STEPHEN N. HAUGHEY

shaughey©fbttaw.com

(513) 651-6127

FROST ROWN TODD

LLC

2200 PNC Center

201 E. Fifth Street

Cincinnati, Ohio 45202-41 82

(513) 651-6800

Facsimile (513) 651-6981

www.frostbrowntodd .com

RECEIVED

ci n~’c

~

MAR 2 8 2002

SlATE

Pollutton Control Board

March 25, 2002

Dorothy Gunn, Clerk

Illinois Pollution Control Board

100 W. Randolph Street, Suite 11-500

Chicago, Illinois 60601

Re:

R 02-Oh/Rulemaking/Water/Comments ofDeKalb Sanitary District

Dear Ms. Dunn:

Enclosed please find for filing in the above-referenced docket the original and ten copies

of the comments of DeKaib Sanitary District in support of IEPA’s proposed triennial review

rulemaking. Please stamp the original and all copies and return the extra copy to me in the

enclosed, self-addressed, postage-prepaid envelope for our files. Thank you for your assistance

in this matter. Please contact me if there are any questions about this filing.

Very truly yours,

FROST BROWN TODD LLC

cc:

Marie E. Tipsord, Hearing Officer (w/encl.)

Mathew Dunn, Illinois Attorney General’s Office (w/encl.)

Office ofLegal Services, IDNR (w/encl.)

IEPA, Division ofLegal Counsel

Service List ofParticipants (w/encl.)

ClNlibrary/1

162595.1

Enclosures

)IZ9~4f~f

Stephen N. Haughey

Outside Counsel to the District

OHIO

•

KENTUCKY

•

INDIANA

•

TENNESSEE

---

ORIGINAL

RECEIVED

BEFORE THE ILLINOIS POLLUTION CONTROL BOARMAR 282002

In the matter ofi

)

)

WATER QUALITY TRIENNIAL REVIEW:

Amendments to

35

Ill. Adm. Code 302.208(e)-(g),

302.504(a), 302.575(d),

303 .444, 309.14 1(h), and

Proposed 35 Ill. Adm. Code 301.267, 301.3 13,

301.413, 304.120, and

309.157

STATE OP IWNOIS

R

02~011!OIlUtIOndontroi Board

(Rulemaking

—

Water)

In Support OfIEPA ‘s Proposed

Water Quality Standards Rulemaking

TO:

Dorothy Gunn

Clerk

Illinois Pollution Control Board

100 W. Randolph Street, Suite 11-500

Chicago, Illinois 60601

Mathew Dunn

Illinois Attorney General’s Office

Environmental Control Division

James R. Thompson Center

100 W. Randolph Street

Chicago, Illinois 60601

Marie E. Tipsord

Hearing Officer

Illinois Pollution Control Board

100 W. Randolph Street, Suite 11-500

Chicago, Illinois 60601

Office ofLegal Services

Illinois Department ofNatural Resources

524

5. Second Street

Springfield, Illinois 62701-1787

Illinois Environmental Protection Agency

do Division of Legal Counsel

1021 N. Grand Avenue East

P.O. Box 19276

Springfield, Illinois 62794-9276

Please take notice that the DeKaib Sanitary District (hereinafter “DeKaib” or the

“District”) hereby offers the following comments in support of IEPA’s proposed revisions to

Illinois’ water quality standards. The proposed revisions are part of the State’s triennial review

)

)

)

)

)

of its water quality standards as required by the federal Clean Water Act.

---

Comments ofLIeKalb Sanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page

2

A.

BACKGROUND.

Among the revisions proposed by the JEPA to Illinois’ water quality standards and

implementing rules is the proposed conversion of Illinois’ water quality criteria for certain

hardness-based metals, including copper, from the present total recoverable form ofthe metal to

a dissolved form ofthe metal, and proposed rules establishing a procedure for developing water

quality based permit limits for these metals in the federally-required total recoverable form using

default or site-specific conversion factors for the proposed dissolved metals criteria. IEPA’s

rules presently allow the use of default or site-specific conversion factors from total recoverable

to dissolved form of the metals only in the Lake Michigan basin.

See

35 Ill. Adm. Code

309.141(h)(3). The proposed revisions would expand the use of these factors statewide. The

proposed revisions are set forth in Title

35,

Subtitle C, Chapter I, Parts 301, 302, 304, and 309 of

the Illinois Administrative Code. The specific sections affected are 301.267, 301.313, 301.413,

302.208, 302.504, 309.141, and 309.157.

B.

NATURE

OF DEKALB’S INTEREST.

The DeKaib Sanitary District was incorporated in 1928. Its service area is a population

of approximately 45,000 people in the City of DeKaib, the Town of Cortland, nearby Northern

Illinois University, and a small group of commercial and industrial customers. The District

employees 15 people. The District owns and operates a medium-sized, trickling filter, activated

sludge biological publicly-owned treatment works (“POTW”) that has a mean annual average

daily flow ofapproximately

5.95

mgd and a design average daily flow capacity of8.63 mgd.

Treated effluent from the POTW is discharged into the South Branch of the Kishwaukee

River pursuant to the terms and conditions of an NPDES permit issued in 1995 by the Division

of Water Pollution Control for IEPA, which permit was renewed in November 2000. The

2

---

Comments ofDeKaib Sanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page

3

renewal permit contains daily maximum and 30-day monthly average, water quality-based

effluent limitations (“WQBELs”) for copper of 0.056 mg/i

56

parts per billion (“ppb”) and

0.034 mg/i 34 ppb, respectively.

In early 2000 the District undertook an evaluation of the sources of copper in its

collection system and determined that more than

95

of the copper coming to the POTW for

treatment was from leaching of residential copper plumbing lines due in part to water softening

systems used to offset exceptionally hard water found in the area. Because of the aesthetic

problems associated with hard water, a substantial percentage of homes in the service area have

private water softeners. These softeners tend to increase the corrosivity of the treated water

during the removal of calcium, magnesium, and sodium-based minerals that contribute to the

hardness of the water and its aesthetic problems. The historic use of private water softeners in

the service area makes it very difficult for the District to limit copper from entering the sanitary

collection system for transport to the District’s POTW.

As part of its efforts to consistently achieve its stringent copper WQBELs, the District

engaged its engineering consultants to investigate the potential costs of modifying the

wastewatertreatment processes at the POTW to remove additional copper from the influent. The

District’s engineering consultants estimated that the costs to modify the POTW in order to

remove additional low ppb levels of copper from the influent would be more than $500,000 in

capital cost, with annual O&M costs ofmore than $600,000.

Based on the primary source of the copper and the substantial expense to consistently

remove low ppb levels of copper from the effluent produced by a conventional biological

trickling filter, activated sludge plant, the District and its engineers decided, with the support of

representatives for JEPA’s Division of Water Pollution Control, to submit a petition to the IPCB

3

---

Comments ofDeKalb Sanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page

4

for a site-specific, adjusted water quality standard for copper using U.S. EPA’s dissolved metals

translator guidance entitled: “The Metals Translator: Guidance for Calculating a Total

Recoverable Permit Limit from a Dissolved Criterion,” EPA 823-B-96-007, June 1996

(hereinafter “U.S. EPA’s Metals Translator Guidance”). Thatpetition was filed with the IPCB in

mid August 2000, and is captioned as

“In the matter of Petition of DeKaib Sanitary Districtfor

an Adjusted Standard from 35 Iii. Adm. Code 302.208(E),

Docket No. AS 01-3 (Adjusted

Standard

—

Water).” The petition is pending before the IPCB.

The portion ofIEPA’s proposed rulemaking relating to dissolved metals is premised upon

the same U.S. EPA guidance document that was used by the District to support its petition for

adjusted standard. If the IEPA’s proposed rulemaking is approved by the IPCB, it would moot

DeKaib’ s petition in that the District would no longer need to complete a formal petition process

before the 1PCB in order to obtain adjusted WQBEL5 for copper based on the total to dissolved

ratio of copper in its treated effluent, but would instead, after the IPCB approves the proposed

revisions, file a request with the IEPA to modify the District’s permit based on the new rules.

Thus, DeKaib has a substantial interest in the outcome ofthe IEPA’s proposedrulemaking.

C.

THE APPROVAL OF IEPA’S PROPOSED

RULEMAKING

IS IN THE BEST

INTEREST

OF ILLINOIS’

SEWER AUTHORITIES GENERALLY.

As a general rule, biological activated sludge wastewater treatment works are designed

primarily to remove the conventional pollutants found in domestic sewage,

i.e.,

ammonia,

organic loading in the form of CBOD, suspended solids, and fecal coliform bacteria. A well-

designed and operated POTW with at least secondary treatment technology typically removes

greater than

90-95

ofthe influent loadings for these parameters. On the other hand, biological

wastewater treatment works typically remove no more than an average of 60-70 ofthe influent

4

---

Comments ofDeKaIb Sanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page

5

loadings of most metals, including copper, and the removals thereof are almost exclusively

through adsorption or uptake by the biomass of the metals and transfer thereof into the sludge

generated at the works, rather than through direct treatment processes.

See

Exhibit A hereto.

The District’s average removal rate for copper is higher than the average at 78, but even at that

level the District has difficulty consistently meeting the current WQBEL5 for copper in its

permit.

In order to meet low ppb WQBEL5 for metals, including copper, a POTW will typically

look first to the possibility of reducing the influent loadings by imposing pretreatment limits on

the largest categories ofcustomers discharging copper into the collection system. Such step can

often be successful in reducing influent copper loadings when the primary sources ofcopper are

industrial and commercial dischargers. These sources can usually be retrofitted cost-effectively

with small treatment systems such as dissolved air floatation (DAF) units, chemical

additionlprecipitationlsettling units, and ion exchange colunms. However, the area served by

many POTWs, including DeKaib, often has a very small industrial and commercial base,

typically contributing less than

5

of the copper influent loading to the POTW. Where the

predominant source of copper is leaching from residential plumbing, enhanced by water

softening units used to combat naturally occurring hard water, as is the case in DeKalb, the sewer

authority cannot readily impose pretreatment limits and control systems on its local residents.

Without a readily available means to significantly reduce the influent loadings ofcopper,

many POTWs in the same situation as DeKalb face the unpleasant alternative of incurring

substantial capital expenditures and large annual O&M costs to modify existing wastewater

treatment equipment in order to remove additional low ppb levels of copper from the final

effluent. The District believes that such additional costs are unnecessary and unjustified to

5

---

Comments ofDeKaIb Sanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page 6

remove such low concentrations of copper that are already well below applicable state and

federal drinking water standards and that, for the reasons demonstrated below, do not present a

threat to aquatic species.

D.

U.S. EPA’S

DISSOLVED METALS

TRANSLATOR GUIDANCE

OR ITS

EQUIVALENT IS WIDELY ACCEPTED

IN

THE SCIENTIFIC COMMUNITY,

HAS BEEN ADOPTED BY ALL NEIGHBORING STATES,

AND

CURRENTLY

IS IN USE IN ILLINOIS AS PART OF THE EXISTING NPDES PERMIT RULES

FOR POINT SOURCE DISCHARGES INTO THE LAKE MICHIGAN BASIN.

On October 1, 1993, U.S. EPA’s Office of Water issued a policy memorandum entitled

“Office of Water Policy and Technical Guidance on Interpretation and Implementation of

Aquatic Life Metals Criteria.” A copy of the policy memorandum is enclosed as Exhibit B

hereto. In that policy memorandum, U.S. EPA stated as follows:

It is now the policy of the Office of Water that the use of dissolved metal to set

and measure compliance with water quality standards is the recommended

approach, because dissolved metal more closely approximates the bioavailable

fraction of metal in the water colunm than does total recoverable metal.

This

conclusion regarding metals bioavailabiity is supported by a majority of the

scientific community within and outside the Agency.

One reason is that the

primary mechanism for water colunm toxicity is adsorption at the gill surface,

which requires metals to be in the dissolved form.

See

Exhibit B at p.2 (emphasis added). The policy memorandum states that, although the

particulate form of metals may contribute somewhat to the toxicity of metals, the toxicity of

particulate metals is dramatically less than that of the dissolved form of the metal.

Id.

The

policy memorandum further states that the toxicity tests used to develop metals water quality

criteria (Illinois developed its state water quality criteria for metals based on the federal water

quality criteria) involve the addition of simple metal salts to pure water, resulting in overly

conservative water quality criteriabased on the dissolved form ofthe metal salts in the water.

Id.

In addition, due to the presence of significance concentrations of binding agents in

6

---

Comments ofDeKaibSanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page

7

virtually all treated discharges and in most ambient waters

(e.g.,

calcium, magnesium, and

sodium minerals found in hard water, and naturally occurring humic compounds found in all

receiving waters), the metals in effluent discharges and in ambient waters are much less

bioavailable to aquatic species then the conditions under which the applicable water quality

criteria were developed.

Id.

The policy memorandum concludes by recommending that states’ water quality standards

for metals be based on the dissolved criteria, and that where states’ existing water quality

standards are based on the dissolved form of the metal that a total recoverable to dissolved

conversion factor be employed.

See

Exhibit B at p.3. The October 1993 policy memorandum

was subsequently incorporated within and replaced by U.S. EPA’s June 1996 Metals Translator:

Guidance, a copy ofwhich was submitted by IEPA in support ofits proposed rulemaking.

The 1996 Metals Translator Guidance or its functional equivalent has been adopted by all

states that border the Great Lakes as part ofthe implementation ofthe

1995

federal Great Lakes

Initiative. In Illinois, a rule permitting the use of dissolved metals translators has already been

promulgated for point source discharges to streams flowing into Lake Michigan.

See 35

Ill.

Adm. Code 309.141(h)(3).

IEPA’s proposed rulemaking seeks merely to expand the use of

dissolved metals translators from the Lake Michigan basin to all basins in the State ofIllinois.

Compare

proposed amended 35 Ill. Adm. Code 309.141 (h)(3) (deleting its limited application to

the Lake Michigan basin)

with

proposed new 35 Ill. Adm. Code 309.157 (transferring its

application to the general statewide NPDES rules). In other Great Lakes states, including

Michigan, Wisconsin, Ohio, and New York, the use of dissolved metals translators has already

been adopted for the entire state. Upon information and belief, the federal guidance or its

equivalent has been adopted in many other non-Great Lakes states.

7

---

Comments ofDeKalbSanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page

8

E.

CONCLUSION.

The DeKaib Sanitary District strongly supports JEPA’s proposed expansion of the use of

default and site-specific dissolved metals translators from the Lake Michigan basin to all basins

in the State of Illinois. The applicability of dissolved metals translators was developed by U.S.

EPA, the agency that developed the recommended metals water quality criteria upon which

Illinois has modeled its current metals water quality criteria. U.S. EPA’s recommendation to the

states to allow the use of such dissolved metals translators has been widely supported in the

scientific community, and is already incorporated within Illinois’ water quality standards,

albeit

at the present time only forpermit holders discharging into the Lake Michigan basin. There is no

logical reason to reject the expansion of the use of such translators to all permit holders in

Illinois.

Respectfully submitted,

DEKALB SANITARY DISTRICT

By___

FROST BROWN TODD LLC

201 B. Fifth Street, Suite 2200

Cincinnati, OH 45202-4182

(513)

651-6127

Counsel to the DeKaib SanitaryDistrict

8

---

Comments ofDeKaib Sanitary District

R 02-011 (Rulemaking

—

Water)

March 25, 2002

Page 9

CERTIFICATE OF SERVICE

In accordance with 35 Iii. Adm. Code 101.304 and 102.400, I hereby certify that a copy

ofthe foregoing comments was served by first class regular U.S. mail to all participants in this

proceeding, based on the current service list as ofMarch

25,

2002 (copy attached).

9

---

SERVICE LIST OF “PARTICIPANTS” AS OF MARCH 25, 2002

Lname

fname

company

address

citystate

Zip

Callahan

Mike

Bloomington Normal Water Reclamation District

P0 Box 3307

Bloomington, illinois

61702-3307

Cox

Larry

Downers Grove Sanitary District

2710 Curtiss Street

Downers Grove, Illinois

60515

Daffleld

Dennis

Department of Public Works City ofJoliet

921 E. Washington Street

Joliet, Illinois

60433

Ettinger

Albert

Environmental Law & Policy Center

35 E. Wacker Drive, Suite 1300

Chicago, Illinois

60601-2110

Frede

Lisa M.

Chemical Industry Council

9801 W. Higgens Rd, Suite

515

Rosemont, Illinois

60018

Gunn

Dorothy

Clerk, Pollution Control Board

100 West Randolph, Suite 11-500

Chicago, fflinois

60601

Barrington

James T.

Ross & Hardies

150 North Michigan, Suite 2500

Chicago, illinois

60601

Bill

Ron

Metropolitan Water Reclamation District

100 East Erie

Chicago, fflinois

60611

Nodge

Katherine

Hodge Dwyer Zeman

3150 Roland Avenue, P0 Box 5776

Springfield, fflinois

62705-5776

Howard

Margaret P.

Hedinger & Howard

1225 5. Sixth Street

Springfield, Illinois

62703

Messina

Robert A.

Illinois Environmental Regulatory Group

215 East Adams Street

Springfield, Illinois

62701

Muth

Tom

Pox Metro Water Reclamation District

682 State Route 31

Oswego, fflinois

60543

Polls

Irwin

Metropolitan Water Reclamation, Dist. Of Chicago

6001 West

Cicero, Illinois

60804

Sofat

Sanjay

Illinois Environmental Protection Agency

1021 North Grand Ave. East

Springfield, Illinois

62794-9276

Tipsord

Marie

Attorney, Pollution Control Board

100 West Randolph, Suite 11-500

Chicago, fflinois

60601

ClNlibrary/l 161370.1

---

Water Research

Vol

15. pp 1053

to

1065. 1981

Pnnted n Great Britarn. All rtghls reserved

ORIGINAL

0043-135481 091053.13502.000

Copynghc

C

1981 Pergamon Press Ltd

THE PHYSICOCHEMICAL SPECIATION OF Cd,

Pb, Cu, Fe AND Mn IN THE FINAL

EFFLUENT OF A SEWAGE TREATMENT

IN THE RECEIVING RIVER

DUNCAN P. H.

LAXEN

and Roy M.

HARRISONt

Department of Environmental Sciences, University of Lancaster, Lancaster LA!

4YQ.

England

(Received July

1980)

Abstract—A scheme for the speciation of metals in freshwaters has been applied to the analysis of the

final effluent from a sewage treatment plant and to the receiving river upstream and downstream of the

effluent outfall. The treatment plant was selected because of the high influent and effluent concentrations

of Cd. The metal speciation patterns in the effluent are interpreted primarily in terms of organic

interactions, which appear to be exerting a solubilizing effect on Cd and Cu. but not on the Pb and Fe

which are principally associated with the particulate size fraction (12

pm).

The Influx of metals with

the sewage effluent alters the speciation pattern in the river. A large part of the Cd is added to the

smallest size fraction (0.015 pm). However, the major part of each metal, with the exception of Mn. is

associated with the colloidal and particulate size fractions, thus minimising the immediate toxic signifi-

cance to aquatic life.

INTRODUCTION

Sewage treatment plants receive effluents from both

domestic and industrial sources, as well as intermit-

tent influxes of stormwater runoff from highways and

urban areas (Harrison & Laxen, 1981; Atkins & Haw-

ley. 1978). All three sources produce effluents con-

taminated with heavy metals and some part of these

metals will pass through the treatment plant to be

discharged to surface waters (Lester er at., 1979).

There is a trend towards the greater use of sewage

treatment plants to treat both industrial effluents and

stormwater runoff. On the one hand, this develop-

ment requires a better understanding of the role of

treatment processes irs metal abatement (Brown &

Lester. 1979). whilst on the other, it focuses attention

* Present address: Grant Institute of Geology, Univer-

sity of Edinburgh. West Mains Road, Edinburgh EH9

3JW. Scotland.

t

Author to whom correspondence should be addressed.

on the effluents from sewage treatment plants as point

sources of metal pollution (Harrison & Laxen. 1981).

The efficiency with which a sewage treatment plant

retains influent metals will depend upon the physico-

chemical forms of the metals and their response to the

physical and biological treatment processes. In all

treatment plants the initial stage is primary sedimen-

tation, whereby larger solids are allowed to settle out.

This is followed by either a biological filter or acti-

vated sludge process in order to digest the organic

matter, succeeded by passage through a settling

(humus) tank to reduce the residual solids loading

before discharge. Typical data for retention of metals

during these different processes are summarised in

Table 1. It would appear that actual retention varies

from plant to plant and within a plant from day to

day. The recent work by Stoveland

et

at. (l979a)

emphasises the importance of the metal speciation,

and their findings may account for some of the varia-

bility in the retention efficiencies. They found that the

Table 1. Removal efficiency for selected metals as a function of the treatment process

Removal (O/)*

Metal

Primary sedimentation

(Lester

et

at.. 1979)

Activated sludge

(Brown & Lester, 1979)

Biological filter

(Stones, 1977)

Cd

Cu

Pb

Fe

Mn

72(60—831t

70(50—95)

73(66—82)

50(11—80)~

71 (55—93)

67(43—89)

86 (72—97)

17(6—28)

59

70

62

* The data from the different references are not strictly comparable.

t

Mean and range for 4 values.

Mean and range of averages from 3 to 6 different studies.

1053

tSR

54

ii

---

1054

DUNCAN

P. H.

LAXEN

and Ro~M.

HARRISON

‘p

presence of the detergent builder, NTA. in the influent

can reduce the efficiency of metal removal in the acti-

vated sludge process. Furthermore, they found that

shock loading of NTA could mobilise some of the

metals previously retained within the sludge (Stove-

land

et

a!., 1979b). Such effects as these are important

in relation to how much metal will find its way to the

final effluent of a sewage treatment plant, but equally

important, in terms of defining the impact of the dis-

charge on the receiving water will be the speciation of

the metals within the final effluent. This will deter-

mine the metal speciation in the receiving water,

which in turn will determine both the immediate and

potential toxic significance to aquatic life, as well as

the mobility of the metals in the waterbody (Andrew

et al.,

1977; Davies

et

a!., 1976; Waiwood & Beamish,

1978; Wagemann & Barica, 1979: Harrison & Laxen,

1980).

In spite of the now well recognised importance of

defining the speciation of metals within a waterbody,

in relation to both their mobility and their toxic

effects, there is still considerable uncertainty as to the

actual metal forms and their various interactions

(Laxen & Harrison. 1977; Florence & Batley. 1980).

This can be accounted for by a number of factors,

including the wide range of possible physicochemical

forms, the difficulty of defining the forms analytically

and the low concentrations of the metals under con-

sideration. A number of studies have examined cer-

tain aspects of metal speciation in freshwater samples

(e.g. Chau & Lum-shue-chan, 1974; Bene~& Steinnes,

1975; Gardiner, 1976: Bene~

er

a!.. 1979). but only one

reasonably comprehensive speciation scheme has

been reported (Batley & Florence, l976a.b). This

scheme was developed principally for use with sea-

water samples and has unfortunately been shown to

be of limited applicability to the study of most fresh-

waters (Laxen & Harrison, l981a). We have however,

recently developed a speciation scheme for fresh-

waters, based upon an initial size fractionation of the

metal species, along the lines of the classification orig-

inally suggested by Stumm & Bilinski (1973) (Laxen

& Harrison, 198 Ia) (Fig. 1).

In this paper we describe the application of our

speciation scheme to the final effluent from a sewage

treatment plant which uses the biological filter pro-

cess, as well as to samples taken upstream and down-

stream of the effluent outfall. The particular sewage

treatment plant examined was chosen because of the

high cadmium loading in the raw sewage, a conse-

quence of discharges from a plating works. In all, five

metals were studied. Cd. Pb, Cu. Fe and Mn. The

latter two were included principally because of their

potential importance as metal scavengers, due to their

tendency to form colloidal/particulate hydrous oxides

(Stumm & Morgan. 1970; Mill, 1980), which have

strong adsorption affinities for certain metsis (Gadde

& Laitinen, 1974).

EXPERIMENTAL

Special ion scheme

The detailed development and testing of the speciation

scheme is described elsewhere (Laxen & Harrison. 1981a).

a

N

p

I

n

I

I

U

Fig. 1. Speciation scheme.

I

U

a

1

I

a

a

a

---

The physicochemical speciation of Cd. Pb. Cu. Fe and Mn

1055

The size fractionation of the metal species was performed

with Nuclepore filters of 12. 1.0. 0.4. 0.08 and 0.015 pm

rated pore size in Amicon stirred cells (Model 52) and an

Amicon PMIO ultrafilter (nominal molecular weight

10.000) in an Amicon Model 202 stirred cell, under nitro-

gen pressure. The filters and stirred cells were cleaned im-

mediately before use with dilute HNO3 and pre-treated

with a CatNO3)2 solution to minimise adsorption losses.

Tests showed such losses to be negligible.

Total

metal concentrations were determined by flameless

atomic

absorption (Fe and Mn) and by ASV

after

u.v.-acid

digestion (Cd, Pb and Cu). Mon-filterable metal collected

onandtheHCIO12pm4

digestion.filter

wasAlldeterminedconcentrationsafter successivewere

determinedHNO3

by the method of standard additions.

Further analyses were performed on certain filtrate frac-

tions as follows: ASV-Iabile metal was determined at the

natural pH of the sample, buffered with a C02:N2 mix-

ture. Analyses were performed with a hanging mercury

drop electrode using a fresh drop for each deposition/strip-

ping cycle. The metals Cd, Pb and Cu were deposited at

—9SOmV (vs SCE) and stripped at 5 mV s’ in the differ-

ential pulse mode. The samples were titrated with a mixed

metal standard to quantify the labile response and provide

information on the state of metal complexation.

Chelex-labile metal was determined using Chelex.lOO in

the calcium form in a batch technique. 2 ml resin to 60 ml

of sample for 48 h. The non-Chelex-labile metal was deter-

mined. as for total metals above, and the Chelex-labile

fraction determined by difference.

Filtrates were exposed to u.v. irradiation at their natural

pH for 8—12 h in quartz tubes and the ASV-labile metals

then re-measured. Prior to u.v. irradiation the sample pH

was adjusted to approx. pH 5.5—6 with HCIO4 to compen-

sate for the increase in pH that occurred during ir-

radiation. A brown precipitate was generally found after

the u.v. irradiation, which is thought to be due to precipi.

tation of organically stabilised hydrous iron oxide colloids

(Laxen & Harrison. 1981a). There was a simultaneous loss

of ASV-labile metal. It is thus not possible to quantify the

non-labile organically bound metal fraction. The u.v. ir-

radiation procedure does however provide qualitative in-

formation on the nature of any metal complexation (Laxen

& Harrison. 1981a).

Standard methods were used to analyse the samples

for their bulk constituents

(Standard Methods,

1976).

Suspended sediment concentrations refer however to

material retained by a 0.4 pm Nuclepore filter. Total

organic carbon was estimated from the u.v. absorbance at

254 tim an filtered samples, according to the relationship

derived by Dobbs

era!.

(1972). The pH and alkalinity were

measured on unfiltered portions of sample immediately on

return to the laboratory. i.e. within 2—3 h of taking the

sample.

Samples

The samples were collected in 5 I. polythene containers

pre-cleaned with l0~HNO3 )Laxen & Harrison. l9Blb)

and rinsed with sample before filling. The sewage final ef-

fluent was taken from Barrowford sewage treatment plant

at Nelson in Lancashire, on 18 February 1980. from the

overflow of the final sedimentation (humus) tank. The flow

of sewage through the treatment plant is split into two

after the primary sedimentation and the sample was taken

from what is known as line B. The effluent is discharged

into a small river named Pendle Water. Samples were

taken from this river on 4 March 1980, whilst the river flow

was low following a dry spell. The samples were taken

about 30 m upstream and approx. 100 m downstream of

the effluent outfall at a point where mixing was complete.

The flow at the downstream site was relatively turbulent in

contrast to that at the upstream site.

*

U

U

RESULTS

Sewage

effluent

The sample of final effluent was taken from a sew-

age treatment plant involving both primary sedimen-

tation and secondary treatment by biological filtra-

tion. Following passage through the primary sedi-

mentation tanks the effluent stream is split to pass

through two sets of biological filters, A and B, each

with their own final sedimentation (humus) tanks. The

quality of the final effluent from the two streams A

and B differs slightly (Table 2). due to a poorer per-

formance of the final sedimentation tank B. The

present sample was taken from this stream. It is

apparent from the results in Table 2 that the sample

taken for this study is representative of the final efflu-

ent from tank B (cf. pH. suspended solids, cadmium).

The speciation results for the effluent sample are

presented inTable...3. A major part of the Cd

(45°ij

is

associated with the smallest size fraction (0.015 pm)

and most of this is probably both ASV and Chelex-

labile. It is also apparent that some of the labile Cd

must be derived from the colloidal size fraction. The

ASV titration provides no indication of a residual

non-labile complexation capacity for Cd. consistent

with the strong labile signal in the original sample.

There was no change in the stripping peak potential

(Ep) following u.v. irradiation, the values remaining

between —575 to —585 mV. A further feature of note

is the association of a significant portion of the Cd

with the 0.08—0.4 pm colloidal size fraction. This par-

ticular size fraction is also emphasised in the case of

Cu and Fe but not for Pb. The Pb shows a particu-

larly strong association with the larger size fraction

material:

Y7°,,

is greater than 0.4 pm. and only a small

portion. lO°. is found in the smallest size fraction

(0.015 pm). There was a small ASV Pb signal at

410—430 mV in the various filtrate fractions. However.

this was not quantified as there was a shift in the peak

potential (Ep) during the titrations to 395—405 mV.

This is indicative of a labile Pb complex in the orig-

inal sample, although caution must be exercised in the

interpretation of Ep shifts due to interference from

organics (Brezonik

et at..

1976: Batley & Florence.

1976a,b). particularly in an organic-rich sewage efflu-

ent. The

ASV

titration curves for the three filtrate

fractions (1 pm, 0.08 pm and PMlO) were similar.

showing a possible small residual complexation ca-

pacity for Pb (Fig. 2). However, after the addition of

approx. 150 pg I - Pb the sensitivity was progress-

ively reduced.

The iesijlts for Cu reveal a strong association (47’~’,,)

with the smallest size fraction 0.015 pin) and much

of this metal would appear to be Chelex-labik. The

Cu is not, however, ASV-labile and thereis a

strong

residual complexation ~a~ia.it~v for the

Cu.

which is

undiminished even in the PMIO ultrafiltrate. suggest-

iflg that ligands of relatively low molecular weight

(10.000 MW~are responsible for the non-labile

complexation. The complexation capacity was almost

---

1056

DUNCAN P. H.

LAXEN

and Roy M. HARRISON

* North West Water Authority data,

effluents A and B.

t Mean and range.

Filtered sample.

~Total metal concentrations.

entirely lost following u.v. irradiation providing evi-

dence that the complexation was due to organic com-

pounds of relatively low molecular weight.

The Fe is present

principally in

association, with the

particulate size fraction 12 pm 4.58), the remainder

being fairly evenly distributed among th.oth.er. size

fractions. Only 30 pg 1’ (4) of the Fe is found in the

smallest size fraction (0.015 pm) suggesting that it is

not complexed to any significant extent with the

lower molecular weight organics (less th~i approx.

300,000 MW). The Mn in found almost entirely (83)

in association with the smallest size fraction

(0.015

pm),

but not all of this is Chelex-labile sug-

gesting the existence of some very strong complexes.

* 1 pm filtrate.

t

NQ

=

not quantifiable.

River

samples

The river samples were taken during a period of

low flow, following 6 days of dry weather. Hence

there was a rather small dilution capacity for the

influx from the sewage treatment plant, of the order of

10—20-fold. The analyses of the two samples taken

upstream and downstream of the sewage effluent out-

fall are shown in Table 4, together with the results

from two stations operated by the North-West Water

Authority (NWWA) 1.5 km upstream and down-

stream respectively of the Barrowlord sewage treat-

ment plant outfall. It should be noted that there is an

additional effluent outfall about 500 m downstream of

the sampling points used in this study, discharging

Table 2. Analyses of Barrowford sewage treatment plant effluents

Parameter

Final effluent B

18.2.80

10.30 h

Treatment plant average, Oct—Dec 1979*

Crude Sewage

Final Effluent A

Final Effluent B

Flow (m3s

‘)

0.051 (0.028—0.081)t

pH

6.9

7.0(6.4—8.2)

6.7(6.4—7.0)

6.8(6.5—7.1)

Temp (rC)

7.5

Specific conductance (pS)

510

Suspended solids (ing I~)

23

122(44—224)

18(10—29)

22(10—34)

TOC(mgl~)

l3~

Alkalinity (as CaCO3)mgl~

88

Ca(mgl~)

40

Mg(mgl’)

3.6

Na(mgl~)

4.9

K(mgl’)

9.1

SO~(mgl’)

56

Cl (mgl~)

59

PO~— P (mgl~)

1.5

NO~— N (mgl~)

0.6(0.2—1.8)

6.8(2.3—14)

5.4(2.3—11)

COD (mgl1)

304(106—580)

77(43—138)

87 (37—134)

Cd

(pgl’)

139*

280(70—1010)

80(30—120)

110(50—170)

Pb(pgl~)

12

10-100

10-40

10—40

Cu

(pgl’)

36

80(10—290)

10—50

10—80

Fe(pgl~)

693

Mn(pgl’)

77

N

L

based on hourly samples. bulked for analysis, see text for difference between

a

U

Table 3. Speciation of Cd, Pb, Cu, Fe and Mn in the sewage treatment

plant final effluent. Sampled 18.2.80

Size

fraction

(pm)

Concentration

l ‘)

Cd ()

Pb ()

Cu ()

Fe ()

Mn ()

12

28(20)

6.3(52)

11.5(32)

405(58)

10(13)

1—12

5(4)

0.4(3)

—

58(8)

—

0.4—1

—

2.7(22)

—

43(6)

—

0.08—0.4

•34(24)

1.6(13)

5.1(14) 101 (15)

—

0.015—0.08

10(7)

—.

2.2(6)

56(8)

3(4)

0.015

Sum total

~ZA~

139

i,2j~.)

12.2

17.0(47

35.8

~QJ~)

693

~1~J

77

ASV~labile*

80(58)

NQt

NQ

Chelex~labiles

86(62)

2.9(24)

16.6(46)

31(4)

31(40)

p

p

p

p

U

---

The physicochemical speciation of Cd, Pb, Cu, Fe and Mn

1057

$00

Fig. 2. ASV titration of 1

p~

filtrate of the sewage effluent.

Result for Pb.

effluent from the somewhat larger Nelson sewage

treatment plant. This will be reflected in the analyses

for the downstream NWWA sampling station (Table

4). The samples taken for this study were clearly fairly

typical of the conditions found in Pendle Water. The

low suspended solids concentrations are consistent

with the low flow conditions, which resulted in a

fairly slow moving body of water. In general terms,

the water is one of moderate alkalinity and pH, with a

relatively high organic loading. The influx of effluent

from the Barrowford sewage treatment plant has a

marked effect on the Cd concentration in the river,

increasing the total concentration from 0.3 to

9.0 pg I

There is also a significant increase in the

Cu concentration. However, apart from these

changes, there is not a great deal of difference between

the composition of the two samples.

The speciation results for the two samples are set

out in Tables 5 and 6. There is no obvious similarity

in the distribution patterns of the five metals. Further-

more, there are significant differences between the

metal distribution patterns above and below the out-

fall. Thus the speciation of each metal will be con-

sidered in turn.

Cadmium

Cd is present at a low concentration in the

upstream sample and is found mostly in the 1—12 pm

particulate size range (58). A significant fraction is

also associated with the smallest size fraction

(0.015 pm) and most or all of this is probably Che-

lex-labile. The ASV Cd signal in the original sample

was more cathodic at —635 to —650 mV than the

stripping potential developed towards the end of the

titration, of —575 mV (after the addition of

160 pg l’ Cd). The ASV titration also indicated a

small residual complexation capacity in the 1 pm fil-

trate (Fig. 3) which was absent in the PMIO ultrafil-

trate. The ASV labile Cd signal in the u.v. irradiated

samples was constant throughout the titration

between —585 and —590mV. The ASV behaviour

Table 4. Analyses of Pendle water upstream and downstream of Barrowford sewage treatment plant

outfall

Parameter

Sampl

Upstream

10.05 h

ed 4.3.80

Downstream

10.00 h

Pendle water analyses, 1976-1980’

Upstream (approx.

1.5 km)

Downstream

(approx. 1.5 km)

Flow (m3s~)

pH

Temp (°C)

Specific conductance (pS)

Suspended solids (mgi_i)

TOC (mgi -

i)

Alkalinity (as CaCO3)(mgl~)

Ca(mgl~)

Mg (mg I

~‘)

Na(mgl’)

K(mgr’)

SO~(mgl’)

CV (mgl’)

PO~

—

P (mgl~)

NO; — N

(mgl~)

COD (mgV’)

Cd (pg I~)

Pb(pgl~)

Cu (pg l1)

Fe(pgl~)

Mn(pgl~)

7.7

3

367

4.6

6.0*

90

44

5.0

24

3.6

56

39

0.2

0.3l~

1.8

3.8

830

116

7.6

3

380

5.4

6.4

91

46

8.5

26

4.1

51

34

0.2

9.0

2.1

6.5

841

113

7.5 (7.0—8.4)t

31(5—285)

82(40—160)

39(12—70)

0.35 (0.05—1.1)

2.2(1.1—3.8)

28(13—60)

3.1 (0.5—t5)~

7.4(6.9—8.2)

33(5—309)

79 (43—129)

44(15—116)

1.1(0.1—3.5)

3.7(1.3—6.8)

37(18—74)

*

North West Water Authority data, 27 and 28 spot samples respectively.

t Mean and range.

~Flow measured 3 km downstream.

~Filtered sample.

¶ Total metal concentrations.

04

4

1.

a

0.2

a

/X

/1

200

400

600

Lead addition

(pig

rI

---

* 1 pm filtrate.

t NQ

=

not

quantifiable.

suggests the presence of a labile organic complex,

together with a residual non-labile complexation ca-

pacity associated with the colloidal size range.

In the downstream sample a major part of the Cd

(44) is associated with the smallest size fraction. A

significant amount is also associated with the

0.015—0.08 pm colloidal size fraction, whereas this size

fraction had no Cd in the upstream sample. The pre-

dominance of the 1—12 pm fraction is lost, but this is

not due to a loss of Cd from this size fraction but

rather to the addition of Cd to the other size fractions

(Fig. 4). It is evident from Fig. 4 that the Cd influx

with the sewage effluent has a major impact on the

smallest size fraction. Not all of this Cd in the smal-

lest size fraction is labile. The ASV titration of the

1 pm filtrate provides no indication of any residual

complexation capacity for Cd.

Lead

In the upstream sample a smaller proportion of the

Pb (18°;)is found in the smallest size fraction than is

the case with Cd (26°c) and Cu (30). The major

* 1

pm

filtrate.

t

NO =

not

quantifiable.

proportion of the Pb is in the particulate (46) and

the 0.08 to 0.4 pm colloidal (29) size ranges. None of

the lead appears to be Chelex-labile. There is however

a small ASY-labile response at —480 to —490 mV,

which is 60—70 mY more cathodic than the Pb re-

sponse towards the end of the titration. The ASV

titration indicates a residual complexation capacity

for Pb (Fig.

5),

which is largely removed by filtration

through the PMIO ultrafilter.

The distribution pattern below the outfall shows an

increase in the lead concentration in the colloidal size

range, but significantly not in the 0.08—0.4 pm fraction

which predominated in the upstream sample (Fig. 6).

The Pb concentration below the outfall is only

slightly increased but there is a significant shift from

the 1—12 pm size fraction to the

12pm particulates.

The ASV titration revealed a similar residual corn-

plexation capacity to that in the upstream sample.

There is a small amount of Chelex-labile lead (7) in

this sample, and the small ASV-labile response was

only 20 mV more cathodic than the response towards

the end of the titration. The lead in both upstream

and downstream samples thus appears to be strongly

1058

L~UNCAN

P. H. LAXEN and Ro~M.

HARRISON

Table

5.

Speciation of Cd. Pb, Cu. Fe and Mn in Pendle water upstream of the sewage

outfall

a

11

U

Size

fraction

(pm)

Concentration (pg 1-’)

Cd ()

Pb ()

Cu ()

Fe ()

Mn ()

12

0.01 (3)

0.40(22)

0.15(4)

59(7)

2(2)

1—12

0.18(58)

0.44(24)

0.89(24)

235(28)

—

0.4—1

0.02(6)

0.08(4)

0.28(7)

180(22)

2(2)

0.08—0.4

0.02(6)

0.53 (29)

0.71(19)

342 (41)

4(3)

0.015—0.08

—

0.03(2)

0.63(17)

—

4(3)

0.015

0.08(26)

0.32(18)

1.12(30)

J,4j~)

104(90)

Sum total

0.31

1.80

3.78

830

116

ASV~Iabile*

NQt

NQ

NQ

Chelex-labile’

0.09(29)

0.00(0)

0.56(15)

80(10)

110(95)

Table 6. Speciation of Cd, Pb, Cu. Fe and Mn in Pendle water downstream of the sewage

outfall

1.

U

U

U

N

m

a

U

U

U

a

a

f

Size

fraction

(pm)

Concentration (pg l

1)

Cd ()

Pb (°‘)

Cu ()

Fe ()

Mn ()

12

1.2(13)

0.80(37)

1.26(19)

96(11)

2(2)

1—12

1.3(14)

0.08(4)

1.04(16)

200(24)

3(3)

0.4—1

0.2(2)

0.16(7)

0.18(3)

143(17)

—

0.08—0.4

0.8 (9)

0.54(25)

0.66(10)

368(44)

6(5)

0.015—0.08

1.5(17)

0.23(11)

1.28(20)

14(2)

5(4)

0.015

4.0(44)

0.33(15)

2.08(32)

2Q(2)

~21~ft1

Sum total

9.0

2.14

6.50

841

113

AS V-labile’

2.9(32)

NQ1~

NQ

Chelex-labile’

3.1(34)

0.16(7)

0.86(13)

91(11)

105(93)

---

The physicochemical speciation of Cd. Pb. Cu. Fe and Mn

1059

complexed and there is a residual complexation ca-

pacity associated with the colloidal size fraction.

Copper

The copper in the upstream sample is fairly evenly

distributed across the size fractions. There is, however,

a significant fraction (17)

associated with the

0.015—0.08 pm colloidal size range in contrast to the

other metals and a fairly high proportion (30) in the

smallest size range. Not all of the Cu in this smallest

size range is Chelex-labile suggesting some of the Cu

to be strongly complexed. There was a significant resi-

dual complexation capacity for Cu (Fig. 7) of approx.

28 pg l 1(0.4 x 10~M), which was undiminished by

filtration through the PMIO ultrafilter. A small ASV

labile response at —135mV may have been due to a

Cu complex. This peak was 160 mY more cathodic

than the Cu response after the complexation capacity

Fig.

5.

ASV titration of I pm filtrate of the Pendle Water

upstream sample. Result for Pb.

had been exceeded. The ASV behaviour following u.v.

irradiation of the sample was somewhat unusual. A

strong initial Cu signal in the 1 pm filtrate at + 20 mY

declined in size rapidly to be replaced by a peak more

cathodic at approx. —60 mY. There is no immediate

explanation for this behaviour.

The speeiatimtpattetii of~ttfeCu below the outfall

shows that tb

ajtn~addthoiiof the Cu jflftLlx~fr~

the sewa e effluent ~

the~smaIlestsizv~fructionand

to thv particulate sire ~fraction’~’+2pm: although

there is also a significant increase

in the

0.015—0.08 pm colloidal size fraction (Fig. 8). The ASV

titration indicated a similar complexation capacity to

the upstream sample, of 27 pg 1~ (0.4 x 10_6 M).

which was also undiminished by passage through the

PMIO ultrafilter. There were ASV signals, possibly

due to labile Cu species at —l9OmV in the 1pm

4

a

Fig. 3. ASV titration of 1 pm filtrate of the Pendle Water

upstream sample. Result for Cd.

0.2

‘4

a.

c

0l

U

a

/

Cadmium addition lpg

i~)

00

200

Lead addition t pg rI

1.

C

0

a

C

0

0

10

00

000

0,000

~

Size (nm)

Fig. 4. Change in concentration of Cd in the different size fractions below the sewage effluent outfall.

---

DUNCAN

P. H.

LAXEN

and RoY M.

HARRISON

filtrate and —250 and — 150 mY in the PM1O filtrate,

which would be indicative of strong complexation.

(The response towards the end of the titration was at

+

3Sto

+

40 mV.) These peaks were absent following

u.v. irradiation.

iron

The predominant size fraction foz Fe” in ihe

upstream. sample is the 0.08-’-0~4’pm colloidal size

range (41 ). Altogether; 98.of. the~Feis

itt

the col-

loidal and-particulate size. range&. The Chelex-labile

fraction (10)

is greater than the Fe in the smallest

size fractions (2) indicating a contribution from the

colloidal size range. A similar Fe distribution is found

in the downstream sample with only slight changes in

the pattern (Fig. 9). There is no major addition to the

total Fe burden of the river due to the effluent dis-

charge.

Manganese

The distribution of Mn in both samples is similar,

with virtually all of the metal being found in the smal-

lest size fraction (86—90)

in a Chelex-labile form.

00

200

Copper addition (pg rI

Fig. 7. ASV titration of 1 pm filtrate of the Pendle Water

upstream sample. Result for Cu.

The sewage effluent has no significant impact in terms

of both the total concentration and the speciation

pattern.

DISCUSSION

Sewage. effluent.- ~oniains high

eoncernrations

.

of

both - particulate ‘and dissolved. organic n~ues

(Hunter, 1971). It is thus reasonable initially to con-

sider the speciation results in terms of the potential

influence of organic interactions. The-results forcop-

p~~ the Barrowford sewage efflu’ent provide fairly

clear~evidea~e conpleiation. with Jew. meleuular

weight organics.410.000MW)...

This

complexation

may’well aucount for

the-fairly

high

proportion of Cu

found. in- the smallest size fraction (1)1)4

S’pm)-

i.e.

comptexation with dissolved organics is exerting a

solubilizing effect. Bender

et a!.

(1970) found that. Cu

added to a

secondary

sewage effluent was complexed.

with two distinct molecular weight fractions. Part of

the Cu was associated with approx. 10,000 MW frac-

tion, but the major part was found in a 500—1000 MW

fraction, which the authors ascribed to complexation

with synthetic chelators. In this context, Gardiner

(1976) measured concentrations of the

chelating agent

EDTA ranging from 0.3 x 10_6 to 4

x l0~’M

in the

final effluent from several sewage treatment plants in

England. The higher concentrations were associated

with plants treating industrial as well as domestic ef-

fluent. C-u~has a particularly strong-affinity for EDTA

with a stability constant of the order of lO’~to

IO’~M’ (Gardiner, 1976), hence Cu complexation

with this chelating agent would be-expected to occur

if it is present in the sewage effluent. However, the

results do not suggest a strong interaction between

Cu and EDTA in the present sample. Under the con-

ditions of the ASV analysis, where a plating potential

of —950 mY was used for all three metals, the Cu-

EDTA complex would be directly reduced (Figura &

McDuffie, 1979). During the stripping the Cu orig-

inally associated with the EDTA would be revealed as

1060

Size (nm I

Fig. 6. Change in concentration of Pb in the different size fractions below the sewage effluent outfall.

U

U

a

U

U

p

p

a

N-

a

p

I

a

a

a

a

C

4

a.

C

U

2

0

/X

/

J

/

/

---

The physicochemical speclation of Cd, Pb, Cu, Fe and Mn

1061

a.

S

Fig. 8. Change in concentration of Cu in the different size fractions below the sewage effluent outfall.

a labile Cu peak at the potential (Ep) for uncom-

plexed Cu (Gardiner, 1975; Schonberger & Pickering,

1980). There is no evidence for such behaviour in the

present sample. The results provide evidence of a par-

tially labile residual complexation capacity (Fig. 10)

and such ASV behaviour is more consistent with Cu

complexation with humic acids. The Cu-humic acid

response is partially reversible (Ernst

et a!.,

1975),

hence the limited response to copper additions prior

to

the satisfaction of the complexation capacity. Cu

complexation with humic acids also results in a cath-

odic shift in the stripping potential (Ernst

et a!.,

1975)

as was evident during the early part of the titration of

the present sample.

The precise shape of the titration response is a re-

flection of both the conditional formation constant of

the complex and the concentration of the ligand (Shu-

man & Woodward, 1977). The concentration of

igand in the present sample is approx. 2 x 10_6 M,

based on the inflexion in the titration curve. Compari-

son of the shape of the titration curve with theoretical

curves presented by Shuman & Woodward (1977)

a.

S

2

C

would suggest a conditional formation constant of the

order of iOo_107 M ‘for a 1:1 complex. Such a value

is reasonable for Cu-humic acid complexation (Ernst

et a!.,

1975; Shuman & Woodward, 1977; Mantoura

et

a!., 1978). Inflexions in the titration curves for Cd

and Pb, due to complexation with humic acids, are

less likely to be observed as their formation constants

with humic acids are lower than those of copper.

Furthermore the mixed metal titration would lead to

a preferential complexation of Cu at the expense of

any cornplexation with Cd and Pb. However, a catho-

dic shift in the Ep would be expected and this was

found in the case of Pb. There might also have been a

marginal cathodic shift in the Ep of Cd of the order of

5

mY.

If humic acids are assumed to be present, then it is

likely that a significant amount of the Cd in the orig-

inal sample would exist as Cd-humic acid complexes,

given its high concentration relative to Cu and the

significant excess complexation capacity for the Cu.

However, in these circumstances, the presence of the

strong complex former, Cu, in the mixed metal litrant

Size (nm)

Size (nm)

Fig. 9. Change in concentration of Fe in the different size fractions below the sewage effluent outfall.

---

1062

DUNCAN

P. H.

LAXEN

and Ro~M. HARRISON

4

a.

C

U

0

Fig. 10. ASV titration of I pm filtrate of the sewage efflu-

ent. Result for Cu.

might lead to the competitive displacement of Cd

from complexation sites. There is indeed some evi-

dence of such a substitution process occurring during

the titration (Fig. 11). The sensitivity to added Cd is

reduced once the Cu complexation capacity has been

fulfilled. The initial labile

Cd response may thus rep-

resent both the added Cd

plus

Cd released from par-

tially labile complexation sites. It is possible therefore

that humic acids are responsible for exerting a solubi-

lizing effect on the Cd as well as the Cu, hence the

high proportion of Cd in the smallest size fraction.

Murray & Meinke (1974) have demonstrated a similar

ability of sewage effluent to retain Cd in solution.

The residual Cu complexation capacity measured

in the present sample of approx. 2 x 10_6 M, is simi-

lar to that measured by Lewin & Rowell (1973) of

1 x 10~M, using a similar mixed metal titration on

the final effluent from the Oxford sewage treatment

plant. They also found a small residual complexation

capacity for Pb of 7.7 x lO~M, but none in the case

of Cd. In the present sample there was a similar small

residual complexation capacity for Pb of approx.

2 x i0~M, but none in the case of Cd. However,

only a small

part of the Pb is in the

smallest size

fraction (10) suggesting that very little lead is being

solubilized by low molecular weight organics and that

other mechanisms are operating to preferentially as-

sociate the Pb with other size fractions. Pb has a

stronger adsorption/ion exchange affinity for clays

and hydrous iron oxides than either Cu or Cd (Rama-

moorthy & Rust, 1978; Farrah & Pickering. 1977).

Hence the different distribution pattern for Pb than

Cd or Cu, in particular the emphasis on the 0.4—1 jim

fraction may reflect the greater affinity of Pb for in-

organic solids.

The speciation results for the sample of final efflu-

ent provide a clue to the differing efficiencies of metal

removal during treatment (Table I). Factors which

retain the metals ~in the colloidal and soluble

(0.Ol5jLm) size fractions will make it more likely

that the metals will pass througfr the plant, whilst

association ofthe metals-with the particulate size frac-

tionwiI1~ensare’a greater- loss of metal during the

passage through the

final

sedimentation tank. The

speciation results (Table 3) would suggest that the

efficiency of removal would be in the order

Fe Pb Cu

Cd Mn, which bears a reason-

able correspondence to the results for actual plants

presented in Table 1. These are not, however, the only

factors to take into account, as the removal efficiency

will also depend upon the affinity of the metals for the

floc in the biological filter or activated sludge plant,

as they pass through the treatment works.

The metal speciation patterns in the river differ

from those in the sewage effluent. A major proportion

of all metals, except Mn, is found in the colloidal and

particulate fractions. This is particularly evident in

the case of Fe where 98 is 0.015 jim. This is prob-

ably due principally to the hydrolysis of Fe to ferric

hydroxide species. X-ray diffractometry performed on

the solids

0.4 pm revealed the presence of goethite

(cz-FeOOH). together with quartz, calcite, truscottite,

kaolinite and muscovite. Amorphous iron oxide

would not have shown up but may also have been

present.

Gardiner (1974) has examined the association of Cd

with river muds and various components thereof and

concluded that the Cd is probably associated with

particulate humic substances. Suzuki

et a!.

(1979) have

similarly suggested that Cd is probably associated

with organic matter in river suspended sediments. It is

thus possible that the colloidal and particulate Cd in

Copper addition (pg ~‘)

a

4

a.

C

C

U

0

Cadmium addition (pg I’)

Fig. 11. ASV titration of 1 ~zmfiltrate of the sewage efflu-

ent. Result for Cd.

a

a

a

N

a

Ii

---

Method

Chele

Pb

x-labile

(

Cd

metal

Cu

pH

TOC

(mgl’)

Sample

Reference

Batch

0

90

23

7.7

6.0

-

Pendle Water (upstream)

This work

Batch

15

49

21

7.6

6.4

Pendle Water (downstream) This work

Batch

18

33

42

7.5

6.3

River Irwell (upstream)

Laxen & Harrison (1981c)

Batch

36

43

48

7.4

6.2

River Irwell (downstream)

Laxen & Harrison (l981c)

Batch

67

94

22

7.7

24

Mordialloc Creek

Hart & Davies (1976)

Batch

55

91

58

7.1

NA

Yarra River

Hart & Davies (1977)

Column

67

94

22

6.5

NA

Susquehanna River

Figura & McDuffie (1977)

Column

100

92

66

6.1

4.5

Woronora Reservoir

Florence (1977)

Column

78

100

49

6.1

2,7

Woronora Reservoir

Florence (1977)

Column

71

100

36

6.0

1.6

Woronora Reservoir

Florence (1977)

the present samples is associated with organics. Mur-

ray & Meinke (1974) have shown that Cd adsorption

onto sediments (0.45 jim) in unpolluted river water

is strongly pH dependent, increasing from 5 to 95

over the pH range 7—8. The pH of the present

samples, pH 7.7 and 7.6, is commensurate with signifi-

cant Cd adsorption.

Copper shows a stronger affinity for adsorption

onto solids than does Cd and- it also shows a strong

affinity towards-

humic acids. Thus if Cd is associated

with colloidal and particulate organics then Cu is

likely to be similarly associated. There is certainly a

similarity in the speciation patterns for Cd and Cu,

especially in the downstream sample. Lead, on the

other hand, as has already been noted, shows a

stronger adsorption/ion exchange tendency towards

inorganic solids, and is thus more likely to be associ-

ated with the clay minerals and

hydrous ferric oxides

in the samples. Adsorption or coprecipitation with

hydrous ferric oxides probably does not account en-

tirely for the Pb distribution as the Pb and Fe

pat-

terns are not wholly similar.

In

terms of the biological significanca. the

most irn~

portant metal species are likely to be those

within

.Lke

smallest size fractioty (0.Ol5prn~. Comp1exation’ will

generally reduce the

bioavailability of thc metals, thus

the Chelex labile fractinn-isprobably;an .oveu.cioser

index

of biological significanca. Indeed, not~al~oi~the

Pb and Cu in the sma4les~sizg.fragtion- was Chelex-

labile indicating a certain arnou~n*of stron~complex-

ation. The ASV resutts-ulse~indicatea significant corn--

plexation of all three

metatrCdç Pb and

Cu

-together

with a residual oomp!~atinn..capaci1~’,further sag’

gesting a reduced biological.~signiflcancs~

In Table 7

the fraction of each metal which is Chelex-labile in

relation to the total metal in the fractions 0.4 jim is

compared to other reported data. The fraction of

labile Cd in the upstream sample is fairly typical of

the results obtained in most other studies.

However,

this fraction is reduced in

the

downstream sample.

This is probably a reflection of the greater proportion

of colloidal (0.015—0.4

jim)

Cd in the downstream

sample. 36

as against 20.

The order of Chelex-

lability of the metals in the present samples is

Cd Cu Pb, which differs from the results for the

River

Irwell

samples, where the order was

Cu Cd Pb, which in turn differed from the gen-

eral pattern found in other studies Cd Pb

Cu.

This merely reflects the high degree of variability in

metal

speciation,

which will be highly dependent

upon the composition of the

particular

water under

study. It also warns against the use of sweeping gener-

alisations in discussing speciation results, which refer

essentially to a particular sample from a particular

water body.

The source of the relatively strong complexation of

the metals in Pendle Water is uncertain. In the

upstream sample there

was a

residual complexation

capacity for Cd which was associated with the col-

loidal size range, with a similar behaviour for Pb in

both samples. The residual complexation capacity for

Cu, on the other hand, was associated with the lower

molecular weight material

(10,000MW)

and was

probably organic in nature. The existence of

these

residual complexation capacities in the upstream

sample would

suggest that addition of metals to the

river would result in their uptake by the components

contributing to the complexation capacity. Hence the

Cd and Pb in the sewage effluent that

are

available

for complexation would be

directed towards the col-

loidal size range upon mixing with the river water,

whilst the Cu would be taken up by the smallest size

fraction. A key word here is “available”. During the

ASV titration the metals are added as simple in-

organic labile metals, whereas the metals in the sew-

age effluent

are

being added in forms which are to a

large extent already complexed. A further compli-

cation is that the ASV titration

was performed

on the

1 pm filtrate

and

hence took no account of the poss-

ible residual complexation capacity of the particulates

(1 pm). For instance Barnes

et

at. (1975) found that

particulates in the 2—16 pm

size

range in a lake water

had a significant coinpiexation capacity for lead (ASV

titration). Nevertheless, the additions of the metals in

I

N

N

N

The physicochemical speciation of Cd, Pb,

Cu,

Fe and Mn

1063

Table 7. Chelex-labile metal as a proportion of total filterable (0.4

pm) metal in various freshwaters

* NA not available.

---

1064

E~UNCAN

P. H.

LAXEN

and Roy M. HARRISON

U

Acknowledgements—We

thank the Ribble Division of the

North West Water Authority (NWWA) for their permis-

sion to sample the Barrowford sewage effluent and for pro-

vision of analytical data for the effluent.

In addition, we

thank the NWWA for providing us with analytical data for

Pendle Water. We also thank Mrs J. Walsh for her assist-

ance with the major ion analyses for this study. Finally we

thank the Science Research Council for their support of

this project.

-

REFERENCES

Andrew R. W.. Biesinger K. E. & Glass G. E. (1977) Effects

of inorganic complexing on the toxicity of copper to

Daphnia maqna. Water Res.

Ii,

309—315.

Atkins E. D. & Hawley

J.

R. (1978) Sources of metals and

metal levels in municipal wastewaters. Research Report

No. 80.

Ontario Ministry of the Environment, Toronto,

Ontario.

Barnes D.. Cooksey B. G.. Metters B. & Metters C. (1975)

Interferences in the electroanalytical determination of

copper and lead in water and wastewater.

Wat. Treat.

Exam.

24, 318—328.

Batley G. E. & Florence T. M. (1976a) A novel scheme for

the classification of heavy metal species in natural

waters.

Analvt. Lett. 9,

379—388.

Batley G. E. & Florence T. M. (1976b) The effect of dis-

solved organics on the stripping voltammetry of sea-

water.

J. Electroanalyt. Chem.

72, 121—126.

Bender M. E., Matson W. R. & Jordan R. A. (1970) On the

significance of metal complexing agents in secondary

sewage effluents.

Envir. Sd. Technol.

4, 520—521.

Benel P. & Steinnes E. (1975) Migration forms of trace

elements in natural fresh waters and the effect of the

water storage.

Water Res.

9, 74 1—749.

Benei P.. Koc J. & Stulik K. (1979) The use of anodic

stripping voltammetry for determination of the concen-

tration and forms of existence of lead in natural waters.

Water Res.

13, 967—975.

Brezonik

P. L.. Brauner P. A. & Stumm W. (1976) Trace

metal analysis by anodic stripping voltammetry: Effect

of sorption by natural and model organic compounds.

Water Res.

10, 605—612.

Brown M. J. & Lester J. N. (1979) Metal removal in acti-

vated sludge: The role of bacterial extracellular poly-

mers.

Water

Res.

13, 817—837.

Chau Y. K. & Lum-shue-chan K. (1974) Determination of

labile and strongly bound metals in lake water.

Water

Res.

8, 383—388.

Davies P. H.. Goetth J. P. Jr. Sinley J. R.

& Smith N. F.

(1976) Acute and chronic toxicity of lead to rainbow

trout

Salmo gairdneri,

in hard and soft water.

Water Res.

10, 199—206.

Dobbs R. A.. Wise R. H. & Dean R. B. (1972) The use of

ultra-violet absorbance for monitoring the total organic

carbon content of water and wastewater.

Water Res. 6,

1173— 1180.

Ernst R.. Allen H. E. & Mancy K. H. (1975) Characteriz-

ation of trace metal species and measurement of trace

metal stability constants by electrochemical techniques.

Water

Res. 9,

969—979.

Farrah H. & Pickering

W. F. (1977) Influences of clay-

solute interactions on aqueous heavy metal ion levels.

Wat. Air Soil Pollut.

8, 189—197.

Figura P. & McDuffie B. (1977) Characterization of the

calcium form of Chelex-lOO for trace metal studies.

Ana-

1w. Chem.

49, 1950—1953.

Figura P. & McDuffle B. (1979) Use of Chelex resin for

determination of labile trace metal fractions in aqueous

ligand media and comparison of the method with anodic

stripping voltammetry.

Analyt. Chem.

51, 120—125.

Florence T. M. (1977) Trace metal species in fresh waters.

Water Res.

11, 681—687.

Florence T. M. & Batley 0. E. (1980) Chemical speciation

in

natural waters.

C.R.C. Critical Rev.

Analyt. Chem.

9,

2 19—296.

Gadde R. R. & Laitinen H. A. (1974) Studies of heavy

metal adsorption by hydrous iron and manganese

oxides.

Analyt. Chem.

46, 2022—2026.

Gardiner J. (1974) The chemistry of cadmium in natural

water—Il. The adsorption of cadmium on river muds

and naturally occurring solids.

Water

Res.

8, 157—164.

Gardiner J. (1975) The complexation of trace metals in

natural waters.

Symp. Proc. tnt. Conf Heavy Metals Env.,

Vol. I, pp. 303-318. 27—31 Oct., Toronto.

Gardiner J. (1976) Complexation of trace metals by ethyl-

enediaminetetraacetic acid (EDTA) in natural waters.

Water

Res.

10, 507—514.

Harrison R. M. and Laxen D. P. H. (1980) Metals in the

environment—I. Chemistry.

Chem. Br.

16, 316—320.

Harrison R. M. & Laxen D. P. H. (1981)

Lead Pollution:

Causes

and

Control.

Chapman & Hall, London.

Hart B. T. & Davies S. H. (1976) Physico-chemical forms

of trace metals and the sediment-water interface.

Proc.

Conf Interact. Sed. Fresh Water,

pp. 398-402. 6—10 Sept.

Amsterdam.

-

Hart B. T. & Davies S. H. (1977) A batch method for the

determination of ion-exchangeable trace metals in

natural waters.

Aust. I. Mar. Freshwat. Res.

23, 397—402.

Hunter J. V. (1971) In

Organic

Compounds in Aquatic En-

vironments

(Edited by Faust S. D. & Hunter J. V.), pp.

5 1—94. Dekker, New York.

Laxen D. P. H. & Harrison R. M. (1977) The highway as a

source of water pollution: An appraisal with the heavy

metal lead.

Water Res.

11, I—lI.

the sewage effluent to the differeat size fractions

(Figs 4, 6 and 8) is in broad agreement with the ident-

ified source of complexation capac1ty. In particular,

the lead is taken up by the colloidal size fraction,

whilst little is added to the smallest size fraction. Suf-

ficient Cd is added by the sewage works effluent to

roughly equal the original residual complexation ca-

pacity hence the absence of any residual complexation

capacity in the downstream sample.

There are some marked changes in the metal size

distribution patterns below the outfall (Figs 4, 6, 8

and 9) which are not merely the result of the addition

of the sewage effluent distribution pattern to that of

the river. However, the size fractions within which

there is a major increase in concentration coincide

broadly with those size fractions which are empha-

sised in the sewage effluent. The altered metal specia-

tion patterns below the outfall correspond with only

minor changes in the bulk chemistry of the river

(Table 4), further emphasising the importance of the

pre-existing speciation in the effluent. The fraction of

Chelex-labile metal below the outfall was not substan-

tially altered, but in the case of Cd the absolute

amount increased from 0.09 to 3.1 pg 1’. Iw’general,

however, the majos part-of the metaI~addition. fame

the effluent became assaciated *ith the colloidal-, and

particulate size fractions;.55 for Cd, 65 for Cu and

97 for Pb;-Hence-a laege.part,o(.each metal added

to the river is unlikely to be-of immediate significsnce~

although. the -speciation’ pattern-may ‘well- change

downstream possibly rendering the metals~more~bio-

logically -available.

U

U

U

U

a

U

U

a

N

p

1

N

I

I

---

The physicochemical speciation ofCd, Pb, Cu, Fe and Mn

065

Laxen D. P. H. & Harrison R. M. (198 Ia) A scheme for the

physico-chemical speciation of trace metals in freshwater

samples.

Sci.

Tot. Envir.

In press.

Laxen D. P. H. & Harrison R. M. (1981b) Cleaning

methods for polythene containers prior to the determi-

nation of trace metals in freshwater samples.

Analyt.

Chem.

53, 345—350.

Laxen D. P. H. & Harrison R. M. (l98lc) The physicoche-

mical speciation of selected metals in the treated effluent

of a lead-acid battery manufacturer and its effect upon

metal speciation in the receiving water.

Enuir. Sci. Tech-

nol.

In press.

Lester J. N., Harrison R. M. & Perry R. (1979) The balance

of heavy metals through a sewage treatment works. 1.

Lead, cadmium and copper.

Sci. Total Envir.

12,

13—23.

Lewin

V.

H. & Rowell M. J. (1973) Trace metals in sewage

effluent.

Effi. Wat. Treat. J.

13,

273—277.

Mantoura R. F. C., Dickson A. & Riley J. P. (1978) The

complexation of metals with humic materials in natural

waters.

Estuar. Coast. Mar. Sd. 6,

387—408.

Mill A. J. B. (1980) Colloidal and macroniolecular forms of

iron in natural waters. 1. A review.

Envir. Technol. Len.

1, 97—108.

Murray C. N. & Meinke S. (1974) Influence of soluble

sewage material on adsorption and desorption behav-

iour of cadmium, cobalt, silver and zinc in sediment-

freshwater, sediment-seawater systems.

J. Oceanog. Soc.

Japan

30, 216—221.

Ramamoorthy S. & Rust B. R. (1978) Heavy metal

exchange processes in sediment-water systems.

EnCir.

Geol.

2, 165—172.

Schonberger E. A. & Pickering W. F. (1980) The influence

of pH and complex formation on the ASV peaks of Pb.

Cu and Cd.

Talanta

27, 11—18.

Schuman M. S. & Woodward 0. P. Jr (1977) Stability

constants of copper-organic chelates in aquatic samples.

Envir. Sci. Technol.

Il,

809—813.

Stones T. (1977) Fate of metals during sewage treatment.

Effi. Wa:. Treat.

J.

17, 653—655.

Stoveland S., Lester J. N. & Perry R. (l979a) The influence

of nitrilotriacetic acid on heavy metal transfer in the

activated sludge process—I. At constant loading.

Water

Res.

13, 949—965.

Stoveland S., Perry R. & Lester J. N. (1979b) The influence

of nitrilotriacetic acid on heavy metal transfer in the

activated sludge process—lI. At varying and shock load-

ings.

Water Res.

13,

1043-1054.

Stumm W. & Bilinski H. (1973) Trace metals in natural

waters; difficulties of interpretation arising from our

ignorance of their speciation.

In

Advances in Water Pal-

lution

Research

(Edited by Jenkins S. H.), pp. 39—49.

18—23 June. Jerusalem. Pergamon Press. Oxford.

Stumm W.

&

Morgan .1. .1.

(1970)

Aquatic Chemistry.

Wiley/Interscience, New York.

Suzuki M.. Yamada T., Miyazaki T. & Kawazoe K. (1979)

Sorption and accumulation of cadmium in the sediment

of the Tama river.

Water Res.

13,

57—63.

Wagemann R. & Barica J. (1979) Speciation and rate of

loss of copper from lake water with implications to tox-

icity.

Water Res.

13, 515—523.

Waiwood K. G. & Beamish F. W. H. (1978) Effects of

copper. pH and hardness on the critical swimming per-

formance of rainbow trout

(Salmo gairdneri

Richardson).

Water Res.

12, 611—619.

---

ORIGINAL

UNiTED STATES ENVIRONMENTAL PROTECTION

/

WASHINGTON, D.C.

20460

OCT 1~93

T~fEMORANDUM

SUBJECT:

Office of Water Policy and Technical Guidance on interpretation and

implementation of Aquatic Life Metals Criteria

FROM:

Martha G~Prothro

~fl

Acting Assistant Administrator for Water

TO:

Water Management Division Duectors

Environmental Services Division

Directors

Regions I-X

Introduction

The implementation of

metals criteria is complex due to the site-specific nature of

metals toxicity. We have undertaken a number of activities to develop

guidance in this aira,

notably the Interim Metals

Guidance,

published

May 1992, and

a public

meeting of experts

held in

Annapolis,

MD, in January 1993. This memorandum

~ansrnitsOffice

of

Water

(OW) policy

and

guidance on the

interpretation and

implementation of

aquatic life criteria

for

the

management

of metals

and supplements my April 1, 1993, memorandum on the same

subject, The issue

covers

a number of areas including the

expression

of

aquatic life criteria;

total

maximum

daily, loads (TMDLs), permit~,.effluent.monitoring,and compliance; and

ambient monitoring. The memorandum covers

each in turn. Attached

to this policy

memorandum are three guidance documents with additional technical

details. They are:

Guidance Document on Expression of Aquatic Life Criteria as Dissolved Criteria

(Attachment h2), Guidance Document on Dynamic Modeling and Translators (Attachment

#3), and Guidance Document on Monitoring (Attachment #4). These will be supplemented

as addidonal’data become available. (See

the

schedule in Attachment #1.)

Since metals toxicity

is

significantly affected by site-specific factors,

it presents a

number of programmatic

challenges. Factors that must be considered in the management of

metals in the

aquatic environment Include: toxicity specific to effluent

chemistry; toxicity

specific to ambient water

chemistry; different

patterns

of toxicity for

different

metals;

evolution of the state

of the science of metals

toxicity, fate, and transport; resource

limitations for monitoring, analysis, implementation, and research functions; concerns

regarding some of the analytical data currently on record due to possible sampling and

analytical contamination; and

lack of

standardized protocols for clean and ultraclean metals

analysis.

The States have

the key role in the risk management process of balancing these

factors in the management of water programs. The site-specific nature

of

this issue could ~

perceived as requiring a permit-by-permit approach to implementation. However, we bel iC~.

e

CFFCEcc

WATER

---

I

2

j

• that this guidance can be effectively implemented on a broader level, across any waters with

roughly the same physical and chemical characteristics, and recommend that we work with

the States with that perspective in mind.

-

-

-

Expression of Aquatic Life Criteria

o

Dissolved vs. Total Recoverable Metal

-

A major issue is whether, and how, to tise dissolved metal concentrations (“dissolved

metal”) or total recoverable metal concentrations (“total recoverable metal”) in setting State

waxer quality standards. In the past, States have used both approaches when applying the

same Environmental Protection Agency. (EPA) criteria numbers.

..

Some older criteria

documents may have facilitated these different approaches to interpretation of the criteria

• because the documents were somewhat equivocal with regards to analytical methods. The

May 1992 interim guidance continued the policy that either approach was acceptable.

ItisnowthepolicyoftheOfficeofWaterthattheuseofdissolvedmetaltosetand

measure compliance with water quality standards is the recommended approach, because

dissolved metal more closely approximates the bioavailable fraction of metal in the water

column than does total recoverable metal. This conclusion regarding metals bioavailabilhty is

supported by a majority of the scientific community within and outside the Agency. One

reason is that a primary mechanism for water column toxicity is adsorption ax the gill surface

which requires metals to be in the dissolved form.

The position that the dissolved metals approach is more accurate has been questioned

because it neglects the possible toxicity of particulate metal. It is true that some studies have

indicated that particulate metals appear to contribute to the toxicity of metals, perhaps

because of factors such as desorprion of metals at the gill surface, but these same studies

indicate the toxicity of particulate metal is substantially less than that of dissolved metal.

Furthermore, any erroL incurred from excluding the contribution ofparticulate metal

will generally be compensated by other factors which make criteria conservative. For

example, metals in toxicity tests are added as simple salts to relatively

elean

water. Due to

andthelikelyambientpr~ncewaters,

ofmetals

asignificantintoxicityconcentrationtests

would

of

generallymetalsbindingbeexpectedagentstoinbemanymoredischarges

I

bioavailabile than metals in discharges or in ambient waters.

If total recoverable metal is used for the purpose of water quality standards,

compounding of factors due to the lower bioavailability ofparticulate metal and lower

bioavailability of metals as they are discharged may result in a conservative water quality

standard The use of dissolved metal in water quality standards gives a more accurate result

However, the majority of the participants at the Annapolis meeting felt that total recoverable

measurements in ambient water had some value, and that exceedences of criteria on a total

recoverable basis were an indication that metal loadings could be a stress to the ecosystem.

particularly in locations other than the water column.

I

-

---

3

•

The reasons for the potential consideration of total recoverable measurements include

risk management considerations not covered by evaluation of waxer column toxicity. The

• ambient water quality criteria are neither designed-nor intended -to protect sediments, or to

prevent effects due to food webs containing sediment dwelling organisms. A risk manager,

however, may consider sediments and food chain effects and may decide to take a

conservative approach for metals, considering that metals are very persistent chemicals. This

conservative approach could include the use of total recoverable metal in water quality

standards. However, since consideraton of sediment impacts is not incorporated into the

criteria methodology, the degree of conservatism inherent in ~he total recoverable approach is

unknown. The uncertainty of metal impacts in sediments Stein from the lack of sediment

criteria and an imprecise understanding of the fate and transport of metals. EPA will

continue to pursue research and other activities 10 close these knowledge gaps.

Until the scientific uncertainties are better resolved, a range of different risk

management decisions can be justified. EM recommends that State water quality standards

be based on dissolved metal. (See the paragraph below and the attached guidance for

technical details on developing dissolved criteria.) EPA will also approve a State risk

management decision to adopt standards based on total recoverable metal, if those standards

are otherwise approvable as a matter of law.

o

Dissolved Criteria

In the toxicity tests used to develop EPA metals criteria for aquatic life, some fraction

of the metal is dissolved while some fraction is bound to particulate matter. The present

criteria were developed using total recoverable metal measurements or measures expected to

give equivalent results in toxicity tests, and are articulated as total recoverable. Therefore,

in order to express the EPA criteria as dissolved, a total recoverable to dissolved correction

factor must be used. Attachment #2 provides guidance for calculating EPA dissolved criteria

from the published total recoverable criteria. The data expressed as percentage metal

dissolved are presented as recommended values and ranges. However, the choice within

ranges is a State risk management decision. We have recently supplemented the data for

copper and are proceeding to further supplement the data for copper and other metals. As

testing is completed, we will make this information available and this is expected to reduce

the magnitude of the ranges for some of the conversion factors provided. We also strongly

encourage the application of dissolved criteria across a watershed or wazerbody, as

technically sound and the best use of resources.

o

Site-Specific Criteria Modifications

While the above methods will correct some site-specific (actors affecting metals

toxicity, further refinements are possible. EPA has issued guidance (Water Quality

Standards Handbook, 1983; Guidelines for Deriving Numerical Aquatic Site-Specific Water

Quality Criteria by Modifying National Criteria, EPA-60013-H4-099, October 1984) for three

site-specific criteria development methodologies: recalculation procedure, indicator species

procedure (also known as the water-effect ratio (WER)) and resident species

procedure.

Only the first two of these have been widely used.

---

4

~nthe National Toxics Rule

(57

FR 60848, December 22, 1992), EPA identified the

WER ~san optional method for site-specific criteria development for certain metals. EPA

committed in the NTR preamble to provide guidance on determining the WER. A draft of

this guidance has been circulated to the :States and Regions for review and comment. As

justified by water characteristics an~das recommended by the WER guidance, we strongly

encourage the application of the WER across a watershed or waterbody as opposed to

application on a discharger by discharger basis, as technically sound and an efficient use of

resources.

•

In order to meet current needs, but allow for changes suggested by protocol users,

EPA will issue the guidance as interim.v EPA will accept WERs developed using this

guidance, as well as by using other scientificaliy.defensible protocols. OW expects the

•

interim WER guidance will be issued in the next two months.

Total Maximum Daily Lcads (TMDLsI and National Pollutant Discharge Elimination System

(NPDES~Permits

•

•

•

o

Dynamic Water Quality Modeling

Although not specifically part of the reassessment of water quality criteria for metals,

dynamic or probabilistic models are another useful tool for implementing waxer quality

criteria, especially for those criteria protecting aquatic life. These models provide another

way to incorporate site-specific data. The 1991 Technical Support Document for Water

Quality-based Toxics Control (~SD)(EPA/505/2-90-OO1) describes dynanuc, as well as static

(steady-state) models. Dynamic models make the best use of the specified magnitude,

duration, and frequency of waler quality criteria and, therefore, provide a more accurate

representation of the probability that a water quality standard exceedence will occur In

contrast, steady-state models make a number of simplifying, worst case assumptions which

makes them less complex and less accurate than dynamic models.

Dynamic models have received increased attention over the last few years as a result

of the widespread belief that steady-stale modeling is over-conservative due to

environmentally conservative dilution assumptions This belief has led to the misconception

that dynamic models will always lead to less stringent regulatory controls (e.g., NPDES

rn

effluent limits) than steady-state models, which is not true

in

every application of dynamic

models. EPA considers dynamic models to be a more accurate approach to implementing

water quality criteria and continues to recommend their use. Dynamic modeling does require

commitment of resources to develop appropriate data. (See Attachment #3 and the TSD for

derails on the use of dynamic models.)

o

Dissolved-Total Metal Translators

Expressing water quality criteria as the dissolved form of a metal poses a need to be

able to translate from dissolved metal to total recoverable metal for TMDLi and NPDES

permits. TMDLs for metals must be able to calculate: (1) dissolved metal in order to

ascertain attainment of water quality standards, and (2) total recoverable metal in order

to

achieve mass balance necessary for permitting purposes

---

5

EPA’s NPDES regulations require that limits of metals in permits be stated as total

recoverable in most cases (see 40 CFR

§

122.45(c)) except when an effluent guideline

specifies the limitation in another form of the metal, the approved analytical methods

measure only dissolved metal, or the permit writer expresses a metals limit in another form

(e.g., dissolved, valent, or total) when required to carry out provisions of the Clean Water

Act. This is because the chemical conditions in ambient waters frequently differ substantially

from those in the effluent, and there is no assurance that effluent particulate metal would not

dissolve after discharge. The NPDES rule does not require that State water quality standards

be expressed as total recoverable; rather, the rule requires permit wnters to translate between

different metal forms in the calculation of the permit limit so that a total recoverable limit

can be established. Both the TMDL and NPDES uses of water quality criteria require the

ability to translate between dissolved metal and total recoverable metal. Attachment #3

provides methods for this translation.

Guidance on Monitoring

o

Use of Clean Sampling and Analytical Techniques

In assessing waterbodies to determine the potential for toxicity problems due to

metals, the quality of the data used is an important issue. Metals data are used to determine

attainment status for water quality standards, discern trends in water quality, estimate

background loads for TMDLs, calibrate fate and transport models, estimate effluent

concentrations (including effluent variability), assess permit compliance, and conduct

research. The quality of trace level metal data, especially below 1 ppb, may be

compromised due to contamination of sampies during collection, preparation, storage, and

analysis. Depending on the level of metal present, the use of clean” and ultraclean

techniques for sampling and analysis may be critical to accurate data for implementation of

aquatic life criteria for metals.~

The magnitude of the contamination problem increases as the ambient and effluent

metal concentration decreases and, therefore, problems are more likely in ambient

measurements.

CIean techniques refer to those requirements (or practices for sample

collection and handling) necessary to produce reliable analytical data in the part per billion

(ppb) range. Ultraclean techniques refer to those requirements or practices necessary to

produce reliable analytical data in the part per trillion (ppt) range. Because typical

concentrations of metals in surface waters and effluents vary from one metal to another, the

effect of contamination on the quality of metals monitoring cata varies appreciably.

We plan to develop protocols on the use of clean and ultra-clean techniques and are

coordinating with the United States Geological Survey

(USGS)

on this project, becatice USGS

has been doing work on these techniques for some time, erpecially the sampling procedures.

We anticipate that our draft protocols for clean techniques ‘viii be available in late calendar

year 1993. The development of comparable protocols for ultra-clean techniques is underway

and will be available in 1995. In developing these protocols, we will consider the costs of

these techniques and will give guidance as to the situations where their use is necessary.

Appendix B to the WER guidance document provides some general guidance on the use of

---

6

clean analytical techniques. (See Attachment #4.)

We

recommend that this guidance be used

by States and Regions as an interim step, while the clean and ultra-clean protocols are being

developed.

o

Use of Historical Data

The concerns about metals sampling and analysis discussed above raise corresponding

concerns about the validity of historical data. Data on effluent and ambient metal

concentrations are collected by a variety of organizations including Federal agencies (e.g.,

EPA, USGS), State pollution control agencies and health departments, local government

agencies, municipalities, industrial dischargers, researchers, and others. Th&data are

collected fora variety

of

purposes as discussed above.

Concern about the reliability of the sample collection and analysis procedures

is

greatest where they have been used to monitor very low level metal concentrations.

Specifically, studies have shown data sets with contamination problems during sample

collection and laboratory analysis, that have resulted in inaccurate measurements.

For

example, in developing a TMDL for New York Harbor, some historical ambient data showed

extensive metals problems in the harbor, while other historical ambient data showed only

limited metals problems. Careful resampling and analysis in 199211993 showed the latter

view was correct.

The

key to producing accurate data is appropriate quality assurance

(QA)

and quality control (QC) procedures.

We

believe that most historical data for metals,

collected and analyzed with appropriate

QA

and QC at levels

of

1 ppb or higher, are

reliable.

The

data used in development of EPA criteria are also considered reliable, both

because they meet

the

above test and because the toxicity test solutions are created

by

adding

known amounts of metals.

With respect to effluent monitoring reported by an NPDES peimittee,

the penn tree is

responsible for collecting and reporting quality .data. on

a

Discharge Monitoring Report

(DMR) Permitting authorities should continue to consider

the

information reported to be

true, accurate, and complete as certified by the pennittee. Where the pernuttee becomes

aware of new information specific so the effluent discharge that questions the quality of

previously submitted DMR~datathe pennuttee must promptly submit that information to the

permitting authority The pernumng authority will consider all informat’on submitted by the

perrmttecin determining appropriate enforcement responses to monitoring/reporting and

effluent violations (See Attachment #4 for additional details)

Summary

I

The

management of metals in

the

aquatic environment is complex. The science

supporting our technical and regulatory programs is continuing to evolve, here as in all

4~.-eas.

The

policy and guidance outlined above represent

the

position of

OW

and should be

incorporated into ongoing prugram operations. We do not expect that ongoing operations

would be delayed

or

deferrod because

of this

guidance.

I

---

7

If you have questions concerning this guidance, please contact Jim Hanlon, Acting

Director, Office of Science and Technology, at

202-260-5400.

If you have questions on

specific details of the guidance, please contact the appropriate OW Branch Chief. The

Branch Chiefs responsible for the various areas of the water quality program are: Bob April

(202-260-6322, water quality criteria), Elizabeth Fellows (202-260-7046, monitoring and data

issues), Russ Kinerson (202-260-1330, modeling and translators), Don Brady (202-260-7074,

Total Maximum Daily Loads), Sheila Frace (202-260-9537, permits), Dave Sabock

(202.260-1315, water quality standards), Bill Telliard (202-260-7134, analytical methods)

and Dave Lyons (202-260-8310, enforcement).

Attachments

---

AUACHMENT #1.

TECHNICAL GUIDANCE fOR METALS

Schedule of Upcoming Guidance

-

Water-effect Ratio Guidance

-

September 1993

Draft “Clean” Analytical Methods

-

Spring 1994

Dissolved Criteria

-

cuiTently being done; as testing is completed, we will release the

updated percent dissolved data

Draft Sediment Criteria for Metals

-

1994

Final Sediment Criteria for Metals

-

1995

---

GUIDANCE DOCUMENT

ON DISSOLVED CRITERIA

Expression of Aquatic Life Criteria

October 1993

AUACHMENT #2

---

10—1—93

Percent Dissolved in Aquatic Toxicity Tests on Metals

The attached table contains all the data that were

found

concerning the percent of the total recoverable metal that was

dissolved in aquatic toxicity tests.

This

table is intended to

contain the available data that are relevant to the conversion of

EPA’S aquatic life criteria for metals from a total recoverable

basis to a dissolved basis. (A factor of 1.0 is used to convert

aquatic life criteria for metals that are expressed on the basis

of the. acid-soluble measurement’ to criteria expressed on the

basis of the total recoverable measurement.) Reports by

Grunwald

(1992) and Brungs .et. al. .(1992) provided references to many of

the documents in which pertinent data were found. Each document

was obtained and examined to determine whether it contained

useful data.

“Dissolved” is defined as metal that passes through a 0.45—pm

membrane filter. If otherwise acceptable, data that were

obtained using 0.3-pm glass

fiber

filters and 0.1—pm membrane

filters were used, arid are identified in the table; these data

did not seem to be outliers.

Data were used only if the metal was in a dissolved inorganic

form when it was added to the dilution water. In addition, data

were used only if they were generated in water that would have

been acceptable for use as a dilution water in tests used in the

derivation of water quality criteria for aquatic life; in

particular, the pH had to be between 6.5 and 9.0, and the

concentrations of total organic

carbon

(TOC) and total suspended

solids (TSS) had to be

below

5 mg/L. Thus most data generated

using river water would not be used.

Some data were not used for other reasons. Data ‘presented by

Carroll et al. (1.979) for cadmium were not used because 9 of the

36 values were above 150. Data presented by Davies et al.

(1976) for lead and Holcombe and

Andrew

(1978) for zinc were not

used because “dissolved” was defined on the basis of

polarography, rather than filtration.

Beyond this, the data were not reviewed for quality. Horowitz et

al. (1992) reported that a number of aspects of the filtration

procedure might affect the results. In addition, there might be

concern about use of “clean techniques” and adequate QA/QC.

Each line in the table is intended to represent a separate piece

of information. All of the data in the table were determined in

fresh water, because no saltwater data were found. Data are

becoming available for copper in salt water from the New York

1

---

Harbor study; based on the first set of tests, Hansen (1993)

suggested that the average percent of the copper that is

dissolved in sensitive saltwater tests is in the range of 76 to

82 percent.

-

A thorough investigation of

the

percent of total recoverable

metal that is dissolved in toxicity tests might attempt to

determine if the percentage is affected by test

technique

(static, renewal, flow—through), feeding (were ‘the test animals

fed and, if so, what

food

and how

much),

water quality

freshwater

-

criteria

I

‘concentrations (cxc and CCC) because percentages for total

recoverable concentrations much (e.g., more than a factor of 3)

above or below the

cxc

and CCC are likely, to be less relevant.

When a’criterion is expressed as a hardness equation,

the

range

given extends from a hardnes. of 50 mg/L to a hardness of 200

mg/L.

-

The following is a

summary

of the available information for each

-

metal:

Arsenic(III)

II,

Th.

data available indicat, that the percent dissolved is about

100, but all

the

availabl, data are for concentrations that are

much higher than

the

CMC

Cadmium

Schuytema

et al. (1984) reported that

“there wers

no real

differences” between measurements of total

and

dissolved

cadmium

at concentrations of 10 to 80 ug/L (pH

—

6.7 to 7.8, hardness

-

25 mg/L, and alkalinity

—

33 mg/L); total

and

dissolved

concentrations were said to be “virtually equivalent”.

The

CHC and CCC

are

close

together

and only range from 0

•

66 to

8.6

uq’j~L~

The only available data that are known to be in the

rang. of the

cxc

and CCC were determined with a

glass

fiber

filter. Th. percentages that are probably most relevant are

75,

92, 89, 78, and 80.

1

Chromium C

III~

The percent dissolved decreased as

the

total

recoverable

concentration increased, even thougn the highest concentratIonS

reduced the pH substantially. The percentages

that

are probablY

2

---

most relevant to the CMC

are

50-75,

whereas the percentages that

are probably most relevant to

the CCC are

86 and 61.

Chromium CVfl

The data available indicate that

the

percent dissolved is about

100, but all the available data are for concentrations that are

much higher than the

CMC

CoDDer

Howarth and Sprague (1978) reported that the total and dissolved

concentrations -of copper-were “,litt1e~different” except when

the

total copper concentration was above 500 ug/L at hardness

—

360

mg/L and pH

=

8 or 9. Chakoumakos et al. (1979)

found

that the

percent dissolved depended more on alkalinity than on hardness,

pH, or the total recoverable concentration of copper.

Chapman (1993) and Lazorchak (1987) both found that the addition

of daphnid food affected the percent dissolved very little, even

though Chapman used yeast-trout chow-alfalfa whereas Lazorchak

used algae in most tests, but yeast—trout chow—alfalfa in some

tests. Chapman (1993) found a low percent dissolved with and

without food, whereas Lazorchak (1987)

found

a high percent

dissolved with and without food. All of

Lazorchak’s

values were

in high hardness water; Chapman’s

one value in high hardness

water was much higher than his other values.

Chapman (1993) and Lazorchak (1987)

both

compared the effect of

food on the total recoverable LC5O with ths effect of food on the

dissolved LC5O. Both authors found

that

food raised both the

dissolved LC5O and the total recoverable LC5O in about the same

proportion, indicating that food did not raise the ~total.

recoverable LC5O by sorbing metal onto food particles; possibly

the food raised

both

LC5O5 by (a) decreasing the toxicity of

dissolved metal, (b) forming nontoxic dissolved complexes with

the metal, or

(C)

reducing uptake.

The

cxc.,

and

CCC are close together and only range from 6.5 to 34

ug/L. The percentages that are probably most relevant are 74,

95, 95, 73, 57, 53, 52, 64, and 91.

Lead

The data presented in Spehar et al. (1978) were from Holcomb. et

al. (1976). Both Chapman (1993) and Holcombe et al. (1976) found

that the percent dissolved increased as

the total recoverable

concentration increased.

It

would

seem reasonable to expect more

precipitate at higher total recoverable concentrations and

3

---

I

therefore a lower percent dissolved at higher concentrations.

The increase in percent dissolved with increasing concentration

might be due ‘to a lowering of the pH as more metal is added if

the

stock solution was acidic.

The

percentages

that

are

probably

most relevant to

the

cxc

ar. 9,

-

that are probably most relevant to the CCC are 9 and 10.:

TheMercurYonly percentage that is available is 73, but it is for a

I

concentration that is much higher

than

the

cxc.

Nickel

The percentages that are probably most relevant to the

cxc

are

88, 93, 92, and

100,

whereas

the only

percentage that is probably

relevant to the CCC is 76.

-

No data are available.

Silver

1

F

There

is a

cxc,

but not a CCC. The percentage dissolved seems

to

be greatly reduced by the food used to

feed

daphnids, but not by

the

food

used to feed fathead minnows.

The

percentages that are

probably meat relevant to the

cxc

are

41, 79, 79, 73, 91, 90, and

93.

-

Zinc

The

cxc

and CCC are close together and only range from 59 to 210

ug/L.

The

percentages that are probably most relevant are 31,

77,

77, 99,

94, 100, 103, and 96.

a

I

I

I

---

Arsenic (III)

Cadmium

Chromium (III)

Copper

Silver

95

85

85

95

85

50

9—95

85

73’

85

NA’

85

85

95

85

85

-

61—86

95

‘

85

25

9—10

NA’

85

76

NA’

NAC

yyD

85

31—103

A

The recommended values are based on

current

knowledge and are

subject to

change as more data

become. available.

‘

All available data are for concentrations that are much higher

than the

cxc.

C

NA

—

No data are available.

D

yy —

-

A

CCC

is not available, and therefore

cannot be

adjusted.

‘

NA

—

Bioaccumulative chemical and not appropriate to adjust to

Considered Most Relevant in Fresh Water

Metal.

5

---

?

44 43 7.4 Spehar and Fiandt 1986

Concn.A

I

ua

IL)

Percent

Diss.’

FM

______

_____

th

SDeCiesD

SRF’

Food Hard.

Alk.

p11

ARSENICI 1111 (Fr.jh~ater:CCC

—

190 ug/L; CMC

—

600—15000 104

12600

100

-

3

CADMIUM

(Freshwater:

CCC

—

0.66

to 2.0

ug/L;

CMC

—

1.8

to 8.6 ug/L)~

0.16

0.28

41

75

?

?

DM

DM

R

Yes

Yes

53

103

46

7.6

7.9

Chapman 1993

Chapman 3.993

-

0.44.0

920

?

CS

F

No

21

19

.

7.1 Finläyson and Vérrue 1982

13

89

3

FM

F

No

44 43

7.4 Spehar and Fiàndt 1986

-

15—21

42

96

84

8

4

FM

FM

S

S

No

NO

42

45

31

43.

7.5

7.4

Spehar and Carison 1984

Spehar

and Carison 1984

10

35

78

77

59

?

?

ON

ON

DM

S

S

No

No

51

105

209

38

7.5

8.0

Chapman 1993

Chapman 1993

Chapman 1993

450—6400

80

9O~

70

8

s

5

?

?

FM

S

F

F

No

?

No

47

46

202

44

42

157

7.5

7.4

Pickering and Cast 1972

6

---

a~. — —

—~

CHROMIWII1111 (Freshwater:

CCC

=

120 to 370 ug/L; CMC

=

980 to 3100 u4/L)’

5-13

94

-

?

SG

F

?

25 24 73 Stevens and Chapman 1984

19—495

86

?

-~

SG

F

?

25 24 7.2 Stevens and Chapman

1984

1100

50-75 ?

SG

F

No

42

54

DN

R

Yes 206 166 8.2 Chapman

114

61

DM

R

Yes 52

45

16840

26

?

ON

S

No 51

9 6.3’ Chapman

26267

27416

58665

-

27

23

?

?

014

ON

OH

S

S

No

No

110

96

190

9

10

6.7

60’

Chapman

Chapman

Chapman

1993

1993

CHROM1UH(V11 (Freshwater CCC

=

11 ug/L; CMC

=

25,000

100

43,300

99.5

FM,GF F

FM

F

No

44 43 7.4 Spehar and Fiandt 1986

COPPER

(Freshwater: CCC

—

6.5 to 21 ug/L; Cxc

=

9.2 to 34 ug/L)’~

10—00

CT

F

10-30

74

?

CT

F

No

27

40200 78

?

CT

F

No 154 20 6.8

30—100 79

?

CT

F

No

74 23 7.6 Chakoumakos

300—200 82

CT

F

No 192 72 7.0

Chakoumakos

20200 86

CT

F

NO

31 78

8.3

40-300 87

CT

F

No

83 70

7.4 Chakoumakos ,et al. 1979

89

7

---

7.0 Chakoumakos e~i~. 1979

8.5 C~akousakos et a.. 1979

‘~

~

—

~

——— ~

—

__

F No

F

No

R

Yes

R

Yes

S

No

R

No

S

No

R

No

31 38 7.2 Carlson et a.. 1986a,b

31 38 7.2 Carison et a.. 1986a,b

52 55 7.7 Carison et a.. 1986b

31 38

52 55 7.7 Carison et a.. 1986b

31 38

7.2 Carlson et al. 1986b

12—91’

18—19

50

5—52

6—80

6.7

13

16

32

33

25—84

17

28—58

26—59

F

Yes’ 47

43 8.0 Lind et al.

F

No

21 19 7.1 Finlayson and Verrue 1982

?

CT

?

CT

2

CD

CD

2

Ok

1

Ok

2

2

FM

FM

?

CS

DM

014

DM

DM

DPI

DPI

014

014

14

F14,GM

6

DPI

14

SC

19

BC

6

7

2

79—84

95

96

91

8

2K

83°

57

73

57

53

52

96

91

88

74

80”

85

79

S

S

R

Yes

R Yes

R Yes

S

No

S

No

S

No

S

No

S

No

S

S

49 37

48 39 7.4

211 169 8.1 Chapman

51 44 7.6 Chapman 1993

52 45

105 79

106 82

52 43 7.3

48 47 7.3

F

Hammermaisteret a.. 1983

Hammermeister et a.. 1983

Hasmermeister et a.. 1983

47 7.7

Call

43 7—8 Benoit 1975

R

R

YesTM

168

117

8.0

Lazorchak

Lazorchak

1987

R

y55N

168

117 8.0 Lazorchak 1987

8

---

iii

FM

F

No

FM

S

CR

F

No

CCC~ 1.3 to

p7.7 ug/L;

CMC

?

OH

R

Yes

ON

DPI

R

Yes

DPI

S

No

DM

DPI

S

EZ

R

F

Yes

F

Yes

F

Yes

F

Yes

F Yea

F

F

S

No

S

No

?

?

BT

?

BT

?

BT

?

BT

?

BT

7

44 43 7.4 Spehar and Fiandt 1986

17 13 7.6 Rice and Harrison 1983

=

34 to 200 ug/L)’~

102

151

86

7.6

7.8

chapman

Chapman

Chapman

1993

1993

100

150

——

——

——

---

——-

———

Chapman

Chapman

Chapman

1993

1993

22

44

43 7.2 Holcombe et a.. 1976

44 43 7.2 Holcombe’ ct al. 1976

44 43 7.2 Holcombe

44 43 7.2 Holcombe et al.

44 43 7.2 Holcombe et al. 1976

44

43

7.2

Holcombe

73

1

FM

F

No

96

86

160

94

1

230—3000

69—79

LEAD (Freshwater:

181

612

952

1907

7—29

58

119

235

474

4100

2100

9

18

29

33

—38

62”

68~

750

81”

82”

79

--

---

220—2700

96

14

580

95

14

SC

49 44 7.2 Hammermeister et al. 1983

51 48 7.2 Hasmermeister et a.. -1983

9

---

NICKEL (Freshwater: CCC

—

21

81

DPI

R

Yes

51 49

7.4 Chapman 1993

150

578

76

87

?

?

DPI

DII

R

Yes

107

205

87

7.8

Chapman

Chapman

1993

645

88

?

DPI

S

No

1809

93

?

014

S

No

51

44 7.7 Chapman 1993

1940

92

?

DII

S

No 104 84

2344

100

?

DII

S

No

100 84

7.9 Chapman 1993

4000

90

PK

R

No

21

--

---

JRB

Associates 1983

SELENIUM (FRESHWATER:

CCC

—

5 ug/L; CMC

—

No data are available.

SILVER (Freshwater:

Cxc

—

0.19

74

DPI

S

No

9.98

13

DII

S

Yea

47 37 7.5 Chapman

1993

4.0

41

DII

S

No

36

25 7.0

4.0

it

DPI

S

Yes 36

25 7.0

Nebeker et al.

1983

3

2—54

79

79

?

?

FM

FM

S

No

Yes°

51

49

49

8.1

7.9

UWS1993

UWS

1993

2—32

73

?

FM

S

No

50 49 8.1

UWS

1993

4—32

91

?

FM

S

No

48

49 8.1

UWS1993

5—89

90

?

FM

S

No 120 49 8.2

UWS

1993

6—401

93

?

FM

S

No 249

49 8.1

UWS

1993

10

~

U

a UU

---

ZINC

(Freshwater: CCC

=

59 to 190 ug/L; CMC 65 to 210 ug/14F

104

54

105

196

31

DM

R

Yes

77

DII

R

Yes

77

?.~:-

DPI

R

Yes

74

DM

S

No

78

OH

S

No

76

OH

S

71—129 2

CD

R

Yes

81—107 2

CD

R

99

2

CD

R

No

94

1

CD

S

100

2

FM

R

No

100

1

FM

S

950

CS

F

100

AS

F

No

83

AS

F

90

?

FM

F

No 204

70

FM

F

103

13 FMGM,DM S

No 52

96

13

SG

S

No

A

Chapman 1993

Chapman 1993

Chapman 1993

83 7.8

47 7.5

85

153

62

356

551

741

18—273’

167’

18 8—393’

40—500

1940

4000

160—400

7.6

8.1

38 7.2

31 38 7.2 Carison et al. 1986b

3.

38 7.2 Carison et al. 1986b

43 7.5 Hammermeister et a.. 1983

46 7.2 Hammermeister et a.. 1983

11

---

C

D

The abbreviations used are:

AS

—

Atlantic salmon

DM Da~hniamauna

ST

—

Brook trout

EZ

—

£1~m~

zonatum

CD

=

ceri~aphnia dubia

FM Fathead minnow

CR

—

Crayfish

GF

=

GOldfish

Cs

—

Chinook salmon

GM

-

Gammanid

CT

—

Cutthroat trout

PK

—

Palaeppnetes kadiakensiB

DA

—

Daphnids

SG

—

~

uairdneri

!

Tha abbreviations used are:

-‘

S

—

static

R

—

renewal

F

—

~ The two numbers are for hardnesses of 50 and 200 mg/L,.respectjvely.

‘~

A O.3pm gj~j~

fiber

filter was used.

“

-‘

H

A O.lO-pS membrane filter was used.

The pH was below 6.5.

‘

The dilution water was a clean river water, with TSS and

TOC below

5

ag/L.

k

Only limited information is availabl, concerning this value.

‘

It is

assumed

that the solution that was filtered was from the test chambers that

contained fish and food.

N

The food was algae.

“

The

food

was yeast-trout chow-alfalfa.

o

The

food

was frozen adu.t brine shrimp.

12

—

—

————— _, _

———

~

—————

---

I.

Geckler,

J.R.,

W.B. Horning, T.M. Neiheisel, Q.H.

Pickering, E.L.

Robinson, and C.E. Stephan.

-

1976. Validity of Laboratory Tests

for Predicting Copper Toxicity in Streams. EPA-600f 3—76-116.

National Technical Information Service, Springfield, VA. Page

118.

-

-

Grurwald, D. 1992.

-

Metal Toxicity Evaluation: Review, Results,

and Data Base Documentation.

Hammermeister, D., C. Northcott, L. Brooks, and D. Call. 1983.

Comparison of Copper, Lead and Zinc Toxicity to

Four Animal

Species in

Laboratory

and ST.

Louis

River Water. University of

Wisconsin—Superior, Superior, WI.

Hansen,

D.J.

1993. Memorandum to

C.E. Stephan. April 15.

Holcombe, G.W., D.A. Benoit, E.N. Leonard, and 3.14. McKim. 1976.

Long-Term Effects of

Lead Exposur.

on Three Generations of Brook

Trout

(Salvelinus .tontLna.Zis). 3.

Fish. Res. Sd. Canada 33:1731-

1741.

-

Holcombe, G.W~, and R.W.

Andrew.

1978. Tb. Acute Toxicity of

Zinc to Rainbow and Brook Trout. EPA—600/3-78—094. National

Technical Information

Service,

Springfield, VA.

Horowitz, A.J., K.A. Elrick, and M.R. Colberg. 1992.

The

Effect

of Membrane Filtration Artifacts on

Dissolved

Trace

Element

Concentrations. Water Res. 26:753-763.

Howarth, R.S., and 3.5. Sprague. 1978.

Copper Lethality to

Rainbow Trout in Waters on

Various Hardness and pM. Water Rca.

12:455—462.

-

JRB Associates. 1983. Demonstration of the Site-specific

Criteria Modification Process: Selser’. Creek, Ponchatoula,

Louisiana.

DaøhniaLazorchak,~

3.14.Straus

1987.

During

TheAcute

Toxicity

Tests

Weight

with

Loss

copper.

of

I

“

:1

Ph.D.

Thesis.

-

Lima,

Northcott,

A.R.,

L.T.

C. Curtis,

Brook..

Mark.s,Toxicities

C.E.

of

I

Arsenic(III) to Fathead Minnows, Flagfish, Daphnids,

and an

Amphipod. Arch.

Environ. Contam. Toxicol. 13:595-601.

Lind, D, K. Alto, and S. Chatterton. 1978. Regional Copper-

,.,

Nickel Study. Draft.

I

Mount, D.I. 1966. The Effect of Total Hardness and pH Acute

Toxicity of Zinc to Fish. Air Water Pollut. mt. 3. 10:49-56.

1

14

I

---

FL -.

Nebeker, A.V., C.K. McAuliffe, R. Mshar, and D.G. Stevens. 1983.

‘Toxicity of Silver to Steelhead and 1~ainbowTrout, Fathead

Minnows, and

DaphnLa

magna.

Environ. Toxicol. Chem. 2:95-104.

Pickering, Q.P., and M.I. Cast. 1972. Acute ‘and Chronic

Toxicity of Cadmium to the: Fathead Minnow

(Pimepha.L.s prom..Zas).

3. Fish. Rca. 3d. Canada 29:1099—1106.

Rice, D.W., Jr., and F.L. Harrison. 1983. The Sensitivity of

Adult, Embryonic, and

Larval

Crayfish

Procamba.rus cla.rk.Li

to

Copper.

NUREG/CR-3

133 or UCRL-53048. National Technical

Information Service, Springfield, VA.

Schuytema, G.S., P.O. Nelson, K.W. Malueg, A.V. Nabeker, D.F.

Krawczyk, A.K. Ratcliff’, and J.H. Gakstatter. 1984. Toxicity of

Cadmium

in Water and’ Sediment’ Slurries to

Daphnta magna.

Environ. Toxicol. Chem. 3:293-308.

Spehar, R.L., R.L. Anderson, and.J.T. Fiandt. 1978. Toxicity

and Bioaccumu.ation of Cadmium and

Lead

in Aquatic Invertebrates.

Environ. Pollut. 15:195—208.

Spehar, R.L., and A.R. Carlson. 1984. Derivation of Site—

Specific Water Quality Criteria for Cadmium and the St. Louis

River Basin, Duluth, Minnesota. Environ. Toxicol.

Chem.

3:651-

665.

Spehar, R.L., and J.T. Fiandt. 1986. Acute and Chronic Effects

of Water Quality Criteria-Based Metal Mixtures on Three Aquatic

Species. Environ. Toxicol. Chem. 5:917-931.

Sprague, 3.3. 1964. Lethal Concentration of Copper and Zinc for

Young Atlantic Salmon.

3.

Fish. Res. 3d. Canada 21:17—9926.

Stevens, D.G., and G.A. Chapman.. 1984.. Toxicity of Trivalent

Chromium to Early Life Stages. of Steelhead’Trout. Environ.

Taxicol. them. 3:125—133.

University of Wisconsin—Superior. 1993. Preliminary data from

work assignment 1—10 for Contract No. 68-C1-0034.

15