Agricultural and Biosystems Engineering
Publications
Agricultural and Biosystems Engineering
3-1994
Occurrence of Atrazine and Degradates as
Contaminants of Subsurface Drainage and Shallow
Groundwater
K. Jayachandran
United States Department of Agriculture
T. R. Steinheimer
United States Department of Agriculture
L. Somadunaran
Du Pont Agricultural Products
Thomas B. Moorman
United States Department of Agriculture
Ramesh S. Kanwar
Follow
and additional
works at: http://lib.dr.iastate.edu/abe_eng_pubs
Iowa Statethis
University,
[email protected]
Part of the Agriculture Commons, Bioresource and Agricultural Engineering Commons,
Entomology Commons, and the Water Resource Management Commons
See next page for additional authors
The complete bibliographic information for this item can be found at http://lib.dr.iastate.edu/
abe_eng_pubs/586. For information on how to cite this item, please visit http://lib.dr.iastate.edu/
howtocite.html.
This Article is brought to you for free and open access by the Agricultural and Biosystems Engineering at Iowa State University Digital Repository. It
has been accepted for inclusion in Agricultural and Biosystems Engineering Publications by an authorized administrator of Iowa State University
Digital Repository. For more information, please contact [email protected].
Authors
K. Jayachandran, T. R. Steinheimer, L. Somadunaran, Thomas B. Moorman, Ramesh S. Kanwar, and Joel R.
Coats
This article is available at Iowa State University Digital Repository: http://lib.dr.iastate.edu/abe_eng_pubs/586
Published March, 1994
Occurrence
of Atrazine
and Degradates as Contaminants
Shallow Groundwater
of Subsurface
Drainage
and
K. Jayachandran, T. R. Steinheimer,* L. Somasundaram,T. B. Moorman,R. S. Kanwar, and J. R. Coats
Atrazine is a commonlyused herbicide in corn (Zea mays L.) growing areas of the USA.Because of its heavy usage, moderatepersistence,
andmobility in soil, monitoring of atrazine movement
underfield conditions is essential to assess its potential to contaminategroundwater.
Concentrationsof atrazlne, deisopropylatrazine (DIA), and deethylatrazinc (DEA)were measured in subsurface drainage and shallow groundwater beneath continuous, no-till corn. Water samples were collected
from the subsurface drain (tile) outlets andsuction lysimeters in the growing seasons of 1990 and 1991, and analyzed for atrazine andtwo principle degradates using solid-phase extraction and HPLC.In 1990, atrazinc concentration ranged from 1.3 to 5.1 pg L-I in tile-drain water and
from 0.5 to 20.5 ~tg L-~ in lysimeter water. In general, concentrations
of parent and degradates in solution were atrazine > DEA> DIA. Lesser
levels of atrazine were measuredin 1991 from Plots 2 and 4; however,
greater concentrations of atrazine (6.0-&4 I~g -1) were measured from
Plot 5. Throughoutthe two growing seasons, atrazine concentration in
Plot 5 tile-drain water wasgreater than that of Plots 2 and 4, suggesting
a preferential movementof atrazine. Concentrations of DIAand DEA
rangedfrom 0.1 to 2.2 and 0.9 to 3.2 ~tg L-1, respectively, indicating that
the degradation products by themselves or in combination with parent
atrazine can exceed the maximum
contaminant level (mcl) of 3 Ixg -~
even though atrazine by itself maybe <3 ~tg L-~. The deethylatrazineto-atrazine ratio (DAR)is an indicator of residence time in soil during
transport of atrazine to groundwater. In Plots 2 and 4, DARvalues for
tile-drain water rangedfrom 0.43 to 2.70 and 0.50 to 2.66, respectively.
By comparison, a DARof 0.38 to 0.60 was observed in Plot 5, suggesting
less residence time in the soil.
G
agricultural
chemicals
as becomea growing concern
in the USA
because
40 to 50%of domestic drinking water is pumpedfrom
groundwater
resources (Jury, 1982; Hallberg, 1986). Several studies suggestthat waterpollutionis the mostdamaging and widespreadenvironmentaleffect of agricultural
production(Cohenet al., 1984: Hallberg, 1989;Ritter,
1990). As a result, numerousmonitoringprogramshave
beenconductedto determinethe presenceof agricultural
contaminantsin surface water andgroundwater.Thepotential of pesticides to contaminateour groundwater
has been
well documented(Hallberg, 1989; Rostad et al., 1989;
Spaldinget al., 1989).
Atrazine (2-chloro-4-ethylamino-6-isopropylamino-striazine) is a preemergent
herbicidecommonly
usedin corn
(Zea maysL.) growingregions of Iowa, wherenearly 3.2
x 106 kg a.i. is appliedannually. Atrazineis also one of
the commonly
detected compounds
in groundwaterin Iowa
RhOUNDWATER
POLLUTION by
K. Jayachandran, T.R. Steirdaeimer, and T.B. Moorman,USDA-ARS,
National Soil Tilth Lab., 2150 PammelDr., Ames, IA 50011; L. Somasundaram, DuPont Agricultural Products, Wtlmington, DE 19880; and R.S.
Kanwar,Dep. of Agric. and BiosystemsEng., and J.R. Coats, Dep. of Entomology,IowaState Univ., Ames,IA 50011. Joint contribution of the USDAARSand Journal Paper no. J-15295 of the Iowa Agric. and HomeEcon.
Exp. Stn., Ames, IA. Project no. 2306. Received 23 Mar. 1993. *Corresponding author.
Published in J. Environ. Qual. 23:311-319(1994).
311
andthe Midwestregions (Hallberg, 1989;Spaldinget al.,
1989;Hartzler and Jost, 1990;Stoltenberget al., 1990;
Goolsbyet al.; 1991). Thefrequencyofatrazine detection
in groundwaterresults from its intense usage, moderate
persistence, andmobilitythroughthe soil profile (Spalding et al., 1989).
Asatrazine movesthroughthe soil profile, several factors includingmoisture,temperature,aeration, soil chemical properties, andmicrobialactivity influencethe fate
of this compound.Atrazine is degradedboth chemically
and biologically. Dealkylationof atrazine is the mainbiotic degradationpathwayin the soil environment.Several
speciesof bacteriaandfungi are effective at deethylation,
formingdeethylatrazine (2-chloro-4-amino-6-isopropylamino-s-triazine; DEA),and somespecies of microorganismsare moreproficientat deisopropylation,
formingdeisopropylatrazine(2-chloro-4-ethylamino-6-amino-s-triazine;
DIA) (Kaufman and Kearney, 1970; Behki and Khan,
1986). Hydrolysisof atrazine to hydroxyatrazine(2-hydroxy-4-ethylamino-6-isopropylamino-s-triazine)
is the major abiotic degradativepathway(Skipperet al., 1967;Obien
and Green,1969). TheN-dealkylationof atrazine by soil
microorganismsis potentially moreimportant for water
quality studies, becausethe resulting compounds
are more
water-soluble and moremobile in the soil profile than
hydroxyatrazine.Hydroxyatrazineis morestrongly bound
to the soil than atrazine, DEA,and DIA,and therefore
is less mobilein soil (Russellet al., 1968;Helling,1971),
so that the possibility of leachinginto groundwater
appears
to be minimal.
Mostof the groundwater-monitoring
programsfocus only
on the parentcompound
andvery little informationis available on the potential of pesticide degradationproducts. A
few studies havedetected the presenceof atrazine degradation products, DIAand DEA,in groundwater(Muir and
Baker, 1976; Rostad et al., 1989; Adamsand Thurman,
1991).Becausemetabolitesare also toxic froman exposure
point of view (Hallberg, 1989), measuringthe environmentalfate of pesticide degradationproductsis increasingly important. Both atrazine metabolites are primary
aminoderivatives of an azaarenering system.Thesetypes
of organic compoundsare knownmammalian
carcinogens
(Hayesand Laws,1991). Thus,their occurrencein water
resources drawnfor domesticsupplyconstitutes a potential risk to humanhealth. The current maximum
contaminantlevel (mcl) for atrazine is 3 ~tg -~ in drinking
water; however,
this doesnot considerthe degradationproducts (Bellucket al., 1991).
Soil microorganisms
convert significant quantities of
atrazine to DEA.Basedon this metabolicprocess, a deAbbreviations: DEA,deethylatrazine; DIA,deisopropylau-azine; mcl, maximumcontaminantlevel; DAR,deethylatrazine-to-atrazine ratio; PVC,polyvinyl chloride; TBA,terbuthylazine; SPE, solid-phase extraction; CH,cyclohexyl; DIAR,deisopropylatrazine-to-atrazine ratio.
312
J. ENVIRON.
QUAL.,VOL.23, MARCH-APRIL
1994
ethylatrazine-to-atrazine ratio (DAR)has been hypothesized as an indicator of hydrogeologic and geochemical
conditions existing in shallow groundwater (Adamsand
Thurman, 1991). The DARalso may be an indicator of
residence time in the soil during transport of atrazine.
Prolongedresidence time in soil during atrazine transport
should result in large DARvalues, whereas, mixing with
atrazine-contaminated groundwater from external pointsources or originating from preferential flow processes
would be expected to yield small DAR(<0.5) values.
This study wasinitiated to monitoratrazine and its degradation products at the Till HydrologySite experimental
plots, 13 km west of Ames, IA, in which corn was continuously grown under no-till management. Water sampies were collected from tile-drains and suction lysimeters
during the growing seasons of 1990 and 1991 following
herbicide application to determinethe concentration of atrazinc and its degradation products that movedinto the subsurface drainage system.
MATERIALS AND METHODS
Site Description
The research field is located at the Agronomy
and Agricultural EngineeringResearch Center, 13 krn west of Ames,IA.
Theexperimental
plots are locatedon a Nicollet loamsoil (fineloamy, mixed, mesic Aquic Hapludolls) with a maximum
slope
of 2 %.This soil series consists of deep, moderateto poorly
drainedsoils formedin loamyglacial till underprairie vegetation. Thegroundwatersystemconsists of an unconfinedaquifer
within a highly weathered
glacial till and includesa perchedwater table varyingfroma depthof<l mto >3 m. The4.0-ha field
is divided into 10 experimentalplots in whichcorn was grown
continuouslysince 1984using no-till and conventionaltillage
practices. Threeno-till plots (Plots 2, 4, and 5) wereselected
for studying the movement
of atrazine and its degradatesinto
shallowgroundwater.The area of the experimentalplots were
3360m2 for Plot 2, 4050for Plot 4, and 2310for Plot 5.
Eachexperimental
plot is drainedby a separatetile-drain (10.2
cmdiam.)spacedat 36.6 mapart. Tile drains wereinstalled in
this area in 1961at a depth of 122 cm.Details on installation
of deepsumpsto intercept tile lines andmeasurement
of tile flow
rates are described elsewhere(Kanwaret ai., 1988). A set
two polyvinylchloride(PVC)suction lysimeters were installed
in twolocations in eachplot at 90- and 150-cmdepthsto collect
water periodicallyfor pesticide analysis. Installation of these
devices includeda soil slurry packover the porouscups. Aseries of five piezometers(3.8 cmdiam.plastic pipe with an open
bottom)werealso installed in Plot 4 at depthsof 120, 180, 240,
300, and360cmto collect water fromthese depthsfor pesticide
analysis. Piezometerswere constructed from PVCcasings and
not polyethyleneor polypropylene;furthermore,all wells were
purgedprior to sampling.Therefore,all atrazine loss fromadsorption is assumedto be minimal.
Thepesticide application history for the experimentalfield
is listed in Table1. Atrazine wasbroadcast applied as a preemergentherbicide at a rate of 1.68 kg ha-1 1and corn, Pioneer
hybrid no. 3475, was planted on 2 May1990. Plot 5 received
atrazine broadcastat a rate of 2.24 kg ha-~ on 14 Nov.1990for
a rainfall simulation study (Table 1). Noatrazine was applied
in 1991.
Table1. Summary
of pesticide use historyin the experimental
plots
at the Till Hydrology
Site (1986-1991).
Pesticide? Application
Year
Crop
Variety
applied
rate
-1
kgha
1986
Corn
Pioneer3475
Alaclflor
2.80
Cyanazine
2.24
Carbefuran
1.45
1987
Corn
Pioneer3475
Atrazine
2.24
Metolachlor 2.80
Chlorpyrifos 1.45
Alachlor
2.24
1988
Corn
Pioneer3475
Cyanazine
2.24
Chlorpyrifos 1.45
Glyphosate
1.68
1989
Corn
Pioneer3475
Alachlor
2.24
Cyanazine
2.24
Chlorpyrifos 1.45
Glyphosate
1.68
1990
Corn
Pioneer3475
Atrazine
1.68
Metolaehlor 2.24
¯ 1.12
2,4-D
Glyphosate
1.12
Terbufos
1.45
1991
Corn
Pioneer3475
Alaeldor
2.24
Cyanazine
2.24
2,4-D
1.12
Glyphosate
1.12
Terbufos
1.45
Additional
atrazine(2.24kgha-1) wasbroadcast
appliedonPlot5 in
November
1990.
Sample Collection
Watersamplingwas started in early May1990after planting
andherbicide application, and continueduntil 20 August.At the
beginningof this field study, samplingsequenceswere planned
and executed on a weeklyinterval basis; later, as the study
progressed, the samplingsequencesweremodifiedaccording to
the tile-flow events. Samplingwas terminatedafter 20 August
(110d after atrazineapplication)becausetile flowceased.Asimilar situation occurredin the followingseason(1991),wheresampling was started immediatelyafter planting 10 May1991and
continueduntil 5 June 1991.
Watersamples were collected in acid-washe~500-mLglass
bottles. Duplicatesampleswerecollected fromthe overflowon
a V-notchweirlocatedin a manhole
at the tile-drain outlet. Sampies weretransported to the laboratory and immediatelystored
in the dark at 4 °C until analysis. Thevolumeof each sample
collected fromthe tile-drain outlet rangedfrom250to 500mL,
with an averagevolumeof abut 400mL.Precipitation data (Fig.
1Aand 2A)wererecordedat a meteorologicalstation at the research center. Daily tile flow rate measurements
(Fig. 1B and
2B)wereusedto calculate the amountof atrazine drainedthrough
the tile lines. Basedonthe daily tile floweventsandanalytical
data points, a linear regressionequationwasusedto extrapolate
fromlate MaythroughAugustfor both growingseasons to estimatethe loss of atrazine and degradatesin the tile-drain water.
The stage recorder and calibrated V-notchweir were installed
in Plot 5 in 1991,so tile flowdata werecollectedonlyfor Plots
2 and 4 in 1990.
Watersampleswere also collected for analysis from suction
lysimeters and piezometersfor analyses. About24 h before sampiing, standingwater waspumped
and discardedfromthe piezometer and a vacuum
wasapplied to each lysimeterto collect fresh
~ Mention
of a tradename,proprietaryproduct,or specificequipment samples.Samplecontainers and storage procedurewere as described earlier. Watersamplesfrompiezometerswere collected
doesnotconstitutea guarantee
or warranty
bythe USDA
anddoesnotimply
only in the monthsof Mayand June 1990. Samplesfrom lysiits approval
to the exclusion
of otherproducts
that maybe suitable.
JAYACHANDRAN
ET AL.:
313
CONTAMINANTSOF DRAINAGE& GROUNDWATER
A
5
5
1 O0
120
140
160
180
200
220
240
100
120
Time (Day of Year)
0.30
140
160
180
200
220
240
Time (Day of Yeer)
0.20
B
,:~ Plot 2
¯
Plot ~.
0.25
,, Plot 2
v
~>, 0.16
~ Plot 5
~ 0.12
o.15
--j
v
0.08
,,
~ 0.04
0.05
o.oo .....
140
160
0.00
180
200
220
240
Time (Day of Yeor)
100
120
140
160
1BO
Time (Day of Yeer)
Fig. 1. (A) Daily precipitation events from April to August1990 at the
Till Hydrology Site, and (B) daffy tile-flow measurementsfrom May
to August 1990. Arrows indicate sampling period.
Fig. 2. (A) Daily precipitation events from April to August1991 at the
’l~ Hydmk~ly $~e ~d (~) ~ly file-flow me~’ement~ from Ap~ql
to June 1991. Arrows indicate sampling period.
meterswerecollected after sufficient precipitation. Thevolume
of samplescollected fromeach lysimeter varied from20 to 400
mL,with an average volumeof about 300 mL.The lack of soil
moistureprevented samplecollection from lysimeters at some
samplingperiods.
tions were controlled from an HPChemStationrunning PASCALSeries software (Rev. 5.21). Liquid chromatography
separations wereaccomplished
using acetonitrileowatergradients on
LiChrospher100 RP-18(Hewlett-PackardCompany,
Little Falls,
DE)column(150 by 4 mm,5 ~tmspherical packing) at 40
The analytical columnwas preceded by a guard columnof the
samepacking (4 by 4 mm).Identifications were madeby peak
comparisonswith spectra stored in a user-generatedlibrary of
referencespectra. Six-pointexternal-standardquantitative calibration plots weredevelopedfor eachanalyteby injection of 25lxL volumesof serially diluted four-component-standard
mixtures. Correlationcoefficients (r:) for the standardplot of DEA,
atrazine, and TBA
were0.999 and 0.996for DIA,all within the
concentrationrange of 0.25 to 2.50 mgL-1. Analytical puritycertified reference standards were providedby the Pesticide
Repository, USEPA,
Research Triangle Park, NC, and by the
Riedel deHaen(Crescent ChemicalCompany,Hauppage,NY).
A parallel Quality Assuranceand Quality Control program
was implementedto assure the precision and accuracy of the
results. Checksampleswere preparedby spiking surface water
representative of agricultural land drainage. Recoverystudies
on grab samplestaken during base-flowfromthe south fork of
the SkunkRiver (near Ames,IA) showedDEA,atrazine, and
TBA
wereconsistently >90%.Recoveryefficiency for DIA,about
Sample Analysis
Sampleswere analyzedby the methodof Steinheimerand Ondrus (1990) as amendedby Steinheimer (1993). Test sample
volumesof 100to 500 mL(typically 250mL)were spiked with
a surrogate compound,terbuthylazine (2-(tert-butylamino)-4chloro-6-ethylamino-s-triazine;TBA),at 1 Ixg -1. Three analytes together with the surrogate wereremovedfrom water by
solid-phase extraction (SPE) on cyclohexyl (CH)cartridges
(AnalytichemBondElut, VadanSamplePreparation Products,
HarbourCity, CA)and followedby elution with pesticide-grade
MeOH
(Burdick and Jackson, Muskegon,MI). Final extract volumewas reduced to 0.5 mLby concentration under N~stream
at 45 °C.
Instrumental determination was achieved on a Model1090M
Seres II liquid chromatograph
(Hewlett-Packard
Company,
Little
Falls, DE)equippedwith an autosampler,a thermostatedcolunm
compartment,and a photodiode-arraydetector. Systemopera-
ENVIRON.QUAL., VOL. 23~ MARCH-APRIL
1994
Days after planting
40
60
80
Days after
100
40
1.0
Plot 2 o
Plot 4 ~
Plot 5
5
60
80
100
~
t-
~ ~, .
~;
_
0,8
Planting
/-~
Plot
,5 ~
~.j 0.6
<
0.4
0.2
0.0
160
180
40
200
220
160
240
180
200
220
Time (Day of Year)
Time (Day of Year)
Days after planting
Days after planting
60
80
40
100
9.0
4.0
C
Plot 2
Plot 4
Plot 5
3.2
D
o
¯ T
60
I , , , I ,
80
240
100
, , | , , , I
Plot 2
Plot 4
Plot 5
T
$
/±X
~A
3.0
0.8
1.5
0.0
0.0
160
180
200
220
240
Time (Day of Year)
60
180
200
220
240
Time (Day of Year)
Fig. 3. Concentrationsof (A) atrazine, (B) deisopropylatrazine (DIA), (C) deethylatrazine (DEA),and (D) atrazine + degradates in tile-drain
during 1990 wowingseason. Error bars represent standard error of mean.
60%,required a matrix-spikerecovery correction for the analyte. Eachinstrumental sequenceconsisted of 10 to 12 sample
extracts sandwichedbetweentwo methanolblanks and two standard mixtures treated as unknowns,both at the beginningand
at the end of the sequence.All injections wereduplicated, with
coefficients of variation calculatedto be 7.5, 3.3, 2.7, and1.4%
for DIA,DEA,atrazine, and TBA,respectively. Meanrecovery
of TBAfor the entire samplingand analysis regimewas 97 %.
Method
detection limits are estimatedat 0.03, 0.05, and 0.10 ~tg
L-~ for DIA,DEA,and atrazine, respectively.
RESULTS AND DISCUSSION
Occurrence of Atrazine Congeners in Tile-Flow Drainage
The concentrations of atrazine, DIA, and DEAdetermined during the two growing seasons of 1990 and 1991,
axe presented in Fig. 3 and 4. Because each experimental
plot was different in tile-drainage pattern (Fig. IB and 2B)
water
and transport of atrazine and degradates into groundwater
(Table 2), results from each plot are presented separately.
The overall concentrations of atrazine in tile-drain water
during the period from 43 to 110 d after herbicide application ranged from 1.9 to 2.9 ~tg L-t in Plot 2, 1.3 to 2.0
ttg L-~ in Plot 4, and 3.1 to 5.1 gg L-~ in Plot 5 (Fig. 3A).
Theseresults are consistent with other reports of similar
or greater concentrations of atrazine found in shallow
groundwatersystems (Muir and Baker, 1976; Hall and Hartwig, 1978; Spalding et al., 1979; Wehtjeet al., 1984; Hallberg et al., 1985; Pionkeet al., 1988; Isensee et al., 1990;
Kanwar,1991; Gaynor et al., 1992). Throughoutthe sampiing periods in 1990, atrazine concentrations in Plot 5
were two times greater than in Plots 2 and 4. Although
there is a slight lateral gradient in groundwaterflow as a
result of <2%relief, topography does not significantly
influence the water table in any of the three plots. There-
JAYACHANDRAN
ET AL.:
CONTAMINANTSOF DRAINAGE & GROUNDWATER
Days after planting
10
0
10
20
Plot 2
Plot 4 ¯
3 15
Days after planting
30
0
2.5
,
B
T
2.0
8
,
Plot
Plot
Plot
10
20
I , , I I ,
2 ,’
4 ¯
/~
~
5 ~
3O
1.5
1.0
0.5
0.0
130
140
150
30
160
Time (Day of Year)
0
, ,
10
I ,
20
, , I ,
150
6O
Time (Day of Year)
Days after planting
3.5
140
Days after planting
3O
10
, , I
, ,
Plot 2 ,,
Plot ¢ ¯
14
3.0
,
20
I ,
T
4_
12
2.5
~ 10
1.5
÷
~ 4
30
, I
,
T
I
I
1.0
0.5
130
140
150
160
Time (Day of Year)
30
140
150
160
Time (Day of Year)
Fig. 4. Concentrationsof (A) atrazine, (B) deisopropylatrazine (DIA), (C) deethylatrazine (DEA),and (D) atrazine + degradates in water
during 1991 growing season. Error bars represent standard error of mean.
fore, responsesto groundwater
level changesin Plot 5 are
no different than Plots 2 and4. Preferential flow through
macroporescould be of greater importancein this plot.
Preferential flow of atrazine throughmacroporeshas been
well documented
(Isensee et al., 1990; Kanwar,1991).
Table 2. Atrazine and degradates loss with subsurface drainage water
during the growing seasons (May-Aug.) of 1990 and 1991.
Year
Plot no. Atrazine
DIA
DEA
Totalt
5.24
2.63
ND
2.39
1.38
1.88
17.61
8.95
ND
3.94
3.02
8.16
-I
gha
1990
1991
2
4
5
2
4
5
9.83
5.40
ND¢
1.11
0.69
5.40
Total = DIA + DEA+ atrazine.
ND= not determined.
2.54
0.92
ND
0.44
0.95
0.88
In the followingseason (1991), averageconcentrations
of 1.0 and 0.6 gg L-1 atrazine were measuredin Plots 2
and4, respectively(Fig. 4A).Theseresidues probablyrepresent carryoverfrompreviousyear’s application. In Plot
5, greater concentrationsof atrazine weremeasured
in tiledrain water (Fig. 4A) than in drainage from Plots 2 and
4. In 1991,there wasno spring atrazine application; however, Plot 5 receivedan additionalapplicationof atrazine
at 2.24 kg ha-1 in November1990. The average concentration of atrazine in Plot 5 was6 to 8 timesgreater than
Plot 2 and 10 to 14 times greater than Plot 4. Theearly
seasontile-flow pattern (Fig. 2B) in Plot 5 further supports the preferentialfloweffect on this plot. Intenserainfall events anda correspondingincrease in tile-flow has
beendocumented
by Muirand Baker(1976), Kanwar(1991),
andKanwar
et al. (1992); however,in our study there was
noimmediate
effect of heavyrainfall on flowrates of drain
316
J. ENVIRON.QUAL., VOL. 23, MARCH-APRIL
1994
Days after
40
60
, , , I
2.4
planting
(1990)
80
1 O0
, , , I
, ,
Plot 2
Plot 4
Plot 5
A
2.0
~ I
Days after
0
, , ,
o
*
v
3.75
I
planting (1991)
20
10
,
,
,
I
i
~
~
I
,
,
30
,
Plot 2
Plot 4
"1 B
~
3.00
1.6
2.25
1.2
1.50
0.8
0.75
0.4
0.0
0.00
160
180
200
220
240
130
Time (Day of Year)
40
Days after
60
planting
80
(1990)
C
Plot 2
Plot 4
Plot 5
0.5
150
160
Days after planting (1991)
10
20
3O
, , ~ I, , ~ I ,,
Plot 2
D
Plot 4
100
0.6
140
Time (Day of Year)
o
¯
~
0.4
5
4
0.3
0.2
2
0.1
1
0.0
Plot
5
0
60
180
200
220
240
Time (Doy of Year)
30
140
150
160
Time (Day of Year)
Fig. 5. (A) and (B) deethylatrazine-to-atrazine ratio (DAR)for We-drainwater collected during the growing seasons of 1990 and 1991, respectively,
and (C) and (D) deisopropylatrazine-to-atrazine ratio (DIAR)for the same samples of 1990 and 1991, respectively. Error bars represent standard
error of mean.
waterexceptin the early part of the growing
seasonof 1990
(Fig. 1 and 2). This was partly because 1988and 1989
wereextremelydry years anddepthof water table wasabout
3 m, in the beginningof 1990.Thetile-flow rates in our
samplingperiod are relatively small comparedwith the
early seasonrate. This could be due to the growingcorn
crop at the surface, whichtakes up a considerableamount
of water fromthe profile.
Twomajor degradates of atrazine, DIAand DEA,were
measuredin all water samplescollected during both seasons. Deethylatrazine
concentrationsin all three plots were
greater than those of DIA(Fig. 3B, 3C, 4B, 4C); however,
there wasa greater concentrationof DIAthan DEA
in Plot
4 on the 30 May1991 sampling. A large standard error
(0.642) associatedwith this data suggeststhat this may
an anomaly.Skipperand Volk(1972) showedthat microbial
removalof the ethyl side chainis 8 to 12 timesmorerapid
than for the isopropylside chain. Usingradiotracer techniques on similar soil associations taken from the same
field, Krugeret al. (1993a) showedthat DEAand fully
dealkylatedatrazine weremajordegradationproducts. They
also studied degradationand movement
in undisturbedsoil
columns(Krugeret al., 1993b).In both studies, DEA
was
a major degradationproduct followedby DIA.Others have
founda similar relationship (greater concentrationof DEA
than DIA)betweenthese two degradatesunder both field
and laboratory conditions (Sirons et al., 1973; Muirand
Baker, 1976; Schiavon, 1988; Adamsand Thurman,1991;
Gaynor, 1992).
Total atrazine residue (atrazine + DIA+ DEA)concentrations are important,becausethe principal degradates
are assumedto be at least as toxic as atrazine (Belluck
JAYACHANDRAN
ET AL.:
317
CONTAMINANTSOF DRAINAGE & GROUNDWATER
et al., 1991), andtheir presenceis of concernat greater
levels. Total atrazine load exceeded
the presentmclin sampies from three plots in both growingseasons. Theone
exceptionwasin Plot 4, in whichthe total load wasless
thanthe mcl.Sincethe s-triazine ring is resistant to cleavage with only small amounts14CO2being liberated from
ring-labeleds-tdazines in soil (KaufrnanandKarney,1970;
Skipper and Volk, 1972; Wolfand Martin, 1975), atrazine’s degradatesare expectedto persist in someformin
soil as well as in groundwater.Bellucket al. (1991)suggests that the total atrazine residue, atrazineplus its degrades, can exceedatrazine regulatory limits eventhough
atrazine does not.
Lossesof AtrazineCongeners
throughTile-FlowDrainage
Quantitativeloss of atrazineanddegradatesin tile-drain
water is presentedin Table2. In the 1990growingseason
(May-August),
the loss of atrazine anddegradatesis twice
as great for Plot 2 as for Plot 4 (Table2). Lossof atrazine
and degradatesfromPlot 5 for 1990wasnot calculated,
becauseonly limitedtile-flow data on that plot wereavailable. Calculations on Plots 2 and 4 showedthat 0.30 to
0.60%of the applied atrazine appearedin the tile-drain
water as the parent compound.
Thetotal atrazine loss was
1.05 and 0.50%of that applied to Plots 2 and4, respectively. Gaynor
et al. (1992)reporteda similar loss of atrazine combinedwith DEA
(1.2-1.4 %)in the tile-drain water from a Brookstonclay loam under continuous-corn
production.However,Muirand Baker(1976) showeda loss
of only 0.15%of appliedatrazineandits degradationproducts in the tile-drain water.
Loss of atrazine and degradates were calculated from
all three plots in 1991.Thoughatrazine wasnot applied
in Plots 2 and 4, measurablequantities of atrazine and
degradatesweredischargedthroughtile-drainage flow(Table 2). Agreater quantity(0.14%)of atrazinewaslost from
Plot 5 than fromPlots 2 (0.07%)or 4 (0.04%). However,
consideringthe additional applicationof atrazine (2.24 kg
ha-1) on this plot, the percentageloss maynot be greater
than that whichoccurredin the previousgrowingseason.
This suggeststhat the amountof rainfall governstransport
of atrazine in the tile-drain water.Thecumulative
seasonal
rainfall in 1990(73.2 cmfrom Mayto August)wastwice
the cumulativerainfall in 1991.
Atrazine Congenersin Lysimeterand PiezometerSamples
Duringthe early portion of the 1991growingseason,
lysirneters yielded recoverablesample.However,
later in
the growingseasonreducedrainfall resulted in soil conditions in whichno sampleswerecaptured. Amaximum
atrazine concentrationof 20.5 ~tg L-1 wasmeasuredat a 90cmdepth in Plot 2 on the 18 June 1990sample(Table 3).
Hall and Hartwig(1978)reported a similar level of atrazinc fromsuctionlysirneters at a 120-cm
depthin silty clay
andclay loamsoils. Theconcentrationof atrazine in Plot 2
gradually declined to 2.4 lxg L-~ by 20 Aug.1990(Table
3). At the 150-cmdepth, the concentration of atrazine
wasless than at the 9O-cmdepth, but followeda similar
trend as seen at the 90-cmdepth. In general, Plot 2
showedgreater concentrationsof atrazine whencompared
"Pable 3. Concentrations of atrazine and its degradates in suction
lysimeter water at the 90 and 150-cm depth (1990-1991).~"
Date of
collection
14 June 1990
Plot
Atrano. Depth zine DIA DEA TotalS:
cm ~ ttg
L -~
-2
4
5
18 June 1990
2
4
5
20 June 1990
2
4
5
6 July 1990
2
4
5
14 Aug. 1990
2
4
5
20 Aug. 1990 2
4
5
12 June 1991
2
4
5
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
90
150
NS#
NS
NS
1.8
4.0
NS
20.5
3.6
NS
1.3
9.7
4.7
13.9
4.2
0.7
NS
NS
2.8
8.9
2.6
NS
0.4
4.2
1.5
5.3
1.7
NS
0.6
2.6
1.2
2.4
NS
0.6
NS
NS
NS
3.7
2.0
NS
0.2
18.0
NS
NS
NS
NS
0.2
0
NS
0.3
0.3
NS
0.1
0.4
0.5
0.6
0.3
0.2
NS
NS
0.5
1.3
0.4
NS
0.2
0.2
0
1.2
0.5
NS
0.2
0.3
0
0.3
NS
0
NS
NS
NS
0.3
0.6
NS
0.7
1.7
NS
NS
NS
NS
0.4
1.1
NS
3.0
0.7
NS
0.1
3.5
0.9
2.5
0.8
0.3
NS
NS
0.6
3.3
0.9
NS
0.2
2.4
0
3.9
0.9
NS
0.6
2.7
0
1.0
NS
0.5
NS
NS
NS
3.1
3.2
NS
0.6
6.0
NS
NS
NS
NS
2.4
5.1
NS
23.8
4.6
NS
1.5
13.6
6.1
17.0
5.3
1.2
NS
NS
3.9
13.5
3.9
NS
0.8
6.8
1.5
10.4
3.1
NS
1.4
5.6
1.2
3.7
NS
1.1
NS
NS
NS
7.1
5.8
NS
1.5
25.7
NS
DIAR§ DAR¶
NS
NS
NS
0.10
NS
0.02
0.09
NS
0.13
0.06
0.12
0.06
0.08
0.30
NS
NS
0.24
0.18
0.20
NS
0.68
0.06
0.28
0.35
NS
0.33
0.16
0.17
NS
NS
NS
NS
0.09
0.34
NS
4.53
0.12
NS
NS
NS
NS
0.27
0.33
NS
0.17
0.24
NS
0.13
0.41
0.21
0.20
0.23
0.46
NS
NS
0.26
0.42
0.39
NS
0.63
0.67
0.85
0.57
NS
1.13
1.19
0.46
NS
1.00
NS
NS
NS
0.97
1.83
NS
3.56
0.38
NS
Asingle sample wascollected fromeach plot and depth and analyzed. Each
valueis a meanof replicate injections.
Total = DIA + DEA+ atrazine.
DIAR= deisopropylatrazine-to-atrazine ratio.
DAR= deethylatrazine-to-atrazine ratio.
NS= not sampled due to inadequate soil moisture.
with Plots 4 and 5, whichis in contrast to the observation
madeon tile-flow analysis, wherePlot 5 atrazine concentrations werealwaysgreater than those in Plots 2 and 4.
In 1991, the Plot 5 lysimeter sampleat the 90-cmdepth
hadgreater concentrationsof atrazine than Plot 2, due in
part to the additionalapplicationof atrazine on this plot.
Deisopropylatrazineand DEAwere detected in most sampies, with the exceptionof the fewat whichthe concentration wasnear the quantitation limit. Throughout
the sampiing periods, DEAconcentrations at both depths were
greater than DIAconcentrations.Thetotal atrazine residue (atrazine + DIA+ DEA)concentrations were greater
at a 90-cmdepth fromPlot 2 comparedwith the other two
plots.
Theaverageatrazine and degradates concentrationsin
the individualpiezometersamplestakenat different depths
318
J. ENVIRON. QUAL., VOL. 23, MARCH-APRIL 1994
Table 4. Concentrations of atrazine and its degradates in piezometer
water collected in the months of May and June 1990 from Plot 4.t
Date of
collection
18 May 1990
12 June 1990
DBA Total* DIAR§ DAR1
Depth Atrazine DIA
cm
120
180
240
300
360
120
180
240
300
360
, _i
Ug L
5.6
13.8
0.3
1.8
0.1
6.4
13.1
0.1
0.1
-
0.5
1.0
0.3
_
0.2
0.4
0.8
0.2
_
-
0.7
0.9
0.1
—
0.5
1.1
—
_
-
6.8
15.7
0.7
1.8
0.3
7.3
15.0
0.3
0.1
-
0.11
0.09
1.08
_
2.64
0.08
0.07
1.92
_
-
0.14
0.07
0.41
_
-
0.09
0.10
—
_
-
t A single sample from each depth was collected and analyzed. Each value
is a mean of replicate injections.
t Total = DIA + DBA + atrazine.
§ DIAR = deisopropylatrazine-to-atrazine ratio.
1DAR = deethylatrazine-to-atrazine ratio.
during May and June of 1990 are presented
in Table 4.
The average concentration of 6 ng L"1 atrazine was detected at the 120-cm depth, whereas at the 180-cm depth,
a concentration of 13.5 (ig LT1 was measured. The atrazine concentration was significantly decreased below the
180-cm depth, suggesting that atrazine is moving at a much
slower rate beneath that depth. The distribution of degradates at different depths in the piezometers followed a pattern similar to atrazine, although concentrations were diminished.
Geochemical Implications of Parent to Degradate Ratios
For the soil metabolism of atrazine, a parent-to-metabolite ratio has been proposed as an indicator of biogeochemical processes occurring in the soil, which influence groundwater quality. In both growing seasons the DAR values in
Plot 5 did not change during the sampling periods (average DAR 0.55 in 1990 and 0.41 in 1991), suggesting a possible role for preferential flow through macropore environments, thus allowing less time for deethylation of atrazine
(Fig. 5A and 5B). In Plots 2 and 4, the DAR values during
1990 were at 0.43 and 0.60 in the early part of the growing
season and gradually increased to 1.95 and 1.27 during
mid to late summer, indicating that the biodegradation process correlated directly with residence time. A smaller DAR
at the beginning of the growing season followed by a gradual
increase in the later part of the season was also reported
by Adams and Thurman (1991).
Because DIA was also detected in almost all samples,
and microbial removal of the isopropyl side-chain is well
documented, a deisopropylatrazine-to-atrazine ratio (DIAR)
was also calculated and correlated with DAR values. In
general, the DIAR values were2 smaller than DAR values.
The correlation coefficients (r ) between DAR and DIAR
were 0.65 for 1990 and 0.47 for 1991 data. The majority
of the DIAR values were less than 0.5 and remained relatively constant. Following formation in the soil, DIA is
either rapidly degraded into other chemical congeners or
is rapidly leached through the soil profile by virtue of its
enhanced hydrophilicity compared to DBA. Therefore, the
use of DIAR value to explain the movement of atrazine
does not offer the same utility as the DAR values.
The DIAR and DAR values for lysimeter and piezometer samples are presented in Tables 3 and 4. In lysimeter
samples, particularly at a 90-cm depth from the plot where
a greater concentration of atrazine was detected, the DAR
value was 0.17. Where lesser atrazine concentration was
measured, greater concentration of DEA was observed,
producing DAR values as great as 0.85. This suggests
prolonged residence tune for atrazine in the surface soil,
resulting in greater microbial degradation.
CONCLUSIONS
Atrazine, DIA, and DEA were consistently detected in
both shallow groundwater and subsurface tile-drainage beneath no-till continuous corn plots in central Iowa during1
2 yr. Atrazine concentration exceeded the mcl of 3 ng L"
in a significant percentage (>40%) of samples. Variability
in concentration of atrazine and degradates was associated
with characteristics of the individual plots (atrazine
0.58.4, DIA 0.1-2.2, and DEA 0.9-3.2 ug LT1). The ratio
of deethylatrazine-to-atrazine (DAR) suggested that shorter
residence time in the soil allowed minimal microbial contact in Plot 5, where the greatest atrazine concentrations
were determined in tile drainage. Atrazine was detected
in both shallow groundwater and hi tile drainage several
months following the most recent application. Because most
tile drainage is discharged to the surface, contamination
with atrazine and degradates represents a risk to surface
water quality. However, the occurrence of these contaminants in the shallow groundwater system does not preclude movement into deeper aquifer environments.
ACKNOWLEDGMENTS
The authors are grateful for financial support from the Leopold
Center for Sustainable Agriculture, the USDA North Central Region Pesticide Impact Assessment Program, and the USDA
(ARS/CSRS) Management Systems Evaluation Area (MSEA)
program. We also acknowledge Dr. E.M. Thurman and Dr. W.C.
Koskinen for their critical analyses of our paper.
JAYACHANDRAN ET AL.: CONTAMINANTS OF DRAINAGE & GROUNDWATER
319