPA (US Environmental Protection Agency). Methodology for assessing health risks associated with multiple pathways of exposure to combustor emissions. Office of Health and Environmental Assessment. Cincinnati, OH, EPA/600/R-98-137; 1998. Eschenroeder A, Lorber M. An evaluation of EPA’s ISCST-Version 3 Model: Part. I. Air dispersion of dioxins. Organohalog Compd 1999;41:547 – 52. Katsumata PT, Kastenberg WE. On the impact of future land use assumptions on risk analysis for Superfund sites. Air Waste Manag Assoc 1997;47:881 – 9. Kaupp H, McLachlan MS. Atmospheric particle size distributions of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polycyclic aromatic hydrocarbons (PAHs) and their implications for wet and dry deposition. Atmos Environ 1999;33:83 – 95. Lee RGM, Jones KC. Gas-particle partitioning of atmospheric PCDD/Fs: measurements and observations on modeling. Environ Sci Technol 1999;33:3596 – 604. Llobet JM, Domingo JL, Bocio A, Casas C, Teixido´ A, Mu¨ller L. Human exposure to dioxins through the diet in Catalonia, Spain: carcinogenic and non-carcinogenic risk. Chemosphere 2003;50: 1193 – 200. Lorber M, Pinsky P. An evaluation of three empirical air-to-leaf models for polychlorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 2000;41:931 – 41. Lorber M, Cleverly D, Schaum J, Philips L, Schweer G, Leighton T. Development and validation of an air-to-beef food chain model for dioxin-like compounds. Sci Total Environ 1994;156:39 – 65.
489
Lorber M, Eschenroeder A, Robinson R. Testing the USA EPA’s ISCSTVersion 3 model on dioxins: a comparison of predicted and observed air and soil concentrations. Atmos Environ 2000;34:3995 – 4010. Meneses M, Schuhmacher M, Domingo JL. A design of two simple models to predict PCDD/F concentrations in vegetation and soils. Chemosphere 2002;46:1393 – 402. Nessel SC, Butler JP, Post GB, Held JL, Gochfeld M, Gallo MA. Evaluation of the relative contribution of exposure routes in a health risk assessment of dioxin emissions from a municipal waste incinerator. J Expo Anal Environ Epidemiol 1991;1:283 – 307. Schuhmacher M, Xifro´ A, Llobet JM, de Kok HAM, Domingo JL. PCDD/Fs in soil samples collected in the vicinity of a municipal solid waste incinerator: human health risk. Arch Environ Contam Toxicol 1997;33:239 – 46. Schuhmacher M, Domingo JL, Xifro´ A, Granero S, Llobet JM, de Kok HAM. Presence of dioxins and furans in vegetation samples collected in the neighbourhood of a municipal solid waste incinerator. J Environ Sci Health 1998;A33:195 – 212. Schuhmacher M, Meneses M, Xifro´ A, Domingo JL. The use of MonteCarlo simulation techniques for risk assessment: study of a municipal waste incinerator. Chemosphere 2001;43:787 – 99. van Leeuwen FXR, Younes M. WHO revises the tolerable daily intake (TDI) for dioxins. Organohalog Compd 1998;38:295 – 8. Wagrowski DM, Hites RA. Partitioning of polychlorinated dibenzo-p-dioxins and dibenzofurans between the atmosphere and corn. Environ Sci Technol 1998;32:2389 – 93. Welsch-Pausch K, McLachlan MS, Umlauf G. Determination of the principal pathways of polychlorinated dibenzo-p-dioxins and dibenzofurans to Lolium multiflorum (Welsh ray grass). Environ Sci Technol 1995; 29:1090 – 8. Zemba SG, Green LC, Crouch EAC, Lester RR. Quantitative risk assessment of stack emissions from municipal waste combustors. J Hazard Mat 1996;47:229 – 75.
Health risk assessment of emissions of dioxins and furans from a municipal waste incinerator: comparison with other emission sources Montse Meneses a, Marta Schuhmacher a,b, Jose´ L. Domingo b,* a
b
Environmental Engineering Laboratory, Department of Chemical Engineering, ‘‘Rovira i Virgili’’ University, 43007 Tarragona, Spain Laboratory of Toxicology and Environmental Health, School of Medicine, ‘‘Rovira i Virgili’’ University, San Lorenzo 21, 43201 Reus, Spain Received 20 June 2003; accepted 26 September 2003
Abstract The aim of this study was to calculate the incremental lifetime-risk to dioxins and furans (PCDD/Fs) for the population living in the surroundings of a municipal solid waste incinerator (MSWI), as well as to establish the potential reduction on human health risks as a consequence of the adaptation to the EU legislation on pollutant emissions from the MSWI stack. Analytical and modelled results were obtained. PCDD/F concentrations in environmental media were determined by means of a simple-compartment-multimedia model (air – soil – vegetation model). Predicted and measured PCDD/F concentrations in soils and vegetation were compared, and the effects of MSWI emissions in the environmental media were determined. Human health risks due to PCDD/F emissions from the MSWI were also estimated based on I-TEQ measured and modelled in various environmental media. Cancer risks due to PCDD/F emissions of the plant were 1.07E 07 and 3.08E 09, before and after installation of the clean air system, respectively. On the other hand, cancer risks due to other PCDD/F emission sources in the area were 5.54E 06 and 1.86E 06. Total PCDD/F cancer risks (including those from diet) for the population living in the vicinity of the MSWI were 1.3E 04 and 4.25E 05, respectively (67.6% of reduction). Hazard ratio for total PCDD/F exposure (including diet) decreased during the last 5 years from 1.16 to 0.38. The above data show that other emission sources of PCDD/Fs also have a notable environmental impact on the area under direct influence of the MSWI. D 2003 Elsevier Ltd. All rights reserved. Keywords: Health risks; Municipal solid waste incinerator; Dioxins and furans; Emissions; Air dispersion; Soils; Herbage
1. Introduction Although incineration is an effective way of treating municipal solid waste, the potential public health effects associated with stack emissions has become a major public concern. Some of the chemicals emitted are constituents of the waste, which travel through the combustion chamber and are not captured by pollution control devices. Chemicals emitted into the atmosphere as air emissions are directly transmitted to humans through inhalation. However, these chemicals can also cross environmental media boundaries becoming distributed in different media: soil, vegetation, water, biota, etc. As a result, human health can be indirectly affected through different pathways such as drinking water or groundwater, skin absorption of the chemicals present in water (i.e., bath, river, lake), eating contaminated foodstuffs, and * Corresponding author. Tel.: +34-977-759380; fax: +34-977-759322. E-mail address: [email protected] (J.L. Domingo). 0160-4120/$ - see front matter D 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.envint.2003.10.001
through ingestion and skin absorption of those chemicals adsorbed to soil. Hence, for an accurate health risk estimate, the pollutant concentrations in each environmental media must be determined (Nessel et al., 1991; Zemba et al., 1996). There is clear evidence that in recent years, atmospheric emission reductions of polychlorinated dibenzo-p-dioxins and furans (PCDD/Fs) have been occurring, with PCDD/F concentrations in ‘‘atmospherically impacted’’ media (e.g., vegetation, cow’s milk), human dietary intake and body burdens showing significant declines. It is interesting to note that data from sediment cores and archived samples suggest that peak inputs of PCDD/Fs to the environment occurred probably in the late 1960s/early 1970s (Alcock and Jones, 1996; Duarte-Davidson et al., 1997; Alcock et al., 1998). However, the large-scale efforts at primary source reduction were not initiated until later in Spain, with efforts to tackle emissions from municipal solid waste incinerators (MSWI) as well as other important primary sources.
482
M. Meneses et al. / Environment International 30 (2004) 481–489
Since 1975, a MSWI has been operating in Montcada (Barcelona, Spain). Until recently, an electrostatic precipitator was used as emission control device. In March 1999, a modernization of the flue gas cleaning system was carried out. An acid gas (HCl/SO2) and metal emission limit reduction equipment was installed and an active-carbon adsorption filter was added to the fabric filter. As a consequence of this, PCDD/F emissions were notably reduced. However, the MSWI is located in an active industrial zone with a heavy traffic. Therefore, other PCDD/F emissions additional to the MSWI were expected in the area under direct influence of the facility. These emissions could mask and/or distort the environmental benefits of the pronounced reduction in the emissions of PCDD/Fs from the MSWI. Taking this into account, the main goals of the present study were the following: – to assess the human exposure to PCDD/Fs due to the plant before and after the installation of the new gas cleaning system – to compare the human exposure to PCDD/Fs due to the plant with the exposure due to others emissions sources in the area – to evaluate the total health risks (carcinogenic and noncarcinogenic) for the population living in the vicinity of the MSWI To achieve these objectives, the concentrations of PCDD/Fs in air, soil and herbage due to emissions from the MSWI were estimated by application of multimedia (air – soil – vegetation) models. These levels were determined before and after installation of the new technical equipment. Subsequently, predicted and measured PCDD/F concentrations were compared and PCDD/F exposure and health risks were assessed.
2. Materials and methods
are some small mountains at the south and southeast of the facility. A motorway with a heavy traffic is near to the incinerator. To establish the environmental PCDD/F levels in the area under potential influence of the MSWI, soil and vegetation monitoring was conducted in 1996 (before the installation of the air-cleaning device). Twenty-four soils and 24 herbage samples (Pipatherum paradoxum L.) were collected in a radius of 3 km at the main wind directions in the area under direct influence of the facility. Samples were taken at 100, 250, 500, 750, 1000, 1500, 2000 and 3000 m from the stack in each of the three main historically wind rose directions in the area: S, NW and NE. Details about sample collection and analytical procedures were previously reported (Schuhmacher et al., 1997, 1998). Only a brief summary of these details is here given. About 50 g (dry matter) of soil sample were used for the analysis. The samples were dried and then ground before to be transferred to a Soxhlet thimble. Every 10 samples, a blank was introduced. Herbage samples were dried and cut to pieces less than 5 mm using an electrical cutter. About 100 g of sample were used for the analysis. The sample material was pretreated by immersion in a 1 mol l 1 hydrochloric acid solution in water to release PCDD/Fs enclosed in fly ash particles deposited on the vegetation. PCDD/F analyses were performed by HRGC/HRMS. In March 2000, 1 year after environmental improvements in the plant were implemented, a new further soil and vegetation monitoring was carried out. Twenty-four soil and 24 vegetation samples were again collected at the same sampling points. The results showed a small reduction in the median PCDD/F concentrations in herbage but not in soil samples (Domingo et al., 2001). However, this reduction was not as great as it could be expected according to the very pronounced decreases in PCDD/F emissions from the stack. It was a clear indication that other emission sources of PCDD/Fs (traffic, industrial activities, local fires, etc.) had also a notable impact on the area under direct influence of the MSWI.
2.1. Description 2.2. Air dispersion modeling The current case study is a MSWI that has been operating since 1975 and processing an average of 49,000 tons of MSW per year. Until March 1999, PCDD/F concentrations emitted by the plant were, on average, 111.39 ng I-TEQ N m 3, indicating that the annual emission rate of total PCDD/F equivalents (I-TEQs) was approximately 32.3 mg. In March 1999, technical improvements were carried out in the facility. Measures were taken to reduce PCDD/F emissions and a new gas equipment (semi-dry scrubber and active carbon addition) was installed. As a consequence, PCDD/F emissions dropped to 0.086 ng I-TEQ N m 3 (Abad et al., 2003). The rate of emission from the new situation was 0.032 mg I-TEQ/year, indicating a 99.9% reduction in PCDD/F emissions. The plant is placed in a residential area with an important industrial activity. There
The fate and transport of PCDD/Fs are often described as a function of the number of chlorine atoms. Thus, lower chlorinated compounds are partitioned in more proportion into the vapor phase than the higher chlorinated compounds. If the air concentrations and deposition fluxes of the pollutants cannot be measured, these can be estimated using an air dispersion model, which simulate the atmospheric dispersion using meteorological and topographic information of the area under evaluation. In the current study, the air dispersion of the emitted 17 toxic PCDD/F congeners was simulated for the surroundings of the MSWI using the commercial software ‘‘BEEST for Windows 95’’. This simulation program provides a graphical interface for the ISCST3 (Industrial Source Complex-Short
M. Meneses et al. / Environment International 30 (2004) 481–489
Term, Version 3) model. This model estimates not only the concentration of the different congeners in the particle phase, but also those in the gas phase. ISCST3 is based on a Gaussian plume model. It can compute ambient air concentrations and surface deposition fluxes at specific receptors near a steady-state emission source. Results of the air dispersion model rely on three basic data sets: meteorological conditions (wind speed and flow vector, ambient air temperature, stability class and rural and urban mixing height), facility characteristics and cartographic (length, latitude and height) dates. The Meteorological Service of Catalonia provided hourly meteorological dates from the study area. These data belonged to a meteorological station located inside the municipality of Montcada. The data showed a high percentage of calm hours. About 27% and 40% of all hourly wind speed values did not exceed 0.1 m/s in 1996 and 2000, respectively. The average wind speed was 2.75 and 3.08 m/s in 1996 and 2000, respectively. The characteristics of the gas source (before and after the installation of the new cleaning gas system) needed to run the model are shown in Table 1. The dispersion of PCDD/F emissions in the atmosphere is primarily governed by their gas-particle partitioning. This process has a decisive influence on transport, deposition and degradation processes of PCDD/Fs. Only PCDD/Fs in the gaseous phase are believed to be depleted due to degradation reactions, while the particle properties determine the transport of particle-bound PCDD/Fs away from sources. Although in different proportion, all 17 toxic congeners of PCDD/Fs are adhered to the particulate emissions. Thus, each congener is distributed between the particulate phase and the gas phase (Kaupp and McLachlan, 1999; Lee and Jones, 1999; Lorber and Pinsky, 2000). Therefore, it was necessary to model the air dispersion of particles and gases. Data about emissions from the stack before and after installation of the new cleaning gas system are shown in Table 2. Pollutants are removed from the atmosphere via deposition processes and deposit on terrestrial and water surfaces. Accordingly, the atmospheric concentration decreases. The ISCST3 model includes algorithms that consider deposition and depletion, which accounts for dry and wet plume depletion depending on the topography of
483
Table 2 Air emission concentrations of PCDD/Fs before (1996) and after (2000) the installation of the air cleaning devicea
2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
1996
2000
1.29 15.41 21.10 48.88 55.28 227.66 251.24 94.10 10.81 81.04 121.76 69.21 10.47 135.83 227.06 68.01 198.91 111.39
0.007 0.026 0.032 0.085 0.074 0.533 0.665 0.029 0.010 0.028 0.052 0.050 0.003 0.102 0.231 0.040 0.345 0.086
a Values are expressed in (ng N m 3) and correspond to one sample in 1996 and the mean of four samples in 2000.
the region. A special terrain grid file with the required topographic information was introduced in the model. Thus, plume depletion could be regarded in the air dispersion modeling. Dry deposition of particles is a result of gravitational settling, turbulent and molecular diffusion. For the calculation of the dry deposition, a distribution of the particle sizes and the related mass density of the particle emissions were required. These values were obtained from the literature (Meneses et al., 2002). The mass density was set to 1 g/cm3 for all sizes. Wet deposition is the removal of particles from the plume through precipitation. To obtain these results, specific scavenging coefficients were needed. The scavenging coefficients of particles depend only on the particle size. Corresponding values were taken from the ISC3 User’s Guide. As these specific values were not available, the PCDD/F vapor form was assumed to have a behavior corresponding to extremely small particles. The ISCST-3 model was run in plume depletion mode, meaning that PCDD/Fs were depleted from the plume moving away from the MSWI by an amount equal to the PCDD/Fs depositing by dry and wet particle deposition. 2.3. Vegetation modeling
Table 1 Characteristics of the municipal solid waste incinerator Height (m) Elevation (m) Diameter (m)
Temperature (jC) Volumetric flow rate (N m3/h) Exit velocity (m/s)
48 50 1.77 Without cleaning device
With cleaning device
236.4 33,332 8.0
175.1 42,951 8.9
There are two possible pathways for the accumulation of PCDD/Fs in vegetation: (a) uptake from soil and (b) direct deposition. Apparently, accumulation from soil is a minor pathway (Wagrowski and Hites, 1998). In addition, the lipid content of the plant, the roughness of the leaves, and the orientation of the vegetation to the atmosphere (i.e., horizontal or vertical) may influence the ability of vegetation to collect and retain PCDD/Fs from both the gas and particle phase (Wagrowski and Hites, 1998).
484
M. Meneses et al. / Environment International 30 (2004) 481–489
There are three general forms of atmospheric deposition: dry gaseous deposition, dry particle-bound deposition and wet deposition of dissolved and/or particle-bound substances as rain, snow or fog. Initially, it was thought that dry particle-bound deposition was the dominant atmospheric pathway to plants (Welsch-Pausch et al., 1995). However, further studies revealed that substantial fractions of airborne PCDD/Fs are present in the gas phase during the growing season (Welsch-Pausch et al., 1995). Taking into account the above considerations, vegetation concentrations can be calculated as the sum of four different pathway contributions (EPA, 1990, 1998). The vegetation concentration Cabv is: Cabv ¼ Cva þ Cddp þ Cwdp þ Cur where Cabv: concentration in above-ground vegetation (ppm), Cva: concentration due to vapor-phase absorption (ppm), Cddp: concentration due to dry particulate deposition onto plant matter (ppm), Cwdp: concentration due to wet particulate deposition onto plant matter (ppm), Cur: concentration due to uptake by root onto plant matter (ppm). Particle-borne contaminants released from the MSWI will deposit on vegetation in the surroundings. The potential of local vegetation to intercept and incorporate settling particles depend on a variety of factors, including rate of contaminant deposition, fraction of deposition intercepted (which depends mainly on the cross-sectional area covered by the vegetation), the rate at which degradation and weathering processes remove contaminants, the length of the growing season and the yield of the particular crop. Particle bound contaminants are assumed to deposit into surface soils (Zemba et al., 1996). The vapor-phase absorption assumes equilibrium partitioning of airborne and sorbed pollutants. 2.4. Soil modeling The accumulation of PCDD/Fs in soil is a result of a particle dry and wet deposition, uptake by roots, background concentration, as well as loss of contaminants through processes such as leaching or volatilization. Distribution and mobility of contaminants are affected by soil conditions such as pH, soil structure and characteristics, and water content (EPA, 1990, 1998; Lorber et al., 1994). As for vegetation, soil concentrations were calculated taking into account the contribution of four pathways: Cs ¼ Cbg þ Csddp þ Cswdp Cur where Cs: concentration in above-ground soil (ppm), Cbg: background soil concentration (ppm), Csddp: concentration due to dry particulate deposition in soils (ppm), Cswdp: concentration due to wet particulate deposition in soils (ppm), Cur: concentration due to the uptake by roots onto plant matter (ppm).
The contaminant mass fraction in surface soils was estimated using a simple accumulation model in which pollutants were assumed to deposit, mix and remain within a fixed soil depth. The processes responsible of distributing contaminants throughout the soil include transport by infiltrating rainwater, mechanical mixing by a variety of fauna (earthworms, ants, termites and burrowing animals) and biological decay and tree uprooting. Contaminants may be removed from surface soils as a result of leaching. In addition, biodegradation, hydrolysis and photolysis serve to destroy some chemicals (Zemba et al., 1996). To take into account these loss mechanisms, a first order removal model was used. Details about the modeling algorithms here used for both vegetation and soil were previously reported (Meneses et al., 2002). The soil and vegetation models presented above were previously assessed. Four different pathways of contribution to the vegetation concentrations were taken into account: vapor-phase absorption, dry particle deposition, wet particle deposition and root uptake. The most important pathway was vapor-phase absorption, while the less was root uptake. In the soil model, four pathways were considered: background soil concentration, dry particle deposition, wet particle deposition and root uptake After background concentration, the most important pathway was wet deposition (Meneses et al., 2002). For I-TEQs, a quite good agreement with the measured levels in soil and vegetation was obtained. Although the models predicted higher concentrations for the higher chlorinated PCDD/Fs, the predictions on total I-TEQ concentrations fit very accurately (Meneses et al., 2002). 2.5. Human health risk assessment Human health risk assessment requires identification of the pathways through which people can be potentially exposed to the chemicals of concern (PCDD/Fs in this case). The quantitative estimation of health risk due to a PCDD/F exposure was considered as a combination of five ways: (1) Intake of contaminated soil. Humans ingest small amounts of soil indirectly (hand-to-mouth transfer), which depend on the potential extent of exposure to soil. (2) Ingestion of vegetables grown in the area under evaluation. Although the population may consume vegetables grown in the vicinity of a MSWI, usually most vegetables in an urban diet does not come from the area of potential influence of the plant. Therefore, only a fraction between 1 and 10% of the total ingestion of vegetables from the home-grown was here considered. (3) Inhalation of re-suspended soil particles. Subjects were exposed to air contaminants, assuming that indoor air exposure was equal to outdoor exposure.
M. Meneses et al. / Environment International 30 (2004) 481–489
(4) Inhalation of air. It depends on the contaminant’s concentration in the atmosphere and the inhalation rate. Both vapour and particle air concentration can be inhaled. (5) Dermal absorption. It was assumed to occur only in case of direct contact of contaminated soil to the skin. It depends on the human activity, seasonal differences and skin surface exposed to soil. Exposure was calculated as the average daily intake of I-TEQ equivalents per unit body weight. Details on the parameters and equations used in human exposure calculations were previously reported (Schuhmacher et al., 2001). Because ingestion of PCDD/Fs through the diet is known to be the major pathway of exposure (Schuhmacher et al., 2001), in the present study PCDD/F intakes from foods (1996 and 2000) were calculated for an adult population (Domingo et al., 1999; Llobet et al., 2003). Food samples were randomly obtained from local markets, supermarkets and grocery stores from various towns in Catalonia. From all groups of analyzed samples, the quantity of each food was estimated according to the dietary habits of the population living in the area under assessment population (Domingo et al., 1999; Llobet et al., 2003). Risks for adverse human health effects were estimated depending on they were carcinogenic or non-carcinogenic. For the non-carcinogenic evaluation, in order to determine if the contaminant can possess a risk to human health, the daily intake was compared with the reference dose for chronic exposure (RfD). Reference dose for chronic exposure (RfD) of PCDD/Fs is 1 – 4 pg/kg/day (van Leeuwen and Younes, 1998). The hazard ratio was calculated by dividing the exposure dose by the RfD. In turn, the carcinogenic risk was calculated by multiplying the estimated dose by the cancer potency factor for PCDD/Fs. The carcinogenic potency factor was considered to be 34,000– 56,000 (mg/ kg/day) 1 (Katsumata and Kastenberg, 1997). Details about the modeling algorithms here used were previously given (Schuhmacher et al., 2001).
3. Results and discussion Since the environmental behavior of individual PCDD/F congeners varies substantially, a serious error might be introduced by the 2,3,7,8-TCDD approach. This is especially important for MSWI emissions, for which PCDD/F distributions are frequently dominated by congeners that have transports properties, which differ widely from those of 2,3,7,8-TCDD. EPA guidance emphasizes the need to consider each PCDD/F congener separately, and data to perform congener-specific fate and transport assessment are rapidly becoming available. However, such analysis requires a large number of parameters. Considerable re-
485
search is still needed to characterize all the media-to-media transfer of interest. Another important issue in PCDD/F fate and transport modeling involves the distinction between vapor and particle-bound phase in air. Recent studies have suggested that vegetation readily assimilates vapor phase PCDD/Fs. Although for some congeners a small vapor fraction may contribute, the majority of the pollutant becomes incorporated in leafy tissues (Zemba et al., 1996). In the present study, air concentrations of the 17 toxic PCDD/Fs congeners were simulated at the 24 monitoring points, before and after the new gas cleaning system was installed in the MSWI. Tables 3 and 4 show the simulated mean air concentrations due to PCDD/F emissions for the 24 sampling points, before and after installation of the new equipment. These tables also show the measured air concentrations in the study area (data provided by the Generalitat of Catalonia) in 1996 and 2000. A notable decrease of the I-TEQ values due to reductions in MSWI emissions from 10.33 to 0.014 fg ITEQ m 3 was observed. A comparison of the total (measured) PCDD/F concentration in the air of the area under evaluation, with the concentration due to MSWI emissions, without and with the new cleaning system gas (Tables 3 and 4), shows that PCDD/F emissions from the MSWI have a low contribution to the total air concentration. PCDD/F emissions from the plant before installation of the new equipment contributed with a 4.25% to the total I-TEQ (air), while after the new cleaning system was installed, only a 0.024% of total I-TEQ (air) would correspond to PCDD/F emissions from the MSWI.
Table 3 Emission concentrations of PCDD/Fs (fg m 3) before installation of the new cleaning gas system (1996) in the MSWI of Montcada Cmeasureda
Cestimated 2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
Cgas
Cparticle
Ctotal
0.04 0.52 0.72 1.67 0.58 2.31 2.06 6.85 0.52 3.96 4.93 2.49 0.10 2.74 3.57 0.53 0.65 4.38
0.08 0.90 1.23 2.85 4.44 18.36 20.71 2.09 0.48 3.61 6.37 3.91 0.85 9.68 17.12 5.63 17.31 5.95
0.12 1.42 1.95 4.51 5.02 20.66 22.77 8.94 1.01 7.57 11.30 6.40 0.95 12.42 20.70 6.16 17.96 10.33
40.8 24.9 40.8 89.6 65.63 814.34 300.0 330.34 59.70 127.88 244.0 103.84 8.11 180.03 481.79 74.5 415.27 243.01
a Values correspond to one sample. Data provided by the Regional Government of Catalonia.
486
M. Meneses et al. / Environment International 30 (2004) 481–489
Table 4 Emission concentrations of PCDD/Fs (fg m 3) after installation of the new cleaning gas system (2000) in the MSWI of Montcada Cmeasureda
Cestimated 2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
Cgas
Cparticle
Ctotal
3.81E 04 1.48E 03 1.88E 03 4.89E 03 1.31E 03 9.15E 03 9.23E 03 3.58E 03 7.91E 04 2.35E 03 3.55E 03 3.04E 03 4.94E 05 3.47E 03 6.16E 03 5.27E 04 1.92E 03 0.005
7.49E 04 2.74E 03 3.41E 03 8.92E 03 1.07E 02 7.76E 02 9.90E 02 1.16E 03 7.79E 04 2.28E 03 4.91E 03 5.09E 03 4.28E 04 1.31E 02 3.15E 02 5.95E 03 5.43E 02 0.009
1.13E 03 4.22E 03 5.29E 03 1.38E 02 1.20E 02 8.68E 02 1.08E 01 4.75E 03 1.57E 03 4.63E 03 8.46E 03 8.12E 03 4.78E 04 1.65E 02 3.77E 02 6.48E 03 5.62E 02 0.014
4.1 7.2 7.2 13.6 27.9 70.7 185.0 91.2 10.2 35.8 78.7 29.9 1.8 44.8 131.0 10.5 101.0 58.0
a Values correspond to the mean of two samples. Data provided by the Regional Government of Catalonia.
Figs. 1 and 2 show the I-TEQ congener profiles for air PCDD/F concentrations due to emissions from the MSWI, as well as the measured concentrations before and after installation of the new technical equipment. It can be noted that the congeners showing the greatest contribution to ITEQ were 2,3,4,7,8-PeCDF, TCDF, TCDD and 2,3,4,6,7,8-HxCDF. A source of certain uncertainty of the results can be due to the dechlorination process, which was not here considered (Lorber et al., 2000). This process may occur between
the emission point and the ambient measuring station. Generally, polychlorinated organic compounds easily experience photochemical loss of chlorine atoms. If the higher chlorinated PCDD/Fs are dechlorinated to form lower chlorinated PCDD/Fs in the atmosphere, more lower chlorinated PCDD/Fs would have arrived at the ambient air-monitoring point causing distinct air profile. An evaluation of the ISCST3 model showed the great importance of using onsite data sets in the modeling of air dispersion (Eschenroeder and Lorber, 1999). However, no onsite meteorological data of the MSWI could be obtained. Therefore, information from the meteorological station of Montcada was used. The assumption of equal meteorological conditions in Montcada and the plant leads to an uncertainty in the computed air concentrations. Table 5 summarizes the levels of the 17 toxic congeners of PCDD/Fs in herbage, which were calculated based on the simulated air concentration due to MSWI emissions and measured air concentrations. The calculated herbage concentrations correspond to PCDD/Fs accumulated during a period of 6 months (the lifetime of herbage). It can be seen that emissions from the MSWI contributed with 0.89% and 0.20% to the total vegetation concentrations before and after the installation of the new air cleaning device, respectively. It can be concluded that emissions from the MSWI mean a minimal contribution to the herbage concentrations, less than 1%. The soil concentration was calculated considering a continuous deposition during the period of time in which the MSWI has been operating. Taking into account that the soil is a media where the pollutants accumulate during long periods of time, soil concentrations before installation of the new equipment were calculated estimating a period of accumulation of 22 years. In turn, soil concentrations after
Fig. 1. Profile of air PCDD/F concentrations without the new cleaning gas system in the MSWI.
M. Meneses et al. / Environment International 30 (2004) 481–489
487
Fig. 2. Profile of air PCDD/F concentrations with the new cleaning gas system in the MSWI.
installation of the new equipment were calculated considering the 25 years that the plant was operating without the new cleaning gas system, plus the accumulation during the last year in which the MSWI was operating with the new equipment. The levels of the 17 toxic congeners of PCDD/ Fs in soils are shown in Table 6. Emissions from the MSWI contributed with a 1.59% and 2.62%, respectively, to the ITEQ soil concentration before and after the new cleaning gas system was installed. It can be concluded that emissions
Table 5 PCDD/Fs levels (ng/kg dry weight) in herbage samples before (1996) and after (2000) installation of the air gas cleaning device in the MSWI of Montcada 1996 Csimulated 2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ a
2.1E 04 2.2E 03 7.4E 03 1.7E 02 9.3E 03 5.0E 02 3.9E 01 1.4E 02 1.7E 03 8.2E 03 2.1E 02 1.1E 02 1.1E 03 1.8E 02 4.4E 02 8.4E 03 2.3E 02 1.7E 02
2000 Cmeasureda 1.0E 01 4.4E 01 3.0E 01 2.4E + 00 1.1E + 00 5.2E + 00 1.3E + 01 1.3E + 00 9.8E 01 9.3E 01 1.0E + 00 9.6E 01 3.0E 02 1.3E + 00 4.0E + 00 3.5E 01 1.3E + 00 1.9E + 00
Csimulated 1.1E 04 5.1E 04 1.5E 04 3.1E 04 4.3E 04 3.5E 03 7.6E 03 9.8E 04 7.2E 04 9.7E 04 9.0E 04 6.3E 04 4.7E 05 1.1E 03 2.4E 03 1.8E 04 7.4E 04 1.4E 03
from the MSWI mean also a minimal contribution to total PCDD/F concentrations in soil. Table 7 shows PCDD/F exposure for the population living in the surroundings of the MSWI before (1996) and after (2000) pronounced reductions in the emissions of PCDD/Fs from the facility. Dietary exposure for the two years of study is included. Human carcinogenic and noncarcinogenic risks (hazard ratio), based on the total dose, are also given. It can be observed that PCDD/F exposure due to emissions from the MSWI decreased from 2.37E 06 to Table 6 PCDD/Fs levels (ng/kg dry weight) in soil samples before (1996) and after (2000) installation of the air gas cleaning device in the MSWI of Montcada 1996
Cmeasureda 5.5E 02 2.0E 01 1.0E 01 3.5E 01 2.5E 01 3.1E + 00 7.5E + 00 7.0E 01 4.5E 01 5.0E 01 4.0E 01 4.0E 01 2.5E 02 4.0E 01 1.6E + 00 1.5E 01 1.3E + 00 7.0E 01
Values correspond to the mean of PCDD/F concentrations in 24 herbage samples collected in the vicinity of the MSWI in 1996 and again in 2000.
2,3,7,8-TCDD 1,2,3,7,8-PeCDD 1,2,3,4,7,8-HxCDD 1,2,3,6,7,8-HxCDD 1,2,3,7,8,9-HxCDD 1,2,3,4,6,7,8-HpCDD OCDD 2,3,7,8-TCDF 1,2,3,7,8-PeCDF 2,3,4,7,8-PeCDF 1,2,3,4,7,8-HxCDF 1,2,3,6,7,8-HxCDF 1,2,3,7,8,9-HxCDF 2,3,4,6,7,8-HxCDF 1,2,3,4,6,7,8-HpCDF 1,2,3,4,7,8,9-HpCDF OCDF I-TEQ
2000 a
Csimulated
Cmeasured
4.70E 04 4.97E 03 1.11E 02 2.58E 02 3.95E 02 3.19E 01 5.12E 01 2.20E 02 4.87E 03 3.63E 02 6.13E 02 3.75E 02 7.96E 03 9.18E 02 1.16E 01 3.79E 02 6.59E 02 5.60E 02
1.45E 01 5.90E 01 7.20E 01 1.15E + 00 1.15E + 00 3.00E + 01 1.60E + 02 1.50E + 00 1.20E + 00 1.55E + 00 2.20E + 00 1.60E + 00 1.75E 01 3.25E + 00 1.30E + 01 1.10E + 00 1.45E + 01 3.52E + 00
Csimulated
Cmeasureda
4.72E 04 4.98E 03 1.11E 02 2.58E 02 3.95E 02 3.19E 01 5.13E 01 2.20E 02 4.87E 03 3.63E 02 6.13E 02 3.75E 02 7.98E 03 9.18E 02 1.16E 01 3.79E 02 6.59E 02 5.64E 02
9.00E 02 5.00E 01 5.50E 01 1.00E + 00 9.00E 01 1.35E + 01 6.40E + 01 1.10E + 00 9.00E 01 1.15E + 00 1.45E + 00 1.35E + 00 1.00E 01 1.85E + 00 9.25E + 00 8.00E 01 8.75E + 00 2.15E + 00
a Values correspond to the mean of PCDD/F concentrations in 24 soil samples collected in the vicinity of the MSWI in 1996 and again in 2000.
488
M. Meneses et al. / Environment International 30 (2004) 481–489
Table 7 PCDD/F exposure (ng I-TEQ/kg/day) and health risks for the population living in the vicinity of the plant before (1996) and after (2000) the installation of the air-cleaning device 1996
2000
MSWI emissions Environmental exposure due to MSWI Hazard ratio Cancer risk
2.37E 06 9.48E 04 1.07E 07
6.85E 08 2.74E 05 3.08E 09
Other emission sourcesb Environmental exposure Hazard ratio Cancer risk
1.23E 04 4.92E 02 5.54E 06
4.14E 05 1.66E 02 1.86E 06
Diet Dietary exposure Hazard ratioc Cancer riskd Total exposure Hazard ratio Cancer risk
2.77E 03 1.11E + 00 1.25E 04 2.90E 03 1.16E + 00 1.31E 04
9.03E 04 3.61E 01 4.06E 05 9.45E 04 3.78E 01 4.25E 05
a
a
Values obtained from simulation data. Values obtained from measured data minus simulation data. c RfD = 2.5 pg/kg/day. d Carcinogenic Potency Factor: 25,000. b
6.85E 08 ng/kg/day (97.1% of reduction) throughout the period 1996 – 2000. These values mean 1.93% and 0.17%, respectively, of the contribution to total PCDD/F exposure (excluding diet). Hazard ratio for total PCDD/F exposure (including diet) was reduced during the last 5 years from 1.16 to 0.38 when a RfD of 2.5 pg/kg/day was considered. Cancer risks due to PCDD/F emissions from the plant was 1.07E 07 and 3.08E 09 before and after installation of the clean air system, respectively. On the other hand, cancer risks due to other emission sources in the area were 5.54E 06 and 1.86E 06, respectively. It means that cancer risk contribution of the MSWI with respect to other PCDD/F emissions in the area decreased from 1.93% to 0.17% after the installation of the clean system device. The total cancer risks due to PCDD/F exposure (including diet) for the population living in the vicinity of the MSWI were 1.3E 04 and 4.25E 05 for 1996 and 2000, respectively. Contribution of PCDD/F emissions from the MSWI to total health risks of PCDD/F (including diet) for the population living in the vicinity of the facility were 0.08% and 0.007% before and after the installation of the new cleaning gas system, respectively. Although the reduction of PCDD/F emissions from the MSWI after the new equipment was installed has been very notable (99.9%), the important exposure reduction during the period 1996 – 2000 (67.6%) has been especially due to the decrease in PCDD/F intake throughout the diet (Llobet et al., 2003). Recent emphasis in the risk analysis field has been focused on uncertainty in the risk assessment process.
Some weakness owing the uncertainties in the data and model parameters must be taken into account. The small number of air stack emissions before installation of the new equipment, as well as the lack of consideration of all possible plume depletion mechanisms in the dispersion and deposition modeling (atmospheric degradation), can be considered as a weakness of the model. Congener-specific atmospheric degradation rates, photolytic dechlorination and/or congener-specific soil dissipation rates have not been considered. The number of PCDD/F concentrations measured in air was only one and two for 1996 and 2000, respectively (mean values are here presented). It made the comparison of predicted and measured air concentrations a tenuous exercise. A similar problem was also found with the emissions from the stack: one and four samples, respectively, before and after installation of the new technical equipment. In summary, the present results show that although in general terms, PCDD/F health risks decreased during the examined period (before and after the installation of the new cleaning gas system), this reduction has not been as great as it could be expected according to the very pronounced decreases in PCDD/Fs emissions from the stack. It should be taken into account that these emissions diminished, on average, from 111.39 ng I-TEQN m3 (before) to 0.036 ng I-TEQ/N m3 (after). It clearly indicates that other sources of PCDD/Fs (traffic, industrial activities, local fires, etc.) also have a notable impact on the area under direct influence of the MSWI. Similar conclusions were also reached following a comparison of the congener profile of PCDD/Fs in soil and herbage samples collected near the MSWI with those from samples collected in a close area, which is outside the direct influence of the plant. According to the above, it can be concluded that the health risks due to the current PCDD/F emissions from the MSWI would be of a relatively small significance for the population living in the neighborhood of the facility. References Abad E, Caixach J, Rivera J. Improvements in dioxin abatement strategies at a municipal waste management plant in Barcelona. Chemosphere 2003;50:1175 – 82. Alcock RE, Jones KC. Dioxins in the environment: a review of trend data. Environ Sci Technol 1996;33:3133 – 43. Alcock RE, Gemmill R, Jones KC. Improvements to the UK PCDD/F and PCB atmospheric emission inventory following an emissions measurements programme. Chemosphere 1998;38:759 – 70. Domingo JL, Schuhmacher M, Granero S, Llobet JM. PCDDs and PCDFs in food samples from Catalonia, Spain. An assessment of dietary intake. Chemosphere 1999;38:3517 – 28. Domingo JL, Schuhmacher M, Granero S, de Kok HAM. Temporal variation of PCDD/F levels in environmental samples collected near an old municipal waste incinerator. Environ Monit Assess 2001;69: 179 – 93. Duarte-Davidson R, Alcock RE, Jones KC. Exploring the balance between sources, deposition and the environmental burden of PCDD/Fs in the
M. Meneses et al. / Environment International 30 (2004) 481–489 UK: an aid to identifying uncertainties and research needs. Environ Sci Technol 1997;31:1 – 11. EPA (US Environmental Protection Agency). Methodology for assessing health risks associated with indirect exposure to combustor emissions. Office of Health and Environmental Assessment. Cincinnati, OH, EPA/ 600/6-90/003; 1990. EPA (US Environmental Protection Agency). Methodology for assessing health risks associated with multiple pathways of exposure to combustor emissions. Office of Health and Environmental Assessment. Cincinnati, OH, EPA/600/R-98-137; 1998. Eschenroeder A, Lorber M. An evaluation of EPA’s ISCST-Version 3 Model: Part. I. Air dispersion of dioxins. Organohalog Compd 1999;41:547 – 52. Katsumata PT, Kastenberg WE. On the impact of future land use assumptions on risk analysis for Superfund sites. Air Waste Manag Assoc 1997;47:881 – 9. Kaupp H, McLachlan MS. Atmospheric particle size distributions of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDD/Fs) and polycyclic aromatic hydrocarbons (PAHs) and their implications for wet and dry deposition. Atmos Environ 1999;33:83 – 95. Lee RGM, Jones KC. Gas-particle partitioning of atmospheric PCDD/Fs: measurements and observations on modeling. Environ Sci Technol 1999;33:3596 – 604. Llobet JM, Domingo JL, Bocio A, Casas C, Teixido´ A, Mu¨ller L. Human exposure to dioxins through the diet in Catalonia, Spain: carcinogenic and non-carcinogenic risk. Chemosphere 2003;50: 1193 – 200. Lorber M, Pinsky P. An evaluation of three empirical air-to-leaf models for polychlorinated dibenzo-p-dioxins and dibenzofurans. Chemosphere 2000;41:931 – 41. Lorber M, Cleverly D, Schaum J, Philips L, Schweer G, Leighton T. Development and validation of an air-to-beef food chain model for dioxin-like compounds. Sci Total Environ 1994;156:39 – 65.
489
Lorber M, Eschenroeder A, Robinson R. Testing the USA EPA’s ISCSTVersion 3 model on dioxins: a comparison of predicted and observed air and soil concentrations. Atmos Environ 2000;34:3995 – 4010. Meneses M, Schuhmacher M, Domingo JL. A design of two simple models to predict PCDD/F concentrations in vegetation and soils. Chemosphere 2002;46:1393 – 402. Nessel SC, Butler JP, Post GB, Held JL, Gochfeld M, Gallo MA. Evaluation of the relative contribution of exposure routes in a health risk assessment of dioxin emissions from a municipal waste incinerator. J Expo Anal Environ Epidemiol 1991;1:283 – 307. Schuhmacher M, Xifro´ A, Llobet JM, de Kok HAM, Domingo JL. PCDD/Fs in soil samples collected in the vicinity of a municipal solid waste incinerator: human health risk. Arch Environ Contam Toxicol 1997;33:239 – 46. Schuhmacher M, Domingo JL, Xifro´ A, Granero S, Llobet JM, de Kok HAM. Presence of dioxins and furans in vegetation samples collected in the neighbourhood of a municipal solid waste incinerator. J Environ Sci Health 1998;A33:195 – 212. Schuhmacher M, Meneses M, Xifro´ A, Domingo JL. The use of MonteCarlo simulation techniques for risk assessment: study of a municipal waste incinerator. Chemosphere 2001;43:787 – 99. van Leeuwen FXR, Younes M. WHO revises the tolerable daily intake (TDI) for dioxins. Organohalog Compd 1998;38:295 – 8. Wagrowski DM, Hites RA. Partitioning of polychlorinated dibenzo-p-dioxins and dibenzofurans between the atmosphere and corn. Environ Sci Technol 1998;32:2389 – 93. Welsch-Pausch K, McLachlan MS, Umlauf G. Determination of the principal pathways of polychlorinated dibenzo-p-dioxins and dibenzofurans to Lolium multiflorum (Welsh ray grass). Environ Sci Technol 1995; 29:1090 – 8. Zemba SG, Green LC, Crouch EAC, Lester RR. Quantitative risk assessment of stack emissions from municipal waste combustors. J Hazard Mat 1996;47:229 – 75.