PAGE 3 of 3

5.6.2. Observations Concerning LADD Exposure Estimates

Much of the differences between exposure pathways and scenarios is due to differences in exposure media estimation. Therefore, much of the above discussion is also appropriate for trend analysis of Lifetime Average Daily Dose, LADD, estimates. What will be noted below are unique observations. Unless otherwise noted, these observations pertain to 2,3,7,8-TCDD:

. General:

LADDs over all example compounds ranged from 10-19 mg/kg-day to 10-8 mg/kg-day. The highest exposures were associated with the off-site soil contamination scenario, Scenarios #3. As discussed above, these scenarios had the highest exposure media concentrations for all exposure media. LADDs for Scenario 3 ranged from 10-14 to 10-9 mg/kg-day.

Fish and water ingestion exposures were very similar for Scenarios #1 (background soil concentrations, central exposures), #2 (background soil concentrations, high end exposures), and #6 (effluent discharges, central exposures). Fish ingestion LADDs for these three scenarios were in the 10-10 to 10-12 mg/kg-day range. Water ingestion LADDs ranged from 10-15 to 10-13 mg/kg-day.

However, fish and water ingestion exposures were roughly 3 orders of magnitude lower for the stack emission scenarios, #4 and #5, as compared to these three. Fish and water ingestion LADDs, on the other hand, were about 5 times higher for the off-site scenario, #3, as compared to Scenarios 1, 2, and 6. Inhalation exposures were similar for the stack emission scenarios (#4 and #5) and the on-site soil scenarios (#1 and #2), in the 10-15 mg/kg-day range. LADDs for the soil related exposures - soil ingestion and soil dermal contact - were 3 to 4 orders of magnitude higher for the on-site soil scenarios as compared to the stack emission scenarios.

Soil ingestion for the on-site soil scenarios, #1 and #2, ranged from 10-13 to 10-12 mg/kg-day, and for soil dermal contact the impact was lower with LADDs ranging from 10-14 to 10-13 mg/kg-day. Like other trends, the soil ingestion and soil dermal contact impacts from a nearby site of high soil concentrations, Scenario 3, were the highest at 9*10-10 mg/kg-day for soil ingestion and 7*10-11 mg/kg-day for soil dermal contact.

The stack emission scenario had lower exposures associated with beef and milk ingestion as compared to the on-site scenarios, over an order of magnitude lower. The range for Scenarios #2 and #5 was 10-14 to 10-10 mg/kg-day. The beef and milk ingestion pathways for the off-site scenario, #3, were generally the highest of all example pathways at 10-10 and 10-9 mg/kg-day. For the stack emission scenarios which evaluated 2,3,7,8-TCDD and TEQ exposures, 2,3,7,8-TCDD exposures were about 5% of what TEQ exposures were. This follows from the trends in exposure media concentration estimation as discussed above.

. "Central" versus "High End":

Differences between analogous "central" and "high end" exposures for the on-site soil source demonstration scenarios were near or less than an order of magnitude (inhalation exposure for the central on-site scenario and the inhalation exposure for high end on-site scenario are analogous exposures). This is because the exposure parameters used to distinguish typical and high end exposures, the contact rates, contact fractions, and exposure durations, themselves did not differ significantly, and these were the only distinguishing features for the on-site soil source category.

In the stack emission scenario, placing exposed individuals either 500 or 5000 meters away from the incinerator did significantly impact the results. In this case, the difference was closer to 2 orders of magnitude for all exposures except water and fish exposures, which were not a function of distance from the stack. The order of magnitude difference in distance added about an order of magnitude difference in exposure media concentrations and hence LADD estimates.

The high end scenarios were modeled after a rural farm and did have exposures from home grown beef and milk food products. The central scenarios were modeled after a non-farming rural residence, and did not have beef and milk exposure pathways. Since beef and milk exposure pathways were noted as the highest exposures, along with ingestion of fish (see next bullet), it would be appropriate to conclude that farming families ingesting a portion of their home produced beef and milk are more exposed than non-farming families without these exposures.

. Exposure pathway analyses:

It is inappropriate to compare and rank exposure pathways across all scenarios because the source terms are different. However, relationships between different pathways within each scenario can be discussed. Table 5-6 was constructed by summing the LADDs for all pathways, and then determining the percent contribution by each pathway. Before the summation, LADDs were corrected to account for absorption - all ingestion LADDs assumed 50% absorption and inhalation LADDs assumed 75% (data on bioavailability from animal feeding studies, suggests that the absorption of 2,3,7,8-TCDD is around 50%; 75% for inhalation reflects a general assumption of greater absorption for this pathways; both simple assumptions made only for the purpose of this comparative exercise).

The dermal contact LADD was the only one where absorption was already considered in its estimation: absorbed dose was estimated as 3% of dose contacting the body. Also, this exercise assumes all pathways occur simultaneously, and so on. Table 5-6 was generated only for the 2,3,7,8-TCDD example compound, and the rows are listed generally from the highest to lowest percentage contribution. The following observations are made from that table:. In high end scenarios which assumed exposure to home grown beef, milk, and fish, Scenarios 2, 3, and 5, exposures to these three foods dominated the results. In Scenarios where beef and milk were not considered, but fish was considered, Scenarios 1, 4, and 6, fish exposures dominated. The general dominance of beef, fish, and milk exposures underscores the importance of food chain exposures.

. Milk exposures were lower than beef exposures because of less milk fat ingestion (10.5 g/day milk fat vs. 22 g/day beef fat) and lower concentrations in milk as compared to beef (this was discussed above).

table Table 5-6. Percent contribution of the different exposure pathways within each exposure scenario.* * Assumes exposed individual experiences all relevant pathways and exposures are additive; see text for further explanation.
expand table Table V3 5-6

. Fish was the principal impacted media for the effluent discharge source category, with fish ingestion 19 times higher than water ingestion, the only two pathways considered for the effluent discharge category. Fish was an important route of exposure in Scenarios #1 and #2 evaluating basin-wide low soil concentrations. It explains over half of all exposures when beef and milk are not considered and still dominate when they are. It is also important for the stack emission central scenario, #4.

However, fish is much less important than beef or milk for the high end stack emission scenario which had a beef and a milk pathway, and when a small site of contamination is near a site of exposure (which would be a farm raising a portion of the farming families beef and milk ingestions). The dilution effect of a small site in a basin with regard to surface water impacts as compared to a much less dilution effect of a small site near an exposure site was discussed in Section 5.6.1 above, under the beef and milk bullet.

. Soil ingestion exposures were also noteworthy, particularly in scenarios that did not consider beef and milk, the central on-site scenario, #1, and the central stack emission scenario, #4. Soil ingestion was also the second highest pathway in the scenario evaluating the impact of nearby soil contamination, #3, ranking higher than milk or fish ingestion. Dermal exposures were non-trivial, but ranked behind the four ingestion pathways previously discussed, beef, milk, fish, and soil.

. Inhalation was the highest impact for the stack emission scenario when farm animal products were not considered, in Scenario #4. Fruit and vegetable exposures were noteworthy only in this same scenario. These trends imply that, where farm animal products are not being produced near a stack emission source, fish and vegetative food products still may dominate the overall exposure, but inhalation exposure can become critical.

. Water ingestion exposures were very low in comparison to the other exposures in these scenarios.

The LADD estimates of all example scenarios were derived assuming a limited duration of exposure to the dioxin-like compounds, and also limited contact with exposure media. A pattern of childhood soil ingestion was assumed to occur over a five-year period. The central scenarios assumed a nine-year duration of exposure to the contamination, and the high end scenarios assumed a twenty-year exposure period. The contact with impacted media was only assumed to occur in the home environment - only a portion of an individual's meat, milk, water, and fruit and vegetable ingestion was evaluated.

This is only one approach to scenario development; other approaches might consider the quality of exposure media not associated with the home environment. For example, if the bulk of an individual's ingestion of produce comes from local farms, and local farms may be impacted by an incinerator, then perhaps 90-100% of an individual's fruit and vegetable ingestion, rather than the 20-40% assumed in this assessment, should be considered impacted.

Another issue to consider while interpreting these scenarios is their relation to background exposures. Dioxin-like compounds are commonly found throughout the environment, leading to "background" exposures even in situations where known sources are not present. Since these scenarios are developed around defined sources, they should be interpreted as incremental exposures beyond "background" levels. This interpretation is clearly less satisfying for Scenarios 1 and 2 where soils within a watershed are assumed to be contaminated at 1 ng/kg (ppt) level which, as discussed earlier in Section 5.5, may be close to a background level.

In this sense, Scenarios 1 and 2 are more representative of background exposures than incremental exposures. However, diet fractions were applied indicating that only a portion of the food supply was contaminated. Also, the exposure duration was defined as less than a lifetime. For purposes of a true background exposure analysis, it might be more appropriate to assume 100% for diet fractions and lifetime exposures for example Scenarios 1 and 2.

An exercise was undertaken making the following changes in example Scenarios 1 and 2: 70 years exposure duration was assumed for all pathways except the childhood pattern of soil ingestion (which remained at 5 years), and all contact fractions were set equal to 1.00. All contact rates were unchanged for this exercise. The one contact rate which might be changed is the rate of fish ingestion. For that pathway, the 1.2 (central) and 4.1 (high end) g/day rates were estimated assuming a number of meals and meal sizes that a rural individual would recreationally obtain from a nearby impacted water body.

For background impacts, it may be more appropriate to have an average ingestion rate, and assume that the 0.6 pg/g fish concentration estimated for example Scenarios #1 and #2 is typical of all fish ingested. That was not done for the results of this exercise that are displayed in Table 5-7, but was done separately and will be discussed in the last bullet below. Table 5-7 includes LADDs and the percent contribution from each pathway for the original and the amended example Scenarios #1 and #2, and only for 2,3,7,8-TCDD. Observations from this exercise are:

. All exposures increased except soil ingestion. Mostly the increase in LADD were within an order of magnitude. The LADDs are still about an order of magnitude lower than limited exposures estimated to occur from living near an area of high soil contamination, as can be seen from comparing these lifetime results with those in Table 5-5 for example Scenario #3 (off-site source category).

. The increases to vapor phase inhalation exposures were small relative to other increases. This occurs because average volatilization flux decreases as a function of time - average fluxes over long periods of time are lower than average fluxes over short periods of time (see Section 4.3.2, Chapter 4, for further explanation). This has a rippling effect on fruit and vegetable concentrations as well as beef and milk concentrations. A trend of decreasing volatilization flux is realistic when highly contaminated soils become depleted over time.

This was a key principal of the volatilization flux algorithm. However, this may not be a realistic trend for ubiquitous concentrations which are likely to be replenished in soil over time. Sensitivity analysis in Chapter 6, Section 6.3.3.1 shows that increasing exposure duration from 20 to 70 years decreases average flux by a factor of 2.

table Table 5-7. Exposures to low soil concentrations of 2,3,7,8-TCDD assuming lifetime exposure durations and unlimited contact with impacted media, compared with exposures assuming limited durations and limited contact.
By implication, vapor inhalation, fruit and vegetable ingestion, and beef and milk ingestion, are all underestimated for a lifetime of exposure for example Scenario #2, by around the same factor of 2.

On the other hand, the discussion on sources in Volume II, Chapter 3, does indicate that sources of release have been reduced over time. If so, an assumption of ongoing replenishment to similar levels may not be warranted.

• The relative impact of soil ingestion dropped when assuming lifetime exposures. It is interesting that direct soil impacts, a childhood pattern of soil ingestion and a lifetime of soil dermal contact, still are noteworthy impacts, particularly when not considering beef or milk exposures.

expand table Table V3 5-7

. Fish ingestion exposures were the predominant exposures when beef and milk were not considered and became even more so for a lifetime of exposure in Scenario #1. The same was not true for Scenario #2, where fish dropped in prominence when going from a limited to a lifetime of exposures. The main reason for this trend was that only the increase in exposure duration came into play for Scenario #2 and fish exposures - the rate of fish ingestion was not changed for this exercise. However, both the exposure duration and beef and milk ingestion rates increased for the lifetime exercise for Scenario #2.

. Different ingestion rates of fish were evaluated for example Scenario #2 and a lifetime of exposure (2nd column on Table 5-7). The rate of 4.1 g/day used for the results in Table 5-7 was increased to 6.5, 30, and 140 g/day. The 6.5 g/day was used in the Ambient Water Quality Criteria document for 2,3,7,8-TCDD, and was described as an average daily per capita consumption of freshwater and estuarine fish and shellfish (EPA, 1984).

The 30 and 140 g/day were recommended as 50th and 90th ingestion rates in EPA (1989) for recreational fisherman in an area where there is a large water body present and widespread contamination is evident, and where site-specific information is unavailable. Increasing to 6.5, 30, and 140 g/day increased the percent of impact for Scenario #2, 70-year impacts, to 46, 79, and 95%, respectively. This does indicate that, even when beef and milk are considered in a lifetime assessment in background settings, ingestion of fish in such settings can be equally if not more critical.

An evaluation of total human exposure, and the relative impact of different exposure pathways for 2,3,7,8-TCDD, was also undertaken by Travis and Hattemer-Frey (1991). Their approach was based on modeling, using their Fugacity Food Chain model. This model requires as input emission rates into air, soil, and water; these emission rates are then transformed into concentrations. Those concentrations are used to estimate food crop concentrations, beef, milk, and fish concentrations. In the application of this model by Travis and Hattemer-Frey (1991), emission rates were calibrated to arrive at air, soil, and water concentrations that were supported by the literature. Resulting concentrations in these three media were:

0.02 pg/m3 in air partitioned as 0.016 pg/m3 in particulate form and 0.004 pg/m3 in vapor form, 0.96 pg/g in soil, and 0.003 pg/L in water.

Using the model to estimate exposure media concentrations, they then assumed contact rates as given by Yang and Nelson (1986) to estimate human exposure. Assuming 100% absorption, they concluded that typical human exposure is on the order of 35 pg/day 2,3,7,8-TCDD.

Their analysis will be compared to the analysis of this methodology. The example scenario most like their's was example Scenario #2, which estimated exposures to soil levels of 1.0 pg/g. The assumption used in the above sensitivity exercise of 100% contact fractions rather than partial contact is also the appropriate assumption for comparison. A key difference to keep in mind is that air and water impacts are estimated given soil concentrations in this methodology, whereas air and water impacts are specified with their model. This comparison is given in Table 5-8 below.

While the exposure of 6.3 pg/day estimated in Scenario 2 (assuming 100% contact fractions) is within an order of magnitude of the 34.3 pg/day estimated by Travis and Hattemer-Frey, there are a few critical differences in the two approaches. One was in the contact rates. However, replacing the contact rates used by Travis and Hattemer-Frey with the contact rates used in this methodology would not particularly change the total exposure estimate - it would increase to 11.3 pg/day using the concentrations estimated in this methodology.

The second and more important difference is that air concentrations are derived from soil concentrations in this methodology whereas they are input for the Fugacity Food Chain model. Air concentrations are used for inhalation exposures, to estimate fruit and vegetable concentrations, but most importantly, beef and milk concentrations in a food chain model. The air concentrations used by Travis and Hattemer-Frey, 0.02 pg/m3, is nearly three orders of magnitude higher than predicted by the volatilization/dispersion modeling used in this methodology.

table Table 5-8 Comparison of exposure pathway contributions to total daily exposure as estimated in example Scenario #2 and in Travis and Hattemer-Frey (1991).
Section 7.2.3.9 of Chapter 7 examines air concentrations found in urban and rural environments. It is noted that the compilation of air monitoring data conducted in Volume II of this assessment indicated an average 2,3,7,8-TCDD concentration in urban air of 0.01 pg/m3, half as much as the 0.02 pg/m3 used by Travis and Hattemer-Frey.

These authors do note that their air concentration originates from monitoring of urban air in Germany. Section 7.2.3.7, Chapter 7, discusses the fact that had the models of this methodology predicted urban air concentrations given low soil concentrations perhaps typical of rural environments, the model should be questioned.
expand table Table V3 5-8

On the other hand, the comparison of limited rural air concentrations and the air concentrations predicted by a background soil concentration of 1 ppt is not favorable.

In one literature article measuring concentrations in an area described as a "remote countryside" in Sweden (Broman, et al. 1991), air concentrations of 2,3,7,8-TCDD were measured at 0.0002 pg/m3. This remote countryside air concentration is hypothesized to be lower than what is likely to occur in rural environments where at least some sources of dioxin release relatively nearby are expected to occur. The air concentration of 2,3,7,8-TCDD modeled in this assessment from background soil concentrations of 1 ppt is nearly an order of magnitude lower than that at 0.000032 pg/m3.

This suggests that the models of this assessment underestimate air concentrations resulting from releases from soils - however, this is not a definitive conclusion. As discussed in Volume II, the principal source of dioxin release into the environment which affects the food chain and associated soils are emissions from combustion sources. These sources provide a steady and ongoing input into the soil and food chains. The true test of the soil release and dispersion algorithms of this assessment would be to measure air concentrations of dioxin-like compounds released from soil into residue-free air, and compare that with modeled concentrations. Such data could not be found in the literature. Most importantly, the fact that air concentrations estimated to result from background soil concentrations are much lower than air concentrations surmised to occur in a rural environment does suggest that the on-site soil source category is an inadequate framework to be estimating background exposures.

The air concentration of 2,3,7,8-TCDD of 0.02 pg/m3 was run through the food chain algorithms of this assessment using the procedures outlined in Chapter 7, Section 7.2.3.9 of this volume, the air-to-beef food chain validation exercise. It was found that the model predicted a whole beef concentration of 0.32 ppt (19% beef fat) and a whole milk concentration of 0.10 ppt (4% milk fat), which is actually fairly similar to the 0.20 ppt whole beef and 0.03 ppt dairy concentrations as predicted by Travis and Hattemer-Frey. However, this is more by chance than deliberate, as there are substantial differences in the specifics of the modeling approach and parameter values of this assessment and that of Travis and Hattemer-Frey. Following is a summary of such differences in modeling and 2,3,7,8-TCDD parameterization:

1) Travis and Hattemer-Frey assumed a vapor/particle split of 20%/80%, whereas the models of this assessment assumed a 55%/45%. The air-to-leaf transfer factor, defined identically for both modeling efforts, nonetheless came up with a different value. The Bvpa of this assessment was 100,000, whereas for Travis and Hattemer-Frey, the Bvpa was 9883. The reason for this difference was that they assumed different values for 2,3,7,8-TCDD log octanol water partition coefficient, Kow, 6.85 versus the 6.64 of this assessment, and Henry's Constant, H, 3.6*10-3 atm-m3/mole versus the 1.65*10-5 atm-m3/mole of this assessment.

2) They used a "biotransfer" factor for estimating whole beef concentrations as a function of mass ingestion of 2,3,7,8-TCDD, whereas this model used a "bioconcentration" algorithm which only depends on the concentration in the dietary intake and predicts the concentration in beef fat.

3) The key factors in the Travis and Hattemer-Frey approach are the mass ingestion rates of dry matter in the diets of beef and dairy cattle, whereas the important factors for this approach are the proportions in the dry matter categories. Travis and Hattemer-Frey made different assumptions regarding proportions in vegetative and soil as compared to this assessment. For the beef cattle, they assumed that 35% of their ingestion was in unprotected forages, while 64% was in protected grains and 1% in soil. This assessment assumed equal proportions in pasture grass - the analogy to unprotected forage - and 48% in partially protected vegetations (hay, silage, grains), and 4% in soils. This assessment also assumed a 50% reduction in beef concentrations due to feedlot fattening. While not explicitly stating it, Travis and Hattemer-Frey also appeared to be modeling the feedlot fattening diet, as the majority of beef cattle diet was in protected grains, and their article did describe the diet as representing, "beef cattle destined for slaughter".

Although both modeling approaches appear to perform quite similarly, this simple comparison shows how important the assignment of parameters can become in estimating exposures. If any of the assumptions and/or parameters used by Travis and Hattemer-Frey used were instead used for this assessment (and vica versa), results would be substantially different. Chapter 7 on Uncertainty critically evaluates the modeling approaches and parameters used in this assessment. Chapter 6 contains additional information pertaining to these models, including sensitivity analysis exercises, discussions of model parameters, and discussions on other modeling approaches. Information in both these Chapters should be reviewed when evaluating the validity of the approaches demonstrated in this Chapter.

REFERENCES FOR CHAPTER 5
  • Bacci, E.; Cerejeira, D.; Gaggi, C.; Chemello, G.; Calamari, D.; Vighi, M.; 1992. Chlorinated dioxins: volatilization from soils and bioconcentration in plant leaves. Bull. Environ. Contam. Toxic., 48:401-408.
  • Bacci, E.; Calamari, D.; Gaggi, C.; Vighi, M. (1990) Bioconcentration of organic chemical vapors in plant leaves: experimental measurements and correlation. Env. Sci. Technol. 24:885-889.
  • Broman, D., Naf, C.; Zebuhr, Y. (1991) Long-term high- and low-volume air sampling of polychlorinated dibenzo-p-dioxins and dibenzofurans and polycyclic aromatic hydrocarbons along a transect from urban to remote areas on the Swedish Baltic Coast. Env. Sci. Technol 25:1841-1850.
  • Broman, D.; Naf, C.; Rolff, C.; Zebuhr, Y. (1990) Analysis of Polychlorinated Dibenzo-P-Dioxins (PCDD) and Polychlorinated Dibenzofurans (PCDF) in Soil and Digested Sewage Sludge from Stockholm, Sweden. Chemosphere (21): 1213-1220.
  • Creaser, C.S.; Fernandes, A.R.; Al-Haddad, A; Harrad, S.J.; Homer, R.B; Skett; P.W; Cox, E.A. (1989) Survey of Background Levels of PCDDs and PCDFs in UK soils. Chemosphere 18: 767-776.
  • Gillespie, W.J. (1992) Summary of data reflective of pulp and paper industry progress in reducing the TCDD/TCDF content of effluents, pulps and waterwater treatment sludges. Unpublished report available from the National Council of the Paper Industry for Air and Waste Stream Improvement, Inc., 260 Madison Ave, New York, NY 10016. August 17, 1992.
  • Nestrick, T.J.; Lamparski, L.L.; Frawley, N.N.; Hummel, R.A.; Kocher, C.W.; Mahle, N.H.; McCoy, J.W.; Miller, D.L.; Peters, T.L.; Pillepic, J.L.; Smit, W.E.; Tobey, S.W. (1986) Perspectives of a Large Scale Environmental Survey for Chlorinated Dioxins: Overview and Soil Data. Chemosphere 15: 1453-1460.
  • Reed, L.W.; Hunt, G.T.; Maisel, B.E.; Hoyt, M.; Keefe, D.; Hacknew, P. (1990) Baseline Assessment of PCDDs/PCDFs in the Vicinity of the Elk River, Minnesota Generating Station. Chemosphere 21: 159-171.
  • Stenhouse, I.A.; Badsha, K.S. (1990) PCB, PCDD, and PCDF Concentrations in Soils from the Kirk Sandall/Edenthorpe/Barnby Dun Area. Chemosphere 21: 563-573.
  • Travis, C.C., and H.A. Hattemer-Frey. (1991) Human exposure to dioxin. The Science of the Total Environment 104: 97-127.
  • U.S. Environmental Protection Agency. (1984) Ambient water quality criteria document for 2,3,7,8-tetrachlorodibenzo-p-dioxin. Office of Water Regulations and Standards, Washington, D.C. EPA-440/5-84-007.
  • U.S. Environmental Protection Agency. (1985) Soil Screening Survey at Four Midwestern Sites. Region V. Environmental Services Division, Eastern District Office, Westlake, Ohio, EPA-905/4-805-005, June 1985.
  • U.S. Environmental Protection Agency. (1987) National Dioxin Study. Office of Solid Waste and Emergency Response. EPA/530-SW-87-025. August 1987.
  • U.S. Environmental Protection Agency. (1989) Exposure Factors Handbook. Exposure Assessment Group. Office of Health and Environmental Assessment, Office of Research and Development, Washington, D.C. EPA/600/8-89/043.
  • U.S. Environmental Protection Agency. (1990) USEPA/Paper Industry Cooperative Dioxin Study "The 104 Mill Study" Summary Report and USEPA/Paper Industry Cooperative Dioxin Study "The 104 Mill Study" Statistical Findings and Analyses. Office of Water Regulations and Standards, July 13, 1990.
  • Yang, Y.Y., and C.R. Nelson. (1986) An estimation of daily food usage factors for assessing radionuclide intake the U.S. population. Health Phys., 50: 245-257.