EPA/600/D-85/136
June 1985
REMOVAL OF AGRICULTURAL CONTAMINANTS FROM GROUNDWATER
by
Ben W. Lykins, Jr.
U.S. Environmental Protection Agency
Water Engineering Research Laboratory
Drinking Water Research Division
Cincinnati, OH 45268
Joseph H. Baier
Suffolk County Deparrment of Health Services
225 Rabro Drive East
Hauppauge, New York 11788
Grant
CR 811109
WATER ENGINEERING RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
CINCINNATI, OH 45268
-------
TECHNICAL REPORT DATA
(Pleate read Instruction on iht revtnt before completing)
t. RCPORT NO.
EPA/600/D-85/136
2.
3. RECIPIENT'S ACCESSION NO.
M-f; '•:. : ;22 /AS
4. TITLE AND SUBTITLE
Removal of Agricultural Contaminants from Groundwater
5. REPORT DATE
June 1985
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
B. W. Lykins* and J. H. Baier+
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
*WERL-Cin, OH ^Suffolk County Dept. Health
USEPA 225 Rabro Drive East "
Cincinnati, OH Hauppauge, NY 11788
45268
11. CONTRACT/GRANT NO.
CR811109
12. SPONSORING AGENCY NAME ANO ADDRESS
Water Engineering Research Laboratory-Cincinnati, OH
Office of Research and Development
U.S. Environmental Protection Agency
Cincinnati, OH 45268
1?. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
EPA/600/14
15. SUPPLEMENTARY NOTES
Project Officer, B.W. Lykins, (513) 684-7460;
16. ABSTRACT
As analytical testing of community and individual homwowner wells has intensified,
more drinking water contamination has been identified. In some instances, this
contamination can be attributed to agricultural practices. Of special concern are
those locations where no community treatment system can be provided. Examination of
various treatment methods applicable to both individual and small community
situations that are cost-effective in removing these contaminants will address an
area of concern that is gaining more attention. In situations where multiple
contaminants, expecially low moledular weight compounds, are found, pilot studies
similar to this study with proposed treatment systems are recommended to more
accurately predict the performance of larger units.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (This Report)
UNCLASSIFIED
21. NO. OF PAGES
17
20. SECURITY CLASS (This page I
UNCLASSIFIED
22. PRICE
EPA F«rm 2220-1 (R»v. 4-77) PKIIVIOU* EDITION i» «—•
-------
NOTICE
This document has been reviewed in accordance with
U.S. Environmental Protection Agency policy and
approved for publication. Mention of trade names
or commercial products does not constitute' endorse-
ment or recommendation for use.
11
-------
REMOVAL OF AGRICULTURAL (V>"7i.".lNANTS FROM GROUNDWATER
Ben W. Lykins, Jr.
Drinking Water Research Division
Water Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, Ohio 45268
Joseph H. Baier
Suffolk County Department of Health Services
225 Rabro Drive East
Hauppauge, New York 11788
Introduction
Suffolk County, New York's groundwater has been designated as a sole
source aquifier under provisions of the Safe Drinking Water Act (P.L.
93-523). In recent years there has been increasing concern about the
contamination of this groundwater by agricultural chemicals (ferti-
lizers, insecticides, herbicides, neraatocides, and fungicides). %This
concern expanded when specific chemicals were identified in homeowner
drinking wells.
For over eight years, Suffolk County has examined their groundwater
for agricultural organic constituents and their decay products. During
this testing, 101 agricultural/organic compounds were evaluated with 41
found in the groundwater. Many of these identified contaminants were
present in trace quantities but aldicarh, carbofuran, 1,2-dichloro-
propane, and 1,2,3-trichloropropane were at elevated levels. In addi-
tion, nitrates from fertilizer application were present in quantities
exceeding the primary drinking water standards.
A cooperative agreement was initiated by the U.S. EPA Drinking Water
Research Division, Cincinnati, Ohio to examine the cost-effectiveness
and removal efficiency of certain water treatment systeas to remove
organic and pesticide contamination, in combination with nitrate, from
Suffolk County groundwater under various flow situations. Two parallel
treatment systems, granular activated carbon plus ion exchange versus
reverse osmosis, will be operated at low flow, similar to home usage.
The coses of these two systems, together vith unit operating effi-
ciency, will be established so that a larger public water supply system
c 'n be designed and tested. The results from this study will be applic-
a: le to other areas where multiple contamination of groundwater is
identified, especially in farming communities.
Background
Agriculture has been a major industry in Suffolk County's North Fork
(Figure 1) for over 200 years. Approximately 26,000 acre.* of land in
this area (36%) is cultivated primarily for potatoes.'*' The North
Fork soils, consisting of 6 to 24 inches of loamy top soil over thick
layers of coarse sand and gravel, are particularly suited for growing
potatoes.
Fertilization practices, with as much as 250 Ib of nitrogen per acre
applied, led to widespread nitrate contamination of the shallow
aquifer.(2) The potato plant is susceptible to a number of pests, most
nobably the golden nematode, which attacks the roots, and the Colorado
potato beetle, which eats the leaves. Since the early 1950s, pesti-
*
1
-------
cides containing 1 ,2-dichloropropane have been applied to fields
infesced with golden nemacodes. In 1974, the carbamate pesticide
aldicarb (TEMIK, Union Carbide Corp.) was registered for use on
potatoes and by 1976 was being used at an application rate of about
3 pounds of active aldicarb per acre. (3)
Aldicarb was used extensively for four growing seasons in Suffolk
County before its sale was prohibited. A direct relationship exists
between the proximity of a well to an agricultural field and the
presence of aJdicarb. Based upon the flow pattern of the groundwater,
Suffolk County and Cornell University have estimated that over 100
years will be required to purge the aquifer of aldicarb. O)
Groundwater contamination by agricultural chemicals is not specific to
Suffolk County. The states of Florida, Wisconsin, and Rhode Island
have also identified problems. For instance, Rhode Island found
aldicarb, carbofuran, and vydate up to concentrations of 150, 7, and 2
ug/L, respectively, in the grcundwater. After testing 73 drinking
water wells, 5 contained aldicarb above 10 ug/L and 1A wells had a
Well Contamination
The New York State guideline for aldicarb is 7 ug/L and some repre-
entative concentrations in private wells in Suffolk County are shown in
Table 1.
Testing for 1,2-dichloropropane began in 1980. In one community, this
contaminant was found in 17 of 33 wells with two wells containing
concentrations around the New York State guideline of 50 ug/L. A
second community showed 2 of 9 wells with concentrations of 10-15 ug/L.
and a third area^had a private well with a concentration of 49 ug/L.(*'
Testing is being done for other organic contaminants as new agricultural
pesticides are applied.
A large portion of the aquifer contains nitrate levels that approach or
exceed the U.S. EPA and New York State drinking water standard of 10
mg/L. The major source of the nitrate is fertilizer. Shown in Table 2
are nitrate concentrations from two communites in Suffolk County.
R search Project Objectives
Fo: effective crop production, fertilizers containing nitrates and
che-.icals to control insects, weeds, fungus, etc. will continue to be
applied. Therefore, treatment of groundwater contaminated by this type
of application will be required to produce an acceptable drinking
water.
The objectives for this study were developed to address the removal of
multiple contaminants from a groundwater source. These objectives were
(1) to determine appropriate treatment for removing organics and/or
pesticides in combination with nitrate from a groundwater source and
(2) to determine cost effectiveness and removal efficiencies under
different flow situations simulating homeowner use versus a municipal
water system.
Within the framework of this research initiative, two different sized
treatment systems will be evaluated. One system will consist of using
small-scale flow rates comparable to home treatment systems to evaluate
(1) carbon adsorption (CAC) followed by ion exchange, and (2) reverse
2
-------
osmosis. Operation and maintenance costs along with removal effi-
ciencies will be documented wliile opeiating in contiguous and inter-
mittent modes simulating homeowner use. The second system will con-
sist of che most effective treatment methodology selected from the
sma'l-scale flow rate study with evaluation on a larger scale simula-
ting centralized treatment.
Site Selection
Selection of a well site for the small-scale flow rate research effort
included reviewing data from over 14,000 wells. Water quality repre-
sentative of what homeowners normally encounter rather than excep-
tionally poor quality was one selection criteria. Ten sites were
selected for further consideration based upon water quality (Table 3).
Additional requirements for selection included the following factors:
(1) site accessibility, (2) existing buildings available and amount of
site work and construction required, (3) availability of utilities, (4)
available assistance and cooperation from site owners, (5) use of
effluent, (6) facilities to handle regeneration and reject water, (7)
well capacity, and (8) security. Table 4 shows these factors applied
to the potential research sites.
Also, private wells were eliminated from consideration because of
potential problems that could occur during the research activities.
Another well was eliminated because of seasonal use. Therefore, three
sites presented in Table 3 had the potential required qualities (1, 2,
10). Site 10 was selected because of the availability of a large
building to house the treatment equipment and the presence of typical
water quality.
•*
R. 0. Membrane Evaluation
As the literature was searched and membrane manufacturers contacted,
there was an apparent lack of information on the performance of reverse
osmosis (R.O.) membranes for the combination of contaminants of concern
(aldicarb, carbofuran, 1,2-dichloropropane, 1,2,3-trichloropropane, and
nitrate) at Suffolk County. Therefore, pilot testing of applicable R.O.
membranes was done to evaluate contaminant removal before proceeding
with the planned research effort. Known manufacturers of R.O. membranes
were contacted and asked to supply a commercially available membrane(s)
they thought would remove these contaminants.
Received for parallel bench-scale evaluation were the following types
of membranes that correspond to the R.O. unit sizes on-site.
0 Culligan cellulose acetate tubular, 2 inch
0 Dupont polyamlde spiral-wound, 2-1/2 inch
* Dupont polyamide hollow-fiber, 4 inch
* FilraTec polyamide spiral-wound, 2-1/2 inch
* Hydranautics polyamide spiral-wound, 2-1/2 inch
" Fluid Systems polyamide spiral-wound, 2-1/2 inch
0 FilmTec polyamide spiral wound, 2 inch
All units were operated continuous1, at a pressure of 160-200 psi.
Each membrane was tested from 5 to 24 weeks. The test units were fed
from a common raw water source and sampled twice weekly. Results from
this evaluation are shown in Table 5 and the following observations can
be made.
3
-------
* Cellulose acetate is not satisfactory for certain organlcs
removal, but it was able to remove carbamaCes.
* Polyamide membranes c.in sntisfactorily remove carbaoaCes and
nitrates.
• The polyamide membrane from FtlmTec showed the best removals
during this limited test.
0 Dupone's spiral polyamide membrane did nor. perfona as well as
ics hollow-fiber configuration.
0 Dupont's hollow-fiber performed equally to the FllnTec spiral-
wound polyamide membranes for 1,2,3-trichloropropane and some-
what poorer for 1,2-dichloropropane.
Although recovery v*s not an Issue during this testing, Che hollow-
fiber did consistently recover more water at the same operating pressure
of 160 pst. It also produced no pH decrease in Che effluent, and the
product water conductivities were higher than Che spiral-wound configu-
ration (Table 6). The latter parameters are desirable features in
product water relaClve to corroslvlcy.
As a result of these tests, a performance specification was developed
for a polyamide membrane unit:
Contaminant Rejection
carbofuran 95%
aldlcarb sulfoxide 95%
nitrate 952
aldlcarb sulfone 90%
total dissolved solids 85*
1,2-dichloropropane 672
1,2,3c-crichloropropane 67%
at: 400 psi feed pressure
5.0 gpm product flow rate
55°F.
50% recovery (minimum)
400 hours of use
either spiral-wound or hollow-fiber Is an
acceptable configuration
CAC and Ion Exchange Selection
Manufacturers of granular activated carbons and ion exchange resins
were contacted to determine Che appropriate type of media Co meet Che
research project objectives. Seven different GACs from five manufac-
turers and four Ion exchange resins from four manufacturers were evalu-
ated. Batch Isotherm tests were run Co determine Che capacity of the
CACs and resins as a comparative evaluation for selection of one CAC
and resin for furCher sCudies. Race sCudies were noC performed.
GACs evaluated included: CECA CAC 40 and CAC 1240, ICI Darco H-90 Plus
and Hydrodarco 4000, Calgon Flltrasorb 300, WesCvaco Nuchar WV-G, and
jybron lonac P-50. Rohm and Haas Amber lite XAD-4 adsorpClve resin was
"""also evaluated. The Ion exchange resins rested for nitrate removal
effecclveness Included: Sybron lonac ASB-2, Purolite A-300, Duollte
A-101D, and Rohm and Haas Amberlite IRA-410. All of these are strong-
base type anion exchange resins.
-------
Based on the Isotherm study, ICI Darco H-90 Plus and Calgon Flltrasorb
300 CACs were selected for further study. All of the four Ion exchange
resins tested were comparable in performance and were listed in the bid
specifications.
Pilot Plant^
The existing 2" diameter well at the selected site yielded a flow of
about 15 gpm. From design estimates, a total of 30 gpm will be needed
for the various treatment options. Before replacing the well, a test
well was Installed adjacent to the existing well and sampled at 10
different depths ranging from 46 to 130 feet below grade. The major
contaminants of concern all had different profiles (Table 7).
Total aldicarb increased to a maximum of 91 ug/L at 71 feet and
decreased to <1 ug/L at 112 feet. At 62 feet, 1 ,2-dichloropropane was
present Increasing to a m.iximum of 68 uf ' at 112 feet and still present
at 132 feet. Nitrate was detected at the surface of the water Cable to
a maximum of 13 tng/L at S2 feet with a concentration of 5 mg/L at 132
feet. A screen setring at 70-80 feet below grade surface was selected
as the most appropriate depth to evaluate the contaminants of concern
and be consistent with normal well depth in Suffolk County.
The well will discharge into a 10,000 gallon storage tank so that a
constant supply of water can be ensured for the system and allow for
"spiking" of contaminants should the quality of well water change
(Figure 2). Following the storage and pumping system are in-line
fibrous material filters to remove any particulates, oxidized iron, etc.
After the filters, the flow will be split into the reserve osmosis and
adsorption/ion exchange systems. All discharges (treated, reject,
backwash, and regenerate water) will be directed to a main drain line
that will convey the water to a cesspool located downgradient fron the
well.
Summary
As analytical testing of coraaunlcy and Individual homeowner wells
has intensified, more drinking wa:«r cone jmtnation has been Identified.
In some instances this contamination can b« attributed to agricultural
practices. Of special concern are those locations where no community
treatment system can be provided. Examination of various treatment
methods applicable to both individual and small community situations
that are cost-effective in removing these contaminants will address an
area of concern that is gaining mere attention. In situations where
nultip'.e contaminants, especially low molecular weight compounds, are
four. '. . pilot studies similar to this study with proposed treatment
systems are recommended to more accurately predict the perfornnca of
units.
Acknowledgements
The authors want to ".hank the following for help in completing
this paper: Keith Cnrswell and Carol Ann Fronk of the U.S. EPA for
their technical review and Patricia Pierson and Maura Lilly for typing
the manuscript; Steven Kramer and Tom Martin of Suffolk County for
equipment set-up and data collection.
-------
REFERENCES
1. Baler, Joseph H. and Robbin:;, Sy F., "Groundwater Contamination
from Agricultural Chemicals: North Fork, Suffolk County", in
Proceedings of Che ASCE Specialty Conference, pp. 1-13, 1983.
2. Baler, Joseph H., ami Rykbost, Kenneth A., 1976. The Contribution
of Fertilizer Co the Croundwater of Long Island, published in
Journal of NWUA, November-December 1976.
3. Baier, Joseph H., and Moran, Dennis, 1981, Status Report on
Aldicarb Contamination of Croundwater as of September 1981, Suffolk
County Department of Health Services, Hauppauge, New York.
4. Toxic Materials News, Vol. 11, No. 33, p. 277, August 29, 1984.
-------
fARMI-AND
FIGURE 1, FARMLAND ON SUFFOLK COUNTY, NEW YORK'S NORTH FORK
-------
TABLE 1. ALDICARB CONCENTRATION IN PRIVATE DRINKING WATER
Coramunicy
Number of
Wells Sarapleu
1-7 ug/L
> 7 ug/L
1
2
3
4
5
222
434
2161
1832
3160
18
46
345
256
374
2
43
351
270
359
TABLE 2. NITRATE CONCENTRATIONS IN PRIVATE DRINKING WATER WELLS<2)
Number of
Connunity Wells Sampled 0-5 mg/L 5-10 mg/L > 10 cg/L
639
1121
372
575
167
354
100
19?
-------
TABLE 3. WATER QUALITY RESULTS l-'ROM POTENTIAL STUDY SITES
Piraneter
Free AmonU, ng/L H
Nltr«tea, t>K/L H
r«
Spec. Cond. , u»!io»/cm
Chloride* , «g/U
Sulf*tet, »R/L
Iron, «g/L
Manganese , *g/L
Cooper, •!•/(.
Zinc, «r/l.
SoUiun, »R/L
T. Hirdneti, »g/U C«CO3
T. AIUllnHy. ng/L CaCOj
Aldletrb, OR/L
Cirbofuran, ug/L
Ductlul, ug/t,
1,2-dIctiloropropane, ug/L
1.2,3-Cr£chloroprop»ne, ug/L
5 tie Location
12 V * 5 6 78 9 10
<.04
10. B
6.1
510.
5'..
126.
.12
<.05
<.10
<.*
11.2
196.
12.
11.
—
Nil
NU
HO
<.0'«
13.2
6.3
660.
A3.
201.
.16
<.05
<.10
<.4
24.2
288.
1A.
86.
21.
90.
HI)
NU
<.lKi
7.*
5.0
590.
89.
90.
<.ll)
<.OS
<.«)
C.4
3*.«
176.
12.
1.
nu
HI)
Nil
Nl>
•"i.S
.4
6.9
K30.
125.
105.
>25.10
>7.20
.21
<•«
115.
-272.
388.
HI*
1.
HI)
36.
6.
<.04
*.*
6.3
4UO.
67.
46.
<.to
<.05
<.H)
<.4
21.5
80.
26.
7.
—
ND
24.
HI)
<.04
25.5
5.9
650.
AS.
162.
.IB
<.05
.12
<.4
11.5
324.
16.
84.
8.
HI)
Ml)
NU
<.04
9.0
6.0
475.
33.
100.
.15
<.05
.17
<.4
17.7
~
~
16.
—
NO
NO
NO
8.10
1.7
6.0
430.
30.
108.
1.78
.05
.10
5.8
12.2
. —
—
NU
NO
ND
5.
ND
<.04
13.8
—
—
23.
90.
<.10
<.05
<.10
<.4
7.5
156.
—
9.
16.
ND
ND
ND
.04
9.1
5.9
450.
40.
105.
2.6
.16
.45
<.4
1.2
196.
1C.
33.
9.
ND
37.
20.
ND: Hoc detected
-------
TABLK A. FBATUKKS OF POTENTIAL STUDY SITES
Factors for Consideration
Accessibility
Utilities
Assistance
Discharge
Well Capacity
Security
Housing
Construction
Capability
Site Location
1 2 34 5 6 7 8 9 JO
/
good
good
good
good
400 gpm
poor;
no fence
poor;
sm.bldg.
good
good
good
poor
good
40 gpm
poor;
no fence
poor; in
ground
poor
good
poor; no
electric
good
poor
200 gpm
poor; no
fence
poor;
sm. bldg.
good
good
good
good
good
200 gpm
good
poor; no
bldg.
good
Rood
good
good
good
275 gpm
poor;
no fence
noor;
sm.bldg.
good
good
poor
poor
poor
10 gpm
NA
good;
Ig. barn
NA
good
poor; no
electric
good
poor
200 gpm
poor;
no fence
poor;
sm. bldg.
good
NA
NA
NA
poor
10 gpm
pobr
poor; in
ground
poor
good but
seasonal
good
good
good
100 gpm
good
boiler
room
good
good
good
good
good
15 gpm
good
good ; i n
Ig.bldg
good
NA: information not available
-------
TABLE 5. REVERSE OSMOSIS MEMIJRANE EVALUATION
Type of
Membrane
Cellulose Acetate
Dupont Spiral
Wound
Dupont Hollow
Fiber
Hydranautics
Spiral Wound
FIlmTec Spiral
Wound
FIlmTec Spiral
Wound
Fluid Systems
Spiral Wound
Diameter
of
Membrane
2 in.
2-1/2 in.
4 In.
2-1/2 in.
2 in.
2-1/2 in.
2-1/2 in.
Feed
Flow
Kate
L/mln
(gpd)
2.2
(836)
1.0
(380)
2.6
(988)
1.3
(494)
1.35
(513)
2.2
(836)
1.2
(456)
I
Recov-
ery
6
10
50
13
10
•
5
16
x
Aldlcnrb*
Sulfoxlde
Feed
uR/L
39
39
39
39
V
39
39
39
Perm.
UR/L
<1.0
<1.0
'
<1.0
1.0
<1.0
<1.0
2.0
Aldicnrb*
Su If one
Feed
UR/L
f,7
47
47
47
47
47
47
i'e rm .
UR/L
3.0
1.0
2.0
2.0
-------
TABLE 6. pH OF R.O. PRODUCT WATER
Feedwater
Cellulose Acetate
Dupont Spiral Wound
Dupont Hollow Fiber
Hydranautics Spiral Wound
Filmtec Spiral Wound
Filmtec Spiral Wound
Fluid Systems Spiral Wound
pH
5.6
5.2
5.6
5.5
4.6
4.9
4.8
4.9
Spec. Cond.
umhos/cra
551
43
36
38
23
15
19
30
12
-------
TABLE 7. DATA FOR TEST WELL AT VARIOUS DEPTHS
Well
Depth
feet
45-47
51-53
61-63
71-73
81-83
91-93
101-103
111-113
121-123
131-133
Aldi-
carb
ug/L
<1
3
24
91
26
4
28
<1
<1
«
1,2-
Dichlor-
propane
ug/L
<2
<2
34
19
41
59
50
68
47
9
Nl-
C rates,
mg/L
4.3
4.6
11.9
12.8
13.4
8.0
9.2
6.4
6.2
»•'.
Cnrbo-
furan
ug/L
<1
<1
8
18
12
2
8
<1
<1
a
1,2,3-
Trtcliloro-
propane,
ug/L
<2
<2
20
15
17
15
15
13 '
11
2
PH
6:0
6.0
5.6
5.6
5.8
5.9
6.0
6.1
6.1
6.4
Chlo-
rlden,
mg/L
30
31
34
48
36
24
30
23
24
73
Sul-
fates,
mg/L
130
189
108
138
129
78
111
82
73
53
Iron
mg/L
0.9
0.8
i.2
0.9
0.2
0.3
0.3
2.9
0.3
1.8
Manga- •
nese
mg/L
11.4
10.4
0.17
0.07
<0.05
0.05
0.05
0.26
<0.05
0.21
Sodium
mg/L
30.5
36.4
12.1
11.7
10.5
12.8
10.5
10.8
15.0
15.0
Hardness
mg/L as
CaC03
192
272
196
204
260
140
176
128
112
132
Alkalinity,
mg/L as
CaC03
88
106
8
6
6
8
io
14
8
14
-------
GAG
GAO
GAG
1
o
REVERSE
OSMOSIS
PERMEATE
!
r
r
S..
REGENERAT
STORAGE
TANK
\
REJECT WATER
WELL
DISCHARGE
FIGURE 2, FLOW SCHEMATIC FOR PILOT PLANT - SUFFOLK COUNTY, N.Y.
------- |