Health risk assessment of dioxin emissions from municipal waste incinerators: the Neerlandquarter (Wilrijk, Belgium)

Chemosphere 43 (2001) 909±923

Health risk assessment of dioxin emissions from municipal waste incinerators: the Neerlandquarter (Wilrijk, Belgium) J. Nouwen *, C. Cornelis, R. De Fre, M. Wevers, P. Viaene, C. Mensink, J. Patyn, L. Verschaeve, R. Hooghe, A. Maes, M. Collier, G. Schoeters, R. Van Cleuvenbergen, P. Geuzens Flemish Institute for Technological Research (Vito), Boeretang 200, 2400 Mol, Belgium

Abstract Two municipal waste incinerators in the vicinity of a residential area close to the city of Antwerp caused concern to local habitants. Risk assessment was performed combining chemical, toxicological measurements and model calculations. As the ®rst step in risk assessment an inventory was made of historic emissions from both incinerators with emphasis on dioxins. The operational atmospheric transport and deposition model for priority substances (OPS) was used to calculate the deposition of dioxins in the vicinity of incinerators. The observed soil contamination pattern did not correspond to the calculated deposition pattern, indicating that other sources may contribute at least partly to the local PCDD/PCDF contamination of the area. Dioxin exposure of people in the Neerlandquarter as a function of the food consumption behavior was calculated using a mathematical model (VLIER-HUMAAN) combined with transfer factors. According to the results of these calculations, just residing in the impact area does not result in a meaningful risk. Only if locally produced food was consumed (milk, meat and vegetables), exposure in the Neerlandquarter was enhanced compared to the average dioxin exposure estimated for the Flemish population. Exposure in 1997 was below the exposure in 1980. As a consequence of di€erent eating habits and lower bodyweight, children are subjected to signi®cantly higher exposure than adults. Adverse health outcomes from dioxin exposure in the past cannot be excluded. There was no evidence for enhanced exposure to genotoxicants based on a comparison of chromosomal damage to blood cells of children from the study area to those from a control group. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Risk assessment; Dioxins; Waste incinerator; Modeling; Exposure; Medical investigation

1. Introduction In 1997, about 60% of the municipal refuse in Flanders was disposed of by incineration in 13 operational waste incinerators. However waste incineration has generated considerable public concern, mainly due to the emission of toxic substances, which present potential human health hazards.

*

Corresponding author. Tel.: +32-14-336950; fax: +32-14336988. E-mail address: [email protected] (J. Nouwen).

In recent years residents of the Neerlandquarter regularly complained of di€erent types of health problems. Several children with congenital anomalies were born from mothers who were inhabitants of the area. The inhabitants linked these malformations to toxicants in the stack emissions from two nearby municipal waste incinerators. Particularly emissions of polychlorinated dibenzo-p-dioxins and dibenzofurans (PCDDs and PCDFs, ``dioxins'') caused deep concern (Schoeters et al., 1998). The Neerlandquarter is a well-outlined residential quarter of 2500 inhabitants located between the two incinerators. The waste incinerators were in full operation since the seventies (waste incinerator IHK) and the

0045-6535/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 0 4 5 - 6 5 3 5 ( 0 0 ) 0 0 5 0 4 - X

910

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

eighties (waste incinerator ISVAG). Both waste incinerators were shut down in November 1997 due to exceeding the dioxin emission standards (0.1 ng TEQ=m3 ) and considerations of health. It should be noted that the Neerlandquarter is also surrounded by other potential sources of environmental pollutants, among them being two very busy trac arteries, a metal smelting and re®ning industry and a crematory (see Fig. 1). Emissions from these sources are not taken into account in this study and neither is dispersion of the ash in the environment. In addition to polychlorinated dibenzo-p-dioxins (PCDDs) and polychlorinated dibenzofurans (PCDFs) also other environmental pollutants are emitted from the incinerators and from other sources in the area including heavy metals, polyaromatic hydrocarbons, polychlorinated biphenyls, sulfur oxides and nitrogen oxides (Valberg et al., 1996). Because PCDDs and PCDFs have generated the most public concern, as they are perceived as the most hazardous (Nessel et al., 1991) the present study focuses on these substances. Despite the concern regarding health e€ects of PCDDs and PCDFs and the fact that PCDDs and PCDFs include 210 di€erent congeners, chronic toxicity studies have been conducted on only 2,3,7,8-tetrachlorodibenzo-p-dioxin (2,3,7,8-TCDD) and a mixture of 1,2,3,6,7,8- and 1,2,3,7,8,9-hexachlorodibenzo-p-dioxins (1,2,3,6,7,8- and 1,2,3,7,8,9-HxCDD). Therefore, all data in this article are transformed to 2,3,7,8-TCDD ``toxic equivalents'' by the application of weighting factors (I-TEFs or International Toxic Equivalent Factors) developed by EPA (Ahlborg et al., 1992). This allows conversion of a concentration of any PCDD or PCDF congener into the toxicologically equivalent concentration of 2,3,7,8-TCDD and estimation of risks associated with exposure to mixtures of PCDDs/PCDFs. The major routes of human exposure and the relative contribution of each route for the period 1971±1997 were identi®ed. This process consists of: analysis of the emission data, utilization of deposition modeling, determination of accumulation in speci®c environmental media using transfer factors, exposure evaluation and risk characterization. The paper summarizes the available results of the medical investigations carried out on the population of the Neerlandquarter. This surveillance was executed to complement the ®ndings of the risk assessment. 2. Emissions, dispersion and deposition Emission measurements became only available in recent years due to regulatory obligations in Flanders. However dioxin emissions have been most signi®cant during the ®rst years of exploitation, when precautions for ¯ue gas scrubbing were limited and the combustion process was interrupted regularly. The emissions used in

the calculations are based on the available data as much as possible while the missing data are estimated on the basis of validated emission factors and ®gures on production and refuse quantities (van Jaarsveld and Schutter, 1992). In this way underestimation due to start-up or shut down resulting in a greater emission of dioxins is avoided. An emission factor of 240 lg TEQ/ ton waste or an average ¯ue gas concentration of 40 ng TEQ=Nm3 was used for an incinerator with only electrostatic dust separation (Thome-Kozmiensky, 1994). For an incinerator with full wet treatment of ¯ue gases (Thome-Kozmiensky, 1994), an emission factor of 22 lg TEQ/ton waste or an average ¯ue gas concentration of 3.7 ng TEQ=Nm3 was taken. Both incinerators burned refuse on grates. Ash falls to the bottom and is removed. Gaseous e‚uents and suspended material pass through an electrostatic precipitator and are vented through a stack. The ISVAG municipal waste incinerator, situated to the west of the Neerlandquarter, was in full operation from 1980 until 1993 only using a dry electrostatic precipitator (ThomeKozmiensky, 1994). The ¯ue gas ¯ow expressed as wet gas is estimated to be 200 000 Nm3 =h, temperature at 200°C and the water content 25% based on measurements. In July 1994, the wet ¯ue gas treatment was installed (Thome-Kozmiensky, 1994). The average ¯ue gas temperature lowered to 80°C and the water content increased to 40%. The ¯ue gas ¯ow expressed as wet gas is 260 000 Nm3 =h. In 1980, the average capacity was 55 000 ton/yr. From 1995 onwards the burning capacity of refuse has been doubled. The IHK municipal waste incinerator (capacity 11 000±30 000 ton/yr) is situated to the east of Neerland and was operational between 1971 and 1997. From the start until 1974, the incinerator consisted of a single Claudius-Peeters furnace line (Thome-Kozmiensky, 1994). In 1974 a second line of the same type was installed, resulting in a doubled capacity. The ¯ue gas scrubbing was improved in 1987 by electrostatic dust separation and sorbalite injection (Thome-Kozmiensky, 1994). The results are calculated using a ¯ue gas volume of 5500 Nm3 =ton, augmented with a water content of 8% and a ¯ue gas dilution factor of 1.82 corresponding with an oxygen content of 15.5%. After installation of the energy recovery in 1987, the ¯ue gas temperature was 160°C in the winter and 180°C in summer. The total dioxin emissions of both waste incinerators amounted to 18.9 g TEQ/yr in 1980 and to 3.1 g TEQ/yr in 1997. At present, the oldest incinerator (IHK) has been closed de®nitively and the second one (ISVAG) remained temporarily closed because of crossing the emission limit of 0.1 ng TEQ=m3 . After some technical adaptations, ISVAG restarted its activities since 3 November 1999. However, the discussion on ®nal closure still goes on. The geographical distribution of the total dioxin deposition as a consequence of the emissions of both

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

911

Fig. 1. The Neerlandquarter location situated in relation to the di€erent sources of environmental pollutants (two very busy trac arteries, industrial area, crematory and the two waste incinerators) and the topsoil measurements.

municipal waste incinerators was calculated using the operational priority substances model (OPS-model) for atmospheric transport and deposition (van Jaarsveld, 1989; Slob and van Jaarsveld, 1993). Using the OPS-

model one supposes that dioxins leaving the incinerators stack are bound to dust particles emitted by the municipal waste incinerator. The dispersion of dust particles in atmosphere depends on characteristics of the

912

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

source (stack height, heat content, dust particle-size distribution) and on the meteorological conditions (wind direction, wind velocity, temperature, atmospheric stability). The previously described emission characteristics of both municipal waste incinerators ISVAG and IHK were used to de®ne the emission sources in the model. Hourly meteorological data from observation stations in Antwerp, Ghent and Mol were used to obtain a climatological database for the period 1990±1994. The predominant wind direction in the area is southwest. The transport and deposition of the particle-bounded dioxins are described using ®ve particle-size classes (< 0:95 lm; 0.95±4 lm; 4±10 lm; 10±20 lm; > 20 lm). This is required to account for the di€erent properties of small and big particles (van Jaarsveld, 1989; Slob and van Jaarsveld, 1993). Out of the three standard particlesize distributions (®ne, medium, coarse), the medium size distribution was chosen. Deposition velocities for each of the ®ve classes are taken from the model of Sehmel (1980). Concentrations and depositions were computed for yearly emission periods. The depositions were accumulated over the period 1971±1997. The modeling was performed on a grid domain of 8  6 km2 with a spatial resolution of 200 m. The resulting depositions are surface concentrations which can be converted to soil concentrations by assuming a top layer of 3 cm and an average soil density of 1:5 g=cm3 . Deposition samples were analyzed as described by De Fre et al. (1994). 3. Environmental fate and transport The accumulation of PCDDs and PCDFs in environmental media was calculated based on the results of deposition modeling. Accumulation in water compartment was not relevant the Neerlandquarter. Atmospheric substances are partitioned to soil and vegetation and may bio-accumulate in livestock (cows milk and meat). This results in a number of indirect sources for human exposure (Valberg et al., 1996). Concentrations of dioxins in vegetables, cows milk and meat were calculated using transfer factors according to Prinz et al. (1993). 3.1. Soil Soils are natural sinks for persistent lipophilic compounds such as PCDDs/PCDFs. Once absorbed to the organic material of the soil, they remain quite immobile. Soil is a typical accumulating matrix with long memory. Dioxin inputs received in the past will remain. Due to the very long half-life (12 yr) of PCDDs/PCDFs in soils, there is hardly any clearance (Nessel et al., 1991). Top soil samples (0±5 cm, except when indicated otherwise) were collected by Kerremans and Geukens (1996) at 15 di€erent locations in the Neerlandquarter from intact ground as indicated in Fig. 1 and analyzed

isomer-speci®cally for PCDDs and PCDFs as reported before (Van Cleuvenbergen et al., 1993b). Samples 3, 8, 11 and 16 were taken at the same location but with increasing depth. Sample 19 was taken outside the impact area and used as a reference for normal soil background concentrations in Flanders. The results are expressed using the International Toxicity Equivalence Factors (ITEFs). Exposure is calculated using the highest reported soil concentration (27.2 ng TEQ/kg dm). 3.2. Vegetation Chemical accumulation in vegetation may occur by several routes: (1) Atmospheric deposition of dioxins originating from the waste incinerator or (2) volatilization from the soil surface and deposition on the leaves (3) root uptake from the soil and translocation within the plants (H ulster and Marschner, 1993; Wagrowski and Hites, 1997). Since PCDDs/PCDFs are hardly soluble, translocation to the aerial parts of plants is limited. The only exceptions involve several members of the cucumber family (cururbitaceae) that do take up and translocate PCDDs and PCDFs to the fruit (H ulster et al., 1994). Additionally, only the lower chlorinated congeners volatilize (H ulster, 1993). Consequently, atmospheric deposition is the major process. One of the highest accumulating plants is kale. However Liem et al. (1993) have observed no di€erence in the dioxin levels between samples inside and outside the impact region. Kale is also a minor component in the average Dutch diet. As a consequence, the ``air-to-plant'' transfer factor for kale was not used (Prinz et al., 1993). The aerial deposition was modeled using an air-to-plant transfer factor of 0.2. Translocation to vegetables was calculated based on the following equations (Van Hall Instituut, 1997): Cst ˆ BCFst dmst Csoil ;

…1†

Cr ˆ BCFr dmr Csoil ;

…2†

Cpt ˆ …Cst ‡ Cdep † fst=r ‡ Cr …1 Cdep Cst Cr Cpt BCFr BCFst dmst dmr fst=r

fst=r †;

…3†

concentration in vegetables by deposition in ng TEQ/day kg fresh weight concentration in the stem based on fresh weight in ng TEQ/day kg fresh weight concentration in the root based on fresh weight in ng TEQ/day kg fresh weight concentration in fresh vegetables in ng TEQ/day kg fresh weight bio-concentration factor root here 0.033 bio-concentration factor stem here 0.0057 dry matter stem here 0.202 dry matter root here 0.117 fraction leaves compared to complete plant here 0.4.

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

For roots the bio-concentration for potatoes was taken although the edible parts of potatoes and carrots develop underground and are considered as not signi®cantly exposed to PCDDs/PCDDFs (Abbott et al., 1997). For the stem a bio-concentration factor for vegetables in general was used (Travis and Arms, 1988). It is assumed that TCDD on the plant surface does not degrade. 3.3. Water Fishery in the impact region of both the municipal waste incinerators was not worth mentioning. Therefore exposure to contaminated particulates via sediment and ®sh was reduced to the background exposure of the general population. Based on the extremely low solubility of TCDD in water (log Kow ˆ 6:62) it is assumed that the PCDDs/PCDFs are concentrated in suspended particles and sediment. Therefore, drinking water is not a signi®cant source of human exposure to PCDDs/PCDFs, since these solids are removed during water treatment processes (Abbott et al., 1997). Preliminary calculations by means of the VLIER-HUMAAN-model (Van Hall Instituut, 1997) subscribed to this point of view but were not taken into account in the ®nal modeling. 3.4. Livestock Farm animals can be exposed to PCDDs and PCDFs via inhalation, ingestion of contaminated food and ingestion of contaminated soil. These compounds bio-accumulate in animals. This results in a contamination of meat and milk products which humans may consume. 3.5. Milk The bioaccumulation of PCDDs and PCDDFs in cow milk was estimated in ®rst instance using the chain model of Slob and van Jaarsveld (1993). However, much higher dioxin concentrations in milk for relatively much lower depositions were observed due to the fact that the chain model is calibrated based on calculated depositions whereas we related measured depositions to concentrations in cow milk. Although not proved it might be that the Bergerho€-method for measuring deposition underestimates the real intake into soil (Prinz et al., 1993). Therefore a transfer factor ``fodder-to-cow milk'' of two was used in order to calculate the dioxin concentrations

913

in cow milk. This resulted in a transfer factor ``dioxin-tocow milk'' of 0.4 and showed to be in good agreement with the experimental results (Cornelis et al., 1998). Milk samples were analyzed according to Van Cleuvenbergen et al. (1993a). Samples of cow milk were taken in January 1992 and May 1998 on behalf of the Ministry for Agriculture. Exposure in 1997 is calculated using the highest (11 pg TEQ/g fat) and the lowest (1.5 pg TEQ/g fat) reported concentration for cow milk. For the state in 1980 the result of the model calculation was used. Milk of a nursing mother residing in the Neerlandwijk was analyzed using the same method. 3.6. Meat The concentration of PCDDs/PCDFs in meat was calculated in a manner similar to that in milk for several reasons: (1) there is a good agreement between the transfer factor ``dioxin-to-cow milk'' de®ned by Prinz et al. (1993) and the observed results (2) the transfer factors are based on deposition measurements using the Bergerho€-method. The ``fodder-to-cow meat'' transfer factor was estimated to be 0.35. Expressed in fat content, this transfer factor results in a similar relation as the fodder-to-cow milk transfer factor. On this basis the ``deposition-tomeat'' transfer factor becomes 0.07. The contribution of intake via soil is not considered since the data of Fries (1995) imply that under normal conditions the pathway air-grass-cow is more important than the pathway soilgrass-cow. Additionally, studies on a calf have shown that orally administered dioxins are excreted for 91.93% via urine (10.36%) and via feces (81.57%). Only 0.0002% are accumulated in the muscles (Hakk et al., 1997). By conclusion our approach does not underestimate the dioxin concentration in cow meat. The di€erent transfer factors applied in the modeling are summarized in Table 1. 4. Receptor population 4.1. Exposure scenarios Five scenarios were examined based on di€erent exposure patterns. Although a worst case pattern (scenario I) is highly unlikely as revealed from inspection of the site, this was included to estimate exposure in individuals

Table 1 Overview of the transfer factors used in the model calculations Transfer factor

Units

Prinz et al. (1993)

Deposition ± plant Fodder ± cow milk Fodder ± meat Deposition ± meat Deposition ± cow milk

(ng/kg dm)/(pg/m2 day) (ng/kg fat)/(ng/kg dm) (ng/kg)/(ng/kg dm) (ng/kg fat)/(pg/m2 day) (ng/kg fat)/(pg/m2 day)

0.2 2 0.35 0.07 0.4

914

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

consuming foods (meat, milk and vegetables) produced entirely in the impact area and living in this area. The common case (scenario II) was comprised of individuals whose exposure pattern would be representative of the general population. These individuals would consume products sold commercially which contain only a small percentage of contaminated foods. These people only reside in the impact area. Three intermediate scenarios with increasing consumption of food produced in this area were also included in the assessment. A lower exposure scenario (scenario III) estimated the exposure of individuals living in the Neerlandwijk and consuming 25% of the vegetables originating from gardens at the location. In scenario IV it was assumed that people lived and consumed 25% of the crops and 50% milk produced in the impact area. The highly exposed case (scenario V) was comprised of individuals who lived in the impact area and grew 25% of their own crops and raised livestock with 50% meat and 100% milk from this area. Although it is known that ingestion of contaminated human milk may result in newborn infants in higher exposure to TCDD/PCDDs, this was not taken into account in the di€erent scenarios, because exposure via breast-feeding is limited in time. 4.2. Background exposure There is only limited information on the composition and the PCDD/PCDF levels in a normal Belgian diet. Additionally, the PCDD/PCDF levels in Belgian cow milk grazing in the neighborhood of municipal waste incinerators are comparable to the PCDD/PCDF levels found in Dutch cow milk (Liem et al., 1991; Van Cleuvenbergen et al., 1993b). If one assumes that Dutch and Belgian diets are roughly the same, the daily background exposure to dioxins in Belgium varies from 1 pg TEQ/kg body weight for adults and 2.4 pg TEQ/kg body weight for children. This dietary background exposure is taken into account proportionally in each exposure scenario. Although certain factors need to be veri®ed for Flanders, a decline in the dietary exposure to dioxins since the end of the 1970s is observed (Liem et al., 1997). It is assumed that the receptor population is exposed to emissions for a 70-yr lifetime, divided into two age groups: young children (up to 6 yr old, 15 kg average body weight) and adults (age 15±70 yr old, 70 kg). Exposure from the sources was estimated in each group and a time-weighted average daily dose was calculated from the results.

various environmental media and biota. The exposure pathways considered in this assessment include inhalation; dermal absorption from soil; ingestion of soil, vegetation, beef, water and cows milk. Exposure was calculated as the average daily intake of TCDD equivalents per unit body weight and was estimated by the models presented in VLIER-HUMAAN. Details on the equations and calculation of the time-fraction can be obtained from the Van Hall Instituut (1997). 5.1. Inhalation Inhalation exposure to emissions is calculated by assuming that individuals are exposed to contaminated air 24 h/day and that indoor air exposure is equal to outdoor air exposure. Respiratory bioavailability is conservatively assumed as 100% since no value has been published. The daily dose for adults/children is calculated by the following equation (Van Hall Instituut, 1997): IPad=ch ˆ Vrad=ch Cair fr …Nso;ad=ch ‡ Nwo;ad=ch ‡ Nsl;ad=ch ‡ Nwi;ad=ch ‡ Nsi;ad=ch †=Wad=ch ;

…4†

where IPad=ch fr Vrad=ch Cair Wad=ch Nso;ad=ch ‡ Nwo;ad=ch ‡ Nsl;ad=ch ‡ Nwi;ad=ch ‡ Nsi;ad=ch

inhalation exposure in ng TEQ/kg day for adults/ children The alveolar fraction retained in the lungs is 0.75 Ventilation Rate for adults 20 m3 =day, for a child 7:6 m3 =day the concentration in air expressed in pg TEQ=m3 body weight adult 70 kg, body weight child 15 kg time fraction that adults/children spend at the impact site

5.2. Dermal exposure According to equations of the VLIER-HUMAAN, dermal exposure is calculated for adults/children by Eqs. (5)±(7) (Van Hall Instituut, 1997): DAad=ch ˆ DAsoil;ad=ch ‡ DAdust;ad=ch ;

…5†

5. Human exposure

DAsoil;ad=ch ˆ DAEo;ad=ch DARad=ch fm24Aexpo;ad=ch Nso;ad=ch
…6† ‡ Nwo;ad=ch Csoil =Wad=ch ;

The human exposure to PCDD/PCDF emissions from the municipal waste incinerator was calculated based on the concentration of TCDD equivalents in

DAdust;ad=ch ˆ DAEi;ad=ch DARad=ch fmfrdi 24Aexpi;ad=ch …Nsi;ad=ch ‡ Nwi;ad=ch †Csoil =Wad=ch ; …7†

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

where DAad=ch DAEo;ad=ch DAEi;ad=ch DARad=ch fm frdi (Nso;ad=ch ‡ Nwo;ad=ch ) (Nsi;ad=ch ‡ Nwi;ad=ch ) Aexp;o;ad=ch

Aexp;i;ad=ch Wad=ch Csoil

dermal absorption in ng TEQ/ kg day for adults/children skin coverage with dust outside 0.0375 kg/m2 for adults and 0.0051 kg/m2 for children skin coverage with dust inside 0.00056 kg/m2 for adults and for children 0.01 1/h matrix factor here 0.15 fraction soil in the inside dust, here 0.8 time fraction, 0.158 for adults and 0.151 for children time fraction here 0.458 for adults and 0.306 for children exposed skin surface area forehands and hands, for adults 0:17 m2 and for children (also including legs) 0:28 m2 exposed skin surface, just hands, for adults 0:9 m2 and for children 0:05 m2 bodyweight adult 70 kg, bodyweight child 15 kg concentration in soil in ng TEQ/kg dm

for adults. Eqs. (8)±(10) deliver the ingestion of dioxins via soil particles: DUad=ch ˆ DUsoil;ad=ch ‡ DUdust;ad=ch ;

PCDD/PCDFs may be ingested directly following deposition on soil and vegetation and indirectly through bioaccumulation in ®sh, beef and milk. The bioavailability is always estimated as 100% from particulates such as soil, which deposit on vegetation as well as from other food sources. This is again a worst case approximation. 5.3.1. Soil Humans ingest small amounts of soil indirectly from food and other sources, including the ingestion of inhaled particles that are cleared from the upper respiratory region. In addition small children may ingest soil directly as a normal behavior during childhood or as a result of pica (Hawley, 1985). Pica-behavior was not considered since it only applies to a small fraction of the population and this is during a very limited time-period. Adult soil ingestion is smaller than that of children. Ingestion estimates of 75 mg/day for children and 35 mg/ day for adults are used in the agricultural scenario. For all the living scenarios the exposure is calculated using a soil ingestion of 50 mg/day for children and 20 mg/day

…8†

DUsoil;ad=ch ˆ …AIDad=ch …24=…24 Tsl;ad=ch †† …Nso;ad=ch ‡ Nwo;ad=ch † Csoil †=Wad=ch ;

…9†

DUdust;ad=ch ˆ …AIDad=ch …24=…24 Tsl;ad=ch †† frsi …Nsi;ad=ch ‡ Nwi;ad=ch † Csoil †=Wad=ch ;

…10†

where DUad=ch DUsoil;ad=ch DUdust;ad=ch (Nso;ad=ch ‡ Nwo;ad=ch ) (Nsi;ad=ch ‡ Nwi;ad=ch ) Csoil

5.3. Ingestion

915

Wad=ch AIDad=ch

Tsl;ad=ch frsi

ingestion of contamination via soil and dust by adults/ children in ng TEQ/kg day ingestion of contaminated soil by adults/children in ng TEQ/kg day ingestion of contaminated dust by adults/children in ng TEQ/kg day time fraction for adults is 0.158 and for children 0.151 time fraction for adults is 0.458 for children 0.306 concentration in soil in ng TEQ/kg dm bodyweight adult 70 kg, bodyweight child 15 kg ingestion of soil particles amounts to 2:6  10 5 kg/ day for adults and 6:3  10 5 kg/day for children sleeping time adults 8 h/day, sleeping time child 12 h/day fraction soil in dust inside, here 0.8

5.3.2. Milk PCDDs and PCDFs bio-concentrate in animals due to their long half-life in adipose tissue. This results in an accumulation in the cow milk. It is assumed that children drink more milk than adults do. Additionally, the receptor population shows di€erent consumption behaviors of the exclusively locally produced milk. In¯uence of the di€erent measured milk samples on the ®nal exposure was checked. The exposure via consumption of contaminated milk is given by Eq. (11):

916

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

Mlmi;ad=ch ˆ …Cmi fmi Qmi;ad=ch †=Wad=ch ; Mlmi;ad=ch Cmi fmi Qmi;ad=ch Wad=ch

…11†

quantity of dioxins ingested via milk by adults/children in ng TEQ/kg day concentration of dioxins in milk in ng TEQ/l fraction contaminated milk consumed daily consumption of milk by adults 0.3 l/day and by children 0.5 l/day bodyweight adult 70 kg, bodyweight child 15 kg

5.3.3. Vegetables Homegrown vegetables cultivated in the vicinity of the incinerator may be consumed by the local people in varying quantities depending on the exposure scenario. No distinction between several types of vegetation consumed daily has been made. The exposure via vegetables is calculated using Eq. (12) (Van Hall Instituut, 1997): VIad=ch ˆ Cpt ffv Qfv;ad=ch =Wad=ch ; …12† where Vlad=ch Qfv;ad=ch ffv Wad=ch

quantity of dioxins ingested via vegetables by adults/children in ng TEQ/kg day quantity of vegetables consumed amounts to 0.345 kg/day for adults and 0.150 kg/day for children fraction contaminated vegetables consumed bodyweight adult 70 kg, bodyweight child 15 kg

5.3.4. Meat Similarly as for vegetables, it is assumed that adults consume more meat than the children do. The daily dose of dioxins for adults and children via meat is obtained by (Van Hall Instituut, 1997): Mlme;ad=ch ˆ …Cme fme Qme;ad=ch †=Wad=ch ;

…13†

where Mlme;ad=ch Qme;ad=ch fme

quantity of dioxins ingested via meat by adults/children in ng TEQ/kg day quantity of meat consumed amounts to 0.14 kg/day for adults and 0.07 kg/day for children fraction contaminated meat consumed

5.4. Total exposure Total human exposure to PCDDs/PCDFs emitted from the municipal waste incinerators is calculated as the summation of the daily doses from the individual sources (Van Hall Instituut, 1997).

Total Exposure ˆ DAad=ch ‡ DUad=ch ‡ IPad=ch ‡ Mlmi;ad=ch ‡ Mlme;ad=ch ‡ VIad=ch :

…14†

6. Medical investigations An inventory of the medical cases was made based on the ®ndings of Cremers et al. (1997) who have reported 10 cases of congenital malformations. Via residents, the local action comite against the waste incinerators and medical doctors of the region, all additional cases of congenital malformations were traced. The rates of occurrence of these diseases in the total Flemish population were compared to that of the residents of the Neerlandquarter. Only malformations diagnosed during abortion or during the ®rst year of life were registered. The parents of children with such congenital malformations were asked in a questionnaire to give information on the date of birth of the child, dwelling place at the time of birth, professional exposure and residence time in the area, smoking behavior, diagnosis and information on other medical tests executed on the child and parents such as chromosomal analysis. Available information on family history relating to the reported cases was checked. The cases were classi®ed as either due to a genetic factor or multifactorial in origin. To study the enhanced exposure to genotoxicants in the area, peripheral blood lymphocytes from children were cultured and analyzed for the presence of chromosomal aberrations. A metaphase test (IAEA, 1986) was performed to evaluate structural changes in the chromosomes caused by a recent exposure while the G-banding technique (Seabright, 1971) allows evaluating chromosomal changes, which occurred eventually earlier but remained stable in time. From each child, 200 cells and 100 cells were studied in, respectively, the metaphase test and Gbanding analysis. The investigated individuals were residing in the Neerlandquarter for a suciently long time (most children more than 7 yr) (Schoeters et al., 1998). Blood samples were taken from a group of 24 children from the Neerlandquarter after obtaining informed consent from the parents. The children were mainly selected based on age and habits (smoking behavior and occupational exposure of the parents, medication, diseases in the children). The ®ndings were compared with a referent group of 20 children with similar age (10±12 yr) from a nearby comparable suburban area, which is not located near the incinerators. The results were statistically analyzed using the percentage test, analyses of variance and Wald-Wolfowitz test (Perera et al., 1992).

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

7. Results and discussion A deposition of 1.31 ng dioxins TEQ=m2 yr (dry deposition 1.05 ng TEQ=m2 yr, wet deposition 0.26 ng TEQ=m2 yr) in the Neerlandquarter was measured during April±May 1997 using the Bergerho€-method. Modeling resulted in a maximum deposition ®eld value of 1.21 ng TEQ=m2 yr (dry deposition 0.89 ng TEQ=m2 yr, wet deposition 0.22 ng TEQ=m2 yr, background deposition 0.1 ng TEQ=m2 yr) for the same period. The background deposition that originates from recirculation and other sources than both municipal waste incinerators ISVAG and IHK is estimated to be 0.1 ng TEQ=m2 yr based on calculations. The background deposition was not considered for the situation in 1980 due to lack of information but must have been signi®cantly higher. Starting from the emissions of both the municipal waste incinerators in 1980, a maximum value of 6.62 ng dioxins TEQ=m2 yr (dry deposition 5.18 ng TEQ/m 2 yr, wet deposition 1.44 ng TEQ/m 2 yr) in the Neerlandquarter was obtained. Table 2 summarizes the measured dioxin concentrations in the di€erent topsoil samples. Fig. 1 indicates the sampled locations. The numbers of the samples correspond with the numbers of the spots in Fig. 1. The measured soil contamination pattern is indicated in Fig. 2. This ®gure is a reconstruction based on the measurements as listed in Table 2. Fig. 3 shows the results of the deposition modeling for the Neerlandquarter.

917

Although the municipal waste incinerators are held responsible for the PCDD/PCDF contamination, the deposition modeling results (Fig. 3) do not correspond with the measured soil contamination pattern (Fig. 2). Firstly, lower soil concentrations are obtained via deposition modeling than those experimentally observed. Secondly, the soil concentration measurements do not correspond with the meteorological statistics, which showed a mainly southwesterly wind direction. The deposition modeling showed two maxima for both the incinerators one in the northwest of ISVAG (neighborhood of receptor point 8, see Fig. 2) and one in the northwest of IHK (neighborhood of receptor point 3, see Fig. 2). This deviates signi®cantly from the measured maximum observed at receptor point 7 (27.2 ng TEQ/kg dm, see Fig. 1) in the northwest of IHK and the northeast of ISVAG and the relatively high measured soil concentration at receptor point 15 in the southwest of both the waste incinerators. It indicates that contamination originates at least partially from other sources. The risk-index was calculated by dividing the summation of the background and lifelong exposure by the Tolerable Daily Intake as de®ned by the World Health Organization (WHO). Consequently, a TDI of 10 pg TEQ/kg bw day was used (Hoogenboom et al., 1995). A risk-index < 1 means that the risks are below the TDI. On the contrary, a risk-index P1 indicates a risk since the TDI is achieved or exceeded. Tables 3 and 4 show the calculated exposures for, respectively, 1980 and 1997. A comparison of the state in 1980 with this in 1997 shows a signi®cant improvement

Table 2 Measured dioxin concentration in soil (Kerremans and Geukens, 1996)a;b

a b

Sample no

Description

depth (cm)

ng TEQ/kg dm

1 2 3 3 4 5 6 7 8 9 10 10 11 12 13 14 15 16 17

500 m NE IHK 1000 m NE IHK 2000 m NE IHKa 2000 m NE IHKa 3000 m NE IHK 500 m SE IHK 500 m SW IHK 500 m NW IHK 600 m NE ISVAG 1200 m NE ISVAG 2400 m NE ISVAGb 2400 m NE ISVAGb 3600 m NE ISVAG 600 m SE ISVAG 600 m SW ISVAG Verge Boomsesteenweg Verge Boomsesteenweg 600 m NW ISVAG Reference Kalmthout

0±5 0±5 0±5 >5 0±5 0±5 0±5 0±5 0±5 0±5 0±5 >5 0±5 0±5 0±5 0±5 0±5 0±5 0±5

15.1 4.0 8.5 5.7 4.5 3.9 4.2 27.7 12.2 15.9 12.8 4.6 4.8 5.8 7.5 15.9 13.5 22.4 1.5

Samples taken on the same location but on di€erent depth. Samples taken on the same location but on di€erent depth.

918

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

Fig. 2. Soil contamination pattern of the Neerlandquarter based on the measured PCDD/PCDF concentrations in the topsoil.

Fig. 3. Soil contamination pattern of the Neerlandquarter based on deposition modeling.

of the situation. The lifelong average exposure has even decreased with a factor 3 in scenario I. In¯uences are most apparent in this scenario because all exposure pathways are completely taken into account. The exposure and the concomitant risk-index increases from the scenario II, III, IV, V to I due to the increasing consumption of food produced in the impact area. Just

residing in the impact area does not result in a meaningful risk (scenario II, risk-index ˆ 0.12 for 1980 and 0.11 for 1997). Prior to 1980 the background deposition must have been signi®cantly higher than in 1997, the exposure for 1980 is probably underestimated and consequently could come up to or exceed the TDI in some cases (scenario I).

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

919

Table 3 Calculated exposure in the Neerlandquarter for 1980 (pg TEQ/kg bw day) a

I IIb IIIc IVd Ve

Child + background

Adult + background

Lifelong + background

Risk-index (exposure/TDI)

16.62 2.48 3.37 6.65 12.64

5.09 1.05 1.49 1.70 3.28

6.08 1.17 1.65 2.22 4.09

0.61 0.12 0.17 0.22 0.41

a

Scenario I: living with consumption of exclusive locally produced food (milk, meat and vegetables). Scenario II: living with consumption of only commercially sold foods (milk, meat and vegetables). c Scenario III: living with consumption of only 25% locally produced vegetables. d Scenario IV: living with consumption of 25% vegetables and 50% milk produced in the impact area. e Scenario V: living with consumption of 25% vegetables, 50% meat and 100% milk produced in the impact area. b

Table 4 Calculated exposure in the Neerlandquarter for 1997 (pg TEQ/kg bw day) Ia IIb IIIc IVd;  IVd;  Ve; Ve;

Child + background

Adult + background

Lifelong + background

Risk-index (exposure/TDI)

4.93 2.42 2.68 3.42 8.17 4.11 13.61

1.71 1.02 1.15 1.26 1.87 1.33 2.55

1.99 1.14 1.28 1.45 2.41 1.57 3.50

0.20 0.11 0.13 0.15 0.24 0.16 0.35

a

Scenario I: living with consumption of exclusive locally produced food (milk, meat and vegetables). Scenario II: living with consumption of only commercially sold foods (milk, meat and vegetables). c Scenario III: living with consumption of only 25% locally produced vegetables. d Scenario IV: living with consumption of 25% vegetables and 50% milk produced in the impact area. e Scenario V: living with consumption of 25% vegetables, 50% meat and 100% milk produced in the impact area. * Exposure calculated using cows' milk containing 1.5 pg TEQ/g fat. ** Exposure calculated using cows' milk containing 11.0 pg TEQ/g fat. b

The calculated exposure is always higher for children than for adults. This has to be at least partially attributed to the di€erent consumption behavior (especially milk) and lower bodyweight of the former. Therefore, risks for children are raised and e€ects cannot be excluded, in particular for the highly exposed scenarios (I: 16.62 pg TEQ/kg bw day, V: 12.64 pg TEQ/kg bw day in 1980, IV : 8.17 pg TEQ/kg bw day, V : 13.61 pg TEQ/ kg bw in 1997). This again con®rms that if one is willing to drive back the exposure for children and to a lesser extent for adults, the dioxin content in milk should be properly controlled. Milk samples from January 1992 and May 1998 contained, respectively, 7.9 and 1.5 pg TEQ/g fat, which corresponded well with the model calculations using transfer factors. Recently in November 1997 two isolated cases showed concentrations of 11 pg TEQ/g fat in milk of a Jersey Cow and 10 pg TEQ/g fat in milk of a standard milk cow. The calculations for scenarios IV and V in 1997 with a dioxin concentration of 11 pg TEQ/g fat resulted in an exposure 8.17

pg TEQ/kg bw day for children and 1.87 pg TEQ/kg bw day for adults. It should be noted that these results are not related to a regularly lactating milk-cow and had only implications for a few families. Additionally, these dioxin concentrations in cows' milk are much higher than one should expect based on measured depositions and model calculations using transfer factors. The level of dioxins in the human milk sample of a nursing mother was 22.7 pg TEQ/g fat which is signi®cantly lower than the previously reported results of PCDD/PCDFs in breast milk of Flemish mothers (27.3± 43.2 pg TEQ/g fat) (Van Cleuvenbergen et al., 1994; WHO, 1997). Assuming an intake of 800 ml milk with 3% fat, this would correspond with an average daily intake of 109.0 pg TEQ/kg body weight for a baby weighing 5 kg. This corresponds with the mean levels in industrialized countries (F urst et al., 1992). However, it should not be an argument against breast-feeding since the PCDD/PCDF body burden declines within a few years to the same background levels reached in children

920

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

of the same age who had not been nursed. Additionally, the bene®ts of breast-feeding outweigh the risks associated with increased PCDD/PCDF exposure during nursing (Kreuzer et al., 1997). Tables 5 and 6 represent the daily intake of PCDDs/ PCDFs in the Neerlandquarter via the major pathways and their relative contribution to the exposure scenarios II and V for the reference years 1980 and 1997. Observation of the highly exposed scenario (1980: V and 1997: V ) in Tables 5 and 6 reveals that intake via food represents the most important dioxin exposure pathway. Consumption of milk, meat and vegetables contributes to more than 95%. The relative part of inhalation is less than 2%. Similar ®ndings are valid for the other exposure scenarios in which consumption of food cultivated in the impact area is involved. Although in scenario II inhalation represents one of the major intake routes, this exposure pathway is more or less 100 times smaller than the background exposure. For this most common scenario in the Neerlandquarter, the main route of exposure is not associated with consumption of homegrown vegetables, meat and milk. Consequently, exposure in

this case is only slightly di€erent from that of the general population in Flanders as well as in 1980 as in 1997. It is unlikely that incinerator emissions will have a measurable health impact on the people just residing in the Neerlandquarter. The risk-index for the local population is < 1 in all cases, indicating that there are no risks if the present reference value (TDI ˆ 10 pg TEQ/kg bw day) of the WHO is accepted as the standard. However the WHO intends to lower this value at least by half (1±4 pg TEQ/kg bw, with the recommendation that every e€ort should be made to reduce exposure to the lower end of this range) (van Leeuwen and Younes, 1998; WHO, 1999). In the latter case there might be a serious indication of concern in the highly exposed scenarios (I and V for 1980, and to a lesser extend IV and V for 1997). In order to ®nd the basis for the evidence of increased health e€ects in progeny of local people exposed to incinerator emissions, an in-depth medical investigation was carried out. Thirteen cases of children with congenital malformations were reported for a population of

Table 5 Daily intake of PCDDs/PCDFs in pg TEQ/kg day and relative contribution from major sources in 1980 to the exposure scenarios II and Va Route of exposure

Child scenario IIb

Adult scenario IIb

Child scenario Vc

Adult scenario Vc

Inhalation Dermal contact soil Ingestion soil Ingestion vegetables Ingestion milk Ingestion Meat

6.32 E-2 (43.3) 1.47 E-2 (10.1) 6.81 E-2 (46.6) 0 (0.0) 0 (0.0) 0 (0.0)

3.56 E-2 (70,1) 7.30 E-2 (14.4) 7.87 E-2 (15.5) 0 (0.0) 0 (0,0) 0 (0.0)

6.32 E-2 (0.6) 1.47 E-2 (0,1) 6.81 E-2 (0.6) 8.95 E-1 (8.0) 7.26 (64.4) 3.0 (26.3)

3.56 E-2 (1.3) 7.30 E-2 (0.3) 7.87 E-2 (0,3) 0.4 (16.4) 0.9 (34.6) 1.3 (47.1)

a

The relative contribution (%) of each exposure route is indicated in parentheses. Scenario II: living with consumption of only commercially sold foods (milk, meat and vegetables). c Scenario V: living with consumption of 25% vegetables, 50% meat and 100% milk produced in the impact area. b

Table 6 Daily intake of PCDDs/PCDFs in pg TEQ/kg day and relative contribution from major sources in 1997 to the exposure scenarios II and Va

a

Route of exposure

Child scenario IIb

Adult scenario IIb

Child scenario V ;c

Adult scenario V; c

Inhalation Dermal contact soil Ingestion soil Ingestion vegetables Ingestion milk Ingestion Meat

1.15 E-2 (30.0) 1.47 E-2 (33.7) 6.81 E-2 (36.3) 0 (0.0) 0 (0.0) 0 (0.0)

6.51 E-3 (72.2) 7.30 E-3 (15.6) 7.87 E-3 (12.2) 0 (0.0) 0 (0.0) 0 (0.0)

1.15 1.47 6.81 0.59 1.50 0.54

6.51 7.30 7.87 0.29 0.19 0.23

E-2 (0.4) E-2 (0.5) E-2 (2.5) (21.9) (54.9) (19.8)

E-3 (0.9) E-3 (1.0) E-3 (1.1) (39.8) (26.0) (31.3)

The relative contribution (%) of each exposure route is indicated in parentheses. Scenario II: living with consumption of only commercially sold foods (milk, meat and vegetables). c Scenario V: living with consumption of 25% vegetables, 50% meat and 100% milk produced in the impact area. * Exposure calculated using cows' milk containing 1.5 pg TEQ/g fat. b

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

2500 residents. Five cases were classi®ed as genetic diseases. Six were considered as multifactorial diseases (for 2 cases no classi®cation could be given). In addition 5 cases of medical abortions were reported. Two cases were classi®ed as due to genetic malformations and two cases were attributed to a multifactorial origin (one unknown). The relative high incidence of Spina bi®da/ anencephaly (normal incidence: 1/1000, Neerlandquarter 4/2500), malformations characterized by a chromosomal marker (normally 10% of the malformations, Neerlandquarter 17%), metabolic malformations (17%) and the incidence in the small population of the Neerlandquarter of rare serious malformations such as gastroschisis (normal incidence: 2/10,000) hydrocephaly (normal incidence 5/10,000), incontinentia pigmenti (very rare), encephalocraniocutaneous lipmatosis (very rare) caused further concern. Heritable diseases or infections could not explain the origin of the reported diseases, nor could the exposure to thalidomide or X-rays. The results of the cytogenetic study did not show a signi®cant (P < 0:05) di€erence between the data obtained from children of the Neerlandquarter and children from the control area. The children of the Neerlandquarter showed less chromatid type aberrations (one chromosome arm involved) than those of the control area (0.5% versus 0.7%) while more chromosome type aberrations (both chromosome arms involved) occurred (0.5% versus 0.25%). These di€erences were not statistically signi®cant. There is currently little evidence that dioxin damages DNA and so causes mutations. Therefore the cytogenetic analysis should primarily be seen as a human biomonitoring study aimed at investigating the presence of genotoxins in the environment as a result of discharges from the municipal waste incinerator, other than dioxins, or from any other source (e.g., high density trac in the neighborhood). These genotoxins could eventually contribute to some of the reported diseases. Comparison with the control data available in the literature showed that the results were in line with those reported by other authors. Similarly, the G-banding technique showed no di€erences between the exposed and control population. In one subject from the Neerlandquarter a heritable extranumerary marker chromosome was found but subsequent genetic analysis of this child and his family showed that this was a heritable trait inherited from his grandparents, and hence the incinerator or other source of pollution was not incriminated. In conclusion there is no evidence that children living in the Neerland area over the last ®ve years received high doses of genotoxicants in their body which is in agreement with the risk assessment based on environmental monitoring of dioxin concentrations and modeling using VLIER-HUMAAN and transfer factors. The latter in-

921

dicated that environmental exposure of inhabitants, especially children consuming locally produced foods, to dioxins in the area is enhanced. The results of these medical investigations do not prove that both waste incinerators are or were a signi®cant public health concern. These medical investigations have been the onset to a more profound Health Research Program 'Environment and Health' throughout Flanders, which has been started recently.

8. Conclusions Two municipal waste incinerators in the vicinity of a residential area close to the city of Antwerp caused concern to local habitants. The observed soil contamination pattern did not correspond to the calculated deposition pattern, indicating that other sources may contribute at least partly to the local PCDD/PCDF contamination of the area. Dioxin exposure of people in the Neerlandquarter was calculated using a mathematical model (VLIERHUMAAN) combined with transfer factors. According to the results of these calculations, just residing in the impact area does not result in a meaningful risk. Only if locally produced food was consumed (milk, meat and vegetables), exposure in the Neerlandquarter was enhanced compared to the average dioxin exposure estimated for the Flemish population. Exposure in 1997 was below the exposure in 1980. As a consequence of different eating habits and the lower bodyweight, children are subjected to signi®cantly higher exposure than adults. Adverse health outcomes from dioxin exposure in the past cannot be excluded. Excessive consumption of locally produced food should be avoided. There was no evidence for enhanced exposure to genotoxicants based on comparison of chromosomal damage to blood cells of children from the study area to those from a control group.

References Abbott, J., Baker, S., Coleman, P., Dyke, P., Fernandes, A., Watterson, J., 1997. A study of Dioxins and Furans in Soil around Four Municipal Waste Incinerators in Hampshire. National Environmental Technology Centre, Abingdon OX143DB Culham. Ahlborg, U.G., Brouwer, A., Fingerhut, M.A., Jacobsen, J.L., Jacobsen, S.W., Kettrup, A.A.F., Koeman, J.H., Poiger, H., Rappe, C., Safe, S.H., Seegal, R.F., Tuomisto, J., van den Berg, M., 1992. Impact of polychlorinated dibenzo-pdioxins, dibenzofurans and biphenyls on human and environmental health, with special emphasis on application of the toxic equivalency factor concept. Europ. J. Pharmacol.environ. Toxicol. Pharmacol. Sect. 228, 179±199.

922

J. Nouwen et al. / Chemosphere 43 (2001) 909±923

Cornelis, C., De Fre, R., Nouwen, J., Schoeters, G., 1998.Voorstel van normen voor dioxines in lucht en depositie. VITOReport 1998/DIA/R20, Mol (in Dutch). Cremers, P., Van Laerhoven, B., Willems, R., 1997. Moorddadig milieu in Vlaanderen. Moeder, waarop leven wij? Standaard Uitgeverij nv Belgie ISBN 90-02-20641-0 Antwerpen. De Fre, R., Wevers, M., Van Cleuvenbergen, R., Schoeters, J., 1994. Measurement of deposition of dioxins in Flanders, Belgium. Organohalogen Compounds 20, 9±14. F urst, P., Beck, H., Theelen, R., 1992. Assessment of Human intake of PCDDs and PCDFs from di€erent environmental sources. Toxic Substances J. 12, 133±150. Fries, G.F., 1995. A review of the signi®cance of animal food products as potential pathways of human exposures to dioxins. J. Anim. Sci. 73, 1639±1650. Hakk, H., Larsen, G., Feil, A., 1997. Excretion, tissue distribution, and metabolism of 1,2,7,8-tetrachlorodibenzo-p-dioxin in a calf. Organohalogen Compounds 33, 191±194. Hawley, J.K., 1985. Assessment of health risk from exposure to contaminated soil. Risk Anal. 5, 289±302. Hoogenboom, L.A.P., Kramer, H.J., Jansen, E.H.J.M., Mengelers, M.J.B., van Klaveren, J.D., van Dooren, M., Traag, W.A., Kuiper, H.A., K onemann, W.H., 1995. Risk assessment and risk management of dioxins and PCBs in food. RIVM Report number 613330001, Bilthoven. H ulster, A., Marschner, H., 1993. Transfer of PCDD/PCDF from contaminated soils to food and fodder crop plants. Chemosphere 27 (1-3), 439±446. H ulster, A., M uller, J., Marschner, H., 1994. Soil-plant transfer of polychlorinated dibenzo-p-dioxins and dibenzofurans to vegetables of the cucumber family cucurbitaceae. Environ. Sci. Technol. 28 (6), 1110±1115. International Atomic Energy Association, 1986. Biological dosimetry: chromosome aberration analysis for dose assessment. Technical Report Series 260, Vienna. Kerremans, W., Geukens, M., 1996. Meetcampagne dioxines rond ISVAG en IHK, Stad Antwerpen, Departement voor sociale zaken, volksgezondheid en milieu, Milieudienst, Antwerpen (in Dutch). Kreuzer, P.E., Csandy, G.A., Baur, C., Kessler, W., Papke, O., Greim, H., Filser, J.G., 1997. 2,3,7,8-Tetrachlorodibenzo-pdioxin (TCDD) and congeners in infants. A toxicokinetic model of human lifetime body burden by TCDD with special emphasis on its uptake by nutrition. Arch. Toxicol. 71, 383±400. Liem, A., Hoogerbrugge, R., Cuijpers, C., denHartog, R., Hijman, W., Linders, S., Marsman, J., van der Velde, E., Zomer, B., 1997. Trends in dietary exposure to dioxins and pcbs in the netherlands results from a duplicate diet study. Organohalogen Compounds 33, 112±115. Liem, A., Theelen, R., Hoogerbrugge, R., deJong, A., 1993. Gehalten van dioxinen in boerenkool. RIVM rapportnummer 730501048, Bilthoven (in Dutch). Liem, A., Theelen, R., Slob, W., van Wijnen, J., 1991. Dioxinen en planaire PCB's in voeding. Gehalten in voedingsprodukten en inname door de Nederlandse bevolking. RIVM Report number 730501034, Bilthoven (in Dutch). Nessel, C.S., Butler, J.P., Post, G.B., Held, J.L., Gochfeld, M., Gallo, M.A., 1991. Evaluation of the relative contribution

of exposure routes in a health risk assessment of dioxin emissions from a municipal waste incinerator. J. Exposure Anal. Environ. Epidem. 1 (3), 283±307. Perera, F.P., Hemminki, K., Gryzbowska, E., Motykiewicz, G., Michalska, J., Santella, R.M., Young, T.L., Dickey, C., Brandt-Rauf, P., DeVivo, I., Blaner, W., Tsai, W.Y., Chorazy, M., 1992. Molecular and genetic damage in humans, from environmental pollution in Poland. Nature 360, 256±258. Prinz, B., Krause, G., Rademacher, L., 1993. Standards and guidelines for PCDD/PCDF ± an integrated approach with special respect to the control of ambient air pollution. Chemosphere 27, 491±500. Schoeters, G., Cornelis, C., De Fre, R., Hooghe, R., Collier, M., Maes, A., Mensinck, C., Nouwen, J., Patyn, J., Verschaeve, L., Viaene, P., Geuzens, P., 1998. Studie van de gezondheidsaspecten en gezondheidsrisico's ten gevolge van de milieuverontreiniging in de Neerlandwijk te Wilrijk. VITO-Report 1998/TOX/R/009, Mol (in Dutch). Seabright, M., 1971. A rapid banding technique for human chromosomes. Lancet 2, 71±72. Sehmel, G.A., 1980. Particle and gas dry deposition: a review. Atmos. Environ. 14, 983±1011. Slob, W., van Jaarsveld, J., 1993. A chain model for dioxins: from emissions to cow's milk. Chemosphere 27 (1-3), 509± 516. Thome-Kozmiensky, K.J., 1994. Thermische Abfallbehandlung. EF-Verlag f ur Energie und Umwelttechnik, GMBH, Berlin. Travis, C., Arms, A., 1988. Bioconcentration of organics in beef, milk, and vegetation. Environ. Sci. Technol. 22 (3), 271±274. Valberg, P.A., Drivas, P.J., McCarthy, S., Watson, A.Y., 1996. Evaluating the health impacts of incinerator emissions. J. Hazardous Mater. 47, 205±227. Van Cleuvenbergen, R., Schoeters, J., Wevers, M., Fre, R., Rymen, T., 1993a. Soil contamination with PCDDs and PCDFs at some typical locations in Flanders. Organohalogen Compounds 12, 243±246. Van Cleuvenbergen, R., Schoeters, J., Bormans, R., Wevers, M., DeFre, R., Rymen, T., 1993b. Isomer-speci®c determination of PCDDs and PCDFs in Flemish cow's milk. Organohalogen Compounds 13, 27±30. Van Cleuvenbergen, R., Wevers, M., Schoeters, J., DeFre, R., 1994. Dioxins (PCDDS and PCDFs) in human milk from Flanders, Belgium: concentration levels and congener pro®le. Organohalogen Compounds 20, 215± 220. Van Hall Instituut, 1997. VLIER-HUMAAN, version 1997. Van Hall Instituut, Leeuwarden. van Jaarsveld, J.A., 1989. Een Operationeel atmosferisch transportmodel voor Prioritaire Sto€en; speci®catie en aanwijzingen voor gebruik. RIVM-Report number 2328603008, Bilthoven (in Dutch). Jaarsveld, J.A., Schutter, M.A.A., 1992. Modelling the longrange transport and deposition of dioxins; ®rst results for the North Sea and surrounding countries. Organohalogen Compounds 9, 299±303. van Leeuwen, F.X.R., Younes, M., 1998. WHO revises the Tolerable Daily Intake (TDI) for dioxins. Organohalogen Compounds 38, 295±298.

J. Nouwen et al. / Chemosphere 43 (2001) 909±923 Wagrowski, D., Hites, R., 1997. The accumulation of polychlorinated dibenzo-p-dioxins and dibenzofurans in the food chain. Organohalogen Organohalogen Compounds 32, 233±237. World Health Organization, International Agency for Research on Cancer, 1997. IARC Monographs on the Evaluation of

923

Carcinogenic Risks to Humans: Polychlorinated Dibenzopara-dioxins and Polychlorinated Dibenzofurans, vol. 69. ISBN 92 832 1269 X, Lyon. World Health Organization, 1999. Assessment of the health risk of dioxins: re-evaluation of the tolerable daily intake (TDI).WHO Report EHBI 010201, Geneve.