Skip directly to search Skip directly to A to Z list Skip directly to site content





On-site contamination of air, water, and soil has resulted and could still result from MSPoperations. Stack emissions, fugitive organic vapor emissions from waste handling andprocessing, and blowing soils and dusts are primary examples of potential on-site aircontamination. Rainwater percolating through on-site soils, and present and past storage ofwastes or produced primary and slagged aggregate, could affect the groundwater underlying thesite. Similarly, site soils may be contaminated from past waste storage practices or spills, leaks,or seepage. Rainfall that runs off site may also be contaminated while on site. With theexception of undisturbed, in-situ soils, all other identified contaminated media can be expected tomigrate off site.

During MSP's operation, EPA, MSP, and LDEQ have on numerous occasions sampled themedia identified previously. Throughout this period, there have been process and facilitychanges that would be expected to affect environmental sampling results. Each medium will beaddressed individually to ease understanding of potentially affected exposure pathways. Numerical data presented in both on-site and off-site contamination tables contain, unlessotherwise noted, the maximum contaminant level values found in a search of all data inATSDR's possession. Many of those data points are below levels of health concern; however,they are included to allow the reader to see the full range of values considered by the Agency. Ifthe reader is aware of additional data that represent even higher contamination levels, those datashould be brought to the attention of ATSDR for further consideration.

AIR: The most obvious and commonly recognized potential for contamination of on-site air isthe discharge of process exhaust gases and particulate through the smokestack (R-14, R-44, R-45, R-56, R-73). Since 1985, the degree of contamination of this effluent stream has beencharacterized by several stack tests conducted by facility contractors; some of which weremonitored by EPA. Criteria pollutants such as nitrogen oxides, sulfur dioxide, carbon monoxide,and particulates have been analyzed for. Also, destruction and removal efficiency (DRE) -- ameasure of hazardous waste destruction performance -- of selected species of hazardous materialfed to the process kiln has been analyzed for. Other potential contaminants of concern analyzedfor include arsenic, barium, cadmium, lead, chromium, mercury, hydrogen chloride, dioxins, andpolycyclic aromatic hydrocarbons (PAHs).

Table 1 compares stack emissions of the criteria pollutants in 1985, 1988, and 1990. Table 2compares stack emissions of selected metals in 1985, 1988, and 1990. The significance and fateof these pollutants are discussed in the Environmental Pathways section of this public healthassessment.

Table 1.

Marine Shale Processors, Inc. Stack Emissions of Criteria Pollutants (lbs/hr)

1985 19881990**
Avg.Max.Test 1Test 2Avg.Max.
Nitrogen oxides19.4419.9427.1 37.766.368.0
Sulfur dioxide53.0584.7128.82.5<7.06.6
Carbon monoxide<100 ppm*
*ppm = parts per million as reported
** MSP Compliance Test for Certification with Boiler and Industrial FurnaceRequirements, June 20-23, 1990.

Reference: D-100

Table 2.

Marine Shale Processors, Inc. Stack Emissions of Select Metals (lbs/hr)

Avg.Max.Test 1Test 2Max.*
Arsenic<0.00020NR 0.004<0.0007720.006
* MSP Compliance Test for Certification with Boiler and Industrial Furnace Requirements, June 20-23, 1990.
             NR = Not Reported
      Note: < means that the metal analyzed for was below the detection levelshown.

Reference: D-100

In 1988, MSP conducted stack tests for the DRE of three hazardous organic feed constituents:trichloroethylene, dichlorobenzene, and carbon tetrachloride. The compounds selected for DREtesting are usually representative of feed materials that are most toxic, most difficult to burn,and/or most prevalent in the expected feed mix. If the MSP process kiln were regulated underhazardous waste incineration standards, it would have to meet a DRE of 99.99%, which meansthat less than 0.01% of the hazardous constituent in the feed mix would be allowed in theeffluent stack gases. That regulatory approach to demonstrating adequate hazardous wastedestruction and removal is based upon the assumption, and some observation, that by carefullyselecting worst-case or most challenging chemicals to test for DRE, other less challengingcompounds will also be successfully processed whenever 99.99% DRE is achieved. However,that approach does not necessarily ensure that all other non-feed products of incompletecombustion (PICs) in stack emissions will be insignificant. PICs could include partialbreakdown products of the parent waste feed compounds, as well as recombinant new speciesthat may be as or more toxic than the original waste constituents. See the discussion on dioxinsfor further information.

MSP demonstrated greater than 99.99% DRE during test runs for trichloroethylene anddichlorobenzene, but not carbon tetrachloride. The relatively low levels of carbon tetrachloridein the feed mix and the limits on the level of detection of carbon tetrachloride in the stackemissions could have prevented accurate DRE determination. Difficulty in calculating DREs forlow-level feed constituents has been described in the literature on hazardous waste incineratortesting (R-30, R-31).

A similar DRE calculation dilemma was reported by MSP in 1985. DREs were analyzed for fourPAHs fed at levels ranging from 85 lbs to 250 lbs per hour. DREs for all four PAHs were greaterthan 99.99%. Three other compounds, benzene, methylene chloride, and toluene, were fed atrates ranging from 0.049 lbs to 0.434 lbs per hour. None of the DREs for those compounds metthe 99.99% criterion. In June 1990, DREs were analyzed for monochlorobenzene and carbontetrachloride. All runs of all tests met and exceeded the 99.99% criterion.

When halogenated organic compounds are burned in the MSP kiln, it is expected that hydrogenchloride (HCl), an acid gas, will be formed. In 1985, MSP determined that emission levels ofHCl were less than 2.5 lbs per hour. In 1988, emission levels of HCl ranged from 51.6 lbs to416.3 lbs per hour. That increase was believed to be caused by operating the caustic feed to thescrubber at less than the theoretical amount needed to completely neutralize the acid gases in theexhaust. In April 1990, HCl emissions ranged from 21.6 lbs to 276.0 lbs per hour. In June 1990tests, HCl emissions ranged from 12.6 lbs to 46.5 lbs per hour.

When burning halogenated organic compounds, it can also be expected that various species ofdioxins and furans may be produced as products of combustion. The MSP facility has beenevaluated for emissions of that family of compounds. The most toxic of those species is2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD). To account for the toxicity of other species ofdioxins and furans, the other species are expressed in what is known as toxicity equivalents ofTCDD. In this manner, the total toxicity equivalent expressed as TCDD can be found. For MSP,mass emission rates per unit of time were given for the various measured species of dioxins andfurans. For public health assessment purposes, it is more useful to review concentration data ofpollutant species; consequently, an average effluent rate had to be assumed to determine stackconcentrations of the various species of dioxins and furans. Following discussions with MSPpersonnel, an exhaust rate of 62,000 dry standard cubic feet per minute was used to determine thestack concentrations. The total equivalent TCDD release was calculated to range from 0.00216micrograms per cubic meter (µg/m3) to 0.00310 µg/m3.

PAHs are a class of organic compounds that have been considered as PICs. During the stacktests conducted in 1985, four PAHs -- fluoranthene, naphthalene, phenanthrene, and pyrene --were fed in sufficient quantity in the feed materials to allow a DRE to be calculated for theeffluent gases. The calculated DRE for all four compounds exceeded 99.99%.

The discussion of air emissions associated with MSP plant activities presented in the previousparagraphs focuses on measured stack emissions. Although useful, that discussion is notcomplete for the following reasons:

    o      Stack tests are usually of relatively short duration, which may or may not be representative of long-term average conditions.

    o      Stack tests are not designed to reflect on-site "fugitive air emissions," such as releases of volatile organic compounds (VOCs) produced when blending or shredding feed materials orsimple process leaks of combustion gases.

One way to assess plant emissions, including fugitive emissions, is to monitor ambient air atvarious locations throughout the facility. The results of such monitoring could reflect stackdownwash emissions (under certain atmospheric conditions), emissions of VOCs associated withhandling, blending, or storage of hazardous materials, and background levels of pollutantsmigrating onto MSP property.

Some ambient air monitoring has been conducted at MSP. One source of ambient air qualitydata is MSP's industrial hygiene data used to monitor the quality of workers' breathing air atvarious facility locations. Another source of data is the four-station ambient air monitoringnetwork maintained by MSP at the site perimeter. In addition, EPA has conducted on-site andoff-site ambient air monitoring. Data reviewed by ATSDR from those sources was generatedbetween 1987 and the first quarter of 1990.

Parameters monitored in the ambient air include dichloromethane, chloroform,1,1,1-trichloroethane, trichloroethylene, benzene, tetrachloroethylene, 1,2-dichloroethane, carbontetrachloride, toluene, and trichlorofluoromethane. Much of the data reported by MSP isexpressed in terms threshold limit values (TLVs). TLVs are used to measure occupationalexposures to airborne concentrations of substances. They represent the conditions under which itis believed that nearly all workers may be repeatedly exposed to a substance without sustainingadverse effects. Some of the data are expressed as actual concentrations of pollutant per unitvolume of air (e.g., µg/m3).

In general, the overwhelming majority of data reviewed suggests that levels of organic vaporsemitted on site, at least for the monitored substances(1), are less than levels of concern foroccupational or off-site public health exposure. For example, of the nearly 2,000 data points forall of 1989 workplace air data (MSP), only one data point exceeded the TLV for dichloromethanenear stockpiled product; a few approached the TLV for carbon tetrachloride near the drumprocess area. Most of the on-site samples, including EPA's sampling, are less than TLV/42 orTLV/100, which are values sometimes used by state or local air pollution agencies to assesspublic exposures to such compounds (R-66). The divisor of 42 or 100 is used to adjust forresidential exposure, which may be continuous, versus occupational exposure, which is limited toabout 40 hours per week. No other standards exist. It should be noted that this discussionfocuses on on-site levels, and it is expected that dispersion would further reduce off-siteconcentration levels, if there are no other off-site contributions of the same species of pollutant.

There have been citizen allegations of heavy smoke releases from the MSP operation,particularly during first couple of years of operations at the facility. It is difficult to assess theimpact of those reports because they were not quantified or characterized. It is assumed,however, that the current ambient air network monitoring results reflect current conditions,including any combustion problems that could result in heavy smoke releases at times when thesmoke would circulate to the ground near the stack.

Water: On-site water contamination could exist in surface waters, such as stormwater runoff thatcontacts stored product, feed materials, or potentially contaminated site soils. Rainwaterpercolating through product, feed, or surface soils downward into underlying groundwater also could be a source of on-site watercontamination. LDEQ inspection reports cite a number of on-site material storage conditionsthat could lead to such water and soil contamination. To assess on-site water contamination,environmental regulatory agencies and MSP have undertaken several sampling and monitoringactivities.

The results of surface water sampling by the state in 1986 showed no lead, phenols, cyanides,chlorinated hydrocarbons (including pesticides), polychlorinated biphenyls, phthalates, ornitroaromatics. Zinc and copper were detected at less than part per million (ppm) levels. Nodetails were provided about exact sample conditions or locations.

In July 1989, EPA conducted substantial environmental tests both on and off site. In addition toair, soil, and sediment, EPA tested on-site monitoring and test wells for groundwatercontamination. Barium ranged from below detection levels to a high of 1,830 parts per billion(ppb). Arsenic ranged from none detected to a high of 987 ppb (D-48a). A partial list of organicspecies detected are shown in Table 3.

Table 3.

On-site Groundwater Contamination (Partial List)
CompoundMaximum Detected
Polynuclear Aromatic Hydrocarbons26 ppb
Pentachlorophenol144 ppb
Trichlorophenol47 ppb
4-Methylphenol2210 ppb
Di-n-butylphthalate6 ppb
Reference: D-28

In 1987, MSP monitored on-site groundwater and found naphthalene, phenol, and phenoliccompounds in the vicinity of the well designated MW-4 (D-115). In response to that finding, aRemedial Action Plan (RAP) was begun to address the contamination and to assess the physicalextent of the contaminated groundwater plume. Several additional on-site monitoring wells wereadded in 1988. Data collected by MSP in 1988 and 1989 suggest that the contaminated plumeremains, although the phenol and phenolic compounds were reduced by about 30% in a one-yearperiod (D-127). In other on-site monitoring wells (Resource Conservation and Recovery Act(RCRA) and non-RAP wells) no hazardous substance list (HSL) compounds were found, exceptfor methyl chloride and bis(2-ethylhexyl)phthalate, both of which may be artifacts (both arecommonly associated with plastic pipe well construction). Metals, including lead, chromium,and cadmium were found in RCRA monitoring wells in below-parts-per-million range, in 1988(D-107a). EPA findings for these metals in 1986 groundwater monitoring well data were insome cases somewhat higher than the MSP sponsored data, however, the maximum values werestill under one part per million for all three metals mentioned here (D-48a).

SOILS, SEDIMENTS, AND AGGREGATE: On-site soils and sediments can be contaminatedwhen feed materials or process residuals are stored in a way that allows rain water to leachthrough solid materials into the underlying soils. Liquid feed materials or spills or leaks canpercolate directly into the underlying soils.

In 1989, EPA sampled on-site soils, sediment, and aggregate (D-48a). In on-site surface fillareas, lead ranged from not-detectable up to 454 ppm (TCLP). Cadmium in those areas was ashigh as 4.8 ppm (TCLP). Total lead and cadmium soil concentrations were up to an order ofmagnitude higher than the leachable (TCLP) values. Some fill areas showed no or virtually novolatile organic or semi-volatile chemical contamination. Other fill areas were contaminatedwith PAHs in the 2 - 10 ppm range. Dibenzofuran in the fill areas ranged from 321 ppb to 755ppb. Piles of aggregate sampled on site contained total lead ranging from not detectable to 5590ppm and total cadmium ranging from not detectable to 41.8 ppm. The piles showed virtually novolatile or semi-volatile organic chemical contamination.

The 1987 U.S. Coast Guard on-site soil study did not show lead or cadmium in soils at levelsexceeding EPA leaching (Extraction Procedure Toxicity) criteria. Those criteria estimate theamounts of metals released from the soil and available for absorption if exposure occurs. Themaximum lead and cadmium levels found in that study of on-site soils were 8,363 ppm and51,442 ppm, respectively. Leachable levels were much lower.

Aggregates stored on site include primary kiln aggregates, which are solids discharged from therotary kiln with only the shredded metal drums removed, and slagged aggregates, which are solidmaterials resulting from further heating and slagging primary aggregate and solids captured bythe air pollution control system (e.g., baghouse dust). Primary aggregate that exceeds 80% of theallowable TCLP criteria for any metal is treated in the slagging kiln. MSP reports that slaggedaggregate must pass TCLP criteria to be sold. For this public health assessment, ATSDRreviewed "Comprehensive Analytical Results for MSP Aggregates -- 1989," an MSP report onanalytical results of primary aggregate, slagged aggregate, and baghouse dust. For lead, aprimary pollutant of concern, the maximum reported contamination levels in baghouse dust were29,796 ppm total lead versus 438.6 ppm leachable lead (TCLP). For primary aggregate, themaximum reported leachable lead level was 131.21 ppm (TCLP). For slagged aggregate, themaximum reported leachable lead was 16.21 ppm (TCLP); that was the only data point forslagged aggregate that exceeded the TCLP criterion of 5.0 ppm for lead. Ninety-two percent ofthe data points reviewed by ATSDR for slagged aggregate showed leachable lead under 0.05ppm. A few data points were between 0.5 ppm and 5.0 ppm for leachable lead. It should benoted that those are data reported by MSP to be used for internal quality assurance to determinewhich aggregate can be sold and which must be further processed.


As mentioned in the preceding section, with the exception of in situ soils, any contaminatedmedia originating on site at the MSP facility could migrate off site. These contaminants couldinclude airborne particulates and gases and waterborne substances in groundwater or surfacewater. Also, process residue or aggregate destined for off-site use could be a source ofcontamination if the contaminants are in a biologically or environmentally available form. Thissection reviews the available characterization data regarding off-site contamination. As in thepreceding section, data are reviewed for extreme values. A notable difficulty in evaluatingoff-site data is the confounding nature of other pollution sources in the area, such as automobileemissions, other industrial activities, and old waste disposal sites. A review of the 1987 and1988 Toxic Release Inventory (TRI) confirmed that there are other industrial chemical releases inthe area, which are considered in this public health assessment.

Air: Off-site ambient air monitoring data are rather limited and are particularly difficult to linkto MSP because of other potential area sources of emissions and the lack of clear identification of"signature compounds" that would be expected to be emitted only from MSP. The most recentlarge-scale study of off-site ambient air quality available for ATSDR review is the EPAmonitoring conducted in August 1989. For that study, seven sampling sites were set up at theMSP facility; four more stations were located off site toward and in Amelia. A number oforganic species, including 1,1,1-trichloroethane, xylenes, 2-butanone, vinyl acetate, and2-hexanone were found at higher concentrations at the MSP site than at the off-site locations. Concentrations of those compounds at MSP were typically under 100 µg/m3 (except one pointfor 2-hexanone at 211 µg/m3) which is less than TLV/100 for the compounds reviewed. Cadmium and lead were also measured; there were virtually no differences in lead and cadmiumvalues found at MSP and off site. Lead ranged in concentration from 0.17 µg/m3 to 0.20 µg/m3on site and from 0.18 µg/m3 to 0.19 µg/m3 off site. Cadmium ranged in concentration from0.0013 µg/m3 to 0.0090 µg/m3 on site and 0.0077 µg/m3 to 0.0084 µg/m3 off site. Meteorologicdata were not available for ATSDR review; consequently, upwind/downwind comparisons couldnot be made.

Water: Off-site water contamination could originate from contaminated surface water runofffrom a site, movement of contaminated groundwater off site, spills or leaks to adjacent sites, anddischarges to adjacent receiving waters. All of those mechanisms for off-site watercontamination have been reported or alleged regarding MSP site activities. LDEQ site inspectionreports and EPA studies reviewed site conditions and on-site and off-site waters to evaluate thepotential for off-site water contamination. The other potential for off-site water contaminationassociated with MSP activity is the leachability of contamination (lead and cadmium) from theaggregate.

In July 1989, samples were taken from water in the adjacent Bayou Boeuf and from sitedischarge outfall points as part of the EPA RCRA inspection. No lead or cadmium was detectedin the bayou water samples. Outfall number 3 showed no lead or cadmium and 89 ppb barium. Outfall number 4 showed 226 ppb lead, 15 ppb cadmium, and 443 ppb barium. Most of theoff-site water samples, including two field blanks contained 2 - 5 ppb of di-n-butylphthalate. Outfall number 4 also showed 3.6 ppb chloroform contamination.

Table 4.

Other maximum value outfall discharge data
SubstanceConcentrationOutfallDate Sampled
1,1-dichloroethane 29 ppb412/12/89
chlorobenzene 18 ppb112/06/89
chloroform 2.5 ppb1, 312/06/89
isophorone 47.3 ppb112/12/89
phenol 71.4 ppb312/12/89
naphthalene 34.2 ppb312/12/89
1,1,1-trichloroethane 119 ppb412/12/89
carbon tetrachloride 21 ppb401/20/90
lead 22 ppm103/01/90
cadmium 1.6 ppm103/01/90
toluene 17 ppb103/07/90
Reference: D-126

As mentioned previously, contaminated groundwater beneath the site could migrate off site. In1989, Woodward-Clyde Consultants prepared a groundwater investigation report for MSP. Thereport stated that groundwater beneath MSP persistently "trends" toward Bayou Boeuf, basedupon observations in 1988 and 1989. On-site groundwater quality is described in the On-sitecontamination section. Off-site groundwater quality is more difficult to characterize in this case,because it is believed that site groundwater discharges into Bayou Boeuf, a surface water.

Soils, Sediments, and Aggregate: The greatest off-site contamination potential for those solidmedia would be expected to be associated with placing aggregate on soils in support of newconstruction, roads, and parking lots. Sediment contamination would be expected in bayousediments adjacent to the MSP storage, handling, and processing area. Table 5 summarizes the highest values of the data reviewed for in-situ aggregate lead and cadmium contents. Those data,assembled in 1989, are for aggregate, possibly mixed with some soil, at six off-site locations. People could be exposed to off-site aggregate by inhaling fine, airborne particles or by ingestion(particularly in children). Ingestion of significant amounts of aggregate that has been slaggedwould not be expected because of its generally coarse structure. (See Public Health Implications section.)

Table 5.

Off-site soil/aggregate lead and cadmium results High Values Found (parts per million)
Area SampledTotalEPTox*
Amelia Community Center
Berry Brothers
Cypress Gardens Subdivision
Residence (aggregate used under house)
Background (neighbor)
Lowlands Construction
Ranch (aggregate used in driveway)
ND -- not detected
* -- Extraction Procedure Toxicity - test to determine the amounts of contaminants that will leach under environmental acidity conditions

      Reference: D-28

Other contaminants found at the Amelia Community Center included arsenic at 32 ppm (versus3.8 ppm background); barium at 26,300 ppm (versus 248 ppm background); and chromium 192ppm (versus 25 ppm background). Mercury was not found in either the aggregate or thebackground samples.

Generally, volatile organic and semi-volatile organic compounds were not found in the samples. One sample at the Berry Brothers site contained 0.522 ppm of a phthalate compound. Severalsamples in the Cypress Gardens subdivision contained ppm-range chlordane, and one sample atan aggregate-contaminated fenceline contained 1.20 ppm hexachlorobiphenyl.

Sediments and bottom bayou soils near the MSP facility have also been analyzed over the life ofMSP operations. After one alleged spill in 1986, chromium in sediment was reported by LDEQto range from 51.1 ppm to 91.2 ppm (dry). Later that same year, downstream sediment wasreported to contain 8.33 ppm chromium; sediment at the plant contained 10.33 ppm chromium. In 1988, LDEQ reported that sediment samples taken near a storage barge that allegedly leakedcoal tar creosote wastes contained organic contaminants characteristic of those materials and in aproportion consistent with the wastes. In 1989, EPA further sampled Bayou Boeuf bottom soils;following are the results:

Table 6.

Sediment Samples from Bayou Boeuf
Total Lead53 to 20,900 ppm
Total Cadmiumnot detectable to 351 ppm
Barium3,350 to 26,400 ppm
Arsenicup to 115 ppm
PAHsup to 20 ppm
Reference: EPA RCRA Inspection, July 1989

Few background sediment data were available for evaluation. In 1976, the U.S. Army Corps ofEngineers analyzed Bayou Boeuf sediments sampled 0.5 mile below Highway 90 near Amelia. That part of the Bayou is not in an industrial area. Results showed comparatively low metalslevels, but high levels of organic compounds and low-level polychlorinated biphenyl (PCB)contamination.

Food Chain Data: In addition to potential contamination of off-site media, contaminants from asite such as MSP could bioaccumulate in various plants and animals eaten by people. Thereappears to be little possibility that site contaminants would affect edible plants because of thelack of agriculture near the facility. Local residents are concerned, however, about possibleimpact on peaches and pecans grown near soils where aggregate has been placed. ATSDR hasno data on potential contamination of peaches and pecans; that topic will be discussed further inthe Pathways section of this public health assessment.

Limited data are available on the flesh quality of fish caught near MSP. In 1986, mercury and barium contamination (0.625 ppm to 1.0972 ppm and 2 ppm to 11.4 ppm, respectively) were found in edible parts of fish caught about 75 yards southeast of MSP (D-149). In fish studies by LDEQ in 1987 (see table) there were no remarkable differences between the edible portions of catfish and carp found in Bayou Boeuf near MSP and in the same bayou about one fourth milesouth of Lake Palourde.

Table 7.

FISH DATA - 1987

Near MSPNear Lake Palourde
p,p' - DDEND to 0.013 ppm0.018 ppm
Arochlor 1254 (PCB)ND to 0.10 ppm0.14 ppm
PentachloroanisoleND to 0.004 ppm0.002 ppm
MetalsNot reportedNot Reported
Reference: D-63

In 1989, LDEQ again analyzed for various organic compounds in edible portions of fish caughtnear MSP; no analysis for metals was conducted. Most results were nondetectable or below-detection levels. The following table summarizes the maximum contaminant levels detected:

Table 8.

Fish Data - 1989
butyl benzyl phthalate2.040 ppm
1,2,3,5-tetrachlorobenzene0.0676 ppm
1,2,4,5-tetrachlorobenzene0.0676 ppm
1,2,3,4-tetrachlorobenzene0.0503 ppm
1,2,4-trichlorobenzene0.035 ppm
Reference: D-70

The available fish data do not address shellfish which are an important source of food in theregion.


This public health assessment is based largely on data developed by LDEQ, EPA, and MSP. When QA/QC measures were described, they appeared consistent with measures normally takenduring environmental sampling and analysis. All data were considered as received, regardless ofwhether a QA/QC description was included; however, ATSDR recognizes that QA/QCprocedures and sampling and analysis methods may not be uniform for all data received. AllATSDR-gathered data sources are listed at the end of this document.


The MSP facility is an active industrial operation with the same potential for accidents thatwould be expected at any similar facility. There is clear evidence of an ongoing worker safetyprogram. The facility is fenced to discourage unauthorized entry; access to the active processingarea is through a guarded weighing station.



Air: Contaminated air originating at the MSP site from either fugitive storage, handling andprocess emissions, or as stack emissions disperses to surrounding air depending on localmeteorologic conditions. Effective stack height and exhaust velocity also significantly influencethe dispersion characteristics of stack emissions. To assess the significance of site air emissions,ATSDR must review the hazard properties of each contaminant, note the typical and maximumconcentration levels, and evaluate the locations of potentially exposed people with respect tosource emissions. Wind direction and speed and natural dilution, decay, or contaminanttransformation mechanisms also affect human exposure levels.

From the data reviewed which measured on-site and off-site air contamination, the indicators ofambient air quality (i.e., MSP's industrial hygiene data and ambient air data, and EPA's ambientair data) do not show cause for concern that the health of neighboring communities is beingaffected. That observation is based on the overwhelming number of data points showing thatchemical contamination on or near MSP operations is well under TLVs established for thespecies observed (i.e., less than TLV/42, TLV/100, or even TLV/1,000). Given the naturaldilution of such contamination as the air moves away from the facility, and the presence of othersources of similar pollutants in the area (as reported in the TRI data for the area in 1987 and1988) it may not be possible to detect specific pollutant increases from MSP, especially atdistances to the nearest major populations. Anecdotal reports indicate that winds arepredominantly out of the north, which would tend to carry air pollutants in the direction of theBayou Boeuf, which is uninhabited. Unfortunately, no local long-term windrose was available toconfirm or refute that observation. A one year general wind direction observation at the facility,however, showed that winds tend to blow airborne pollutants away from populated areas abouttwo thirds of the year.

Stack emissions likewise follow the predominant wind direction; however, local terrain coupledwith effective stack height and exit gas velocity can result in variation of ground- and stack-levelemissions. In the next section, Human Exposure Pathways, a number of types of stack emissions will be evaluated to determine their significance to human health.

In general, no ongoing real-time or averaging off-site air monitors are in the area that could beused to assess short- or long-term air emissions from MSP to the nearest populations. Althoughit might be very difficult to establish monitoring systems that would not be confounded by otherair emissions in the area, such systems could help detect a massive release, such as might occur ifa waste tank truck overturned and ruptured while entering the facility. Also, if data werecollected over a sufficient time period and in conjunction with local meteorologic data, additionalinformation might support or refute the hypothesis that it may not be possible to measure MSP'simpact on air in residential areas away from the plant.

Water: Contaminated groundwater and surface water originating at the site and discharges fromsite outfalls all empty into Bayou Boeuf. The bayou generally flows toward the Gulf of Mexico;one citizen stated, however, that under certain tidal conditions the bayou flows up to LakePalourde, which is a source of potable water for the local communities.

The other potential off-site water pollution pathway involves the leach potential of the aggregatein the natural environment. MSP quality assurance data reported that the aggregate passes EPA'sleaching test criteria and consequently should not leach to other ground or surface waters. Thecommunity expressed skepticism about the appropriateness of the EPA leach tests for assessingbioavailability of lead or cadmium in aggregate that might be ingested. That issue will bediscussed further in the Toxicological Implications section of this public health assessment.

Soils, Sediments, and Aggregate: Contaminated site soils at MSP are not likely to migrate offsite, particularly if left undisturbed. Since operations began in 1985, part of the site has beenpaved to improve storage and containment capacity. Windblown soil particles from the sitewould be expected to "fall out" locally and not affect neighboring communities. Sediments in theadjacent bayou might or might not migrate substantially. Factors influencing such migrationcould include major storms that affect tidal flow and deposite new contaminated oruncontaminated sediments that cover and immobilize existing sediments.

Off-site aggregates can be applied or stored where children and adults have direct access to theaggregate; fine particles can become windborne and transported by air. Once aggregate isapplied and covered with asphalt, as in a paved road or parking lot, it is assumed thatcontaminants are relatively immobile, except for relatively slow releases following abrasion andwear.

Food Chain: There are conflicting reports about the consumption of fish from surface watersnear MSP. MSP and LDHH report that Bayou Boeuf is heavily traversed and, consequently,little sport or commercial fishing takes place near MSP. The community, however, says thatthere is a significant amount of fishing in local waters. The food chain pathway is difficult toevaluate because of the limited data available and the confounding effect of other pollutionsources in the area that could affect water and fish quality. Also, the available data, which arelimited, suggest there is no significant difference in the quality of fish examined from BayouBoeuf and those from Lake Palourde. See the Human Exposure Pathways and the Public Health Implications sections of this public health assessment for further discussion of that topic.

Limited data are available to evaluate the possible migration of aggregate contaminants intonearby crops, such as the peaches or pecans. In the absence of data, such as the assay of pecanmeats or peach flesh, ATSDR must base its analysis on the known characteristics of theaggregate and the growing sites in question.

Regarding the aggregate, it has been shown that contamination from organic compounds islargely nonexistent, as would be expected following combustion. It has also been shown thattotal metals, such as lead and cadmium, can be quite high; on the other hand, leachable metalsincluding lead and cadmium as assayed by EP toxicity or the TCLP tests are low.

The following observations were made about the specific sites where aggregate has beendeposited. The aggregate was placed as fill along a fenceline dividing three properties. Some ofthe aggregate spilled or washed across the fenceline into two neighboring properties. Mature,productive peach and pecan trees are in two yards adjacent to areas where the aggregate isreported to have migrated. ATSDR does not know if the roots of those trees extend into soilswhere the aggregate migrated. Using available information, however, it is not likely that eitherthe peaches or pecans would sustain measurable contamination from metals in the aggregate. Water percolating through the soil/aggregate mix is not likely to be a stronger leaching agent thanacid used in the EPA leaching tests; consequently, the metals should remain bound in theaggregate matrix. Assays of the peaches and pecans would help to resolve that issue.

B. Human Exposure Pathways

Air: In the preceding section of this document, a discussion of ambient air releases concludedthat there would not be significant or incremental exposure to fugitive plant emissions by citizensin neighboring communities. That conclusion was based on the low measured concentrations oforganic species near plant operations. Also, limited meteorologic data indicate that suchemissions would tend to move away from populated areas. A small number of MSP employeeslive in mobile homes immediately west of the active facility. Again, the reported data, coupledwith the apparent prevailing wind information, suggest that even that population would notinhale fugitive emissions at levels of health concern.

Many stack emissions data are reported in terms of pounds of pollutant emitted per hour. Toevaluate potential human exposure, that information must be translated into potential exposureconcentration levels. As a part of this public health assessment, ATSDR conducted twoscreening exercises. First, using the reported volumetric flow rate of stack effluent, pollutantconcentrations were determined. Stack concentrations approached or exceeded TLV were thenanalyzed with a mathematical air dispersion model based on worst-case assumptions todetermine maximum ground-level concentrations (MGLCs) of any species of concern (R-76). MGLCs were then reviewed with respect to possible impact on public health. Generally, noenvironmental standards exist for human exposure by way of inhalation to many of the pollutantsreviewed in this manner. This approach is crude and serves only as a screening tool to estimateorder-of-magnitude concentrations and thereby indicate situations that merit closer examination. ATSDR policy stipulates that modeling cannot serve as a proxy for actual measurements ofexisting conditions when public health implications are being determined.

Using a stack gas flow rate of 62,000 dry standard cubic feet per minute, three of seven metals(chromium, mercury, and nickel) exceeded the lowest TLVs for the most toxic species of theparent metals. After calculation of MGLCs, nickel and chromium were found to be about1/160th and 1/80th of their TLVs, respectively; mercury was about 1/8,000th of its TLV. In theabsence of standards for that type of exposure, some regulatory jurisdictions have used 1/42nd,1/100th or 1/1,000th of the TLV as a criterion for acceptability of an emission (R-66). For thisanalysis, nickel and chromium are the metals of concern, and the State of Louisiana hasidentified those two metals as being emitted in excess from the incinerator stack (state stack testannouncement in 1988). The state's use of a more refined model than that used by ATSDRresulted in it finding slightly lower concentrations than ATSDR. The general orders ofmagnitude, however, were in close agreement. The state-generated data also showed thatMGLCs for nickel and chromium were under TLV/42. Maximum chromium and nickelemissions in June 1990 were more than 100 times lower than the maximum values used for thisscreening exercise.

The method used by ATSDR to calculate MGLCs for metals resulted in an approximate10,000-to-1 dilution ratio of the stack gas concentrations. The state's model dilution factor wasabout 30% greater (13,000 to 1). For the remainder of this analysis, the more conservative(worst-case) ATSDR factor will be used to estimate MGLCs for other pollutants from the stack(Table 9). Use of the lower dilution factor results in a higher exposure estimate.

Table 9.

Selected High-Value Stack Emission MGLC Estimates
CompoundsStack ConcentrationsEstimated MGCL
Dioxins 0.0031 µg/m30.00000031 µg/m3
Carbon Monoxide 136,842 µg/m313.68 µg/m3
1,1,1-Trichloroethane2461 µg/m30.25 µg/m3
Trichloroethylene1181 µg/m30.12 µg/m3
Sulfur Dioxide64,650 µg/m36.46 µg/m3
Hydrogen Chloride1,793 mg/m3179.3 µg/m3
Nitrogen Oxides180 mg/m318 µg/m3
Reference: D-100, D-137

The preceding discussion and estimates are used to screen for possible human exposures ofpublic health concern. It must be remembered that they are based on highest reported emissionlevels and a meteorologic scenario that yields the highest ground-level concentrations of apollutant. Even with that bias, most pollutants are well below levels of health concern. Forexample, are below TLV/1000 for worst-case maximum concentrations. HCl is the only speciesthat exceeded TLV/100 in this exercise (179 µg/m3 versus 75 µg/m3). Actual average humanexposures are likely to be considerably less than the modeled values. Values for the pollutantscarbon monoxide, sulfur dioxide, and nitrogen oxides did not exceed National Ambient AirQuality Standards.

Water: Considering the available data, human exposure to water contamination associated with MSP activities and produced aggregate should be inconsequential. Surface waterand groundwater from the site flow into Bayou Boeuf, where they could have a localized impacton water quality. With the considerable dilutional capacity of that body of water, however,ATSDR estimates that the current pollution load is not likely to be detectable above backgroundlevels within several hundred yards from the discharge point. Past alleged discharges may havehad more lasting effects on bayou sediments, which could in turn contaminate the waters above. However, with the normal flow of the bayou toward the Gulf of Mexico (away from the drinkingwater intake), and the distance and dilution that would be experienced during unusual tidalevents (when water would reverse its normal flow away from the Gulf) it is highly unlikely thatdrinking water from Lake Palourde would be significantly affected by this source.

Under normal conditions, aggregate has not been shown to leach significantly, particularly whenassessed using the EPA leachability tests, which should be appropriate for environmental pHconditions. Consequently, water percolating through the aggregate in situ (in its position) shouldnot create the possibility of human exposure to site contaminants when water is ingested.

Soils, Sediments, and Aggregate: Human exposure to contaminated MSP soils or nearby sediments is not expected to be a public health problem. Levels of soil contaminationon site are low, and the site is not particularly convenient as a play area. Similarly, sediments inthe bayou are not normally easily accessed by humans.

Human exposure to aggregate is possible when the aggregate is applied to a surface, but notpaved over or otherwise covered. Possible routes of exposure include inhalation, ingestion ofwindblown aggregate particles, and direct ingestion of aggregate by children playing in or onexposed aggregate. The significance of this route of exposure is discussed in the Public HealthImplications section of this public health assessment.

Food Chain: A potential for human exposure to contaminated fish, and/or peaches and pecans has been identified. Data were too limited to determine potential sources offish contamination; however, in general, there does not appear to be a fish contaminationproblem clearly attributable to MSP. One fish sample in 1986 had mercury levels of possiblepublic health concern; however, the contamination was not shown to be linked to MSP activitiesor site contaminants. Regardless of the source of contamination, it would be wise to performadditional fish assays for metals contamination to better understand whether that one fish samplewas representative of local fish populations. Subsequent fish assays focused on organiccontaminants rather than metals.

ATSDR is not aware of any assay data for the peaches or pecans discussed previously in thisreport; consequently, human exposure potential from the food chain cannot be addressed. It isrecommended that those foods be assayed for metals, particularly lead and cadmium, to resolveconcerns of the owners of the potentially affected trees.


A. Toxicological Implications

Air: Evaluation of data and pathways did not indicate a potential for people to be exposed tolevels of health concern. Under worst-case conditions in the modeling, nickel and chromiummay be emitted from the stack at levels above health comparison values. Nickel is a silverymetal that occurs naturally in the environment. People can be exposed to nickel compounds byway of inhalation, ingestion of food or water containing nickel compounds, and/or skin contactwith many consumer products (R-11). Nickel exposure by inhalation is small compared to theamount consumed in the average diet. Insoluble nickel compounds that are inhaled canaccumulate in the nasal mucosa and lungs, depending on the size of the particles. Tobaccosmoke contains relatively high levels of nickel compounds, including nickel carbonyl. About 5%of people are sensitive to nickel; the most common adverse effect is skin sensitivity (R-4). Individuals not sensitive to nickel have not been shown to have adverse effects from long-termexposure by inhalation (R-4). Using a worst-case scenario, ATSDR considers the nickel levelspotentially emitted from the facility to be below those of health concern.

Chromium is an element found in three valence states in the environment: chromium(0),chromium(III), and chromium(VI). Chromium(III) occurs naturally; the others are produced byindustrial processes. Chromium(III) is believed to be an essential element for the body's use ofglucose. Chromium compounds are used in chrome plating, and as catalysts in steel and otheralloy production, leather tanning, wood and water treatment, and photography (R-68). Humanexposure could result by way of food intake and, to a lesser degree, by way of dust inhalation and water ingestion. People are usually exposed by inhaling the metal and its salts. Insolublechromium compounds are deposited in the lung (R-11). Long-term exposure to chromium(VI)has been found to cause cancer in animals. Exposure to high levels of chromium by way ofinhalation has been associated with lung cancer in workers (R-3). Levels found at the site arebelow recommended health guidelines and should not pose a public health risk.

Hydrogen chloride (HCl), a colorless, water-soluble gas, was also potentially emitted at moderatelevels from the plant when a worst-case scenario was used. In aqueous solution, HCl is known ashydrochloric acid or muriatic acid; it can be corrosive to the skin, eyes, mucous membranes, andrespiratory tract. People with skin, respiratory, or digestive diseases can be more sensitive to theeffects of HCl(R-29). Irritation can occur at exposure levels above 5 ppm. Worst-caseground-level concentrations from MSP activities were found to be less than 1 ppm. Nosignificant effects have been seen following chronic exposure to low levels of the gaseous formof HCl (R-63, R-80).

Water: Past spills of aggregate into a localized area of the bayou raise questions about theleachability of lead and cadmium into surface water. Leachability testing at environmental pHindicates that the spills should not be a problem under normal environmental situations. EPAtests of surface waters in the area did not find significant levels of lead or cadmium. Therefore,their presence is not believed to be of public health concern.

Barium was found in the water of Bayou Boeuf at levels ranging from non detectable to 1830ppb. In its National Interim Primary Drinking Water Regulations, EPA set its allowableconcentration of barium in domestic water supplies at 1000 ppb (R-11). Those regulations applyonly to community drinking water systems. In May 1989, the Proposed Maximum ContaminantLevel (PMCL) for barium, 5000 ppb, was published in the Federal Register. The barium inBayou Boeuf is below levels that pose a threat to public health.

Soils, Sediments, and Aggregate: There is community concern about past discharge of creosote from a storage barge into the bayou. Creosote is a heavy, oily, colorlessliquid that has a characteristic smoky, tarry odor. If contaminants are present creosote is a brownto black color (R-68). Creosote is essentially insoluble in water. The primary route of humanexposure at MSP, therefore, would have been by direct dermal contact and , possibly, exposureby inhalation of creosote stored on the barges. Ingestion of the water from the bayou would nothave been an exposure route. Creosote is a varied mixture consisting of about 85% PAHs and 2 -17% phenolic compounds. PAHs from creosote released into surface water tend to bind to soilsand sediment particulates with low solubility and mobility. PAHs have been found in surfacewaters throughout the United States; concentrations in sediments are generally much higher thanin surface waters. Creosote irritates the skin; long-term exposure to low levels can causereddening, blistering, or peeling. Dermal exposure was a possible route of exposure for workerson the barges and for workers around the barge area when the barges were present. Followingremoval of the barges, and given the limited access to Bayou sediment, human exposure tocreosote in sediments is not likely (R-7).

Two issues have been raised about the safety of the aggregate: leachability of metals from thematrix and the fibrogenic potential (the ability to form a protein clot in the lungs) of smallparticles of the material. The metals of primary concern are cadmium and lead. Children aremore sensitive than adults to effects of lead exposure. The central nervous system is the primarytarget organ for lead toxicity in children (R-50, R-16, R-49, R-58, R-23). Recent studies indicatethat adverse effects on the function of the central nervous system persist into adulthood (R-51). New data also indicate significant adverse effects in children with blood lead levels previouslybelieved to be safe. Current information has shown disturbances in neurobehavioraldevelopment in children with blood lead levels in the range of 10 - 25 µg/dL and in childrenwhose mothers had blood lead levels in that range during pregnancy. There has been someindication that effects occur at blood lead levels below 10 µg/dL (R-9). The Centers for DiseaseControl has recently recommended that the intervention level be established at 10 µg/dl (R-84). The earliest intervention is health care workers counseling of parents and identifying possiblesources of lead exposure for children. Lead has been shown to leach from soils, and blood leadlevels in children appear to be related to the level of contamination in their environment (R-28). Blood lead levels generally rise 3 - 7 µg/dL for every 1,000 ppm increase in soil or dustconcentrations (R-84). Experimental studies relating blood lead levels to dietary lead intake foradults produced estimates of a 0.02 µg/dL increase in blood lead per microgram of lead per dayof total intake (R-8).

Laboratory results on the aggregate at one residence suggested that the leachable leadconcentration was 7 µg/gm. The slagged aggregate cannot be directly compared to soil becausethe particle structure might bind the materials in a manner similar to pottery glazes (i.e., leachingof lead from the aggregate may be similar to lead glazes in ceramic material rather than leachingof lead from soils). The amount of lead leached from ceramic material varies with the percent oflead in the glaze, the other constituents (pigments), and the temperature of the firing (R-27, R-57).

The aggregate produced at MSP contains a high level of lead. However, the firing temperaturemight be high enough to reduce leaching. EP Toxicity and TCLP testing did not indicatesignificant leaching of lead at environmental pH. Further study done at a pH of 1 indicatedincreased amounts of leaching. A pH of 1 is closer to the pH of the stomach and could indicateleaching potential from ingested aggregate. Digestion in the stomach is initiated by release ofHCl. Distention of the stomach by food triggers the release of gastrin, a hormone that controlsproduction of HCl in the stomach. A gastric pH of 3.0 begins to inhibit gastrin production;production ceases at a pH of 1.5. When gastrin production stops, HCl production also slows. The average pH reached in the stomach is 1.4. The results of leaching at pH 1 might indicatepossible exposure if significant amounts of aggregate are ingested. In addition, a bioavailabilitystudy in mice showed a rise in blood lead levels after gavage feeding of the aggregate. Thelevels reached would not pose an acute problem; however, body burden was not examined in thestudy. Those data indicate that the aggregate is bioavailable to some extent at a gastric pH(D-137A); therefore, the possibility of increased body burden of lead from chronic exposure is ofconcern.

The normal maximum soil ingestion rate in children is estimated at 0.5 gram/day. A study usingtracer elements to identify actual ingestion found a median intake of soil, in children one-year tofour-years old, ranged from 0.009-0.096 gram/day. The majority had a median intake of lessthan 0.05 gram/day (R-18).

The average amount of soil ingested by children does not indicate the likelihood of significantacute exposure to lead by aggregate ingestion. Also, slagged aggregate is a coarse ceramic-likematerial, which would not be likely to be ingested in large amounts. In addition, the metalscontent of the aggregate varies widely. Therefore, it is not possible to determine the typicalbioavailability of lead from the entire range of aggregate produced. Analysis of blood lead levelsin children at a residence where aggregate was placed as foundation material did not indicateincreased lead levels; however, the blood lead in one of the children was 8 µg/dL. Aggregate atthe residence was "primary" aggregate and has different properties than slagged aggregate. ATSDR reviewed information on the exposure potential at this residence and data about thecharacteristics of aggregate at the residence. ATSDR concluded that aggregate at the residence isof potential public health concern because there is a potential for lead exposure and there is nodefined threshold for chronic lead toxicity (D-24A). The aggregate at the Amelia CommunityCenter is located under a concrete slab, which provides an effective barrier to human contact;thus, there should be no significant exposure at the community center.

The amount of cadmium in the aggregate may be of concern if a significant part of the aggregateis ingested. The kidney is the primary target organ for cadmium in people (R-77, R-67). Inhalation of cadmium has been associated with increased risk of lung cancer. Smoking is asignificant source of cadmium (R-5). Only about 1 - 5% of ingested cadmium is absorbed intothe body. Long-term intake of 0.005 mg/kg/day (0.075 mg/day for a 15-kg child) is not believedto pose a risk for tissue damage (R-5). Leachability information indicates that a child wouldhave to ingest 25 grams of the aggregate daily to absorb 0.075mg of cadmium. As notedpreviously, the nature of slagged aggregate makes this level of ingestion unlikely for thatparticular material.

Use of the aggregate on roadways and other areas where crushing and dust generation are likelyraises the issue of particle-induced pulmonary (affecting the lungs) toxicity. Numerous types ofparticles have been shown to induce lung disease (R-36, R-62, R-71). Particles with anaerodynamically equivalent diameter of less than 3.5 micrometers can be deposited in theairspaces of the lung (R-34). In a study by Schneider et al, particles with irregular, angularshapes (quartz, volcano ash, heat-treated clay) tended to produce a denser type of fibrosis(overgrowth of connective tissues, particularly in response to an injury) with associateddestruction of the alveoli (air cell of the lung) (R-65).

Particles that have a long lung retention time and that stimulate digestive mechanisms of lungtissue are more likely to produce emphysema (R-39). Emphysema is a lung disease that resultswhen the alveoli are destroyed and the lung loses its ability to expand and collapse, causing air tobe trapped and breathing difficulties.

A scanning electron microscope study of particles from dust generated at a road where aggregatewas placed indicates that respirable-sized particles are generated, some of which are angular inshape (D-21). These angular shapes that are respirable could be fibrogenic if inhaled. Furtherinvestigation of those particles is needed to determine what portion of such particles is actuallyfrom the MSP aggregate before this product can be used safely in areas where dust generationmay occur.

At one site, arsenic was found at levels of 32 ppm (compared to 3.8 ppm background). Arsenicoccurs naturally throughout the environment; most people are exposed to trace levels duringeveryday activities. Food is the major source of intake, accounting for between 25 and 50 µg/day(R-2). Sensitive individuals (adults) can experience symptoms of exposure after ingestion of1,000 to 1,500 µg/day. The primary features of arsenic intoxication are gastrointestinal anddermatological (R-33, R-17). Human exposure to arsenic levels found at the MSP site should notpose a public health threat.

Food Chain: Mercury was detected in 1986 in one sample of fish from the bayou. Mercuryoccurs naturally in the environment in many forms, both organic and inorganic. The types ofadverse health effects seen with exposure depend in part on the form of mercury (R-1, R-75). Inorganic mercury salts are nephrotoxic (causing injury to the kidney) and can produce acutetubular necrosis (death of kidney tubule tissue) (R-77). Methylmercury (organic) is toxic to thebrain, particularly in the developing nervous system (R-23). Mercury has been shown tobioaccumulate in fish (particularly tuna and swordfish) in the organic form of methylmercury. The mercury reported in the 1986 samples of fish was at levels of 0.625 mg/kg to 1.0972 mg/kg. EPA estimates that adults could be exposed to levels of mercury up to 0.021 mg per day in foodor water without harm to their health (R-1). The amount of fish with the mercury levels found inthe Bayou that could be ingested before reaching 0.021 mg is 110 grams per day. The FDAAction Level for methyl mercury in edible portions of fish and shellfish is 1 ppm (mg/kg). Oneof the fish sampled exceeded the FDA action level for mercury. The amount of fish from thebayou ingested by people is not known, and information on analysis in samples after 1986 wasnot available. Further studies should be done to evaluate the significance of this finding.

The community is concerned that possible continued contamination of the bayou from pastcreosote spills could increase the potential for fish contamination. The major constituents ofcreosote are PAHs. Substances that are persistent and not able to be broken down in theorganisms who take them up are concentrated in the tissues. Thus, the organisms at the top ofthe food chain (humans, carnivores) will receive the highest concentrations of these substances. This process is called food chain biomagnification. Uptake of PAHs by aquatic organisms hasbeen demonstrated. These compounds are extensively metabolized by fish and higher organisms,so biomagnification in the food chain is limited. The amount of PAHs found in food also isaffected by the method of preparation (grilling or smoking increases the amount of PAHs infood). The food chain would not appear to be an important source of human exposure to PAHsat this site based on the limited biomagnification of these compounds and the lack of continuousexposure.

B. Health Outcome Data Evaluation

Census data provide some background for evaluation of the health outcome data. Table 10 givesthe 1980 -1986 Census information for St. Mary Parish.

Table 10.

Census Data

United StatesLouisianaSt. Mary ParishMorganCity
1980 Land Area
(square miles)
1980 Population226,546,0004,206,00064,253
1986 Population241,078,0004,501,00064,30016,030
1986 Pop./sq. mile68.1101.1104.9
1984 % White Pop.
       % Nonwhite Pop.

1984 Age Group
    < 5 years
    5 - 14 years
    15 - 24 years
    25 - 44 years
    45 - 64 years
    > 65 years




1984 Births

1984 Total Deaths

Infant Deaths
       @ Rate







1985 Per Capita Income$10,797$8,836$8,768$10,360
1979 Percent Below Poverty




* Per 1,000 resident population            # Per 1,000 resident population
@ Per 1,000 live births

Reference: D-77

Review of the census data suggests there is a younger population in Louisiana and St. MaryParish than in the nation as a whole. St. Mary Parish has a high nonwhite population and largernumbers of people live below the poverty level than in the United States as a whole. MorganCity has a higher per capita income than either the parish or the state and approaches the percapita income of the United States. In 1985, Louisiana ranked 40 among the 50 states in percapita income. The influence of a large minority population and low economic status must beconsidered when examining health parameters for the area. Racial and ethnic groups havedifferent health problems. For example, sickle cell anemia is much more common in individualsof African descent (R-85). Socioeconomic status can affect overall health of a community, thusinfluencing the community's susceptibility to certain diseases. Poor nutrition, crowding, andlimited access to health care are some factors resulting from low economic status that affecthealth.

Numerous community concerns exist about adverse health effects from exposures that mighthave occurred during 1986 and 1987. Tables 11a and 11b review the vital statistics St. Mary parish and Louisiana for those two years.

Table 11a.

1986 Vital Statistics
OutcomesLouisianaSt. MaryMorgan City
Live Births: Total

Rate: LB/1,000 Population







Immature Births: Total

Rate:(Per 1,000 Live Birth)

      (Excludes Stillbirth)

    Rate/1,000 Population







    Cause of Death:
      Heart Disease (*)
      Malignant Neoplasms
      Cerebrovascular Diseases
      COPD & Related Disorders
    Nephrosis, Nephritis, etc
      Chronic Liver Disorders
      Congenital Anomalies

    (*):Rate/1000 Population

12,728 (2.8)
8,036 (1.8)
2,578 (0.57)
2,014 (0.45)
1,017 (0.23)
648 (0.14)
507 (0.11)
622 (0.14)
380 (0.08)
269 (0.06)

151 (2.3)
104 (1.6)
40 (0.62)
42 (0.65)
8 (0.12)
12 (0.18)
4 (0.06)
5 (0.08)
4 (0.06)
3 (0.05)

COPD -- Chronic Obstructive Pulmonary Disease (lung diseases)

Reference: D-75

Table 11b.

Human Health Effects at Various Hydrogen Sulfide Concentrations in Air
OutcomesLouisianaSt. MaryMorgan City
Live Births: Total

      Rate: LB/1,000 Population







Immature Births: Total

Rate:(Per 1,000 Live Birth)
      (Excludes Stillbirth)

    Rate/1,000 Population







    Cause of Death:
      Heart Disease (*)
      Malignant Neoplasms
      Cerebrovascular Diseases
      COPD & Related Disorders
    Nephrosis, Nephritis, etc
      Chronic Liver Disorders
      Congenital Anomalies


13,050 (2.9)
8,119 (1.8)
2,517 (0.56)
1,842 (0.41)
1,073 (0.24)
559 (0.15)
454 (0.12)
570 (0.15)
347 (0.10)
269 (0.07)

162 (2.6)
97 (1.5)
30 (0.47)
27 (0.43)
12 (0.19)
1 (0.02)
10 (0.16)
5 (0.08)
4 (0.06)
6 (0.09)

COPD -- Chronic Obstructive Pulmonary Disease (Lung Diseases)

Reference: D-76

Table 12.

Congenital Malformations
CodeMalformationLouisianaSt. Mary
740Anencephalus1415 710 916 6 1

741Spina Bifida22192028252021

1 1 1
742Other NervousSystem12261417232122

744Ear, Face, andNeck12192520204729

1 5 1
745Bulbus Cordisand CardiacSeptal Closure 5 9 81412 913

1 1 1
746Other Heart15221331262622

1 1 2

747Other CirculatorySystem151510141910 7 1


749Cleft Palate andLip43554649495044 2 2 2
1 1 1
750Other UpperAlimentary Tract11101410 71514

751Other DigestiveSystem13201217111110


752Genital Organs57595662899385 3
2 3 4 8 4
753Urinary System 5131318141622

754CertainMusculoskeletalAnomalies86846666567067 2 3
4 2 1 4
755Other LimbAnomalies173156151148150189172 8 5 4 4 4 4 7
756OtherMusculoskeletal37342928449244 1 1
2 1 8 3
757The Integument(skin)303530504011173 1 1

410 3
758Chromosomal19282927292034 2
759Other andUnspecified15223836483126


743Eye 712 6 419 519

748RespiratorySystem 8171718 71213

Total Malformations59967060466769786474321121317234227
Reference: D-85

Congenital malformations were identified in approximately 1% of live births in Louisiana for theyears 1984, 1986, and 1987. A larger number of children with malformations was identified in1986. St. Mary Parish had 1% of live births identified as having malformations in 1984, 3% in1986, and 2% in 1987. The typical incidence of congenital malformations identified on birthcertificates is 0.5% to 1%. If hospital discharge diagnoses are used, the incidence is between 2%and 4% (R-54). Louisiana and St. Mary had a cause-specific mortality rate for congenitalmalformations of 0.06/1000 in 1986 (R-54). That rate is consistent with the natural average ofcause-specific mortality. The number of malformations in St. Mary Parish was higher in 1986than in previous years. Louisiana also noted a larger number of cases in 1986, but it was not atthe proportion noted at the parish level. Several parishes in the northern half of the state alsoexperienced a rise in the number of malformations in 1986 similar to that seen in the area aroundSt. Mary.

Both Louisiana and St. Mary experienced an increase in anomalies of the skin, ear, face, andneck anomalies; and those classified as other musculoskeletal anomalies. A review of congenitalmalformations in St. Mary showed that 27 of the 42 were minor anomalies such as skin tags, cafeau lait spots, and undescended testicles. The number of malformations identified in both St.Mary Parish and Louisiana was lower in 1987 than in 1986. The reasons for the fluctuations arenot known, but fluctuations are common in this type of data.

St. Mary has had a higher incidence of spina bifida than the national average in the past. Thatexperience is unusual considering that there is a large African-American population in the areacompared to the national average. Spina bifida occurs less frequently among blacks than amongwhites or other races (R-82). ATSDR does not have information on the racial make-up of thechildren with spina bifida. Louisiana has a greater black population and a lower rate of spinabifida than the United States. Several parishes in the Louisiana Health Service Area 2 wereconsidered to have a high incidence of spina bifida, which did not change from 1983 to 1987. The incidence of spina bifida was high in 1983, before MSP began operation. Therefore, the rateof spina bifida in the parish appears to be unrelated to exposures from the facility.

Cancer incidence in southern Louisiana is lower than in the United States as a whole for thecategory of total cancer. The rate of lung cancer is higher than the U.S. rate for males. TheLafayette Region, which includes St. Mary Parish, has lower rates than the overall rates forsouthern Louisiana.

Table 13.

Cancer Incidence
SiteUnited StatesSouth LouisianaLafayette Region
All Sites469.4351.8435.4296.1414.3282.2
Colon **46.936.338.832.535.128.6

Prostate (male)99.7
Breast (female)
Incidence rates per 100,000 population. U.S. rates are for 1983-1986. Southern Louisiana and Lafayette Region rates are for 1983-1986.

** U.S. rates include rectal, which is not included in Louisiana rates.

Reference: D-93, D-94

During 1986 and 1987, an excess of neuroblastoma cases was reported in St. Mary Parish(Expected: 0.4 cases, Observed: 4 cases). A case-control study by the Epidemiology Section ofthe Department of Pathology at Louisiana State University Medical School investigated thecluster. The cases included four children from St. Mary Parish and one child from AssumptionParish, whose mother had worked in St. Mary during her pregnancy. Cases were compared withtwo control groups: a sample of children born in the same hospital as the cases and a sample ofchildren born in Bogalusa, LA. The study focused on exposures before and during pregnancy. Examination of environmental exposures such as drinking water source, residential proximity toindustry, and pesticide treatments did not indicate that cases had increased exposure. Nooccupational or industrial exposures were significantly different between cases and controls. SeeAppendix A for a complete discussion of this study.

It is extremely difficult to identify causative agents in investigations of neighborhood clusters ofdisease, particularly if the cluster occurred during a specific period and is not ongoing. Identifying an environmental exposure that might have caused a chronic disease, such as cancer,is difficult because of the long latency period, and because exposures can occur by way ofenvironmental routes such as air, water, or dust. Because everyone in the area is exposed tosome extent to water, air, or dust, it is difficult to correlate an individual's exposure with aspecific outcome (R-53).

Neuroblastoma is a primitive tumor of the sympathetic nervous system arising most commonly inthe retroperitoneal area (the adrenal medulla is the most common site). The tumors metastasizeearly to the bones, lymph nodes, liver, and subcutaneous tissue. Neuroblastoma accounts for 8%of all childhood cancer in the United States; about 500 new cases are diagnosed annually. Ininfants, it is the most commonly diagnosed malignant. Fifty percent of neuroblastomas arediagnosed during the first two years of life; more than two thirds are diagnosed before age 5. Because the majority occur at such young ages, many researchers believe the tumor originatesbefore birth. It is less common in black children than white children, suggesting that race couldplay a role in the etiology. Multiple cases of the disease in some families suggest there could beinherited susceptibility (R-45, R-42, R-55, R-60, R-37, R-59, R-13, R-41, R-26, R-83, R-20,R-74, R-25).

Many factors have been investigated for links to neuroblastoma, however, no factor has beenconsistently associated with an increase risk of neuroblastoma in those investigations. Birthweight above 4,000 grams has been suggested to be a factor in cancers before age 2 (R-24). Asignificant increase of breast cancer in mothers of children with solid tumors was seen at onereferral unit (R-72). Cytomegalovirus infection has also been suggested as an etiologic agent(R-78). The occurrence of two cases of neuroblastoma in children with fetal hydantoin syndromesuggested a possible link to maternal Dilantin use in pregnancy (R-48). Fetal hydantoinsyndrome is a group of birth defects seen in infants of mothers who used seizure preventionmedications (such as Dilantin) during their pregnancies. A large case-control study in Minnesotadid not identify any associations between neuroblastoma and complications of pregnancy, highbirth weight, or shorter gestational period (R-52).

A study investigating medical and drug risk factors for neuroblastoma found associations withmaternal use of alcohol, neurally active drugs, sex hormones, hair-coloring agents, antinauseamedication, and diuretic use for water retention during pregnancy. Familial factors, parentalcancer, irradiation, and cigarette smoking were not associated (R-41).

Recent epidemiological studies of parental occupation and neuroblastoma have suggested a linkbetween parental employment in occupations with electromagnetic field exposure andneuroblastoma (R-70, R-15, R-81). That association was not significant in a study by Bunin etal. In summary, there has been no consistent link between neuroblastoma and any environmentalcausative agent.

The community has expressed concern about possible exposure to Dinoseb which wasincinerated at the MSP facility in January and July, 1987. The concern is related to the fact thatDinoseb, a nitrophenol agent used as an insecticide and herbicide, was one of four pesticidestargeted for further study in the report on a cluster of childhood cancers in McFarland, California. McFarland is a small town north of Bakersfield in which 13 cases of childhood cancer occurredbetween 1975 and 1989. Only four cases would have been expected during that time periodbased on typical incidence patterns. Only one of the cases in McFarland was a neuroblastoma. To date, no common exposure has been identified (R-19). Dinoseb has been shown to be areproductive toxicant in animals, but not a carcinogen.

Because there was no definite environmental exposure identified for the children withneuroblastoma, it is not likely that any study could make a link to any single facility.

Numerous health parameters of concern have been identified in the community, such as the spinabifida rate elevated above the usual rates, the higher than expected rate of congenitalmalformations, and the previous cancer excess. Most of those parameters, however, have beenelevated for many years, and it is not possible to link them specifically to exposures from MSP. The review of health outcome data for this public health assessment indicates that thecontaminants identified at the facility have not been shown to cause the health problems that thecommunity is concerned about. A more indepth study of the area might provide additionalinformation that would indicate some link between those adverse health effects and anenvironmental exposure.

C. Community Health Concerns Evaluation

1. Safety of the aggregate

As noted in the Toxicological Evaluation section the available information does not suggest thatanyone would be likely to ingest enough of the aggregate to risk intoxication from metals in theaggregate. However, there are very few data on the body burden resulting from long-term, low-level exposure to the aggregate. Specifically, there is insufficient information to determine thepublic health impact of the use of the aggregate on roadways and other areas where crushing anddust generation might occur or the use in areas where children may have chronic contact with theaggregate. ATSDR is concerned about any potential source of environmental lead exposure forchildren. Lead tends to stay in the body a long time once it is absorbed. To determine the safetyof the aggregate requires additional information that considers the fact of increasing body burdenof lead from chronic low-level exposures. That chronic problem is one of the major reasons theU.S. Public Health Service has developed a strategy to reduce exposures to lead from allpotential environmental sources.

2. Contamination of Bayou Boeuf

As discussed in the Human Exposure Pathways section, exposure to site contaminants in thebayou is not likely to be of public health concern. However, ATSDR received limited data forevaluation of this concern.

3. Stack emissions

Based on the information discussed in the Environmental Contamination and Pathways Analysessections, human exposure to stack emissions from the MSP site is unlikely to be of public healthconcern.

4. Clusters of neuroblastomas in the area

The incidence of neuroblastomas and other health concerns were discussed in the HealthOutcome Data Evaluation section. Using available information, it is not possible to link theincreased number of neuroblastomas in the area in 1986 with exposures to contaminants at theMSP site. A community health investigation will be conducted to better assess this adversehealth outcome and other health concerns expressed by the community living near the site.

Next Section       Table of Contents The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #