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



Potential Exposure Pathways

The potential pathways of human exposure to RMA- and non-RMA-related contaminates aresummarized below and in Table 18.

Potential Onpost Surface Soil and Sediment Pathway

A total of 178 soil contamination sites were identified in the SARs and two additional soilcontamination sites were evaluated during the development and screening of alternatives forremediation (EBASCO, 1993b, v. II, p. 1-11). Those sites evaluated were the Basin F Wastepileand RMA "Surficial Soils", including areas outside the boundaries of the SARs sites (EBASCO,1991). In the Detailed Analysis of Alternatives, a third site consisting of the soils beneath threebuildings in the North Plants was added (EBASCO, 1993b, p. 1-11). Of those 181 sitesevaluated, the bulk of the contamination detected in surface soil samples (0-2" depth) is locatedin the central sections of RMA in and around the manufacturing complexes, solid waste disposalareas, and liquid waste basins (EBASCO, 1993b, v. 1, p. 8). The location of that surficial soilcontamination is shown generally in Figure 2.1-1 given in the RISR (EBASC0, 1992, p. 2-11). Specifically, the six most contaminated sections are Sections 25, 26, 35, and 36 (T2S, R67W)and Sections 1 and 2 (T3S, R67W). About half of the acreage of those central zone sections iscontaminated.

For the purpose of this Assessment the data amassed for Onpost sediments has been includedwith the surface soil samples because the contaminants they contain are or could be available to the potential surface soil pathway. The sediments grouped with surface soils include dredged lake-bed sediments and dry or seasonally dry lake beds, streams, or ditch channel segments.

ATSDR reviewed Onpost surface soil and sediment data incorporated in the RMAED andidentified the Organosulfur Compounds (OSCHs) Chlorophenylmethyl sulfide (CPMS), -sulfone(CPMSO2), and -sulfoxide (CPMSO); the Organophosphorous Compound - GB-Agent Related(OPHGBs) Isopropylmethyl phosphonic acid (IMPA); the Organophosphorous Compound -Pesticide Related (OPHP) Atrazine (ATZ); the Organochlorine Pesticides (OCPs) Aldrin(ALDRN), Chlordane (CLDRN), and Dieldrin (DLDRN); and Arsenic (AS), Beryllium (BE),Copper (CU), and Lead (PB) as contaminants of potential concern (see Tables 9 and 10). Thosecontaminants have been detected at the maximum concentrations listed in Tables 9 and 10 withinthe central six sections.

Although contamination of deeper soils (>2") is somewhat more widespread outside of thenucleus of contamination comprised by the six central sections of RMA, surface soil andsediment contamination in the following areas is largely localized as described below:

  • Southern RMA -
      Drainage ditch sediments in the Havana Street Interceptor and the Uvalda Streetstormwater ditches which carry offpost flows on to RMA from southern sources.
  • Northern RMA -
      Sediments of the outfall ditches associated with the RMA sewage treatment plant.
  • Eastern RMA -
      Isolated "hot spots" of surface soil contamination in the new toxic storage yards(Sec. 31, T2S, R66W) and the old toxic storage yard (Sec. 6, T3S, R66W).

Thus, surrounding the surface soil contamination in the center of RMA a 1-2 mile wide zone ofrelatively uncontaminated surface soils exists. Within this external zone or rim of sectionsadjacent to the RMA boundary, scattered occurrences of two contaminants, dieldrin and arsenic,are detected at levels at or above levels of possible health concern. Aldrin is also detected in thisrim of sections, but at levels below health based comparison values. This may be due, at least inpart, to the fact that dieldrin, which has a half-life of 7 years in soils, is also a decompositionproduct of aldrin, which has a half-life of 4 to 5 years in soils (EBASCO, 1988). Table 16 givesthe general location and concentration of those surface soil contaminants.

Conducted during the RI, the surface soil sampling shows that most of elevated concentrations ofdieldrin and, to a lesser extent, aldrin and other organic contaminants, occur in areasimmediately east and southeast of Basins A and F. The concentration of those contaminantsdiminishes rapidly to the west of the basins and more slowly in other directions. ATSDRreviewed the areal distribution of dieldrin in surface soils and found that, except for a fewisolated occurrences, the concentration of dieldrin falls below levels of health concern withinabout 1 mile to the east, 1.5 miles to the southeast, 0.5 mile to the south, 0.25 mile to the west, and 0.5 mile to the north of the principal source areas.

Surface soil contaminants detected in the rim of sections surrounding the central 6-section core of contamination may have migrated to those nonsource areas by three principal mechanisms:eolian (windblown) transport, surface water transport, and direct application (of pesticides). Surface water transport has been an important process for the dissemination of contaminants insediments in Onpost creeks and ditches, overland runoff accounts for only localized erosion andredeposition of surface soil contaminants and does not account for the pattern of surface soilcontamination observed in the samples collected (EBASCO, 1992, p. D1-11). Direct applicationof pesticides has been restricted to specific areas and does not explain the observed distributionof those contaminants outside of the sources areas. The observed pattern of contaminantdistribution is consistent, however, with observed wind patterns and eolian transportation.

EBASCO, 1992, Appendix D). See Fig. 3 for general locations.">

Table 16.

Onpost surface soil contamination above comparison values (DLDRN = 0.04 ppm & AS = 0.40 ppm) in the sections near the RMA boundaries. (RMAED run of 10/15/93; EBASCO, 1992, Appendix D). See Fig. 3 for general locations.
LocationAnalyteMaximum Detect (ppm)Comments
T3S, R66W - Southeast boundary area
Sec. 5DLDRN0.0413Concentration of contaminants decreases towards RMAboundary. Contaminants below levels of concern at theRMA boundary.

DLDRN "hot spots" of contamination in the east half ofSec. 6 in the "old toxic storage yards" and rail spur(max 0.0949 ppm).

Sec. 6DLDRN0.23
Sec. 8DLDRN0.0446Isolated detection.
T3S, R67W - South boundary area
Sec. 11DLDRN0.047Isolated detection along Sand Creek Lateral. Concentration of contaminants decreases towards RMAboundary.
T2S, R67W - West and northwest boundary areas
Sec. 34DLDRN0.0711Maximum concentrations measured just west of thecentral six sections of RMA and rapidly decline belowlevels of concern towards RMA boundary.
Sec. 27DLDRN0.0456
Sec. 23DLDRN0.60Max. just north Basin F. Below concern at boundary.
Sec. 24DLDRN0.11Contaminant concentrations decrease progressivelyfrom RMA center to below levels of concern atboundary areas.
T2S, R66W - Northeast and east boundary areas
Sec. 19AS2.89Isolated "hot spots" of AS contamination. Progressivedecrease of other contaminants from RMA center toboundary areas.
Sec. 20AS2.64
Sec. 30DLDRN0.20Max. values detected in west and levels decreaseprogressively towards RMA boundaries.
Sec. 31DLDRN0.30Max. DLDRN value detected in west and levelsdecrease progressively towards boundary. Isolated "hotspots" of contamination in toxic storage yard.

The wind rose data presented previously in Figure 4 discloses that winds at RMA mostcommonly blow from the south or south-southwest at speed less than about 11.5 miles per hour(mph). Winds less commonly blow from the west and northwest, but about 50 percent of thosewinds exceed 11.5 mph. Many of the stronger west-northwest winds are the "chinook" windswhich occur infrequently during the winter and spring when surface soils tend to be moresusceptible to wind erosion and transport.

The average wind speed at RMA for the period 1975-1979 is about 9 mph. While meteorologicconditions certainly vary from year to year, examination of the Comprehensive MonitoringProgram Air Quality Data Assessment Reports for successive years' data shows little variation. For example, during Fiscal Years 1989 and 1990 the mean and maximum wind speeds measuredat RMA were 8.5 and 8.6 mph and 33.2 and 35.9 mph respectively. However, during thoseyears short-term events such as wind gusts showed greater variation from 72.5 and 52.8 mph(Stollar, 1991).

Wind speeds in excess of 7 mph are necessary to erode silt-sized particles and exponentiallygreater wind speeds are necessary to move fine-grained and larger sand sizes (Chorley andothers, 1984). Therefore, strong winds from the west-northwest and other less-common strongwinds from more northerly or southerly directions may have accounted for most eoliantransportation of contaminants (EBASCO, 1992, p. D1-13).

Prior to the Basin F and Basin A IRAs eolian transport appears to have been an importantmigration pathway for soil contaminants at RMA. The extent of past human exposure Onpost tocontaminated surface soils is unknown, but incidental ingestion, dermal contact and inhalationwas possible for RMA workers from 1942 to the cessation of industrial operations in 1982.

The Offpost Endangerment Assessment (HLA, 1992b, p. I-18, Figs. 16 and 17) presentsevidence suggesting a low level of wind-borne OCPs (chlordane, dieldrin) may have beendeposited in the 96th Avenue area, immediately north of RMA (sections 13 & 14, T2S, R67W). The most widespread and frequently detected OCP is dieldrin. Concentrations of dieldrin inOffpost Study Area surface soils range from 0.002 to 25 ppm. However, the highestconcentrations of those pesticides and arsenic occur in an irregular pattern of "hot spots" that isnot consistent with eolian transportation and deposition alone. Offpost pesticide use, naturalvariability of surface soils, and irrigation with well water appear to be major factors influencingthe occurrence of those compounds in the Offpost surface soils.

Based upon the surface-soil contamination data developed for the Offpost Study Area, ATSDRconcludes that the Offpost surface soils have been contaminated by a combination of sourcesincluding eolian deposition, agricultural application of pesticides, and irrigation practices. Wind-borne transport of contaminants from RMA represents an unknown percentage of the lowlevels of contamination detected. Dieldrin and other OCP surface-soil contamination detected in the 96th Avenue area is found at levels no greater than half of those OCP levels found in theadjacent Onpost soils. Elsewhere, the levels of Offpost Study Area surface soil contaminationare generally tens or hundreds of times lower than detected Onpost (HLA, 1992b, p. I-18). While wind-borne transport of OCP contaminants to these Offpost areas may be a minorpathway of human exposure, the levels of surface soil contamination are, with isolatedexceptions, well below levels of health concern and this potential pathway will not be considered further.

It is unknown what level of Onpost occupation exposure to RMA contaminants those workersmay have experienced in the past but the opportunity for direct exposure to those chemicals inthe workplace may have been much higher than the opportunity for Onpost environmentalexposure to chemicals outside the manufacturing, storage and handling facilities. However, aretrospective follow-up study of the mortality experience of 2,384 workers at Shell's plantbetween 1952 and 1982 found an overall mortality rate similar to the Colorado rate and thatthere was no apparent widespread or potent health threat at the plant (Amoateng-Adjepong,Sathiakumar, and others, 1995).

At this time, IRAs have been implemented or completed that address the most significantsources of windblown contamination originating from dust and have contained and subsequentlydestroyed the volatile liquids previously contained in Basin F. Thus, since the termination of allRMA industrial activities in 1982, the areal distribution of surface soil contamination has notincreased and, because of the natural attenuation of many RMA organic contaminants, thesurface soil concentration of those contaminants has decreased.

Beginning in 1984 the Army initiated the remedial investigation process and since that time thepopulation potentially exposed to RMA surface soil contaminants has been principallycomprised of field- and office-oriented remediation workers and managers and, beginning in1989, office and field personnel of the U.S. Fish and Wildlife Service staffing the RMA NationalWildlife Area (to become a National Wildlife Refuge upon completion of clean-up activities andreal estate transfer by and from the Army). Beginning in November 1989 and continuing to thepresent, the public (school-aged children and adults) have been participating in increasingnumbers in USFWS staff- and volunteer-guided, vehicle-based tours in the rim-sections, largelyexternal to the RMA core 6 sections of surface soil contamination. USFWS records show that,since November 1989, over 100,000 people have visited the Arsenal during 1- to 2-hour (bus)wildlife tours, nature walks, educational programs, and special events. Additionally, school and youth group activities brought about 14,000 students to RMA for supervised tours and othersupervised field program activities in FY 91 and an average of about 3419 people visited theEagle Watch Observation area (eastern boundary RMA, Sec. 5, T3S, R66W) during the FY 90,FY 91, and FY 92 winter season (USFWS, 1993b, p. 101-114).

The USFWS use of the RMA Onpost area is regulated by a site-specific Army policymemorandum (July 17, 1991). That policy sets forth an area-wide system of three mapped zonesbased upon surficial soil and other contaminant data (see Fig. 10):

  • "White areas" - No restriction on access or activity except by general regulations and the USFWS. Visitor guidelines and multiple-use coordination requirements exist. Shown as unshaded areas on Figure 10.
  • "Blue areas" - Limited-use access. Visitor access is not restricted if participating in pre-approved activities. Visitor guidelines and coordination requirements apply as above.
  • "Pink areas" - Restricted access and activities. Visitor access limited to guided bustours on hard surfaced roads and fishing at Lakes Mary, Ladora, and Lower Derby. Other access conditional to the approval of a health and safety plan. Visitor guidelinesand coordination requirements apply.

In addition to the tours and events described above, the Arsenal has operated a popular catch andrelease fishing program at lakes Mary, Ladora, and Lower Derby. Those lakes and the USFWSVisitors Center are located in Sections 1 and 2 (T3S, R67W) just south and southeast of theSouth Plants. Here, within the southern end of the core of more highly contaminated sections,the surface soil samples collected near the lakes and from the fields near the USFWS Visitorscenter are found to contain higher levels of contaminants. Surface soil samples collected aroundLower Derby Lake have found frequent detections of dieldrin (max. 0.94 ppm), less frequentdetections of aldrin (max. 0.0952 ppm), and infrequent detections of chlordane (max. 0.63 ppm)and arsenic (max. 3.32 ppm; local AS background < 2.5 ppm). Around Lake Ladora, LakeMary, and the Visitors Center surface soil samples show the frequent occurrence of dieldrin(max. 0.980 ppm) and isolated detections of arsenic (max. 3.58 ppm; local AS background < 2.5 ppm).

Pathway Summary - Onpost Surface Soils and Sediments

A potential past pathway of human exposure to the surface soil and sediment contaminants ofpotential concern given in Tables 9 and 10 may have existed for RMA workers from sometimeafter 1942 until all industrial operation ceased in 1982 and perhaps, for a lesser number ofworkers until 1988 when the Army placed the Arsenal in the inactive list. The North and SouthPlants and most sources of contamination are located within the 6 central sections of RMA as isthe highest levels of surface soil contamination. The nature of the industrial activities conductedhere suggests that most workers had no more than infrequent, incidental exposure to theenvironmental contaminant and the potential for much more frequent, but unquantified,occupational exposure to those same chemicals. This past pathway is not further analyzed in this Assessment.

A potential present and future pathway of incidental human exposure to the surface soil andsediment contaminants of potential concern listed in Tables 9 and 10 may have existed for RMAremediation workers since 1984. Although IRAs have been completed that minimize thepotential for exposure to those chemicals, future remediation measures may create situationswhere remediation workers may be inadvertently exposed for short periods of time to disturbedsurface soils, subsurface soil, dust, and volatile chemicals released from the soils.

Those potential present and, until final Onpost cleanup activities are completed, future exposuresshould be, at most, infrequent incidental ingestion, inhalation, and dermal exposures becausesuch remediation work is planned and implemented to eliminate or minimize such occupationalexposures. To detect possible environmental exposure to contaminants and to monitor the healthof their employees and volunteers, the USFWS (1993c) has implemented a medical monitoringprogram. Additionally, RMA has established a Contingency Plan (RMA, 1993) designed tominimize hazards to human health and the environment from fires, explosions, or any unplannedrelease of hazardous substances or pollutants to air, soil, surface water, or groundwater. Similarcontingency planning is part of the final cleanup activities at RMA as well.

Distinct from the potential pathway of exposure for remediation workers, a potential present and,until final Onpost cleanup is completed, future pathway of human exposure to dieldrin andarsenic (also to aldrin and chlordane, to a lesser extent) at levels of health concern may exist forUSFWS RMA Wildlife Area workers and volunteers and the public that participate in wildlifetours and other activities at RMA (see Table 16). This potential pathway exists for infrequent,incidental ingestion, inhalation, and dermal exposure. USFWS employees, currently totallingabout 40 of which 14 are field-going, have the greatest potential of incidental exposure. Anestimated 50 percent of the field-going USFWS employee's time is actually spent in the field(EBASCO, 1993, p. B. 2-3) and thus, about 110 days per year is estimated as the maximumannual time period available for incidental exposure to contaminants. For those USFWSemployees conducting intrusive soil activities in "Pink and Blue Areas", the potential forexposure is the greatest. But those employees are required to use personal protective equipmentand thus the actual potential for exposure is low and not of health concern. Throughout thepublic health assessment process and RMA remediation, ATSDR will continue to work with theUSFWS to further evaluate the potential for exposure pathways to their field personnel.

The USFWS part-time volunteers (approximately 60 in FY 93) collectively worked about 3.3work-years (USFWS, 1993b, p. 113). Those volunteers have, by virtue of the nature of theirduties at RMA, a very small opportunity for even incidental exposure to contaminants. TheUSFWS personnel that are largely office-bound, are unlikely , by virtue of the indoor nature oftheir jobs and the office and Visitors Center location, to be exposed to surface soil/sedimentcontaminants. This is true for the office-bound Arsenal employees as well.

And finally, of the public visiting RMA on tours and other permitted activities, the group thathas the greatest potential for incidental (ingestion, inhalation, and dermal) exposure to dieldrinand other contaminants is comprised of the people fishing in the lakes in the southern part ofRMA. A survey of angler use at RMA was conducted for USFWS (1993a, p. 117) found thatArsenal anglers average 12.5 days per year. The angler-day averaged 2.4 hours per visit. Thus,an interval of about 30 hours per year is the interval that RMA anglers might be subject toincidental ingestion, inhalation, or dermal exposure to RMA surface soil and sedimentscontaminants of potential concern. Other members of the public may make more frequent use ofthe open-space and wildlife opportunities that the annual use described for anglers but it is notexpected that such regulated or casual use would be permitted in areas where surface soil orsediment contaminant concentrations are higher than those occurring around the lakes.

As discussed earlier in this Assessment, the Conceptual Remedy (Colorado, 1995) and theresultant Onpost ROD (Foster Wheeler, 1996) incorporates 28 specific components which willgenerally govern the conduct of final Onpost cleanup measures. Seventeen of those componentsdeal directly with the excavation of contaminated soils, demolition of Onpost structures, Onpostlandfilling of those soils and debris, and backfilling, capping, and grading with clean soils. Those components also incorporate provisions for protective controls to minimize emissions andodors and to ensure the protection of the public health of the nearby neighborhoods during theextended period of remediation activities. For these reasons, ATSDR concludes that there is alimited potential for off-site human exposure to RMA contaminants during the final remediationactivities at RMA. Given the location of the contaminated soils and sediments at RMA and thesafe-guards that will be in place during remediation, incidental, short-duration exposure(s) tolow-levels of contaminants is all that can be reasonably foreseen.

Potential Ambient Air Pathway

The Remedial Investigation Summary Report and the 1985-mid 1993 data recorded in theRMAED document on- and off-post air quality monitoring at RMA. ATSDR has reviewed thedata and determined 14 chemicals to be of potential concern. Those compounds are volatilehalogenated organics (VHOs) 1,2-dichloroethane (12DCLE), carbon tetrachloride (CCL4),chlorobenzene (CHC6H5), chloroform (CHCL3), methylene chloride (CH2CL2), andtetrachloroethylene (TCLEE); the volatile aromatic organic (VAO) benzene; the organosulfurcompounds - herbicide related (OSCHs) chlorophenylmethyl sulfone and sulfoxide (CPMSO2 &CPMSO); the organochlorine pesticides (OCPs) aldrin (ALDRN), chlordane (CLDAN), anddieldrin (DLDRN); and the elements arsenic (AS), mercury (HG), and cadmium (CD). Of thosecontaminants, only 5 compounds (12DCLE, CHCL3, C6H6, AS, and CD) have been detected inthe off-post air at levels potentially of human health concern (see Table 11).

The Onpost remedial investigation data (EBASCO, 1992) reviewed by ATSDR suggests non-RMA sources may be wholly or partially responsible for the contaminants detected. The TRI(1994) database records that, in 1988, 30 facilities in close proximity to RMA reported 96chemical releases. The compounds released included VHOs, VAOs, PAHs, and metals. Thedata recorded in the most-recent available 1991 inventory differs very slightly in detail from the1988 data, with the exception that the total volume of releases to the atmosphere is reduced toabout 54 percent of the 1988 emissions. The TRI chemical inventory data establishes that thereare nearby, off-site sources of atmospheric contaminants which correspond to all of thosechemicals noted as contaminants of potential concern in the Offpost Study Area (see Table 11),with the exception of the VHOs 12DCLE, CLC6H5, CH2CL2, and TCLEE.

Prior to the Basin F and Basin A IRAs, volatilization from surface water and soils accounted forlow but detectable concentrations of volatiles and semivolatiles in air peripheral to these basins.IRAs have been instituted to mitigate these low-level releases (EBASCO, 1992, Appendix C),and they have addressed the most significant sources of windblown contamination originatingfrom dust and have contained volatile liquids from Basin F. Although past disturbances ofcontaminated soils and releases of VOC soil contaminants occurred during implementation ofIRAs and may occur as a result of future remedial actions, future remediation activities mayresult in an increased potential for volatilization, suspension, and transport of contaminationaway from source areas. Those effects can be controlled through the use of adequateengineering and management measures. Air monitoring, which is currently part of the CMP, willcontinue to assess air impacts at RMA.

Public concern about off-post air pollution was generated during the initial phases of the Basin FIRA (May 1988 - December 1988) and subsequently with the testing and March 1993 start-up ofthe Submerged Quench Incinerator (SQI) designed to destroy Basin F Liquids. Air monitoringfor potential off-post human exposure (Irondale community) to airborne Basin F contaminantsand subsequent health studies has been inconclusive. Comparison of the quantitative datagathered Onpost with largely qualitative data gathered in the Irondale and Commerce City areasyields a plausible explanation for the odors noted and physical symptoms experienced by someIrondale residents. However, the data will only support the possibility of short-term (less than24 hour) episodes of acute exposure to organochlorine pesticides (OCPs) and lesser quantities ofother VOCs (ATSDR, 1989). ATSDR-funded health studies by Colorado Department of PublicHealth and Environment and others were initiated following this possible Basin F exposureincident. The findings of these ongoing health studies of this potential past pathway of humanexposure to airborne contaminants is discussed in the Background and Public Health Implicationsections. Because health studies of this past pathway are ongoing, this pathway will not befurther analyzed in this Assessment.

Pathway Summary - On- and Offpost Air

A potential pathway of human exposure to air contamination can be attributed to both RMA andnon-RMA sources. In the past, chemical manufacturing activities conducted at RMA, includingthe storage and disposal of contaminants and the subsequent remedial measures taken to controlthose contaminant releases, may have resulted in levels of air contamination and consequenthuman exposure to those contaminants at levels that may be approximated by the levels given inTable 11. The available data presented in the Remedial Investigation reports and in the HealthConsultation prepared by ATSDR (1989) strongly suggests that, aside from unknown levels ofRMA occupational exposure, past short-term human exposure to airborne contaminants ofconcern may have occurred during the 1988-89 Basin F remediation episode. During thisepisode, remediation workers may have been exposed to contaminants at levels approximated bythose values given in Table 11.

Offpost residents in the Irondale area may have been exposed to unquantified levels of thoseairborne contaminants during the Basin F remediation interval. Given the prevailing windpatterns and the available monitoring data of contaminant concentrations at the RMA northernboundary, it appears that those off-site residents may have experienced unquantified, short-term,acute exposure to organochloride pesticides (OCPs) and additional exposure to the quantifiedlevels Offpost contaminants of concern identified in Table 11. Offpost sources of aircontamination may be responsible for some the of the detected contamination.

The documented performance of the now completed SQI remediation project indicate that nopathway of Onpost or off-post human exposure to the air contaminants from SQI was created bythis project. Continuous (24-hour) monitoring of SQI is was conducted at the Arsenal andremote monitors were installed at the offices of CDPHE and EPA to ensure safety and oversightof this incinerator.

A potential future pathway of human exposure to airborne contaminants may arise from the soilexcavation, structure demolition, and Onpost landfilling activities to be conducted at RMA. Inaddition to the human health protective provisions of the RMA Contingency Plan (RMA, 1993),the final cleanup activities at RMA will be implemented with safe-guards that will prevent allbut incidental exposure to contaminants at levels that would result in adverse health effects. Thepotentially exposed population is comprised of Onpost remediation workers, USFWSemployees, and the public using the RMA National Wildlife Area. The observed patterns ofsoil contamination at RMA strongly suggest that air-borne contaminants accidentlyreleased during final site cleanup activities are not likely to result in offsite contaminationof nearby neighborhoods at levels that could result in harmful health effects.

Both the Onpost and Offpost RODs (Foster Wheeler, 1996; HLA, 1995) incorporaterequirements for environmental monitoring and requirements to protect public health in nearbyneighborhoods during remediation.

Potential Foodchain Pathway

Game and Fish

Contamination of resident game species on RMA is documented in many sources including theBiota RI (ESE, 1989b) and the Integrated Endangerment Assessment/Risk Characterization(EBASCO, 1994). Investigation of Offpost Study Area biota contamination detected low levelsof contamination in both wildlife and farm animals sampled (see Table 12A).

The movement of game species of wildlife from RMA into off post areas comprises a potentialpathway of migration for the higher-levels of biota contamination found on RMA into the offpost environment. Off post hunting of pheasants is documented north of RMA, hunting ofmourning doves southeast of RMA, and movement of deer across the north, and east boundariesof RMA is also recorded. Generally, the quality of the wildlife habitat is better Onpost than innearby areas beyond the RMA boundary fence (ESE, 1988; HLA, 1992a; EBASCO, 1993). Thus, species that do not range far may have little incentive to leave the confines of RMA.

Information on the home range size (land area occupied by an individual animal or group ofthose animals that has the habitat components necessary to sustain the animal) of pheasants andcottontail rabbits was determined. Although the sample sizes were small (14 pheasants and 16cottontails) and the duration of the radio transmitter monitoring study was only about one year,the data obtained on the home ranges size observed for RMA pheasants (11 to 278 ac.) andcottontails (2.3 to 12.6 ac.) is within the range of the home range sizes reported from studiesconducted elsewhere in the nation (ESE, 1988, p. 7-1 - 7-10).

Pheasants were occasionally observed flying to and from RMA along the northern boundary, andhunters were observed along the northern boundary in November, 1985. No seasonal patternwas apparent upon examination of the fall and winter telemetry data. The greatest distancetraveled by monitored pheasants was 1.4 miles but the average distance traveled seemed toincrease with increasing numbers of observations. For pheasants with more that 20 observationsthe average distance traveled was 0.86 miles. However, a local hunter did return a RMApheasant leg band which he claimed was from a pheasant he'd killed about 7.5 miles northeast ofRMA (sec. 13, T66W, R1S). It was concluded that the pheasant home range size determinedprobably represents the minimum home range size (ESE, 1988). The home range size anddistance traveled is large enough to permit pheasants exposed to known sites of Onpostcontamination to move into off-post areas where they could be hunted. If the true home rangesize is relatively small (about 1.5 miles across) then RMA-resident pheasants found in off-postareas must reside in the outermost, least contaminated rim of section, just inside the RMAboundary. Alternatively, if local area pheasants occupy a large home range (about 7.5 milesacross), the individual birds could access the most contaminated sites at RMA, but may spendless time in the contaminated portion of their larger habitat.

None of the cottontail rabbits that were equipped with radio transmitters during the Offpost RI study were observed in the off-post areas even though the rabbits selected for study were alllocated within 0.75 miles of the north RMA boundary. The small home range size determined(ave. 6.5 ac) and the short distance traveled (ave. 0.17 mi) suggest that only individualsinhabiting the least-contaminated border area of RMA might move into the off-post areas.

Of the 27 mule deer and 10 white-tailed deer collected by April 1991 for sampling on RMA, the only detectable levels of any of the contaminants for which analyses were conducted wasdieldrin (ESE, 1988; EBASCO, 1993). Three detections of dieldrin in mule deer liver arereported (0.101 - 0.281 ppm) in the results amassed in the Offpost RI (ESE, 1988), the OffpostRI Addendum (HLA, 1992a), and the Onpost Integrated Endangerment/Risk Characterization(EBASCO, 1994). Those reported levels are below the 0.3 ppm FDA Action Level. The muledeer with the highest concentration of dieldrin in the liver tissue was collected between Basin A and the South Plants near the center, and most contaminated portion of the Onpost area (ESE,1988, p. 7-11).

The USFWS conducted a deer-health/contaminant study in 1993 and published their findings inthe 1994 Annual Progress Report (USFWS, 1995). Eighteen radio-collared deer were collectedbased upon their home ranges. Although dieldrin was detected in fat (organ fat), liver, and braintissue, no dieldrin was reported in muscle tissue, even from deer that frequented the South Plant and Basin A areas.

RMA has had an active fishing program since the 1960s. Currently fishing permits are sold on arandom lottery (prior to 1993, on a first-come, first-serve) basis for 500 anglers. Spouses andfamily members under the age of 18 are able to fish on one permit. Because contaminants havebeen detected in RMA fish (see Table 12A), fishing is catch and release only and that regulationis strictly enforced. Fishing season dates are set yearly and depend upon the bald eagle usage ofthe Arsenal lakes, but generally run from early April through late September or early October. Additional permits are issued to military and Arsenal personnel, senior citizens, and physically-challenged individuals (eg. 143 permits in 1991). During FY 91, about 2012 people fished atRMA (USFWS, 1993). Despite this active sport-fishing program, no pathway of humanexposure is likely because the catch and release fishing program (no consumption) is effectivelyimplemented and aggressively enforced .

Tri-County Health Department personnel conducted a survey wild game hunting and hunter-success in the one-mile-wide belt of private lands surrounding the Arsenal on the northwest,north, and east boundaries (TCHD, 1994). Those surveys were conducted during the Fall of1988 and the Winter of 1989 disclosed that pheasant hunting to the northeast of RMA resulted inonly 7 birds taken for a success ratio of about 1 bird per 21 hunters. To the east of RMA thesuccess ratio was of about 1 bird per 10 hunters. However, one hunter was identified thatfrequented this area and had harvested about 12 pheasants per season. He was also responsiblefor taking over 100 doves, one deer, and one antelope. This individual and his family had thegreatest potential for consumption of any RMA foodchain contaminants that may be present inthese off-post sites.

The data given in Table 12A indicates that edible tissue of water fowl and mourning doves has,in the past, carried elevated levels of the pesticides aldrin, dieldrin, and endrin. The results ofthe 1994 Biomonitoring Program (USFWS, 1995) are given in Table 12B. Those data indicatethat in 1994 the edible-tissue body burden of those pesticides was within FDA Action Levels.Therefore, human consumption of those species with those contaminant levels should not resultin adverse health effects. The USFWS is continuing this monitoring program in their continuingeffort to monitor the general health of the RMA Refuge ecosystem.

Vegetable Produce

ATSDR initially disclosed in the public comment draft of this public health assessment(ATSDR, 1995, p. 145) that a potential pathway of human exposure to vegetables grown incontaminated surface soils and irrigated by contaminated groundwater may have existed in thepast and may exist again in the future on certain lands within the Offpost Study Area. Of thelands originally thought to have the highest future potential were the Offpost lands bounded bythe Burlington Ditch/O'Brian Canal and the north RMA boundary in the First Creek area. Thoselands are characterized by the combination of both higher levels of surface soil and groundwatercontamination. Acquisition of those rural, residential properties by Shell Oil Company in 1989-90 and the conversion of those lands to non-residential uses ended the possibility of humanexposure to vegetable produce grown on those lands and potentially intention to manage thoselands so as to preclude any potential future residential use of those land was affirmed in aSeptember 10, 1991 letter (Shell, 1995). The subject properties were previously served byArapahoe Aquifer wells and thus, groundwater was never a factor in past potential foodchaincontamination on these lands.

Elsewhere in the Offpost Study Area surface soils are generally uncontaminated but, AlluvialAquifer wells are used for garden irrigation, non-food crop irrigation, or livestock watering. A summary of Offpost irrigation well samples which, as of April 1994, had elevated contaminantlevels is given in Table 17. Data has been presented previously in this Assessment that shows declining trends for many groundwater contaminants.

In October and November 1995 samples were collected of vegetables grown in the OffpostStudy Area of potential concern and irrigated with well water produced from the AlluvialAquifer. Those vegetables were analyzed for their DIMP content to evaluate the potentialuptake of that contaminant in area vegetables and to provide data to compare with thecontaminant levels predicted in the endangerment assessment prepared for the Offpost OperableUnit (HLA, 1992a). The samples were collected from the garden of private residence north ofRMA on October 12, 1995 and from two large vegetable farms on

Table 17.

Detections of irrigation well contaminants at levels of potential concern:
Alluvial/Unknown Aquifer, Offpost Study Area, as of April 1994 (a).
Well # Detected Value
> CV (b)
Sampling Date
Well Use
(GI, NFI, L)(d)
Volatile Halogenated Organics - (VHOs)
1,2-Dichloroethane (12DCLE) (CV = 0.4 ppb CREG; 5 ppb MCL)
1068A4.5893273GI, L
Tetrachloroethylene (TCLEE) (CV = 0.7 ppb CREG; 5 ppb MCL)
1068A14.093723GI, L
1189A14.792122GI, NFI
Organochlorine Pesticides (OCPs)
Dieldrin (DLDRN) (CV = 0.002 ppb CREG)
1189A0.06292122GI, NFI
ICP Metals / Anions & Cations
Arsenic (AS) (CV = 0.02 ppb CREG; 50 ppb MCL;
          Natural Background Level <2.5 ppb)
Nitrate (NIT) (CV = 10,000 ppb MCL)
986A12,00091289GI, NFI
    (a) RMAED database runs of 04/26/94 and 05/04/94; and Tri-County Health Dept. database queries of12/20/93 through 05/11/94).

    (b) > CV = greater than or equal to the Comparison Value for drinking water (see the EnvironmentalContaminants and Other Concerns section of this Assessment for further discussion).

    (c) yr/day = the year followed by the chronologic, numerical date (eg. "94109" is April 19, 1994).

    (d) The uses of wells in the RMA Offpost Study Area were inventoried by the Tri-County Health Dept. andincorporated in their database of Offpost wells. The symbols used are "GI" for garden irrigation, "NFI" for non-food irrigation, and "L" for livestock watering.

November 10, 1995. The private garden was selected by the Army because it is irrigated onlywith water from well 1178B which has higher concentrations of DIMP remaining than any otherOffpost area well. The concentration of DIMP in that well have ranged from about 150 to 950ppb. The commercial farms were selected because they utilize a combination of groundwater(DIMP <20 ppb) and surface water (uncontaminated by DIMP).

The vegetables were analyzed using a recently developed method to determine DIMPconcentrations in vegetables (detection limit = 20 ppb). The concentrations of DIMP detected inthe private garden cucumbers ranged from 340 to 920 ppb and the tomatoes from that gardencontained DIMP at an unquantifiable level below the detection limit. The vegetable samples(corn, onion, chile pepper) taken from the large vegetable farms contained no detectable DIMP.

Comparison of this 1995 garden vegetable data with the calculation of potential risk given in theOffpost Endangerment Assessment (HLA, 1992b) indicates that even under a "worst-case "exposure scenario, the consumption of vegetables irrigated with DIMP-contaminatedgroundwater is not expected to result in adverse health effects. This conclusion is basedupon the estimation of the potential DIMP concentration in lettuce. Lettuce was selected toevaluate the potential risk associated with consumption of Offpost-grown vegetables becausestudies have shown that plant leaves have the highest potential to accumulate DIMP. Other"worst-case" assumption included in this evaluation were a 900 ppb concentration of DIMP ingroundwater, a high ingestion rate 0.125 kg/day), and daily ingestion for 30 years. Based uponthese conservative assumptions the calculated associated risk, expressed as a hazard quotient, is 0.2 (level of concern is 1.0).

The provisions of the Conceptual Agreement (Colorado, 1995) and Offpost and Onpost RODs(HLA, 1995; Foster Wheeler, 1996) specify that the existing drinking water wells in the OffpostStudy Area within the "DIMP footprint area" (DIMP >0.392 ppb) will be replace by municipalwater system or a suitable alternative. Thus, the use of small irrigation wells may also be greatlyreduced or eliminated. Moreover, the continued functioning of the BCSs and the groundwaterintercept and treatment system north of RMA will continue to lower contaminant levels in theOffpost Study Area groundwater. In addition, institutional controls will be utilized to preventdomestic use of Offpost contaminated groundwater and to discourage the development of newdomestic wells in areas of contaminated groundwater.

The provisions of the RODs coupled with the low-levels of DIMP found in the vegetablesamples analyzed in 1995 eliminate a present or possible future vegetable produce pathway fromfurther consideration.

Pathway Summary - On- and Offpost Foodchain

A potential pathway of off post human exposure exists for the consumption of pheasants huntedin the nearby off post areas to the north and east of RMA. Pheasant in those areas are known tofrequent or inhabit RMA. Dieldrin has been detected in pheasant tissue (0.02 - 5.95 ppm). TheFDA Action Level for dieldrin is 0.3 ppm. Although dieldrin has not been detected in pheasantsamples collected in the Offpost Study Area, the size of the pheasant's home range is sufficientto permit pheasants to be exposed to RMA sources of contamination and to travel to off postareas where they may be legally hunted and consumed. The home range data gathered suggeststhat pheasants frequenting the off post areas adjacent to RMA may also be those pheasants thatinhabited the outer rim of least-contaminated sections at RMA. Thus, while a potential pathwayexists for the consumption of dieldrin-contaminated pheasant tissue, it is likely that if dieldrin ispresent in the tissue of some unknown percentage of pheasants hunted near RMA, it is present atthe lower end of the observed range of values detected thus far. The study conducted for theOffpost RI concluded that both waterfowl (eg. mallard ducks) and mourning doves migratedwidely and thus, even if those species were contaminated by RMA sources, their number wouldbe so diluted by the greater number of birds not frequenting RMA, that the potential for multiple exposure by consumption would be small (ESE, 1988, p. 2-30).

Historically, dieldrin is detected in mallard tissue in the range of 0.01 to 4.53 ppm, whichexceeds the 0.3 ppm FDA Action Level. However, sampling conducted by USFWS in 1994found a maximum of only 0.068 ppm dieldrin in the edible muscle tissue of the duck and Canadagoose samples analyzed (see Table 12B; USFWS, 1995). Qualitative data presented in theOffpost RI reports suggests that the local area duck population may be large enough that it islikely that a hunter would only take an occasional RMA bird and thus, consumption ofwaterfowl does not appear to be a significant off post pathway of human exposure to RMAcontaminants.

Samples of off post mourning dove have not been collected but those samples collected Onpostindicate the doves are frequently burdened with high levels of a number of pesticides andmercury. As shown in Table 12A, mourning doves sampled have been found to contain mercury(0.05-0.40 ppm), aldrin (0.0227-1.3 ppm), dieldrin (7.8-32.0 ppm), endrin (0.14-1.3 ppm), DDT(0.04-0.308 ppm), DDE (0.04-0.942 ppm), CPMS (0.3-1.46 ppm), and CPMSO2 (0.22-2.53ppm). Of those detections, the ranges of values reported for the organochlorine pesticides(OCPs) aldrin, dieldrin, and endrin all exceed the respective FDA Action Level of 0.3 ppm foreach pesticide. In order to more fully evaluate contaminant uptake in mourning doves and thepotential health and ecosystem risk that uptake might pose, the USFWS collected additionalcontaminant accumulation data on this species during 1994 (USFWS, 1995, p. 1-91). A total of12 birds were collected and samples of brain, liver, and muscle tissue analyzed for OCPs,chlordane, DDE, AS, and HG. There were no detections of aldrin, endrin, isodrin, chlordanes,or arsenic. The results of the other tests are summarized in Table 12B.

Dieldrin was detected in 6 of 12 doves tested, but of those positive contaminant detections onlyone sample of liver tissue at 0.533 ppm dieldrin was above the FDA Action levels. Ediblemuscle tissue of those dove tested did not contain contaminant concentration above FDA ActionLevels and the indication is, based on the 1994 data, that human consumption of mourningdoves taken in the off post areas is not likely to pose a health risk.

The nearby Denver International Airport opened in 1995 and increasing levels of urban andcommercial development are already beginning in the areas north, east, and southeast of RMA. As these areas develop, it is likely that off post habitat quality for species such as pheasants maybegin to decline, as will hunting pressure. On this basis, it is likely that a potential present orfuture pathway for the hunting and consumption of contaminated pheasants will not be a viablepathway for long.

Potential Groundwater Pathways

Potential Irondale/Western Plume Group Pathway

Potential Nitrate Contamination

In addition to the contaminants of potential concern identified in Table 7, ATSDR considersnitrate (NIT) to be a contaminant of potential concern in the Irondale/Western Plume Grouppathway. The results of routine inorganic water quality sampling of SACWSD wells showed ahigher than expected level of nitrate concentration in District Wells Nos. 2, 3, 5, and 17. Because nitrate was detected as high as 9.7 ppm (10 ppm is the EPA MCL for this compound) inthose samples, SACWSD initiated a study to evaluate the possible source(s) and concentration ofnitrate in the upgradient plume of groundwater flowing towards the system in theIrondale/Western paleochannel. In a report prepared by HRS Water Consultants (HRS, 1993) acomplex plume of nitrate contamination was delineated that flowed onto RMA from the south. Nitrate concentrations at the boundary between RMA and the former Stapleton InternationalAirport exceed 30 ppm and concentrations in the range of 15-20 ppm are interpreted in the mainpaleochannel and within 1 mile of the of the affected municipal wells.

ATSDR reviewed a compilation of nitrate analyses conducted by the Klein Water TreatmentFacility between January 4, 1993 and August 22, 1994. That summary forwarded to ATSDR byRMA personnel (Kilgannon, 1996, personal communication) documents that during thespecified time interval, nitrate levels in wells numbered 2, 3, 5, 14, 16-18, 21, and 47 fluctuated between 4.83 to 12.9 ppm.

Evaluation of the alternative measures to ensure that water supplied to the SACWSD municipalsystem remains below the EPA drinking water standard of 10 ppm is underway.

Thus, a potential pathway of human exposure to groundwater contamination in excess of thefederal drinking water standards exists for the users of the SACWSD municipal water systemsupplied by water finished by the Klein Water Treatment Plant. This pathway will becomecomplete for the more than 30,000 users of this municipal water system if the levels of nitrate in groundwater in the Irondale/Western Plume Group continue to increase and the operators of the treatment facility cannot maintain the level of nitrate below the EPA MCL of 10 ppm.

Potential VHO contamination

A potential present and future pathway of human exposure to the contaminants of potentialconcern listed in Table 7 may exist for the people at the 10 or fewer residences or businesses inthe Commerce City "EPA Study Area" south of 88th Avenue that continue to use privateAlluvial Aquifer wells for domestic, tap water uses and for persons at residences that continue touse the Alluvial Aquifer wells for garden and lawn irrigation. Because current usage of thoseAlluvial Aquifer drinking water wells in the EPA Study Area is not verified, current exposure tocontaminants cannot be verified and this pathway is relegated to a potential pathway of humanexposure. The concentration of contaminants to which those people may be exposed is less thanor equal to the maximum concentrations expressed in Table 8. The total duration of exposurefor those people is unknown but may have begun as soon as occupancy was established in thearea and may continue to the present time.

Because most, if not all, groundwater contamination found south of 88th Avenue is not fromRMA sources this potential pathway will not be analyzed further in this Assessment.

Potential Denver Formation Aquifer Pathway

Contaminants have been identified in the confined aquifer zones of the Denver Formation,although contaminant plumes have not been delineated. The contaminants have been identifiedin Denver Formation monitoring wells drilled the area overlain by the Northern and First Creekpaleochannels of the Northern Plume Group. Unlike the unconfined flow system, pathways inthe confined flow system in this aquifer are difficult to identify and generally consist of fine-grained, highly-discontinuous sand lenses and lignites separated by fine-grained siltstones andclaystones. Permeability and corresponding groundwater flow rates are significantly less than in the alluvium of the unconfined flow system (EBASCO, 1992, p. 2-30). Both upward anddownward groundwater flow have been detected in monitoring wells penetrating the DenverFormation in the Offpost area immediately north of the NBCS and the RMA boundary (ESE,1988, p. 3-45). This indicates that, at least locally, contaminants in the Alluvial Aquifer canmigrate into the more permeable beds of the Denver Formation. The dominant horizontalgroundwater flow direction in the Denver Aquifer is to the north-northwest (ESE, 1988, p. 3-35).

The compilation of previously collected data from 16 Denver Formation wells summarized inthe RISR (EBASCO, 1992) indicates that the Denver Formation detections do not appear to bespatially related. However, confined flow system detections generally correspond spatially withplumes in the overlying unconfined flow system. These data suggest that, locally, contaminantsmigrate from the unconfined flow system to the confined flow system. The presence ofdetections at varying depths in the Denver Formation indicate that vertical migration ofcontaminants does occur within this aquifer. Although contamination has migrated from theunconfined flow system to the confined flow system within the Denver Formation, the degreeand extent of confined flow system contamination, and its rate of migration, generally aresubstantially less than in the unconfined flow system. In isolated locations, however,contaminant concentrations in the confined flow system may approach or exceed those in theoverlying unconfined flow system.

It was speculated (EBASCO, 1993, p. 2-31) that some organic compounds such as benzenecould be naturally occurring compounds that have migrated from coal or hydrocarbon depositslocated elsewhere in the Denver Formation.

Fourteen Offpost monitoring wells located north of the NBCS in section 13 and 14 (T2S, R67W)were sampled for the CMP during the Fall 1989 and Spring 1991 sampling rounds (see HLA,1992b, Fig. 15). The organic compounds detected were benzene, chlorobenzene, chloroform,DIMP, DBCP, phenol, and 111TCE. The most frequently detected organics were DIMP (0.443-46.0 ppb; 26 % of samples collected), chloroform (0.631-1.30 ppb; less than 10 % of samplescollected), and chlorobenzene (1.10-51.5 ppb; less than 10% of samples collected).

Pathway Summary - Denver Formation Confined Aquifer

Based upon the available data, contaminants, at levels of potential concern, have not beendetected in Offpost Study Area, Denver Formation domestic wells. The geologic data compiledduring the preparation of the Offpost RI (ESE, 1988, p. 3 - 34-48) infers that a very limited areaof the Offpost Study Area, largely in the area surrounding the Offpost groundwater intercept andtreatment facility (IRA-A) is potentially suitable for the development of a Denver Aquifer well. If any new wells are drilled in this area, it is likely that they would be drilled to the ArapahoeAquifer to assure adequate water volumes for domestic use. Thus, no future Denver Aquiferwells are foreseen for the area.

A significant percentage of this area is currently owned by Shell Oil Company and subject totheir controls and the institutional control provisions of the Offpost ROD (HLA, 1995). Thosecontrols include the provision that will prevent the use of groundwater with contaminant levelsgreater than the remediation goals adopted in the ROD. On this basis, no pathway of humanexposure is identified and it is very unlikely that a pathway of human exposure to groundwatercontaminants in the Denver Formation will be created in the future.

Table 18.

Potential Present and Future Exposure Pathways.
Pathway NameCompoundsExposure Pathway ElementsTime
SourceMediaPoint of
Route of
Analyte Groups:
* Onpost: Core
* Onpost: Outer
- OCPs, AS
Onpost soils in
the vicinity of:

- South Plants
- North Plants
- Basins
- Ditches, etc.
Onpostsurfacesoils &
Surface soils
in the "Pink
Zone" (see
Fig. 10)
inhalation, and
dermal exposure
* Workers:
- USFWS (~40
people) &
* People
recreating at
RMA Refuge

(until finalcleanup isdone)

* Confined Denver
Aquifer pathway

* Irondale/Western
Plume contamination

* Denver Aquifer
- VHOs
* Irondale/Western
- VHOs
* Offpostgroundwater inDenver Fm.
* Largely non-RMA sources
* unknownwells
* Pvt. wellsin EPA SA& poss.Klein Plant
Ingestion,inhalation, anddermal exposure* Unknown

* <10 pvt. wells& poss.municipalwater; ~30,000people

Analyte Groups:
- VHOs
- VAOs
- OCPs
- Anions/metals
Windblowndust andvolatilesgeneratedduringremediationmeasuresAir (?)UnknownInhalation,

* Peoplevisiting Refuge
FOOD CHAINAnalyte Groups:
* Hunting
- OCPs
* Offpost
* Fish/
* Hunting onnearbyoffpost areasIngestionUnknownUnknown

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. #