Sunteți pe pagina 1din 15

.., JAMES M INHOFE, OKLAHOMA.

CHAIRMAN

JOHN W.WARNER,VIRG!NIA JAMESM JEFFORDS. VERMONT


CHRISTOPHER S. BOND, MISSOURI MAX BAUCUS. ,'JiONTANA
GEORGE V VOINOVICH, OHIO HARRY REID. NEVADA
MICHAEL D. CRAPO,IDAHO BOB GRAHAM. FLORIDA
LINCOLN CHAFEE, RHODE ISLAND JOSEPH I. LIEBERMAN. CONNECTICUT
JOHN CORNYN, TEXAS BARBARA BOXER. CALIFORNIA
LISA MURKOWSKI, ALASKA RON WYDEN. OREGON
CRAIG THOMAS, WYOMING THOMAS R. CARPER. DELAWARE
WAYNE ALLARD, COLORADO HILLARY RODHAM CUNTON. New YORK
COMMITTEE ON ENVIRONMENT AND PUBLIC "VORKS
ANDREW WHEELER, MAJORITY STAFF DIRECTOR WASHINGTON, DC 20510-6175
KEN CONNOlLY,MINORITY STAFF DIRECTOR

September6, 2006

The HonorableStephenL. Johnson


Administrator
U.S. EnvironmentalProtectionAgency
Ariel Rios Building
1200PennsylvaniaAvenue,NW
Washington,DC 20460

Dear AdministratorJohnson:

Weare writing to you regarding the results of recent collaborative studies conducted by
the City of Austin, Texas, and the United StatesGeological Survey (USGS) that have identified
coal-tar based sealcoat-the black, shiny surface often applied to asphalt pavement -as a
significant and previously unrecognized source of extremely elevated concentrations of
polycyclic aromatic hydrocarbons (PARs) in streams.PARs are suspectedhuman carcinogens
and are toxic to aquatic life. We believe that the findings of these studies have major
implications for the City of Austin and the rest of the country becausethese sealantsare used
nationwide.

The studies show that runoff from parking lots sealed with coal tar-based sealanthad
PAH concentrations 65 times higher than concentrations in runoff from unsealedparking lots. In
the greater Austin area, an estimated 660,000 gallons of coal-tar sealantare applied annually.
PAHs are increasing in Town Lake sediments, which receives drainage from many of Austin's
urban creeks. (The attached articles contain specific information about the levels of PAHs
detectedin selected Austin streams.) Biological studies conducted by the City indicate that these
sealantsare toxic to aquatic life at levels found in local waterways and are degrading the health
of Austin's creeks, as indicated by a loss of speciesand decreasednumbers of organisms. As a
result of these findings, the City banned the use of coal-tar sealants in November 2005.

According to studies conducted by the USGS of38 reservoirs and lakes sampled in 18
metropolitan areasacrossthe nation from 1996 -2001, PARs in lakes and rivers are increasing.
The biggest increasesare in areaswith watershedsaffected by urban sprawl. For example, PARs
increasedten-fold in Lake in the Hills (suburban Chicago, Illinois) as the watershed rapidly
developed. This infonnation raises important local and national policy questions about the use of
sealantsand methods to prevent contaminated runoff from reaching urban water bodies.

In 1992, the EPA excluded coke product residues,including coal tar, from classifications
as hazardouswastes if they are recycled. As a result, under the Resource Conservation Recovery

PRINTED
ON RECYClED
PAPER
Act (RCRA), these coal-tar sealantsare consideredproducts that contain recycled coal tar and are
therefore not regulated. In the July 26, 1991 notice of proposed rulemaking (56 FR 35758) that
discussedthis recycling exemption for coke product residues that are blended with coal tar
product that is sold, EPA stated: "For thesereasons,the Agency believes that reinsertion of these
residuals into coke ovens and mixing of theseresiduals with coal tar to be sold as a product are
recycling practices that do not increase the levels of hazardousconstituents in the final coke by-
product, and therefore do not pose any significantly increased risk to human health and the
environment." In light of the studies conducted by the USGS and the City of Austin that show
that coal-tar based sealantscontribute to PAH contamination in urban and suburban water bodies,
we request that the EPA revisit this rulemaking and determine whether coke product residues
blended with coal tar and sold as a product should be regulated under RCRA.

Additi~nally, we would like your responses


to the following questions:

1. What action is the EPA taking to investigate the application of coal-tar based sealantto
asphaltas a major source ofF AH contamination in water bodies acrossthe country?

Pleasespecify the EPA's current strategy for controlling PAHs in urban environments.
How will identification of this new source ofF AH contamination influence future
strategies?

3. How is the EPA working with its Regional Offices, the States, stonnwater liaisons at the
federal and local levels, and trade associationsto advise communities about this research,
its potential implications for aquatic wildlife, and the existence of safer sealant
alternatives?

We urge the EPA to perfonn a national study and report its findings on how coal-tar
based sealantsincrease PAHs in water bodies, and the effects of these PAHs on human health
and the environment. We appreciate your immediate responseto this matter.

Sincerely,

~ ~ 2.s~~~~~~::~~-
"'-
JohnWarner

Attachments:
(1) The Coal Tar Facts,www.citvofaustin.orl!
(2) Van Metre,P.C.;Mahler, B.J.; Bashara,T.J.; Wilson, J.T.; Johns,D.A. ParkingLot Sealcoat:
An UnrecognizedSourceof UrbanPolycyclic AromaticHydrocarbons.
Environ. Sci. Techno/.2005,39, 5560.
(3) Parking-LotSealcoat:A Major SourceofPAHs in Urban and SuburbanEnvironments,U.S.
GeologicalSurvey,CongressionalBriefing, December2,2005.

2.
What are coal tar sealants? What are the alternatives?
Coal tar sealants are surface finishes for parking lots, Asphalt based sealants are a comparable alternative offering
driveways and airports. They contain varying concentrations a level of protection similar to coal tar sealants with lower
of coal tar depending on product and formulation. I~vels of PAHs. Sealant products* that do not contain coal
tar include:
What's the problem?
Coaltar sealants contain extremely high levels of PAHs RETAIL:
(PolycyclicAromatic Hydrocarbons). PAHsare a group of Henry PM2000 Premium Driveway sealer/Filler
chemicals formed during the incomplete burning of coal,
Henry Elastomeric Emulsion Crack Filler
gasoline, wood, garbage or other organic substances,such
astobacco and charbroiled meat. Becausethe sealants
COMMERClAUWHOLESALE:
wear off the asphalt surface,recommendations call for
reapplying them every two to three years. An estimated Paveshield
660,000gallons are applied annually in the Austin area. As Jennite Asphalt Emulsion Pavement Sealer
the sealantswear off pavement surfaces,stormwater washes Gilsonite Asphalt Driveway Sealer
the particles into local waterways. PAHsare increasing in
Town lake sediments,where many of Austin's urban creeks For updates see:
drain to. City staff has documented that coal tar sealantsare www.cit-/ofaustin.org/watershed/coaltar_altproducts.htm
toxic to aquatic life at levels found in local waterways and
*There may be other sealant products available that do not contain coal
are degrading the health of Austin'screeks.
tar so please read labels carefully. Listing of a specific product trade name
does not constitute an endorsement of its use.
PAHTrends in Town Lake Sediment (Core)Samples
1960-2000 PAH Concentration in Asphalt Based
and CoalTar Based Sealants
(Note: Percentagesare on a Jog scale)

Toxicity of CoalTar Sealants in Sediments to Aquatic Organisms


(Hya/feJaazteca) at three treatment levels (low, med, high)
100

81
~
~
~
~-;;
For additional infonnation and data see
5 www.cit'jofaustin.org/vvatershed/coaitar_main.htm
C/)
"E 40
<lJ 29
U
<i;

t(I
a.
13

,. .1J]"".~~;".,j
JjJ~l ;.!.'i~ J"..,1~".+;",.,
..;""""'J'~ 512 - 974 - 2550
.., '. ~~~~~~ DataRr~vided~y UnitedStates
~ Cow Med High
..!J",,3.,:;j,.ill.""-i 4;#1.".] Geological Ser.llce
(17ppm) (36ppm) (200ppml
Of~{j
..~~\)..,f",
;::;;,~/j}
;~ ..
science ii,r a changing world ~~,
Oll_1t_D~1?"

Prepared ill cooperation with the City of Austin, Texas

Collaborativestudiesbythe City of Austin and United StatesGeologicalSurvey(USGS)


have Sealcoatis usedcommercially
identifiedcoal-tarbasedsealcoat-the black,shinyemulsionpaintedor sprayedon asphaltpave- andby homeownersacrossthe Nation.
It commonlyis appliedto parking lots
mentsuch as parkinglots-as a majorand previouslyunrecognized
sourceof polycyclicaromatic
associatedwith commercialbusinesses
hydrocarbon(PAH)contaminationin someAustin areastreamsediments.PAHsare suspected (including strip malls andshopping
humancarcinogensand are toxic to aquaticlife. The studiesshowthat runofffrom coal-tarbased centers);apartmentand condominium
complexes;churches,schools,and busi-
sealcoatedparkinglots hasconcentrationsof PAHsthat are about65 timeshigherthan concen-
nessparks;andresidentialdriveways.
trations in particleswashedoff parking lots that have not beensealcoated.Biologicalstudies, The City of Austin, Texas,estimatesthat
conductedbythe City of Austin in the field and in the laboratory,indicatethat localizedPAH about600,000gallonsof sealcoatare
appliedeveryyear in the greaterAustin
levelsin sedimentscontaminatedwith abradedsealcoatare toxic to aquaticlife and are degrad-
area.
ing aquatic communities,as indicatedby loss of speciesand decreasednumbersof organisms. Nationalusenumbersare not avail-
Identificationof this sourcemayhelp to improvefuture strategiesfor controllingPAHsin urban able; however,use-patternssuggestthat
asphalt-based sealcoatis more commonly
water bodiesacrossthe Nationwhere parking lot sealcoatis used. usedon the WestCoastand coal-tar
basedsealcoatis morecommonlyusedin
the Midwest,the South,and on the East
Coast.

What are PAHs, coal tar, and


sealcoat:?

Polycyclic aromatichydrocarbons(or
PARs)area groupof organiccontam-
inants that form from the incomplete
combustionof hydrocarbons,suchascoal
and gasoline.PARsarean environmental
concernbecausethey aretoxic to aquatic
life andbecauseseveralaresuspected
humancarcinogens.
Coal tar is a byproductof the coking
of coal, and cancontain50 percentor
more PARs by weight.
Sealcoatis a blackliquid thatis
sprayedor paintedon asphaltpave-
mentin an effort to protectand beautify
the asphalt.Most sealcoatproductsare
coal-taror asphaltbased.Many coal-tar
sealcoatproductscontainas muchas
30 percentcoaltar by weight. Product
analysesby the City of Austin indicated
thatcoal-tar sealantproductshadmedian Runofffrom coal-tar based sealcoatedparking lots has concentrations of PAHsthat are
concentrationsof total PARs about70 about 65 times higherthan concentrations in particles washed off parking lots that have not
times higherthan asphalt-based sealants. been sealcoated.
How does sealcoat get
from parking lots into the ments.PARstendto attachto sediments;
environment? the ProbableEffect Concentration
(PEC)-a widely usedsediment-quality
Vehicletires abradeparking lot sealcoat guidelinethatis the concentrationof a
into smallpieces.Thesesmallparticles contaminantin bed sedimentexpected
are washedoff parking lots by precip- to adverselyaffectbenthic(or bottom-
itation into stormsewersand streams. dwelling) biota-is 22.8milligrams per
Sealcoat"wear and tear" is visible in high kilogram (mg/kg)for totalPAR. Stud-
traffic areaswithin a few monthsafter ies by USGSand City of Austin did
application.Sealcoatmanufacturers rec- not evaluatehuman-healthrisk from
ommendreapplicationevery2 to 3 years. exposureto sealcoat.Human-healthrisk
from environmentalcontaminants is often
evaluatedin termsof exposurepathways.
What are environmental For example,peoplecould potentially
be exposedto PAHsin sealcoatthrough Concentrationsof total PAHsin particulates
and human-health concerns skin contactwith abradedparticlesfrom
in runofffrom sealed parking lots greatly
associated with PAHs? exceeded concentrationsfrom unsealed
parking lots, inhalationof wind-blown
parking lots. The bar on each graph is the
particles,and inhalationof fumes that
PAHsaretoxic to mammals(includ- mean concentration.
volatilize from sealedparkinglots. PARs
ing humans),birds, fish, amphibians, in streamsandlakesrarelyposea human-
invertebrates,andplants.Possibleeffects healthrisk via drinking waterbecause
of PAHs on aquaticinvertebratesinclude of their tendencyto attachto sediment What concentrations of PAHs
inhibited reproduction,delayedemer- ratherthandissolvein water.In addition, are in runoff from sealed and
gence,sedimentavoidance,and mortal- becausePARsdo not readilybioaccu- unsealed parking lots?
ity, andpossibleadverseeffectson fish mulate within the food chain,possible
include fin erosion,liver abnormalities, human-healthrisksassociatedwith con- Concentrationsof PAHswere much
cataracts,andimmune systemimpair- sumptionof fish arelow. higherin runoff from parking lots sealed
with coal-tarbasedsealcoatthanfrom
all othertypesof parking-lotsurfaces.
Specifically,the averageconcentration
in runoff from coal-tarsealedlots was
How did USGS study parking-lot runoff? 3,500mg/kg, about65 times higher than
the averageconcentrationin particles
USGSresearchers sampledrunoff at 13 parking lots washedoff parking lots thatwere not
representinga rangeof different sealantr,ypesin sealcoated(54 mg/kg).The average
Austin. Theyalso took scraping-samples of parking concentrationin particleswashedoff
lot surfacesto comparesourcematerialsto wash-off parking lots sealedwith asphalt-based
particulates.Sourcematerialsand wash-off particu- sealcoatwas 620mg/kg, about6 times
lateswereanalyzedfor a suiteof PAHs,majorele- less thancoal-tarbasedsealcoat,but still
ments,andtraceelements.USGS researchers sprayed 10 times higherthanthe concentration
wateron four differenttypesof parking-lotsurfaces from unsealedparking lots.
in Austin: lots sealedwith coal-tarbasedsealcoat Concentrationsof PAHs in particles
(top photo),lots sealedwith asphalt-based sealcoat, washedoff eachof the differentsurface
unsealedasphaltlots,andunsealedconcretelots. The types-including the unsealedparking
runoff wascollectedbehind spill berms,pumpedinto lots-exceeded the PEC of 22.8 mg/kg.
containers(middle photo)andfiltered throughTeflon This is not surprisingbecauserunoff from
filters to collectthe particulatesfor analysis(bot- parking lots is likely to containPAHs
tom photo).The particulates,the filtered water,and from manysources,including leaking
samplesof sealcoatscrapedfrom the parking-lotsur- motor oil, tire particles,vehicle exhaust,
faceswere analyzedfor PAHs at the USGSNational andatmosphericdeposition.However,
WaterQualir,yLaboratory.Concentrationsandyields the largedifferencesbetweenconcentra-
(the amountof PAHscomingoff eachlot) wereused tions associatedwith sealedand unsealed
to determinelevelsof contaminationin runoff from parkinglots indicate thatabradedsealcoat
eachtype of lot andthe importanceof sealedlots asa is a majorand previouslyunrecognized
sourceof PAHsto urbanstreams. contributorto PAH contaminationin
urbanand suburbanwaterbodies.
.50
E~
.,
t;c:
~
0 0
"0
S
E
~
.,
50
t;
Co
.:!
~
..2
..E 100
..
";:
~
0)
0
'0 150
~
.c
E
~
z 200
0 500 1,000 1,500 2,000 2,500
Changein total PAHs,in microgramsper gramorganiccarbon
(upstreamto downstream)

Field assessmentsin selected Austin streamsshowed loss of species (taxa)and decreasesin the number of aquatic organismsdownstream
of coal-tar sealed parking lots that can be,in large part, explained by increasesin total PAHs.

How do F-AHsfrom sealcoat their currentloadsif all of the parking cantbiological degradationin responseto
lots wereunsealed. sealcoatadditions.Finally,field assess-
impact t~le quality and biology Studiesby City of Austin biologists mentsin selectedAustin streamsshowed
of streamls? showedthat sealcoatparticlesentering loss of speciesanddecreasesin the
streamscould be adverselyaffecting numberof aquaticorganismsdownstream
Studiesby USGSscientistsdemonstrated aquaticcommunities.Specifically,toxic- from inflows of runoff from sealedpark-
possibleconnectionsbetweenPARs in ity testingof organismsin the laboratory ing lots. The impactscoincidedwith
particleswashedfrom sealedparking showedlargeincreasesin mortalityas increasesin concentrationsof PARs in
lots and PARs in suspendedsedimentin sealcoatamounts/concentrations were streamsedimentsbelowsealedparking
four streamsin Austin and Fort Worth, increased,and thatcoal tar sealantsin lots. Overall,City of Austin scientists
Texas.FindingsshowedthatPARs in sedimentsweretoxic to aquaticlife at havereportedPAR contaminationat lev-
suspendedsedimentsin the streamswere PAHconcentrationsobservedin Austin els predictedto betoxic to benthicinver-
chemicallysimilar to thosein runoff waterways.Controlledexperimentsthat tebratesin over13 percentof sampled
from parkinglots sealedwith coal-tar usedaquariumswith diversenatural Austin creeks.
basedsealcoat.Analysis of the total mass biological communitiesshowedsignifi-
of PARsexpectedto washoff sealed
parking lots andthe total massof PARs
measuredin suspendedsedimentsin the
How did City of Austin scientists conduct biological studies?
four streamsafterrainstormsindicated
thatrunoff from sealedparking lots could
City of Austin biologistsconductedlaboratoryandfield studiesto evaluatethe
accountfor the majority of PAR loadsto effectsof sealcoatedparking lots on aquaticcommunitiesin areastreams.These
the streams. studiesincludedtoxicity testingin controlled laboratoryexperimentsthatexposed
Apart from the sealcoatingitself, organismsto sedimentsspikedwith coal-tarand asphalt-based sealcoat(left photo);
unsealedand sealedparking lots receive
controlledexperimentsthatusedaquariumswith diversenaturalbiological commu-
PARs from the sameurbansources- nities to which sealcoatwasadded(middle photo); andfield assessments of aquatic
tire particles,leaking motor oil, vehicle communitiesin streamsupstreamanddownstreamfrom inflows of runoff from
exhaust,andatmosphericdeposition-
sealedparking lots (ri~t photo).
yet the averageyield of PARs from
sealedparking lots was50 times greater
than that from unsealedlots. Whatwould
be the effect on PAR loading to the
streamsif parking lots were not sealed?
Estimatesfrom the USGS studyindicate
that total loadsof PARs coming from
parking lots in the studiedwatersheds
would be reducedto aboutone-tenthof
How do these findings apply to thoughtto be dominatedby numerous links to related publications,
urban liakes and reservoirs? nonpointsources,suchasleaking motor
data and maps
oil, tire wear,vehicularexhaustand
PARs in lakesand reservoirsacross atmosphericdeposition.Suchsourcesare
City of Austin Coal Tar Sealant
the Nationare increasing,asindicated difficult to quantify or controlbecauseof
Information-
their diffuse,nonpointnature.In contrast,
by USGSstudiesof 38 reservoirsand http..//www.ci.austin.tx.us/watershed!
lakessampledin 18 metropolitanareas sealedparking lots contributeto urban
stormwaterrunoff (seephoto below), bs_coaltar:htm
acrossthe countryfrom 1996to 2001.
andthe useof sealcoatis voluntaryand USGS frequently asked questions-
Sedimentcores(vertical tubesof mud)
werecollectedfrom reservoirandlake controllable.Possiblealternativesto http..//water;usgs.gov/nawqa/asphalt-
coal-tarbasedsealcoatingof parkinglots sealers.html
bottoms(seephotobelow); analysisof
thesecoresprovidesa reconstructionof includethe useof concreteandunsealed
Basic information on the toxicity of
historical waterquality overtime, much asphaltpavementandthe useof asphalt-
PARs to biological organisms,
basedsealantsthat contain1.ower levels
like usingtreerings to reconstructhistori- U.S. Environmental Protection Agency
of PAHs.
cal climate.Runoff carriessoil, debris, (USEPA)-
and attachedcontaminantsto lakesand http..//www.epa.gov/R5Super/ecology/
reservoirs,which settleto the bottom;as htmlltoxprofiles.htm#pahs
the sedimentbuilds up,changesin water
General information on PAH
quality arerecordedin the successive
exposure, Agencyfor Toxic Substances
sedimentlayers.
and DiseaseRegistry (ATSDR)-
USGSfindings showthatconcen-
trationsof total PARs in the majority of http..//www.atsdr;cdc.gov/toxprofiles/
lakesandreservoirsin urbanand sub- phs69.html
urbanareasacrossthe Nationincreased
significantly from 1970to 2001.The
References
.increases
" were greatest
in lakeswith Currently,coal-tarbasedsealcoat Mahler,B.J., Van Metre,P.C.,Bashara,
rapidly urban- is not federallyregulated.In 1992,the T.J.,Wilson, J.T.,and Johns,D.A.,
izing water- U.S. EnvironmentalProtectionAgency 2005.Parkinglot sealcoat:An unrec-
sheds(urban (USEPA)excludedcokeproductresidues, ognizedsourceof urbanPARs:Envi-
sprawl); for including coal tar, from classificationas ronmentalScienceand Technology,
example,over hazardouswastesif they arerecycled. vol. 39, no. 15,p. 5560-5566.
j the last 10 Underthe ResourceConservation and
., yearsPARs Van Metre,P.C.and Mahler,B.J., 2005.
RecoveryAct, coal-tarbasedpave-
increasedten- Trendsin hydrophobicorganiccon-
ment sealantsareproductsthat contain
fold in Lake taminantsin urbanandreferencelake
recycledcoal tar and,therefore,are not
in the Hills sedimentsacrossthe United States,
regulated.Furtherstudiesareneededto
1970-2001:EnvironmentalScience
(suburban evaluatepotentialimpactson the aquatic
Chicago,illinois) asthe watershedrap- and Technology,vol. 39,no. 15,p.
environmentin otherpartsof the country.
idly developed.Furtherstudyis needed 5567-5574.
to assessdirectlinks betweenthe use of Contactsfor additional City of Austin. 2005.PAHs in Austin,
sealcoatand PAH trendsin theselakes. Texas.(http://www.ci.austin.tx.us/
information
watershedldownloads/coaltar _draft-
Peter Van Metre and Barbara Mahler
What are the implications of pah_study.pdf)
U.S. GeologicalSurvey
these studies? 8027ExchangeDrive
Austin,Texas78754-4733
The studyof parking-lotsurfacesby the
USGSand the City of Austin showthat (512) 927-3506or pcvanrnet@usgs.gov
abradedsealcoatcould bea majorsource (512)927-3566or bjrnahler@usgs.gov
of PAHsto urbanandsuburbanwater Mateo Scoggins
bodiesin watershedswheresealcoatis
used.Suchfindings mayhaveimpli- City of Austin,
cationsthatextendbeyondTexasas Watershed Protectionand Development
sealcoatis usednationwide.Identification Review Department
of this major newsourcemay influence 505 BartonSpringsRoad,11thFloor
future strategiesfor controlling PAHs in ..Austin, Texas78704
urbanenvironments.In the past,sources (512) 974-1917or
of PAHs in urbanwatershedswere mateo.scoggins@ci.austin.tx.us
Environ. $ci. Techno/.2005, 39, 5560-5566

underlying asphalt pavement and enhanceappearance.The


two primary sealcoat materials on the market are refined
coal-tar-pitch-based emulsion and asphalt-basedemulsion.
Although similar in appearance (glossyblack), coal tar and
asphalthave different molecular structures stemming from
their origins: coal tar is a byproduct of the production of
BARBARA J. MAHLER, ..t coke from coal,whereas asphaltis derived from the refining
of crude petroleum. Coal tar, a known human carcinogen,
PETER C. VAN METRE,t
is 50% or more PARs by weight (2); the predominant
THOMAS J. BASHARA,*
JENN[FER T. WILSON,tAND constituents of asphalt are bitumens, complex mixtures of
DAVID A. JOHNS* hydrocarbonsthat include asphaltenes,saturates,aromatics,
and resins (9). Coal-tar-emulsion- and asphalt-emulsion-
United States Geological Survey,8027 Exchange Drive,
based sealcoatstypically contain 20-35% of the emulsion.
Austin, Texas 78754, and City ofAustin Watershed Protection
Department, P.O. Box 1088, Austin, Texas 78767 Parking lot sealants are used extensivelyin the United
States and Canada. Although national use figures are not
available, the Blue Book ofBuilding and Construction (10),
a directory for the construction industry, lists over 3300
Polycyclic aromatic hydrocarbons (PAHs)are a ubiquitous pavementsealantcompaniesin 28 U.S.states.One company
contaminantin urban environments. Although numerous advertisesthe application of 1.7billion liters to dateworldwide
(11),and anotherreportshaving sealedover33 million square
sources of PAHsto u..rbanrunoff have been identified, their
meters (12). The City of Austin, population 650000 (2000
relative importance remains uncertain. We show that a census),estimates that about 2.5 million liters of sealcoatis
previously unidentified source of urban PAHs,parking lot used annually in this city (13).
sealcoat,maydominateloadingof PAHsto urbanwater bodies Sealcoatabrades from the parking lot surface relatively
in the United States. Particles in runoff from parking lots rapidly,and reapplicationis recommended everytwo to three
with coal-tar emulsion sealcoathad mean concentrationsof years (14). In 2003, the City of Austin identified abraded
PAHs of 3500mgjkg, 65 times higher than the mean parking lot sealcoatas a possible source of high concentra-
concentration from unsealed asphalt and cement lots. tions of PARsin streambed sediment (15).Here we present
Diagnostic ratios of individual PAHs indicating sources are evidence suggesting that parking lot sealcoat could indeed
similar for particles from coal-tar emulsion sealed lots be the dominant source of PARs to watersheds with
residential and commercial development.
and suspended sedimentfrom four urban streams.
Contaminantyields projected to the watershed scale for Experimental
Section
the four associated watersheds indicate that runoff from
SampleCollection. We compared concentrations and yields
sealed parking lots could accountfor the majorityof stream of particulate PAHs in simulated runoff from parking lots
PAH loads. sealed with coal-tar-based sealcoat, from lots sealed with
asphalt-based sealcoat, and from unsealed asphalt and
cementlots. Thirteenurban parking lots,representinga range
Introduction of sealanttypes that are currently in use in Austin, TX, were
sampled (Table 1). In addition, four test plots, each about
Concentrationsofpolycyclicaromatichydrocarbons(PAHs)-
120 m2,were sampled. Three of the test plots were sealed
a group of widely recognized aquatic contaminants (1)
just prior to testing, and one was left unsealed (asphalt
comprising numerouscarcinogens(2)-have beenincreasing
surface).The test plots are at the Robert Mueller Municipal
in recent decadesin many urban lakes,particularly in areas
Airport, Austin, TX,which has been closedsince 1999.A full
undergoing rapid urban growth (3). PARs adversely affect
description of the sampling is given in ref 16. In brief, 50 m2
mammals (including humans), birds, fish, amphibians,
areas of each parking lot and the test plots were sprinkled
invenebrates, and plants; in the aquatic environment, the
with 2 mm of distilled water (100 L over a 50 m2 area) to
effectsofPAHs on invenebratesinclude inhibited reproduc-
simulate a light rain, and concentrations of PAHs were
tion, delayedemergence,sedimentavoidance,and monality,
analyzed in particles filtered from the runoff. The study
and the effectson fish include fin erosion,liver abnormalities,
focused on the particulate fraction, as PAHs in urban runoff,
cataracts,and immune systemimpairments (4). Numerous
particularly those of higher molecular weight, are mostly
sources of PARs to urban runoff have been identified,
associatedwith particulates (7,17); for selectedsamples (test
including automobile exhaust,lubricating oils, gasoline,tire
plots and sevenparking lots), the dissolved phase also waS
particles, erosion of street material, and atmospheric depo-
analyzed.The testing followed a minimum of 5 days with no
sition (5-8), but uncenainty remains as to their relative
rainfall. The parking lots were sampled once, and the test
imponance. Investigations of urban sources of PARs have
plots were sampled three times over a 6 week period. Water
thus far overlooked a potentially major source: parking lot
was sprayed from a plastic hand-held sprayer at a rate of
sealants,also called "sealcoat". Our objective in this study
about 7 L/minfrom a height of abouto.75 to 1 m. Spill berms
wasto evaluatethe contribution ofPAHs from sealedparking
were used at the down-slope end of the delineated area to
lots to urban streams. gather water, which was then pumped into high-density
In the United Statesand Canada, sealcoatis applied to
polyethylene (HDPE) containers (Figure SI, Supporting
many parking lots and drivewaysin an effon to protect the Information). Recovery of water and observations about
lossesof water to wetting and leakageunder the berms were
.Corresponding author phone: (512) 927-3566;fax: (512)927.
noted. The water was returned to the laboratory, poured
3590;e-mail: bjrnahler@usgs.gov.
t United StatesGeological Survey. into a 50 L chum to keep the sample well mixed, and fIltered
* City of Austin Watershed Protection Department. through 0.45 !tm pore size PTFE filters. The filters were
5560. ENVIRONMENTAL SCIENCE & TECHNOLOGY I VOL 39. NO. 15. 2005 10.1021/esO501565 CCC: $30.25 @ 2005 American Chemical Society
Published on Web 06/22/2005
TABU:1. SamplingSite Characteristics
site name surface type date of sealantapplication samplingdate study component
CTTP1 coal-tar emulsion sealant Aug 5-6, 2003 8/21/2003 test plot
CTTP2 coal-tar emulsion sealant Aug 5-6, 2003 8/21/2003 test plot
ASTP asphalt emulsion sealant Aug 5-6, 2003 8/21/2003 test plot
UNSASTP unsealed asphalt pavement Aug 5-6, 2003 8/21/2003 test plot
CTTP1 coal-tar emulsion sealant 9/9/2003 test plot
CTTP2 coal-tar emulsion sealant 9/9/2003 test plot
ASTP asphalt emulsion sealant 9/9/2003 test plot
UNSASTP unsealed asphalt pavement 9/9/2003 test plot
CTTP1 coal-tar emulsion sealant 9/26/2003 test plot
CTTP2 coal-tar emulsion sealant 9/26/2003 test plot
ASTP asphalt emulsion sealant 9/26/2003 test plot
UNSASTP unsealed asphalt pavement 9/26/2003 test plot
ASPL1 asphalt emulsion sealant June 2003 9/7/2003 parking lot
ASPL2 asphalt emulsion sealant June 2003 9/7/2003 parking lot
ASPL3 asphalt emulsion sealant July 2003 9/28/2003 parking lot
CT PL1 coal-tar emulsion sealant March 2003 9/7/2003 parking lot
CT PL2 coal-tar emulsion sealant July 2003 9/28/2003 parking lot
CT PL3 coal-tar emulsion sealant July 2003 9/28/2003 parking lot
CTPL4 coal-tar emulsion sealant July 2003 9/30/2003 parking lot
CT PL5 coal-tar emulsion sealant July 1999 9/30/2003 parking lot
CT PL6 coal-tar emulsion sealant Nov 2000 9/30/2003 parking lot
UNSASPL 1 unsealed asphalt pavement 9/8/2003 parking lot
UNSASPL2 unsealed asphalt pavement 9/30/2003 parking lot
UNSCONpL 1 unsealed concrete pavement 9/8/2003 parking lot
UNSCONpL2 unsealed concrete pavement 9/8/2003 parking lot

massagedinside locking bags to remove retained particles, particleswas associatedwith the watervolume lost to surface
as described in ref 18, and the recovered particulates were wetting and that water leaking past the bemls had the same
submitted as chilled slurries in clean glassvials to the U.S. SSCand contaminant levelsasrecoveredwater. Forunsealed
GeologicalSurveyNational WaterQuality Laboratory (NWQL) asphalt lots, the loss to wetting was estimated as36 L for a
for analysis. In some casesthe filtrate also was shipped, in maximum potential recovery of 64 L. Thus, to estimate the
chilled and clean amber glass bottles, to the NWQL for total yield of sediment from each lot, SSCwas multiplied by
analysisof dissolvedPAR.One or more samplesof unfiltered the assumedmaximum recovery (82L for sealedand cement
water were collected from the churn for measurement of lots and 64 L for unsealed asphalt lots) to account for
suspendedsediment concentration (SSC),usedto determine recovered water and leakage past the bemls.Yields ofPAH
the mass of sediment recovered during eachtest. Although were estimated by multiplying the total yield of sediment
the 2 mm of simulated rain was not enough to wash off all times particle concentrations.
of the mobile sediment, the recovered water was visibly Chemical Analysis. Sampleswere prepared by extracting
clearertoward the end of eachapplication. In samplesfrom about 0.5 g dry weight of sample using pressurized solvent
the five sitesin which SSCwas measured in the fIrst 50L and extraction at 120 and 200 °C with a mixture of water and
final 50 L of water, SSCdecreasedby a mean of 65% (range isopropyl alcohol. The samples were extracted at each
of 39-84%). We therefore assumed that the testsrecovered temperatureata pressureof 13800kPa.Surrogatecompounds
most of the sediment that would be mobilized from the were addedto the sample prior to extractionto verify method
parking lot surfacesby a rain event,regardlessof magnitude. recoveries. The extract was cleaned up using polystyrene
Large,intense storms,however,likelywould generatea higher divinylbenzeneand Florisil solid-phaseextractioncartridges.
yield of sediment. The extract was concentrated, solvent exchanged to ethyl
The test plot and parking lot scrapingswere obtained by acetate,and diluted to 10mL. An internal standard mixture
scraping a small area (lessthan 0.25 mZ)with a metal paint was added to an aliquot of the extract, and the extract was
scraper.The particulates removedwere brushed onto a piece analyzed by full-scan gas chromatography/mass spectrom-
of new cardstockand then into a cleaned glassjar. The paint etry (GC/MS). Difficult sample matrices were diluted before
scraper was cleaned between sites, and a new brush was the full-scan analysis,and diluted surrogateswere estimated
used at each site. Scrapings were examined by light and in the samples.
electron microscopy (Figure S2, Supporting Information), Compound identifications were based on comparison of
and submitted to the NWQL for PAR analysis. peak retention times and mass spectra to those of authentic
Computation of Yields. Losses of water to wetting and standard compounds for the target compounds. Response
losses of water and sediment leaking under the berms were factors were calculated for each compound from a set of
estimated. Recoveryof water ranged ftom 19to 85 L with a calibration standards.Quantitation was carried out following
median of58 L The lowestrecoverieswere from flat, unsealed the methods of Olson et al. (19).For PAHsin the particulate
asphalt lots, and the highestrecoverieswere from sealedlots phase,the estimated method reporting limit (MRL)is 5 !.tgl
and cementlots with gentle slopes.On the basis ofrecoveries kg for a 25 g sample.Aslessthan 25 g was extracted,the MRL
and field observations, it was concluded that about 18 L of wasraisedaccordingly,on a sample-by-samplebasis.In some
water was retained on the surface of sealed lots and cement cases,MRLswereraisedbecauseofbackgroundinterferences.
lots and that the remainder of the water loss was a result of Dissolved-phase samples were analyzed following the
leakage past the berms. It was assumed that no yield of method described by Fishman and Friedman (20),with the
VOL. 39, NO. 15,2005/ ENVIRONMENTAL SCIENCE & TECHNOLOGY. 5561
.
difference that continuous liquid-liquid extraction was
substituted for use of the separatory funnel. In brief, 1 L
samplesfortified with surrogate compounds were extracted
by continuous liquid-liquid extraction for 6 h under acidic
and then basic conditions. Internal standards were added
and sample extracts concentrated to 1 mL. Sampleswere
analyzed by GC/MS in electron impact mode. Sample --c;
~
identifications were made by matching retention times and "OJ
mass spectra with those of standard compounds. Quanti- g
tation involved use of internal standards and calibration cn
curvesgeneratedby standardcompounds of known amounts. J:

Quality control (QC) consisted of environmental and ~


0
internal laboratory samples. Two duplicate enVironmental T-

samples for particulate analysis of PARwere collected. For O


one of the sets of duplicates, 2:PARdiffered by 8%; for the E
::J
second (which had ~AR > 4000 mg/kg), 2:PARdiffered by U)

54%. In the equipment blank analyzed for dissolved PAR,


three parent PARs were detected at concentrations about
half that of the environmental sample with the lowest
concentrations, and less than 1% that of the environmental CT AS US
sample with the highest concentrations.
FIGURE 1. Sumof 10PAHs(fluoranthene,pyrene,benz[a]anthracene,
Laboratory QC samples for particulate PAR analyses
consisted of analysis of spiked samples,blanks, and samples benzo[a]pyrene,benzo[e]pyrene,indeno[12,3-cdJpyrene.chrysene.
benzo[b]fluoranthene,benzo[k]fluoranthene,and benzo[ghl]perylene)
of certified reference material (CRM).The median recovery in particles in runoff from simulated rainfall on coal-tar emulsion
for the six spiked samples was 76%. For the six laboratory sealcoat (CT),asphaltemulsion sealcoat (AS),and Illnsealedcement
blanks, an analyte was detected in 85 of 336 possible cases. and asphalt (US) parking lots (0) and test plots 18). Parking lots
The detected concentrations ranged from 0.1% to 3.5% of were sampled once, and test plots \\rere sampled three times.
that in the enVironmental sample with the lowest concen- Concentrations for other PAH sources reported in the literature
tration for that analyte. For the two analyses of CRM, the also are indicated. These 10 PAHs werle summedfor this graph to
recoveries were within the NWQL-established acceptable facilitate comparison between experimental and reported con-
range for 83% of the cases. centrations.
Three commercially available asphalt-based emulsion sealed lots also were much higher. in most cases by orders
sealcoatproducts and six coal-tar-based emulsion sealcoat of magnitude, than PAH concentrations in other urban
products were analyzed at DHL Analytical, Round Rock,TX, sources such as tire particles, motor oil, and weathered
using EPAmethod SW8270 (21). In each case,the product asphalt (Figure 1; note that this figure uses a different
samplewas taken directly from the container. Concentrations summation of PAll). ~AH concentrations in runoff from
of 16 parent PARswere detemiined. The sealantsanalyzed the sealed test plots were generally lower than those from
were not necessarily the same as those applied to the test the sealed parking lots, but the difference was not statistically
plots or on the parking lots in use,although there was some significant, and concentrations from unsealed surfaces, with
overlap (product ASpAwas used on test plot A5TP;product the exception of one outlier, were similar for test plots and
CTpFwas used on test plot CTTP2)(Table 2). parking lots.
Concentrations of ~AH in the scrapings ranged from
Results 9500 to 83000 mg/kg forcoal-tar-emulsion-sealed surfaces
Concentrations and yields of total particulate PAH and total (including test plots) and from 110to 2000mg/kgfor asphalt-
dissolved PAH in the runoff and total PAHin the scrapings emulsion-sealed surfaces (Table 2). Scrapings of two unsealed
were computed and compared between parking lot surface asphalt parking lots had ~AH concentrations of7.1 and 20
types (Table2). The total particulate PAH (LPAll) concentra- mg/kg. Scrapings were observed under light and electron
tion was computed for each sample as the sum of naph- microscropy (Figure S2, Supporting Information).
thalene, 2-methylnaphthalene, acenaphthylene, acena- Concentrations of total dissolved PAH (~AHdiss, com-
phthene, fluorene, phenanthrene,anthracene,fluoranthene, puted as the sum of the same PAHs as in ~AH excluding
pyrene, benz[a]anthracene, chrysene,benzo[a]pyrene,and 2-methylnaphthalene; nondetections treated as zeros) for
dibenz [ah]anthracene,which are the sameasthose used for the test plots were about an order of magnitude greater in
the consensus-basedsediment quality guidelines of Mac- samples from the coal-tar-sealed test plots than concentra-
Donald et al. (22). For unsealed parking lots (asphalt tions in samples from the asphalt-sealed test plot, which in
pavement and concrete combined), the mean~AH was 54 turn were about an order of magnitude greater than those
mg/kg (rangeof7.2-75 mg/kg),more than twice theprobable from the unsealed test plot (Table 2). Nine of the 16 PAHs
effect concentration sediment quality guideline of 22.8mgl analyzed for were detected (complete data are in ref 16).
~g (22) (Table 2), and in the range of those found by others Higher weight PAHs-benzo[b]fluoranthene, benzo[k]-
in urban and roadwayrunoff (e.g.refs 23-25). However,the fluoranthene, benzo [a] pyrene, indenopyrene, benzo[ghzl-
mean ~PAH concentration from the asphalt-sealedparking perylene, benz[a]anthracene, and dibenz[ah]anthracene-
lots was more than 10 times higher (mean of 620 mg/kg, were not detected at laboratory reporting levels ranging from
range of 250-830 mg/kg) than that from unsealed parking 1.7 to 3.4flg/L. Four PAHs (acenaphthylene, acenaphthene,
lots, and the mean ~AH concentration from the coal-tar- chrysene, and fluorene) were detected only in runoff from
sealedparking lots was 65 times higher (mean of 3500 mgl the coal-tar-sealed test plots; anthracene was detected in
kg, range of 520-9000 mg/kg) (Table 2; complete concen- runoff from all the sealed test plots but not from the unsealed
tration data are given in ref 16). ~AH concentrations in site. A similar suite of PAHs were detected at those parking
runoff from coal-tar-sealedlotswere significantly higherthan lots for which the fIltrate was analyzed (Table 2; complete
in runoff from other surface types (Kruskal-Wallis test of data are in ref 16).
comparisons, hypothesis of no difference between groups Concentrations of ~AH in the commercially available
rejected for p < 0.05). PAH concentrations from coal-tar- sealant nroducts and surface scrapings exceeded those of

5562. ENVIRONMENTAL SCIENCE & TECHNOLOGY I VOL. 39. NO. 15. 2005
TABLE2. Concentrationsof PAHin WashoffSamples, 1,000,000
-0;
:Scrapings,
andUnappliedSealcoat Produc~ ~
'" 100,000
.§.
wash off samples scrapings product c
.2
10,000
I:PAH. I:PAHdi... I;PAH.

.
~PAH(dry), 1§
mg/kg Jlg/l mgikg mg/kg C 1,000
~
c
CTTP1,8/12/03
Test Plots
21 83000
"
0 100
CTpA 34000 :I:
10
CTTP1,8/21/03
CTTP1,9/9/03
1700
530
14 CTpB 113000 ~
~
CTpC 202000 1
CTTP1,9/26/03 460 6.9 CTpD 86000
CTpE 49000
CT TP2, 8/12/03 11 11000 CTpF 61000
CT TP2, 8/21/03 1200 7.3
CT TP2, 9/9/03 4000 ASpA 6600 FIGURE 2. Comparison of I;PAH concentrations in (:ommerciallv
CT TP2, 9/26/03 140 3.8 ASpS 1300
available sealcoat products, scrapings from parking lots, and
ASpC 300
ASTP, 8/12/03 1.3 110 particles in wash oft from parking lots for coal tar (0) and asphalt
ASTP, 8/21/03 96 1.2 (e) based sealants.
ASTP,9/9/03 40
ASTP, 9/26/03 28 0.64 concentrations for the different surface types suggestthat
abraded sealant products are a potentially important (and
UNSASTP, 8/12/03 0.16 heretofore unrecognized)contributor to PAH contamination
UNSASTP, 8/21/03 410 0.34 in urban and suburban water bodies.
U NSASTP, 9/9/03 25
UNSASTP, 9/26/03 14 0.17 Comparison of Medium, Aging, and Vehicle Use on
Concentrations and Yields. For both coal-tar- and asphalt-
Parking Lots emulsion-basedsealants,the ~AH concentration decreased
CTPL1 2000 NA 25000
CT PL2 9000 5.4 15000
from the unapplied sealantproducts to the scrapingsto the
CTPL3 2000 7.1 11000
washoff samples. as did the difference in concentration
CTPL4 1300 12 9500 betweenthe coal-tar-basedand asphmt-basedsealantsamples
CT PL5 520 2.3 9900 (Figure2). The difference in the median ~AH concentration
CTPL6 5900 16 17000 between the coal-tar-based and asphalt-basedsealantswas
70-fold for the products analyzed and decreasedto 40-fold
ASPL1 250 NA 340 for the scrapings and to a factor of about 8 for the washoff
ASPL2 830 NA 2000
samples.Although the chemicalchangesbetweentheproduct
ASPL3 770 5.1 420
pre- and postapplication were not the focus of this study.
UNSCONpL 1 75 NA NA the decreasein ~AH concentrations from the scrapings to
UNSCONpL2 69 NA NA the washoff particulates and the magnitude of the difference
UNSASPL1 64 NA 7.1 between the coal-tar-sealedlots and the asphalt-sealedlots
UNSASPL2 7.2 0.24 20 can be attributed to dilution of abraded particles with less
'Sums are as defined in the text. NA = not analyzed, CT = coal- contaminated street dust and the greater abrasion of the
tar-based emulsion, AS = asphalt-based emulsion, UNSAS= unsealed asphalt-sealedcompared to the coal-tar-sealed surfaces.A
asphalt pavement, and UNSCON = unsealed concrete pavement. simple mass balance. assuming dilution of the coal tar
scrapings(median ~AH concentration of 13000mg/kg) by
street dust (median ~AH concentration of 50 mg/kg) at a
the particulates in the washoff. Concentrations of 1:PAHin proportion of 1 part abraded particles to 7 parts street dust,
commercially available coal-tar-based sealcoat products results in the concentration found in the washoff. If the
ranged from 3.4 to 20 wt %, compared to 0.03 to 0.66 wt % proportion of abraded particles is increased for the asphalt
for asphalt-based~sealcoatproducts analyzed (fable 2; lots on the basis of the increasedyields measuredfor asphalt-
complete data in Table SI, Supporting Information). sealedlots (assuming that the greatermedian particle yield
Yields of 1:PAH(mass of LPAH per unit area of parking of 320 mg/m2 from asphalt-sealed lots versus 200 mg/m2
lot) computed for the simulated rainfall followed patterns from coal-tar-sealed lots results from increased abrasion),
similar to those of concentrations. Complete data for yields the concentration found in the washofffrom asphalt-sealed
can be found in TableS2 in the Supporting Information. As lots is well approximated.
with the concentrations,there was a wide range in the yields The effect of aging of sealantson concentration over the
for a given surface type, in most casesmore than an order short term (7 weeks)was evident at the testplots (Figure 3a).
of magnitude. The mean yield from coal-tar-sealed lots Overall, the concentration of ~AH and ~AHdiss in the
exceededthat from asphalt-sealedlots by more than a factor washoff from eachtest plot decreasedover the 7 weekperiod
of 2, although this difference was not statistically significant following application. In one instance (CTTP2. second sam-
(Kruskal-Wallis test of comparison, p < 0..05).However,the
piing ofwashotl) kPAH exceededthat previously sampled.
mean yield from sealedlots (asphaltand coal tar combined) but in all casesthe concentration at the end of the period
exceededthat from unsealed lots by a factor of 50, and the was less than that at the beginning. The PAH assemblage
difference was statistically significant. changed over the same period as well, as represented by a
comparison of higher molecular weight (MW) to lower MW
Discussion PAHs. In the particulate samples the ratio of higher MW
Runoff from parking lots typically is contaminated with PAHs PAHs (represented by benzo[a]pyrene + chrysene) to the
from leaking motor oil, tire particles, vehicle exhaust,and lower MW PAHs (represented by fluorene + phenanthrene;
atmospheric fallout, and it is not surprising that the mean these two PAHswere chosenas they were detected in most
concentration of LPAH in particles washed off each of the of the samples)increased at all of the sealed test sites.As the
different surfacetypesexceededthe probable effect sediment lower MW PAHs aremore volatile and soluble than the higher
quality guideline. Howexer, the large differences between MW PAHs,volatilization and leaching of the lower MW PAHs

VOL. 39, NO. 15, 2005 I ENVIRONMENTAL SCIENCE & TECHNOLOGY. 5563
1.
~
1.5

1.4

1. " ri ~
~
: 1.0
I, n A
::t"'~:
.6,'
, .
: .
.,
0.6' '-'-. ...I
1.0 1.2 1.4 1.6 1.8 2.0
Fluoranthene:pyrene
.1.8, -I

1.8
N
...1.4
~'1.2

1.0
0.8 ,
...,..-;
.~"
',.+.
~~
.~"":'~;.:~:~~
.,.
"\.
'.',
','" +..' .).
4 x
,
,
:I:

0.6 ",'
" ,
0.4 0.6 0.8 1.0 12 14 1.6 1.8
Benzo[s]pyrene: benzo[e]pyrene
1.6 ,
2,500
2.0 Ii 1.~ ~
~.. I ry-'
1.2
0.0 .0
0 500 1.000 1.500 2.000 1.0 ,

Days since product applied '.


,

\.v x
RGURE3. 1:PAHconcentrations (closed symbols)and PAH ratios , 0.6
0.8lr "'""",:
of higher and lower molecular weight PAHs (open symbols)(a) in
scrapings and particles washed off coal-tar-sealed test plots O.~
(D. 0) and an asphalt-sealed plot (6) and (b) as they relate to the 1.0 1.2 1.4 16 1.8 2.0
Fluoranthene:pyrene
age of coal-tar sealant in samples from parking lots in use.
Samples and ellipse approximate distribution of:

from the newlyapplied sealantmight be responsiblefor some 0 Coal-tar sealed parking lots
of the decreasein concentration. For the parking lots, only ,.,~: Asphalt sealed parking Jots
'..'
coal-tar-sealedlots representeda range of ages,and for these (~) Unsealed asphalt and cement parking lots
lots there was no relationship between concentration or <9 Urban stream sediment
higher MW to lower MW PAH ratio and age of the sealant
(Figure 3b). This might be because the parking lots were FIGURE4. Comparison of indicator ratios of PAils in particles
eachsampled only once; the verywide rangein PAHcontent washed from parking lots with coal-tar emulsion sealcoat, asphalt
between products, even those of a similar kind, may mask emulsion sealcoat, and unsealed asphalt pavement and concrete
the effect of aging when time-series data are not available. pavement, and in suspended sediment collected from four urban
Although the data are limited, they suggest that lots with streams after storms.
older sealant tend to have a higher ratio of higher MW to watersheds is more than 90% urban; full land use is given in
lower MW PAHs,and that that ratio may reach a plateau Table S3, Supporting Information); the streams are ephem-
after a period of time. eral, and urban runoff is assumed to comprise a large
Comparison of the yields from the parking lots to those component of storm flow.
from the test plots, which receive no vehicle traffic, dem- PAHs comprise a large group of compounds, and PAH
onstrates the importance of abrasion of sealcoatby vehicles assemblage is often used to infer PAH sources (27). Differ-
on 1:PAHyield: the mean kPAH yield was 20 and 160times ences in PAH assemblages can be investigated by computing
greater for the coal-tar-sealed and asphalt-sealed parking the ratios of selected PAHs (28, 29). The best indicator ratios
lots, respectively,than for the analogoustestplots. This does of coal tar as a PAH source have been identified as fluor-
not appearto be attributable to use patterns,although traffic anthene:pyrene. indeno [1,2.3-cd)pyrene:benzo [ghll perylene.
counts were not made: the coal-tar-sealed lots are a mix of and benzo [a]pyrene:benzo [e]pyrene (30, 31). In graphs that
lotsin constantuse throughout the day (e.g.,shopping center) combine these ratios, similarities and differences between
and those with all-day parking (e.g., office), which are parking lot and stream samples are apparent (Figure 4): ratios
assumedto receive less use than those in constant use; all in the urban stream sediment group match those in runoff
of the asphalt-sealedlots are all-day parking. from coal-tar-sealed lots more closely than they do those
Environmental Implications. Given the extremely el- from asphalt-sealed lots and from unsealed lots (asphalt and
evated concentrations of PAHs in particles washed from cement). We found these ratios were far more effective at
sealed parking lots, how important is this contribution to distinguishing between the different parking lot samples and
the total mass of PAHs in urban streams?To answer this stream samples than ratios indicative of combustion versus
question,we compared the PAHassemblagesand estimated noncombustionsources,or other approachessuch as
PAHloads associatedwith particulates in parking lot runoff comparison of parent compound distribution (32). Although
to thoseassociatedwith suspendedsedimentcollectedduring ancylated PAH homologues were analyzed (including CI-
storm flow in four streams: Williamson Creek (Austin, TX) C5 homologues of the MW 128, 178,202.228. and 252 PAHs) ,
(18) and influent streams to Echo Lake, Fosdic Lake, and their interpretation did not assist in discriminating between
Lake Como (Fort Worth, TX) (26).These four streams are in PAHs from the different parking lot surfaces.
highly urbanized watersheds (land use for the Austin The relative amount of similarity between groups of
watershed is about 65% urban, and for the three Fort Worth samples, as defined by the ratios, was quantified through

5564. ENVIRONMENTAL SCIENCE & TECHNOLOGY I VOL 39. NO. 15. 2005

-..
"'.
Lake Como Inflow .200

§
"0 1,000
IV
10 !
e~~ .Q
SOD
f~::iJ :I:
c{
Q.
H 600
1 "0
1,000 Q)
"'iU 400
E
:w
100 II)
200
W
--0;
~ 10 0
J: o~ o~ o~ 0"-
~
~ 1 ,~~
~O
,~~
*-0
,~~
*-0
0,,0
O~
'0- 100 00 v~ v~ ~
0 Fosdic Lake Inflow 0 .c, '"
"tJ
m v~*-0 ",ao
"V «0~O" ~".$-~
0
..J
FIGURE6. Comparison of event loads ~If1:PAH fDr four urban
1 watersheds estimated for parking lots in their current (2004)state
(sealed by gray bars and unsealed by white bars) ;~ndprojected
0.1 loads if all existing parking lots were unse,aled(black bars). Loads
10,000 were estimatedon the basis ofthe yieldsfrom the runo:ffexperiments
Williamson Creek I and the area of parking lots in each watershed.
unsealed lots were identified by site inspection. We computed
the hypothetical storm-event load generated by sealed
100 parking lots in each watershed by multiplying the mean yield
for sealed parking lots (coal-tar and asphalt emulsion sealcoat
combined) determined from the runoff experiments by the
10ln
sealed parking lot area of each watershed. We assumed that
StOml events
the 2 rom of water applied for the field tests mobilized all
FIGURE 5. Comparison of estimated event loads of I;PAH from
available particles, and that all runoff from parking lots
sealed parking lots and measured instream storm-event loads for
entered storm sewers and was delivered to the stream.
four urban watersheds. The interquartile range of estimated loads
Although there is substantial variation in event loads for each
is s~lown in gray shading, on the basis of 25th and 75th percentile
stream (18, 26), for all four watersheds the estimated kP AH
yields computed for sealed parking lots; the mean estimated load
is indicated by a dashed line. Measured instream loads for four to
loads contributed by sealed parking lots are similar in
eight individual events are shown as bars.
magnitude to measured stream loads, even though sealed
parking lots cover only 1-2% of each watershed (Figure 5).
These results might explain why an investigation carried out
discriminant function analysis. In discriminant function in Marquette, MI, found that runoff from commercial parking
analysis, each significant independent variable adds to lots contributed 64% of the PAH load to the urban watershed
discrimination between multiple groups. The three ratios studied (33).
(tluoranthene:pyrene, indeno[l ,2,3-cd]pyrene:benzo [ghll- What would be the effect on PAH loading to these
peryiene, and benzo[a]pyrene:benzo[e]pyrene) were entered watersheds if parking lots were not sealed? For each of the
into the analysis as the independent variables, with the four watersheds, we compared the kPAH load contributed
different types of samples (coal-tar-emulsion-sealed lots, by parking lots (computed on the basis of the aerial extent
asphalt-emulsion-sealed lots, unsealed lots, and urban of unsealed and sealed parking lots) to that obtained by
stormtlow stream sediments) defining four groups of de- applying the average yield for unsealed lots to all parking
pendent variables. All three variables were shown to con- lots (Figure 6). We estimate that the kPAH load from parking
tribute significantly to discrimination between the groups (p lots in these watersheds would be reduced to 5-11 % of the
< 0.001).The distancesbetween the centroids of the groups
current loading if all lots were unsealed.
were determined by computing the squared Mahalanobis With the exception of the sealcoat itself, unsealed parking
distance, which is a measure of the distance between two lots receive PAHs from the same urban sources as do sealed
points in the space defmed by two or more correlated parking lots-e.g., tire particles, leaking motor oil, vehicle
variables.The centroid of the group defined by the suspended exhaust, atmospheric fallout-yet the average yield of PAHs
sedimentfrom urban streamsis closestto the centroid of the from sealed parking lots is 50 times greater than that from
coal-tar-based sealant group, next closest to that of the unsealed lots. PAH assemblages and estimated loads further
unsealed lot group, and farthest from that of the asphalt- suggest that sealed parking lots could be dominating PAH
based sealantgroup (squaredMahalanobis distances of 5.7, loading in watersheds with commercial and residential land
13.0, and 25.0, respectively). Thus, on the basis of the three use. The implications of these results extend beyond Texas
ratios diagnostic of coal-tar sources,the PAHassemblageof to the rest of the United States and Canada, where parking
the suspendedsediment from the urban streamsmost closely lot sealcoat is used extensively, and to other countries where
resemblesthat of the coal-tar-basedsealantgroup,supporting sealcoat is being introduced. Previously identified urban
the hypothesis that coal-tar-basedsealantsare an important sources of PAHs, such as automobile exhaust and atmo-
source of PAHs in urban streams. spheric deposition, have been difficult to control or even
Moving to a mass-balance approach at the watershed quantify because of their nonpoint nature. In contrast, sealed
scale for each of the four urban watersheds,we compared parking lots are point sources, and use of the sealant is
measured storm-event stream loads of LPAH to those voluntary and controllable.
estimated to be contributed by sealedparking lots. Digital
land-use maps that included parking lots were provided by Acknowledgments
the Cities of Austin and Fort Worth and were updated using We thank Robert Eganhouse, E. Terrence Slonecker, and three
recent aerial photography and site inspections. Sealedand anonymous reviewers for critical reading of the manuscript.

VOL. 39. NO. 15. 2005/ ENVIRONMENTAL SCIENCE & TECHNOLOGY- 5565
.
.This research was carried out asa cooperative project between (20) Fishman, M. J.; Friedman, L C. Methods for determination of
the U.S. Geological Survey and the City of Austin. inorganic substances in water and fluvial sediments; U.S.
Geological Survey Open-File Repon 93-125; U.S. Geological
SupportingInformationAvailable Survey: Denver, CO, 1993.
(21) Method 8270D: Semivolatile organic compounds by gas
Two figures and three tables. This material is available free
chromatography/mass spectrometry (COMS); U.S. Environ-
of charge via the Internet at http://pubs.acs.org. mental Protection Agency: Washington, DC; http://www.
epa.gov/ epaoswer/hazwaste/test/8_series.htm(accessed March
LiteratureCited 2005).
(22) MacDonald, D. D.; Ingersoll, C. G.; Berger,T. A. Development
(1) u.s. Environmental Protection Agency. The incidence and and evaluation of consensus-based quality guidelines for
severityofsedimentcontamination in surfacewater ofthe United freshwaterecosystems.Arch. Environ. Contam. Toxicol. 2000,
States;EPA823-R-97-006;Washington, DC.
39,20.
(2) Report on Carcinogens,lOth ed.; National Toxicology Program,
(23) Brenner, R C.; Magar,V. S.; Ickes,J. A.; Abboutt, J. E.; Stout,S.
Public Health Service,U.S. Department of Health and Human A.; Crecelius,E. A.; Bingler, L S. Characterizationand FATEof
Services: Washington, DC, 2002. PAH-contaminated sediments at the Wyckoff/Eagle Harbor
(3) Van Metre, P. C.; Mahler, B. J.; Furlong, E. T. Urban sprawl Superfund Site. Environ. Sci. TechnoL2002, 36, 2605.
leavesits PAHsignature. Environ. Sci. Technol.2000, 34,4064.
(4) U.S.Environmental ProtectionAgency.Information on the Toxic (24) Krein, A.; Schorer, M. Road runoff pollution by polycyclic
aromatic hydrocarbons and its contribution to river sediments.
Effects of Various Chemicals and Groups of Chemicals,2003. WaterRes.2000, 34, 4110.
http://www.epa.gov/R5Super/ecology/htrnl/toxprofiles.htrn# (25) Durand, C.; Ruban,V.; Ambles,A.; Oudot, J. Characterization
(accessedJanuary 2005). of the organic matter of sludge: determination of lipids,
(5) Jiries,A. Vehicular contamination of dust in Ammam, Jordan.
hydrocarbons and PAHsfrom road retention/infiltration ponds
Environmentalist 2003, 23, 205. in France. Environ. PoUut. 2004, 132,375.
(6) Takada,H.; Onda, T.; Ogura, N. Determination of polycyclic (26) Van Metre, P. C.; Wilson, J. T.; Harwell, G. R; Gary, M. 0.;
aromatic hydrocarbons in urban street dusts and their source
materialsby capillarygaschromatography.Environ.Sa. Technol. Heitmuller, F. T.; Mahler, B. J. Occu"ence, trends,and sources
in particle-associatedcontaminantsin selectedstreamsand lakes
1990,24; 1179.
in Fort Worth, Texas;U.S. Geological SurveyWater Resources
(7) Hoffman, E. J.; Mills, G. L; latimer, J. S.; Quinn, J. G. Urban
runoffas a sourceofpolycyclic aromatic hydrocarbonsto coastal InvestigationsRepon 03-4169;U.S.GeologicalSurvey: Denver,
CO,2003.
waters. Environ. Sa. Technol. 1984, 18, 580.
(27) Yunker,M. B.; Macdonald,R. W.; Vingarzan,R.; Mitchell, R H.;
(8) Rogge,W. F.; Hildemann, L. M.; Mazurek, M. A.; Cass,G. R.
Sources of fine organic aerosol: Road dust, tire debris, and Goyette,D.; Sylvestre,S.PAHsin the FraserRiverbasin: a critical
appraisal of PAH ratios as indicators of PAH source and
organometallic brake lining dust: roads as sourcesand sinks.
Environ. Sa. Technol. 1993, 27, 1892. composition. Org. Geochem.2002, 33, 489.
(9) Irwin, R. J.;VanMouwerik,M.; Stevens,L; Seese,M. D.; Basham, (28) Eganhouse, R P.; Gossett, R. W. Historical deposition and
W. Environmental Contaminants Enclyclopedia;National Park biogeochemical fate of polycyclic aromatic hydrocarbons in
Service,Water ResourcesDivision: Washington, DC, 1997. sediments near a major submarine wastewater outfall in
Southern California. In Organic Substancesand Sedimentsin
(10) Contractors Register, Inc. The Blue Book of Building and
Water;Baker,RA., Ed; LewisPublishers: BocaRaton,FL,1991;
Construction;JeffersonValley, NY, 2005. p 191.
(11) SealMaster Pavement Products and Equipment. http://
(29) Heit, M. The relationship of a coal fired power plant to the
www.sealmaster.net/ (accessedJune 2005).
(12) New England Sealcoating Sealcoating and Striping. http:// levels of polycyclic aromatic hydrocarbons (PAH) in the
sediment of CayugaLake. WaterAir Soil Pollut. 1985, 24, 41.
www.newenglandsealcoating.com/sealcoating.htrn(accessed (30) Canton, L; Grimalt, J.O. Gas chromatographic-massspectro-
June 2005).
(13) City of Austin Report to Council on Coal Tar. http:// metric characterization of polycyclic aromatic hydrocarbon
mixtures in polluted coastal sediments.]. Chromatogr. 1992,
www.cLaustin.tx.us/watershed/bs_coaltar.htrn (accessedJanu- 607, 279.
ary 2005).
(14) Dubey,G. Sellingsealcoating.Pavement1999,MarchI April, 42. (31) Marvin, C. H.; McCarry,B. E.; Villella, J.; Allan, L M.; Bryant,
(15) Haurwitz, R. Parking Lot Contaminant Theory Explored. In D. W. Chemicaland biological profiles of sedimentsas indicators
Austin American-Statesman;Austin, Texas,Cox Communica- of sources of genotoxic contamination in Hamilton Harbour.
tions: Atlanta,GA, 2003;p A6. PartI: Analysis of polycyclic aromatic hydrocarbonsand thia-
06) Mahler, B. J.; Van Metre, P. c; Wilson, J. T. Concentrations of arene compounds. Chemosphere2000,41,979.
polycyclicaromatic hydrocarbons (PARs)and major and trace (32) Lake, J. L; Norwood, C.; Dimock, C.; Bowen, R. Origins of
elements in simulated rainfall runoff from parking lots, polycyclic aromatic hydrocarbons in estuarine sediments.
Austin, Texas,2003;U.S. Geological SurveyOpen-File Report Geochim.Cosmochim.Acta 1979,43,1847.
2004-1208;U.S. Geological Survey: Denver, CO, 2004; http:// (33) Steuer, J.; Selbig, W.; Hornewer, N.; Prey, J. Sources of con-
water.usgs.govIpubsl of/2004/1208/pdfl ofr2004-1208.pdf(ac- tamination in an urban basin in Marquette,Michigan and an
cessedJanuary2005). analysis of concentrations, loads,and data quality; U.S. Geo-
07) Eganhouse,R. P.; Kaplan, I. W. Extractable organic matter in logical SurveyWater ResourcesInvestigations Repon 97-4242;
urban storrnwater runoff. 1. Transport dynamics and mass U.S. Geological Survey: Denver, CO, 1997.
emission rates. Environ. Sa. Technol. 1981, 15,310. (34) Takada,H.; Onda, T.; Harada, M.; Ogura, N. Distribution and
08) Mahler, B. J.; Van Metre, P. C. A simplified approach for sources of polycyclic aromatic hydrocarbons (PAHs)in street
monitoring of hydrophobic organic contaminants associated dust from the Tokyo Metropolitan area. Sci.TotalEnviron. 1991,
with suspendedsediment- methodologyand applications.Arch. 107,45.
Environ. Contam. Toxicol. 2003, 44, 288. (35) Yang,S.Y.N.;Connell,D. W.;Hawker,D.W.;Kayal,S.I.Polycyclic
09) Olson, M. C.; Iverson, J. L; Furlong. E. T.; Schroeder,M. P. aromatic hydrocarbons in air, soil and vegetationin the vicinity
Methods ofanalysisby the U.S.GeologicalSurveyNational Water of an urban roadway. Sci. Total Environ. 1991, 102,229.
Quality Laborato~Determination of polycyclic aromatic
hydrocarboncompounds in sediment by gaschromatographyI Receivedfor review January 24, 2005. Revised manuscript
mass spectrometry;U.S. Geological Survey Water Resources receivedApril 29, 2005.AcceptedMay 2, 2005:
Investigations Report03-4318;U.S.GeologicalSurvey: Denver,
CO, 2004. E50501565

5566. ENVIRONMENTAL SCIENCE & TECHNOLOGY./ VOL 39. NO. 15.2005

S-ar putea să vă placă și