CHLORIDE BUDGET FOR THE GREAT  LAKES:

                 A CURRENT ASSESSMENT
                                                       OOOR81004
                                            By:

                                            William C. Sonzogni

                                            William Richardson
                                            Paul Rodgers

                                            Timothy J. Monteith
    Great Lakes Environmental Planning Study (GLEPS)
Great Lakes Basin Commission, Ann Arbor,  Michigan  48106
                 Contribution No. 39
                      June, 1981

-------
          CHLORIDE BUDGET FOR THE GREAT LAKES:

                  A CURRENT  ASSESSMENT
                          By:
                  William C.  Sonzogni1
                   William Richardson2
                      Paul Rodgers^
                  Timothy J.  Monteith3
    Great Lakes Environmental Planning Study (GLEPS)
Great Lakes Basin Commission, Ann Arbor,  Michigan 48106
                  Contribution No. 39
                       June,  1981
     ^Great Lakes Environmental Research Laboratory
     National Oceanic and Atmospheric Administration
                  Ann Arbor,  MI  48104
              (GLERL Contribution No. 272)
              2Large Lakes Research Station
          'U.S. Environmental Protection Agency
                 Grosse Isle, MI  48138
                     Lakes Basin Commission

-------

-------
                                 ABSTRACT
                                                        *



     Concentrations of chloride and other conservative  ions  in  the Great




Lakes are of interest, not only as indicators of pollution,  but also because




these ions may affect species diversity of algae and other aquatic orga-




nisms.  As part of an evaluation of the current status  of conservative ion;




in the lakes, a current cl ioride budget was developed and its implications




assessed.  Chloride inputs to the lakes (in 106 MT/YR)  increase in the order




of Superior  (0.3), Michigan  (0.9), Huron  (1.1), Erie  (3.7),  and Ontario




(6.3).  The  Oswego River, draining into Lake Ontario, contributes more




chloride than any other U.S. tributary.  Regarding  specific  sources,




discharges from industrial processes  are probably the most significant.




Road salt contributes an  important, but not necessarily predominant, portion




of the anthropogenic load to the lakes.  Overall, chloride inputs-, espe-




cially industrial discharges, appear  to have decreased  in recent years.




Lake Erie in-lake chloride concentrations have, in  fact, decreased




measurably compared to those in the early 1960's.




     Based on a chloride  model that treats the lakes  as five completely




mixed systems in series,  it  is estimated  that, if current loads are main-




tained, the  average chloride concentrations in Lake Michigan should increase




over the long term from less than 8 mg/L  to nearly  20 mg/L.   Chloride con-




centrations  in the other  Great Lakes  are  predicted  to remain relatively




stable.  Based on analogies  to other  lakes where chloride levels have




increased, it is uncertain whether the expected rise  in the  chloride con-




centrations  in Lake Michigan will result  in a shift in  phytoplankton toward




more nuisance species.  It may well be that increased conservative ion




levels are a factor ecologically, but of  less importance  than other factors,

-------
such as nutrients.  The response of the Great Lakes to the massive




phosphorus reduction program of the 1970's should provide further insight




into the relative importance of chloride and other conservative ions as




pollutants.

-------
                               INTRODUCTION




     Chloride has long been used as a tracer of pollution (Beeton, 1965;




Upchurch, 1976), but has only recently been implicated as a pollutant itself




in the Great Lakes, the worlds largest freshwater resource.  The concern




over chloride (and other conservative ions) stems from its possible role in




changing the species diversity ar J distribution of algae and other organisms




in the Great Lakes.




     Concern over chloride is exemplified by a recent conference held to




help develop a five year federal plan for research and development in the




Great Lakes.  Chloride discharges, which have increased markedly over the




last century, were identified as one of several major problems worthy of




special attention (Beeton et al., 1980).




     In response to renewed interest in chloride, this paper summarizes




current information on chloride  inputs to the Great Lakes from various sour-




ces.  Chloride trends in the lakes are then evaluated with the aid of two




mathematical models.  Finally, management implications of current and future




chloride inputs to the lakes are discussed.  Particular attention is given




to Lake Michigan, especially with regard to modeling the response of lakes




to chloride inputs.









                             Chloride Sources




     Chloride sources include land runoff, base flow (i.e., ground water




inputs), municipal and industrial effluents, the  atmosphere and several




minor contributers.  In the developed portion of  the Great Lakes basin,




point sources and runoff from agricultural and urban land dominate chloride




sources, while in the undeveloped portion, which  includes most of the




Superior and Huron watersheds and part of Michigan's basin, chloride inputs

-------
are mostly derived from natural weathering of chloride containing




minerals.




     Because of its local availability  (large underground deposits withir>




the basin provide an abundant supply of  salt),  sodium chloride is commonly




used for road deicing in the Great Lakes basin.  Salt use has proliferated




in recent years as a "bare pavement policy" has been in  force in many urban




areas.  Highway runoff may enter  tributaries directly or may enter tribu-




taries via storm sewers (separate or combined)  and  can result in chloride




concentrations as high as 10,000  mg/L (Schraufnagel, 1965).




     One portion of the Great Lakes that has been acutely affected by road




salt runoff is Irondequoit Bay, a small embayment near Rochester, N.Y.,  that




is separated from Lake Ontario by a sandbar.  Over  the past 20 years road-




salt runoff to the bay has resulted in  an increase  in _the chloride level to




the point that the drinking water standard of 250 mg/L has been exceeded




(Bubeck et al., 1971).  Further,  a vertical density gradient has formed




which has impeded mixing of the bay, particularly during spring and fall




turnover.  It is speculated that  continued road salt buildup may eventually




cause the bay to become meromictic  (Bannister and Bubeck, 1978).  Although




Irondequoit Bay is a dramatic example of road salt  pollution of a Great




Lakes embayment, it is perhaps a  unique case due to the  very limited mixing




between the bay and Lake Ontario  proper.




     The Pollution from Land Use  Activities Reference Group  (PLUARG, 1977)




estimated that about 2.8 x 10° metric tons of chloride are currently applied




to roads in the Great Lakes basin annually.  However, due to (1) increasing




salt prices, (2) concern over salt  induced corrosion, (3) local water quality




problems caused by salt runoff and  (4)  damage to terrestrial vegetation,

-------
the rate of salt application will probably decline (PLUARG 1977, 1978).




Importantly, the Pollution from Land Use Activities Reference Group Study




(PLUARG, 1978) concluded that, given the current and expected future use of




chloride for deicing, road salt pollution was more of a local problem than .1




Great Lakes problem.






Base Flow                              L




     Ground water inputs to streams (i.e., base flow) are also a source of




chloride in tributaries.  In some areas groundwater chloride concentrations




can be high, particularly in deeper bedrock deposits.  Cummings (1980)




reported the results of a survey of the chemical characteristics of ground-




water deposits within Michigan and found that bedrock deposits had a mean




concentration of 71 mg/L while glacial (surface) deposits had a mean con-




centration of 11 mg/L.  Ground water contributions to streams would likely




be from surface deposits, and thus high chloride levels in streams are not




generally the result of groundwater inputs (Sonzogni et al., 1978).






Point Sources




      Municipal sewage effluent contains substantial amounts of chloride.




Conventional treatment processes do not remove chloride to an appreciable




extent.  Typical chloride concentrations in municipal effluents of plants




draining into the Great Lakes range from 50 mg/L or less to over 160 mg/L.




Major sources of chloride received by  treatment plants include human wastes,




garbage disposal wastes, water softening by-products, urban runoff (for com-




bined sewer systems) and industrial contributions.  Humans contribute about




6 g of chloride per person per day to  sewage (Metcalf and Eddy, 1972).  Home

-------
water softening systems, where numerous, can be responsible for large


chloride inputs to wastewater (Schraufnagel, 1965).


     Whereas most municipal waste treatment processes do not remove chloride,


chemical treatment practices may actually add chloride to effluents.  Over



the last ten years phosphorus removal processes, largely in the form of


chemical precipitation have been implemente  at a  large number of municipal

                                           •J
sewage treatment plants in the Great Lakes basin.  Of the chemicals added to


wastewater for phosphorus removal, ferric chloride is most frequently used



(Monteith et al., 1980).  Since chloride is a by-product of ferric chloride


use, this phosphorus removal practice appears to be adding to the chloride


content of discharge water.  For example, Kenaga (1978) noted that the


chloride concentration in Lansing, Michigan, wastewater effluent increased


about 10 mg/L as a result of phosphorus removal.   Consequently, phosphorus


removal may be contributing measurably to chloride inputs.



     Discharges from industrial processes are probably the most significant


of all chloride sources to the Great Lakes.  Chemical, steel and food


packaging industries are major sources.  A number  of  industries have located


in the basin not only because of the abundance  of  water, but also because of


the large salt deposits found in different parts of  the basin.  For example,


Ownbey and Kee (1967) reported that, at the time of  their work, over half of


the chloride load to 'the Detroit River could be attributed to industrial


sources.  Many chemical firms are, in fact, situated  in the Detroit-Windsor


area in order to use the inexpensive salt brine for  the production of soda


ash and other alkali products (Ownbey and Kee,  1967).  In fact, Ownbey and


Kee noted that Lake Erie's chloride  level first began to increase at the


turn of the century when these industries became established.

-------
     The influence of industrial sources of chloride is further illustrated




by discharges to Onondaga Lake, a lake which drains into Lake Ontario via



the Oswego River.  Onondaga Lake has extremely high chloride concentrations



(seasonal averages have exceeded 1500 mg/L), largely as a result of



discharges from a chlor-alkali manufacturer (Effler et al., 1981).  As will




be discussed, the outflow from Onondaga Lake is a ' ajor reason for the



extremely high chloride load to Lake Ontario from "the Oswego River.






Atmospheric Sources



     Chloride also enters the Great Lakes via rainfall.  Atmospheric




chloride inputs can be an appreciable part of the total load, particularly




for Lakes Superior and Michigan which have large surface areas.  The salt



content of rainfall is largely derived from sea spray (Tiffany et al., 1969).




Other means by which chloride could enter the atmosphere are by wind erosion
                                                                   9


of soil and industrial emissions.  The significance of these sources is



not known, however.






Minor Sources




     Some minor  sources of chloride to the Great Lakes include direct



groundwater inputs, shoreline erosion and vessel discharges.  The quantity



of groundwater which directly flows into the lakes is not known, but is



believed to be a small part of the water budget (Quinn, 1978).




Consequently, chloride load from this source is assumed to  be negligible.



     Regarding shoreline erosion, which can contribute a large amount of




particulate material to the Great Lakes (Monteith and Sonzogni, 1976),




little information exists on chloride inputs.  The Upper Lakes Reference



Group (1977) did estimate a chloride load of about 800 metric tons per year

-------
from shoreline erosion, however.  Since this is a small part of the total


load (as will be shown), it will be ignored as a source.


     Finally, the Upper Lakes Reference Group (1977) estimated that Lake


Superior received a surprisingly high chloride load, about 11,000 metric tons


per year, from vessel discharges (mostly salt water ballast).  However, this


source is likely only to be significant relative ts other sources in Lake

                                                  j
Superior, Lake Superior having the lowest total chloride load of any of the


lakes.  Vessel discharges are not considered further here as an important


source.




                               Chloride Loads


Total Loads


     Chloride loads to each of the Great Lakes are  summarized in Table 1.


These loads represent data primarily from the mid to late 1970's.


     U.S. total tributary (river mouth) loads are the average annual load


over the water years 1975 through 1978, except for  Lake Erie, where data


were not available for 1978.  and Lake Michigan, where  the 1975 load was


excluded as a major shift in industrial chloride discharges occurred during


this year.  Calculations of chloride from gaged tributaries were made using


the ratio estimator method (Great Lakes Water Quality Board, 1976; Sonzogni


et al., 1978).  This method accounts for the importance of flow variability


in estimating annual loads.  Chloride loads from ungaged and/or unmonitored


tributaries were estimated as reported in Sonzogni  et al. (1978).  Primary


sources of data included state surveillance programs, reports of the U.S.


Geological Survey, the Upper Lakes Reference Group  Study of  the Inter-


national Joint Commission, the U.S. Army Corps of Engineers Lake Erie

-------
             Table 1

      Summary of Estimated
Chloride Loads to the Great Lakes
    During the 1970's (mt/yr)

Lake
Superior
U.S.
Canada
Lake
Michigan
Lake
Huron
U.S.
Canada
Lake
Erie
U.S.
Canada
Lake
Ontario
U.S.
Canada
Total
Tributary
195,900
89,500
106,400
598,000
583,400
315,800
267,600
741,100
598,100
143,000
1,725,300
1,365,800
359,500
Direct
Point
Sources
32,600
3,200
29,400
220,400
23,200
15,000
8,200
170,000
168,000
2,000
104,700
23,300
81,400
Input
jj from Upstream
Atmospheric Lake(s)
55,000
82,900
48,700 477,700
20,100 2,759,000
14,000 4,448,100

-------
Wastewater Management Study and other university studies and special state

or federal projects.  A more detailed description of the load calculation

methodology and sources of data may be found in Sonzogni et al. (1978) and

Sullivan et al. (1980).

     U.S. direct point sources inputs were obtained or calculated from

various sources.  Direct inputs to Lake Superior and Lake luron were based
                                                         »*?
on data presented by the Upper Lakes Reference Group (1977).  While this

data represents information from the early to mid 1970's and thus may not

reflect current conditions, inputs (with the possible exception of the

direct point source chloride input to Lake Superior from Canada) are rela-

tively small.  Direct point source inputs to Lake Michigan were based on

recent industrial loads supplied by the U.S.  Environmental Protection

Agency's Great Lakes National Program Office (H. Zar, U.S. EPA Region V,

personal communication, 1980) and direct municipal inputs calculated from

the total annual flow (average of 1975 and 1976 flows) discharged from all

direct municipal dischargers and an assumed chloride effluent concentration

of 160 mg/L.  The use of 160 mg/L was based on the average effluent chloride

concentration for Lake Erie basin municipal treatment plants as determined

by the U.S. Army Corps of Engineer's Lake Erie Wastewater Management Study

(reported in Sonzogni et al., 1978).  Municipal direct point source inputs

to Lake Michigan turned out to be about equal to industrial direct inputs.

U.S. direct point source loads to Lakes Erie and Ontario were estimated

similarly, except that no data on direct industrial chloride inputs were

available.  While these industrial inputs are not likely to be a major com-

ponent of the total chloride budget to these lakes, there are a few indus-

trial operations  (e.g., steel plants) on the lake that could contribute

substantial amounts of chloride and thus should be considered in future work.



                                    10

-------
      Canadian tributary and direct point source loads to Lakes Superior and




 Huron are based on 1973 through 1975 data as reported by the Upper Lakes



 Reference Group (1977).  Canadian total tributary and direct point source



 loads to  Erie and Ontario are from Fraser and Wilson (1981) and Casey and




 Salbach (1974), respectively.



      Atmospheric chloride inputs to each of the lakes were "* ised on Andren
                                                             * /
                                                             s»

 et al.  (1977).  Inputs to Lake Huron from upstream lakes (Lalces Michigan and




 Superior) are from the Upper Lakes Reference Group (1977).  The upstream lake



 contribution to Lake Erie is based on the average annual Detroit River load



 over the  period 1975 through 1978 as supplied by the Michigan Department of



 Natural Resources (J. Hartig, Michigan Dept. of Natural Resources, personal



.communication, 1981).  The upstream (Niagara River) input to Lake Ontario is



 the average of the 1975, 1976 and 1977 annual mean load reported by Chan




 (1979).






 Inputs from Individual U.S. Tributaries



      Chloride loads to each of the Great Lakes from selected major tribu-



 taries are presented in Table 2 for water years 1975 through 1978.  Also



 included  in these tables are the river mouth tributary flows.  These data



 illustrate that large changes in year-to-year chloride loads are often clo-



 sely related to changes in flow.  Finally, the tables contain extrapolated



 total loads and flows for the whole basins.  The methodology used for the




 load determinations were previously described.  Similar information on other



 U.S. tributaries can be found in Sullivan et al. (1980).



      Lake Superior.  The largest U.S. contributor is the St. Louis River,



 which also has the largest flow.  However, the load is high relative to the



 flow, indicating the likelihood of major industrial and municipal chloride




                                     11

-------
                                   Table 2
    Annual Mean Chloride Loads and Flows of Selected U.S. Great Lakes
             Tributaries for Water Years 1975 through 1976
                     Chloride Load (mt/yr)         Flow at Mouth (nrVs)

River             1975    1976    1977    1978   1975   1976   1977 >) 1978
Superior
St. Louis
Nemadji
Ontonagon
Carp
Mineral
Total,
Tributary3

25468
812
3697
1362
21243

92680

14351
486
3381
1764
—

81600

15617
810
2435
922
—

80220

25044
1125
4075
-
—

104060

84.5
12.9
40.5
3.6
—

450.3

47.7
10.3
40.4
4.0
—

379.5

29.9
7.7
29.0
2.3
—

271.9

112.7
15.1
41.9
-
—

566.4
Michigan
Me nominee
Fox
Menomonee
St. Joseph
Kalamazoo
Grand
Manistee
Manistique
Total
Tributary3
3210
51200
10300
78300
60200
171000
163000
4070

775000
3970
55700
10800
86800
57000
150000
85700
3900

711600
5250
36100
5250
68300
48600
95600
74900
2980

490200
7800
53200
-
73100
57000
116000
71800
4200

590300
100.8
118.4
3.3
123.0
68.4
162.4
60.8
61.2

1190.7
95.6
124.2
3.0
143.4
67.1
185.4
66.2
58.3

1276.0
58.8
58.1
1.3
96.1
44.1
73.2
53.3
48.0

732.0
94.6
121.6
3.9
135.9
52.8
102.5
55.5
76.0

1108.4
Huron

  Pine
  Au Sable
  Au Gres
  Saginaw

  Total
  Tributary3
  1100     467     510     689
  9926   10047    6900    7839
  2974    3797    2465    3960
295140  320890  156433  202946
377400  422100  200600  263800   444.6   530.3  232.1  333.6
11.5
59.4
3.5
165.5
9.5
59.5
5.1
216.4
7.7
46.3
1.8
63.0
10.5
48.8
4.3
109.2
                                    12

-------
                            Table 2 (Continued)
                     Chloride Load (mt/yr)         Flow at Mouth  (m^/s)

River            1975    1976    1977    1978   1975   1976   1977    1978
Erie

  Clinton
  Rouge
  Maumee
  Sandusky
  Vermillion
  Cuyahoga
  Chagrin

  Total
  Tributary3

Ontario

  Genesee
  Oswego
  Black
  Oswagatchie
  Raquette

  Total
  Tributary3
17724
273018
46846
4715
110964
21672
44549
161000
25800
16149
132639
20944
56100
73700
123000
29400
5400
71000
15900
                                  17.6
                                 157.0
                                  39.8
                                   8.8
                                  50.5
                                  14.4
        26.7

       165.6
        29.8
         7.1
        39.2
 855600  696900  592200
 13.
 12,
100,
 24,
  7.2
 28.3
        12.5   10.1
608.3  615.0  422.9
 129819  129359  141060  213370   95.1  114.2   94.1
1057788 1386606  965441 1160764  215.7  312.4  214.5
   7548    8606   10050    8537  137.0  194.1  161.9
   4834    7348    5739    5382   75.2  125.5  100.1
   2481    3308    3106    4660   62.8   95.0   75.4
                     126.6
                     305.3
                     168.8
                     112.6
                      91.1
1197900 1607800 1166000 1489200  616.8  873.0  673.4  843.5
alncludea contributions from all U.S.  tributaries,  not  just  those  listed;
    see Sonzogni et al. (1978) and Sullivan et al.  (1980)  for  details.
                                     13

-------
contributions.  The Carp River, which drains a portion of the upper penin-



sula of Michigan, also delivers a high chloride load relative to its flow.



The Carp River receives considerable municipal inputs.  Finally, the aptly



named Mineral River delivers an extremely large chloride load relative  to



its flow.  The large chloride load is apparently caused by the discharge of



brine to the river from local mining operations (Sonzogni et al., 1978).
                                                                            i

                                                                            *t'
     Lake Michigan.  The largest single contributer of chloride was the    **



Grand River, which drains into the eastern part of the lake.  The Grand is



also the largest (in terms of flow) of the Lake Michigan tributaries.   Note



that the Menomonee River, which drains a highly urbanized area within



metropolitan Milwaukee, contributes a large load relative to its flow  (Table



2).  Road salt may have been a major contributer to this load.  The Manistee



River, which drains into eastern Lake Michigan, also contributes a large



amount of chloride relative to its flow.  Discharges to the river from salt



mining operations account for the high load.  The  1975 chloride  load from



the Manistee was particularly high relative to 1976, 1977 and 1978.



Apparently, the reduced annual loads following 1975 reflect abatement  of



some industrial chloride discharges to the river.  Notice that the major



tributaries draining into the lower two-third of Michigan's eastern basin—



namely,  the St. Joseph, Kalamazoo, Grand, Muskegon and Manistee Rivers—



contribute a large proportion of the total tributary chloride load to  the



lake.



     Lake Huron.  Most of the U.S. tributary inputs to Lake Huron is deli-



vered to Saginaw Bay.  The Saginaw River is the principal contributor  to



Saginaw  Bay.  The Au Gres River  (which also drains into Saginaw Bay) also



contributes a large load relative to its flow.  High chloride loads to
                                     14

-------
Saginaw Bay reflect the influence of municipal and industrial sources.  Most




(over 90 percent) of the total U.S. wastewater flow to Lake Huron tribu-




taries is delivered to streams draining into Saginaw Bay.  Chemical




industries and brine wells are also important sources of chloride to Sagint.,




Bay.




     Lake Erie.  In almost all cases, U.S. tributary chloride loads to Lake




Erie tend to be high relative to flows (compared to tributaries with less




developed watersheds).  The Mauraee River, the second largest U.S. tributary




to the Great Lakes, ranks second behind Lake Ontario's Oswego River as the




largest contributer (among tributaries) of chloride to the Great Lakes.




Steel and manufacturing, as well as chemical industries, contribute to high




chloride loads to streams draining the Detroit metropolitan area, such as




the Clinton and Rouge Rivers.  The Cuyahoga River, which drains the




Cleveland.area, also contributes a high chloride load relative to its flow.




Overall, however, the largest input to Lake Erie, as seen from Table 1, is




the input from the channel connecting it with Lake Huron, i.e., the Detroit




River.




     Lake Ontario.  The tributary chloride load to Lake Ontario is dominated




by the Oswego River (Table 2), the largest contributer of chloride of all




the U.S. tributaries.  While  the Oswego has the largest historical annual




mean flow of any U.S. tributary, its chloride load is also high relative to




its flow.  As discussed previously, a chlor-alkali plant on Onondaga Lake,




which drains into the Oswego, is a major source of chloride to the Oswego.




Overall, about 50 percent of  the Oswego's chloride load has been attributed




to point sources (Sonzogni et al., 1978).  Despite the importance of the
                                    15

-------
Oswego as a chloride source, the largest overall input to Lake Ontario  is




the flow from the upper lakes through the Niagara River.








Deicing Salt Useage Versus Loads




     While it is not possible to directly assess the amount of road  salt




that reaches the Great Lakes, some insight on the important of road  salts




can be obtained by comparing application rates with the  total chloride  loads




given in Table 3.  Based on PLUARG (1977), the amount of road salt applied




annually to basins of Superior, Michigan, Huron, Erie and Ontario  total




88,000, 595,000, 370,000, 696,000, and 1,046,000 metric  tons, respectively.




Accordingly, even if all applied reached the lakes (which is not the case),




road salt would account- for less than 35 percent of the  load for all lakes




except Michigan.  In the case of Lake Michigan, the higher ratio of  road




salt applied to total load is perhaps the result of an overestimate  of  the




road salt applied.  Apparently, in determining salt useage for Lake




Michigan, PLUARG (1977) estimates were made on a county  basis.  However,




because large portions of counties adjacent to southern  Lake Michigan (these




counties are heavily populated and presumedly have high  salt useage) do not




drain into the lake, salt useage in  the Michigan drainage basin may  have




been overestimated.




     Thus, it appears that road salt contributes a significant, but  not




necessarily a predominant, portion of the anthropogenic  chloride load to




Great Lakes.  This conclusion is consistent with the results of a  study of




chloride inputs from the City of Buffalo (Meredith and Rumer, 1976), which




indicated that road salts were the source of 36 percent  of the chloride




leaving Buffalo's combined sewer system.
                                     16

-------
                                  Table 3

       Comparison of Estimates of Chloride Loads (metric tons/year)

                            to the Great Lakes
Lake
 Estimated Load for 1960
(O'Connor & Mueller,  1970)
  Estimated Load for
the mid to late 1970'i
    (Present Study)
Superior

Michigan

Huron

Erie

Ontario
          230,000

          927,000

        1,417,000

        4,619,000

        5,944,000
        283,500

        901,300

      1,133,000

      3,690,200

      6,292,100
                                    17

-------
Comparison of Present Chloride Load Estimates With Previous Estimates




     In their now classic water quality modeling paper, O'Connor and Mueller




(1970) estimated 1960 chloride loads for the entire Great Lakes system.




Their estimated chloride loads .to each of the Great Lakes are compared with




the loads from this study (mid to late 1970's) in Table 3.  O'Connor and




Mueller's chloride load includes an "other sources" category, which was




determined by difference (so that inputs balanced storage and outputs).




These "other sources" comprised 22, 23, 48, 17 and 21 percent of O'Connor




and Mueller's loads presented in Table 3 for Lakes Superior, Michigan,




Huron, Erie and Ontario, respectively.  Thus, were it not__£or_ their— "other




sources," O'Connor and Mueller's loads would be considerably less  than those




of this study for all lakes except Erie.




     Ownbey and Kee (1976) presented an assessment of chloride loads from




individual tributaries to Lake Erie.  Their results are compared to estima-




tes from this study in Table 4.  Note that in most cases current estimates




are greater than Ownbey and Kee's.




     The estimates of both O'Connor and Muller (1970) and Ownbey and Kee




(1967) were necessarily based on scant data.  In many cases, loads were




derived from unit area loads or per capita inputs rather than actual




measurements.  Loads from the present study are based on better information




and likely are more accurate.  For example, loads to Lake Erie were based  on




extensive monitoring of streams, especially during the high flow events when




a large portion of the total load may enter the lakes, as part of  the Lake




Erie Wastewater Management Study (U.S. Army Corps of Engineers, 1979).
                                     18

-------
                                 Table 4
                  of
                               to Lake Erie
_ ""
	 '-
— — — — — — — — — ^^^^^
Detroit River
Hur on
Raisin
Moumee
Portage
Sandusky
Black
Rocky
Cuyahoga
Chagrin
Grant " «>hio>
Ash tabula
Conneant
U.S. Direct Municipal
__ 	 — 	 • 	
__
Ownbey and Kee (1967)
._ 	 . 	 • 	 • 	
	 — 	 • 	
2,996,400
14,982
20,884
118,040
5,448
29,510
7,264
19,068
72,640
4,540
635,600
9,080
2,724
Discharges 49,940
__— 	 	 	 	
Present Study*
•••••^
2,759,000
25,650
29,687
185,673
11,542
34,015
16,737
10,800
104,701
19,339

4,553
4,498
143,000
• 	 • —
centration of 160 mg/L chloride.
                                    19

-------
      Long-Tenn Trends in Chloride Loads and In-Lake Concentrations




     While not obvious from the comparison of current loads with 1960 loads




(Table 3), there is evidence that some chloride inputs have decreased, espe-




cially industrial inputs.  As early as 1972, considerable progress was made




in controlling certain Michigan point sources of chloride.  For instance,




the U.S. Environmental Protection Agency (1972) reported that a 40 percent




reduction in the chloride discharge of several major industries would occur




between 1971 and January 1973.  As mentioned previously, a significant




decrease in the chloride load of Michigan's Manistee River, which received




chloride from major salt producers, is attributed to abatement measures




(Little, G., Michigan Department of Natural Resources, personal




communication).  Perhaps the best example of decreasing chloride loads is




the Detroit River contribution to Lake Erie.  Figure 1 shows how the




chloride load has steadily decreased since the late 1960's.  This decrease




is most likely the result of reduced industrial chloride discharges.




Further, comparing the concentration of chlorides at the head of the Detroit




River with those at the mouth (Table 5) suggests that the reduced Detroit




River load is the result of decreased inputs from the Detroit-Windsor




complex.  While the concentration of chloride at the head of the Detroit




River has changed little over the period of record, the concentration at




the Detroit River mouth has progressively decreased.  This indicates that




the observed decrease is probably the result of reduced chloride discharges,




most likely from industrial sources from the Detroit-Windsor area.




     Table 6 shows typical average concentrations (open lake) that were




measured during the early 1900's, the 1960's and 1970's.  While con-




centrations have apparently increased significantly since 1900, the changes
                                    20

-------
                                  Table 5




         Comparison of Annual Mean Chloride Concentrations (mg/L)




               at the Head and Mouth of the Detroit River3
Water
Year
1967
1968
1969
1970
1971
1972
1973
1974
1975
Head
9
10
11
10
9
9
. 9
9
9
Mouth
-
23
18
18
15
17
14
16 •
15
aData from Great Lakes Water Quality Board (1976)
                                    21

-------
                                  Table 6

          Changes in Great Lakes Chloride Concentrations (mg/L)

               (values from Beeton, 1969, except as noted)

Lake Superior
Lake Michigan
Lake Huron
Lake Erie
Lake Ontario
1900' s
1.2
3
5
11
10
1960 's CurrcuL
1 la
6-7 7.7b
7 5.5a
26C 20d
27xr~ 27. 7e
aUpper Lakes Reference Group (1977)

bRockwell et al (1981)

        and Chawla (1969)
^Rockwell, D.C., U.S. EPA, Great Lakes National Program Office, Chicago
     personal communication, 1981)

eSimons (1979)
                                    22

-------
between the 1960's, when chloride inputs from industrial sources were likely




at their peak, and the present are not as obvious.  In fact, current con-




centrations of chloride in Lake Erie appear to be less than reported for the




I960's.  This may be an indication of decreased chloride loads to Lake Erie.




Further, because Lake Erie is much shallower than the other lakes it flushes




more quickly.  Hence, a decrease in chloride concentration as a result of




decreased chloride loads would be noticed in Lake Erie much faster  than in




the other Great Lakes.




     Accordingly, it would appear that, at least in some locations,




industrial chloride inputs are decreasing.  On the other hand, new




industrial operations, such as new plants or new treatment technologies,




could increase inputs.  For example, a new wastewater treatment plant for




the steel industry, located near Gary, Indiana, is expected to increase its




chloride input to Lake Michigan by 49,800 metric tons per year (University




of Michigan Research News, 1978).  Further, as mentioned previously, munici-




palities may be discharging more chloride than previously with the  use of




metal salts for phosphorus removal.  More information is needed on  indivi-




dual point source contributions to evaluate overall trends of chloride




inputs  to the Great Lakes.








                    Chloride Model of the Great Lakes




     Several investigators have presented models relating chloride  inputs to




chloride concentrations in the Great Lakes.  The chloride model of  O'Connor




and Mueller  (1970) is particularly noteworthy, as it considered the Great




Lakes as an integrated system.  Snow (1974) applied O'Connor and Mueller's




approach specifically to Lake Michigan.  Meredith et al.  (1974) and
                                     23

-------
Richardson (1974) applied the same basic model approach to Lake Erie and




Saginaw Bay, respectively.  Other modeling attempts  similar  to O'Connor and




Mueller's include those of Rainy (1967), Dingman and Johnson (1971), and




Butler et al. (1974).




     A chloride simulation (CS) model  similar to that  developed by O'Connor




and Mueller (1970) is used here to reevaluate predictions of future chloride




concentrations using current loading information.  Importantly, the model




considers the effect of changes to one lake  on another by treating the lakes




as five completely mixed systems in series.  The model is used to examine




Lake Michigan in detail because of current concern over the  effects of




chlorides and other dissolved solids in the  lake.






Model Formulation




     The CS model is based on a simple chloride mass balance which reflects




the conservative behavior of chloride.  The  change in  chloride concentration




as a function of time is thus represented as the sum of inputs minus the




outputs, expressed mathematically as:









                              V -j| - ZW - QC                            (1)









where,




           C « in-lake chloride concentration




           V » lake volume




          EW » sum of all chloride loads, including  those from upstream




               lakes




           Q » flow out of the lake




           t - time






                                    24

-------
Equation 1 can be solved analytically, yielding:
                                    -St.

                      C(r) - SL (l-e  M
                      ^^t; " TT— ^i e    ;
where,



    C(t) « Chloride concentration at time t



      Co • Initial chloride concentration, when t = 0






Equation (2) also shows that following a change in the chloride  loading,



equilibrium or steady state conditions are approached exponentially.  At


                          EW
steady-state (t * <»), C - ^— .  That is, at steady-state  the  in-lake con-




centration equals the average concentration of the inflow.  Approximately




95% of  the steady-state concentration is obtained within  three hydraulic


                                                          V
residence "times, the hydraulic residence time defined as  Q-  •  For  detail on



basic model formulation, including elaboration on treating  the lakes in




series, see O'Connor and Mueller  (1970) and Chapra and Sonzogni  (1979).






Model Inputs



     Chloride inputs for the CS model were those summarized in Table 3.  For



lake volumes and flows, values reported in O'Connor and Mueller  (1970) were



used except in the case of Lake Michigan where data summarized in  Rodgers




and Salisbury (1981) was used.  Lake Michigan hydrology is  complicated by



the difficulty of measuring flows out through the Straits of  Mackinac.



Consequently, outflows from Lake Michigan must be calculated  indirectly




using a water budget procedure (Quinn, 1977).



     Table 7.compares water budgets from three sources, including  the Rodgers



and Salisbury (1981) budget used  here.  The tributary discharge  in Rodgers






                                    25

-------
                                 Table 7

                     Water Budgets for Lake Michigan

                                 (m3/7r)


Tributary Discharge
Net Precipitation
(Preceiptation-Evaporation)
Infiltration
Storage
0' Conner
and
Mueller
(1970) Quinn (1977)a
34.8 x 109 33.0 x 109
14.3 x 109 6.3 x 109
_
0 0.4 x 109
Rodgers
and
Salisbury
(1981)
32.3 x 109
12.5 x 109
0.4 x lo'
0
Outflow through Chicago     2.8 x  10'
Diversion Canal
                                    c
Outflow through            46.3 x  10"
Straits of -Mackinaw

Volume (km3)                    4877

Water Residence Time (yrs)      99
 2.9 x ID'
2.9 x 10'
36.0 x 109(70.8xl09b) 42.3 x 109
     4915
     126 (67°)
    4976

    110
aAverage over 1950-1966

"Assumes return flow during  stratified period
                                    26

-------
and Salisbury is based on long-term historical flows recently reported by


Sonzogni et al. (1979) and Sullivan et al. (1980).


     Note in Table 7 that Quinn  (1977) reports two possible hydraulic resi-


dence times.  The shorter estimate, 67 years, incorporates a return flow


through the straits during summer  stratification  (based on limited measure


ments of currents in the straits).  The inclusion of the deep layer return


flow when calculating the hydraulic residence time or as a source of loading


presupposes that these waters are  completely mixed throughout Lake Michigan.


There is some evidence that  this is not case, and that the return flow


creates only a localized cell of mixing.  For instance, Moll et al.  (1976),
using cluster analysis of chemical and biological  parameters, were able  to


identify Lake Michigan waters  in a plume  extending into Lake Huron, but  Lake


Huron waters could not be located west of Bois Blanc  Island  (Lake Huron).
                                                               •
Chloride data from Lake Michigan during 1962-63 and 1976, as reported  in


Rockwell et al. (1980), indicates a  sharp gradient in the area adjacent  to


the straits as compared to  the rest  of Lake Michigan.  Consequently,
                                             %

hydraulic  residence  times on  the order of 100 years appear  to be most  repre-


sentative  for chloride modeling purposes.



Model Results


     Figure 2 depicts the chloride concentrations  changes in each of the


Great Lakes will change over  time given that current  loads  to the Great


Lakes System (Table  1) remain the same.   Projected long-term steady-state


concentrations range from about 4 mg/L in Lake Superior to  about 30 mg/L in


Lake Ontario.  The greatest chloride concentration change is expected  to


occur in Lake Michigan, where the current level of 7.7 mg/L is projected to


rise to nearly 20 mg/L over the next 300  years.



                                     27

-------
     Figure 2 also indicates a relatively small rise of chloride in Lake




Erie.  This increase is mostly in response to the gradual  build-up of




chloride in the upper lakes, which drain into Lake Erie.   The  initial dip  in




concentration in Lake Ontario in Figure 2 reflects a lagged  response to




changes in chloride concentrations in Lake Erie.  Both Lakes Erie and




Ontario are presently close to being in equilibrium with their loads, and  in




fact the present Lake Ontario chloride concentration appears to still be




equilibrating to decreasing Lake Erie chloride concentrations  during the




recent past.  Thus, assuming current loads to the system remain the same,




chloride in the lower lakes will increase slightly as a result of the




buildup of chloride in  the upper lakes according to the CS model.




     Table 8 shows the  reductions in external chloride loads that would be"




needed to maintain current chloride levels in the lakes (initial,conditions




in Figure 2).  Reductions assume that upstream lakes maintain  current con-




ditions.  For instance, to maintain a concentration of 5.5 mg/L, Lake Huron




would require a 32 percent reduction in its  external chloride  load provided




Lake Michigan and Lake  Superior remained at  their present  concentrations of




1.0 and 7.7 mg/L, respectively.  Note that to maintain present levels in




Lake Michigan, a large  reduction in the current  load will  be required.




Also, while Table 8 shows a large percent reduction in the chloride load to




Lake Superior would be  required to maintain  the  very low current chloride




concentration of 1 mg/L, the actual load reduction required  is relatively




small.




     Examining Lake Michigan in more detail, Figure 3  illustrates a range  of




possible equilibrium chloride concentrations bounded by upper  and lower esti-




mates of the current chloride load and hydraulic residence times.  A + 25
                                     28

-------
                                Table 8

              Reduction in External Chloride Load Required

                    to Maintain Current Concentrations
                              % of present                   MT/YR
Lake                          External Load                Reduction
Superior                          77                        218,400

Michigan                          60                        540,800

Huron                             32                        209,700

Erie                               8                        226,200

Ontario                            0                              0
                                    29

-------
percent range for the chloride load was  chosen, reflecting an approximate




error bounds in chloride loading estimates.  Ranges  in hydraulic  residence




times were deduced from Table 7, and  reflect the uncertainty in outflow  from




Lake Michigan.




     Notice in Figure 3 that to achieve  an equilibrium concentration belov?




the current concentration,  the current load to Lake  Michigan would  have  to




be reduced by a factor of 2 or more even under the most  optimistic




(shortest) hydraulic residence time.  From a Lake Michigan management




perspective, then, it is clear that Lake Michigan chloride concentrations




are likely to continue to build-up, even under conservative estimates  of




current loading and hydraulic residence  times.  Subsequently, Lake




Michigan's chloride build-up will  affect Lakes Huron, Erie and Ontario.




     A relevant question is whether the  response time to a Lake Michigan




load reduction could be shortened.  For  example, what chloride load reduc-




tion to Lake Michigan could achieve a desired in-lake chloride concentration




in 30 years as opposed to the normal  response time of 300 years.  In order




to reduce the time required to reach  a target concentration, it can be shown




from a reformulation of equation 2 that  the chloride load would have to  be




reduced to a greater extent than if the  time of response was not  a  cri-




terion.  In other  words, to achieve  a 50 percent reduction in the  chloride




concentration of Lake Michigan at  an  accelerated pace, the load reduction




would necessarily have to be greater  than 50 percent until the target  con-




centration was obtained.  Accordingly, in lakes with long hydraulic resi-




dence times such as the Great Lakes,  the long-term consequence of allowing




conservative substances to  exceed  prudent limits should  be realized in




advance and long-term planning made accordingly.  Managers should therefore
                                     30

-------
realize that recovery periods for  conservative ions will be more prolonged




or require accentuated  levels of treatment as compared  to nutrients which




are subject to losses other  than natural  flushing  (i.e., sedimentation).








Management Considerations




     The CS model results  (Figure  2)  indicate that the  chloride con-




centration in Lake Michigan, and to a lesser extent Lake Huron, may be




expected to increase most  dramatically in the future.   For Lake Michigan,




concentrations are expected  to  increase to about 20 mg/L, an  increase of




greater than 6 fold compared to concentrations at  the turn of the  century.




A salient management question,  then,  is whether the Lake Michigan  ecosystem




will be seriously degraded by the  chloride change.




     Storemer (1978) provides evidence that the less desirable phytoplankton




species that have invaded  the Great Lakes tend to  come, virtually  exclusi-




vely, from saline waters.  The  filamentous blue green alga, Stephanodiscus,




which has been known to decrease filtration time at water treatment plants




as well as cause  taste  and odor problems, has already been observed to  be




increasingly more prevalent  in  southern Lake Michigan (where  industrial




chloride discharges are high).  The  rapid spread through the  Great Lakes of




the marine alga, Bangia atropurpurea, is  believed  to be linked to  increased




chloride levels or increases in dissolved solids.  Eureytemora affinis, a




brackish water copepod, is also now  established in the  Great  Lakes.




Stoermer (as noted in Great  Lakes  Water Quality Board,  1977)  has further




hypothesized that a biological  breakpoint between  7.5 and 10  mg/L  chloride




may exist for Lake Michigan. Beyond this concentration, a major shift  in




phytoplankton toward nuisance  taste  and odor causing blue green algae could




occur.






                                     31

-------
     There is also some evidence  that,  in general, productivity increases




with increases in chloride or  total  dissolved  solids  (Rawson,  1951, 1960;




Northcote and Larkin, 1956; Kerekes  and Nursall,  1966;  Seenayga,  1973).




Robertson and Powers (1967) reported that total organic matter in the Great




Lakes increased in the order of Superior, Huron,  Michigan, Erie and Ontario*




Average chloride concentrations also increase  in  the  same order.   Stoermer




(1978) has thus questioned whether nutrient  control alone will provide




desired improvements (less eutrophic)  in Great Lakes  water quality, which is




often measured in terms of the assembleges of  organisms.  However, as




discussed by Sorrenson et al.  (1978) and Stoermer (1978), the  effects of




chlorides or other dissolved solids  are subtle, are often confounded by




other factors (e.g., nutrient  enrichment) and  are difficult  to directly eva-




luate experimentally.




     Despite the difficulty in directly assessing whether increasing




chloride levels will seriously impair Lake Michigan,  some inferences may be




obtained from other lakes with higher chloride levels.  For  example, Lake




Ontario's average chloride concentration currently is about  four  times




higher than found in Lake Michigan.   Lake Ontario's Irondequoit Bay has




chloride concentrations several times Lake Ontario's  and Onondaga Lake has




even higher chloride levels.




     Undoubtedly, the Lake Ontario ecosystem has  been severely disturbed.




Stoermer et al.  (1975) reported that Lake Ontario's phytoplankton assemblage




is dominated by  species indicative of degraded water  quality,  including spe-




cies of blue-green algae  that  are potential  nuisances.  They note that many




of the taxa commonly found in  the offshore waters of  the more  oligotrophic




upper Great Lakes are absent or rare in Lake Ontario.  Finally,  they report
                                     32

-------
a shrinking abundance of halophilic species in the lake,  such  that  the  domi-




nant and sub-dominant taxa are more commonly found in  brackish and  saline




inland waters.




     Although these changes indicate degraded water quality, it is  not  clear




what impact this has had on uses of the  lake.  Little  information exists =» *,




the economic impacts of these changes or how human perception  of the  quality




of water has changes.  Lake Ontario is supporting a growing  sport fishery




and is still a major source or drinking  water.  Thus,  while  Lake Ontario




has undergone extensive ecological changes, the ecosystem still serves  many




uses.  Accordingly, it is difficult to quantify the practical  impact  that




would result should some of the changes  observed in Lake  Ontario become




manifested in Lake Michigan.




     If an effect of chloride or increased  dissolved  solids  were to be




obvious, it should be so in Onondaga Lake and Irondequoit Bay, where




chloride levels have gotten "extremely high.  Interestingly,  the high  sali-




nity of Onondaga Lake has not resulted in significant  amounts  of non-fresh




water organisms at any trophic level (Effler et al.,  1981).  Irondequoit Bay




was reported to have phytoplankton species  and numbers similar to Onondaga




Lake.  Both of these waters are productive, however,  and  receive large




nutrient inputs.




     In summary, analogies to other lakes where chloride  levels have




increased provides conflicting inferences as to the future of  Lake  Michigan.




It may well be that the influence of conservative  ions is secondary to




nutrient enrichment.  For example, while higher conservative ion levels




may provide a competitive advantage for  halophilic species,  they may  not




flourish without abundant nutrient supplies.  In other words,  increased
                                     33

-------
conservative ion levels may be a factor ecologically, but of  lesser  impor-




tance than other factors such as nutrients.




     The occurrence of halophilic organisms  in the Great Lakes as  opposed to




other inland lakes with high chloride  levels may also be related to  the




direct linkage of the Great Lakes with the sea.  With the construction of




St. Lawrence Seaway system, salt water organisms have had access to  all  the




Great Lakes.  The invasion of non-native  fish in this manner  (for  example,




the sea lamprey and alewife) is well documented.  Further,  ocean going




vessels using the Great Lakes often dump  salt water  ballast into the lakes.




     The uncertainty of the effect of  conservative ions exemplifies  the




need to carefully monitor how the Great Lakes respond to the  massive munici-




pal point source phosphorus control program  enacted  during  the 1970's.   As




discussed in Heidtke et al. (1979), future phosphorus loads to the Great




Lakes from municipal point sources should be reduced over 50  percent (750




metric tons/year) compared to mid-1970 levels.  The  extent  to which  the




lakes actually respond to this reduction  should provide valuable insight




into the importance of conservative ions  as  a pollutant and the potential




need for controlling chloride inputs to  the  Great Lakes, especially  Lake




Michigan.
                                     34

-------
                                REFERENCES






Andren, A., Eisenreich, S., Elder, F., Murphy, T., Sanderson, M. and




     Vet, R. J. (1977).  Atmospheric Loadings to the Great Lakes, Polluti




     from Land Use Activities Reference Group Technical Report,




     International Joint Commission, Great Lakes Regional Office, Windaor,




     Ontario, Canada.






Bannister, T. T. and Bubeck, R.C. (1978).  Limnology of Irondequoit Bay,




     Monroe County, New York, in Lakes of New York State, J. A.  Bloomfield,




     Ed., Academic Press, N.Y., 106-214.






Beeton, A. M. (1965).  Eutrophication of the St. Lawrence Great Lakes,




     Limnol. Oceanogr. 10:240-254.






Beeton, A. M. (1969).  Changes in the Environment and Biota of the Great




     Lakes, in Eutrophication;  Causes, Consequences and Correctives, Nat.




     Acad. Sci., Wash., D.C., p. 150-187.






Beeton, A., Becker, M., Bulkley, J., Cowden, J., Fetterolf, C., Libby, L.,




     Shannon, E. and Moll, R. (1980).  Report of Great Lakes Region




     Conference on Marine Pollution Problems, Working Paper No. 3:  Federal




     Plan for Ocean Pollution Research, Development and Monitoring, FY




     1981-1985, National Oceanic and Atmospheric Administration, Rockville,




     Md.






Bubeck,  R. C., Dimient, W. H., Deck, B. L., Baldwin, A and Lipton, S. D.




     (1971)  "Runoff  of De-icing Salt:  Effect on  Irondequoit Bay, Rochester,




     New York, Science 172:112-1132.
                                     35

-------
Butler, K. S., Gates, W. A. and McCown, B. H. (1974).  A Pollution




     Displacement Model of the Great Lakes System, Proceedings of the 1974




     Winter Simulation Conference, Rockefeller Foundation, Washington, D.C.,




     p. 172-175.






Casey, D. T. and S. E. Salbach (1974).  IFYGL Stream Materials Balance




     Study, Proc. 17th Conf. Great Lakes Res., 668-681.






Chan, C. H. (1979).  Niagara River Chemical Loading, 1975-1977, Scientific




     Series No. 106, Inland Waters Directorate, Burlington, Ontario, Canada.






Chapra, S. C. and__Son_20ignit__W.» C. (1979).  Great Lakes Total Phosphorus




     Budget for the Mid 1970's, JWPCF 51:2524-2533.






Cummings, T. Ray. (1980).  Chemical and Physical Characteristics of Natural




     Ground Waters in Michigan:  A Preliminary Report, U. S.  Geological




     Survey Open-File Report 80-953, Lansing, Michigan, 34 p.






Dingman, S. L. and Johnson, A. H. (19-71).  Pollution Potential of Some New




     Hampshire Lakes, Water Resour. Res. 7:1208-1215.






Effler, S. W., Field, S. D., Meyer, N.A. and Sze, P. (1981).  Response on




     Onondaga Lake to Restoration Efforts, J. Environ. Eng. Div., Am. Soc.




     Civ. Eng. 107:191-210.






Fraser, A. S. and Wilson, K. E.  (1981).  Loading Estimates to Lake Erie




     (1967-1976), National Water Research Institute, Canada Center for




     Inland Waters, Burlington, Ontario, Canada.
                                     36

-------
Great Lakes Water Quality Board (1976).  Appendix B, Surveillance




     Subcommittee Report, International Joint Commission, Great Lakes




     Regional Office, Windsor, Ontario, Canada






Great Lakes Water Quality Board (1977).  Appendix B, Surveillance




     Subcommittee Report, ibid.






Heidtke, T. M., Monteith, T. J., Sullivan, R. A., Scheflow, D. J., Skimin,




     W. E. and Sonzogni, W. C. (1979).  Future U.S. Phosphorus Loadings to




     the Great Lakes:  An Integration of Water Quality Management Planning




     Information, Great Lakes Environmental Planning Study Contribution No.




     11, Great Lakes Basin Commission, Ann Arbor, Michigan.






Kenaga,- D. (1978) Chlorides in Lake Michigan, Report of  the Water Quality




     Division, Michigan Department of Natural Resources, Lansing, Michigan.






Kerekes, J. and Nursall, J. R. (1966).  Eutrophication and Senescence in a




     Group of Prairie - Parkland Lakes in Alberta, Canada, Verh. Internet.




     Verein. Limnol., 16:65-73.






Meredith, D. D., Rumer, R. R., Chien, C. C. and Apmann,  R. P. (1974).




     Chlorides in Lake Erie Basin, Water Resources and Environmental




     Engineering Research Report No. 74-1, Department of Civil Engineering,




     State University of New York at Buffalo.






Meredith, D. D. and  Rumer, R. R. (1976).  Chloride Management in Lake Erie




     Basin, Water Resources and Environmental Engineering Research Report




     No. 76-2, Department of  Civil Engineering, State University of New York




     at Buffalo.
                                     37

-------
Metcalf and Eddy, Inc. (1972).  Wastewater Engineering, McGraw-Hill, New York,




     782p.






Moll, R. A., Schelske, C. L. and Simmons, M. S. (1976).  Distribution of Water




     Masses in and near the Straits of Mackinac, J. Great Lakes Res.




     2:43-59.






Monteith, T. J. and W. C. Sonzogni (1976).  United States Great Lakes




     Shoreline Erosion Loadings, Pollution from Land Use Activities




     Reference Group Technical Report, International Joint Commission, Great




     Lakes Regional Office, Windsor, Ontario, Canada.






Monteith, T. J., Sullivan, R. A. and Sonzogni, W. C. (1980).  Phosphorus




     Control Strategies at Municipal Wastewater Treatment Plants in the U.S.




     Great Lakes Basfn, Great Lakes Environmental Planning Study




     Contribution No. 14, Great Lakes Basin Commission, Ann Arbor, Michigan.






Northcote, T. C. and Larkin, P. A.  (1956).  .Indices of Productivity in




     British Columbia Lakes, J. Fish. Res. Board Can.  15:1515-540.






0'Conner, D. J.  and Mueller, J. A.  (1970).  A Water Quality Model of




     Chlorides in Great Lakes, J. San. Eng. Div., Am.  Soc. Cer. Eng.




     96:955-975.






Ownbey,  C. R. and Kee D.  A.  (1967).  Chloride in Lake  Erie, Proc. 10th Conf.




     Great Lakes Res., 382-389.






PLUARG  (1977).   "Land Use and Land  Use Practices in  the Great Lakes Basin",




     Task B  Joint Summary Report  (United States  and Canada), International




     Joint  Commission, Great Lakes  Regional Office, Windsor, Ontario.






                                     38

-------
PLUARG (1978).  Environmental Management Strategy for the Great Lakes




     System, Final Report, International Reference Group for Great Lakes




     Pollution from Land Use Activities, International Joint Commission,




     Great Lakes Regional Office, Windsor, Ontario, Canada.






Qi inn, F. H. (1977).  Annual and Seasonal Flow Variations Through the




     Stratis of Mackinac, Water Resour. Res., 13, 137.






Quinn, F. (1978).  Hydrologic Response Model of the North American Great




     Lakes, Hydrology. 37:295-307.






Rainy, R. H. (1967).  Natural Displacement of Pollution from the Great




     Lakes, Science 155:1242-1243.






Rawson, D. S.  (1951).  The Total Mineral Content of Lake Waters, Ecology




     32:669-672.                              .               ,  -






Rawson, D. S.  (1960).  A Limnological Comparison of Twelve Large Lakes in




     Northern  Saskatchewan, Limnol. Oceanogr. 5:195-211.






Richardson, W. L.  (1975).  Modeling Chloride Distribution  in Saginaw Bay,




     Proc. 17th Conf. Great Lakes Res., Part I, 462-470.






Robertson, A.  and  Powers, C. F.  (1967).  Comparison of  the Distribution of




     Organic Matter in the Great Lakes, Special Report  No. 30, Great Lakes




     Res. Div., Univ. of Michigan, Ann Arbor, 18 p.






Rockwell, D.   ., DeVault, D. S.  Ill, Palmer, M. F., Marion, C. V.  and




     Bowden, R. J.  (1980).  Lake Michigan Intensive Survey, 1976-1977, Great




     Lakes National Program Office, U. S. Environmental Protection Agency,




     Chicago,  Illinois.






                                    39

-------
Rodgers, P. and Salisbury, D. (1981).  Modeling of Water Quality in Lake




     Michigan and the Effect of the Anomalous Ice Cover of 1976-1977,  Great




     Lakes Environmental Planning Study Contribution No. 44, Great Lakes




     Basin Commission, Ann Arbor, Michigan, 53 p.






Schr  ifnagel, F. H. (1965).  Chlorides, Wisconsin Committee on Water




     Pollution, Wisconsin Department of Natural Resources, Madison, 20 p.






Seenayga, G. (1973).  Ecological Studies in the Plankton of Certain




     Freshwater Ponds of Hyderabad - India, Hydrobiologia 41:529-540.






Simons, T. J.  (1979).  Assessment of Water Quality Simulation Capability for




     Lake Ontario, Scientific Series No. Ill, Inland Waters Directorate,




     Canada Center for Inland Waters, Burlington, Ontario, Canada






Snow, R. H. (1974).  Water Pollution Investigation:  Calumet Area of Lake.




     Michigan, U. S. Environmental Protection Agency Report No.




     EPA-905/9-74-011-A, Chicago, Illinois.






Sonzogni, W. C., Monteith, T. J., Bach, W. N. and Hughes, V. G. (1978).




     United States Great Lakes Tributary Loadings, Pollution from Land Use




     Activities Technical Report, International Joint Commission, Great




     Lakes Regional Office, Windsor, Ontario, Canada






Sonzogni, W. C., Monteith, T. J., Skimin, W. E. and Chapra, S. C.  (1979)




     Critical  Assessment of U. S. Land Derived Pollutant Loadings to the




     Great Lakes, Pollution from Land Use Activities Technical Report,




     International Joint Commission, Great Lakes Regional Office, Windsor,




     Ontario,  Canada.
                                     40

-------
Sorrenson, D. L., McCarthy, M. M., Middlebrooks,  E.  J.  and Porcella,  D.  B.




     (1977) Suspended and Dissolved Solids Effects on Freshwater Biota;  A




     Review, U.S. Environmental Protection Agency Report No.  600/3-77-042,




     Washington, D.C. 65p.






Stoermer, E. F., Bowman, M. M., Kingston, J. C.,  and Schaedel,  A. L.




     (1975).  Phytoplankton Composition and Abundance in Lake Ontario during




     IFYGL, U. S. Environmental Protection Agency Report No.




     EPA-660/3-75-004, Washington, D. C. 373 p.






Stoermer, E. F. (1978).  Phytoplankton Assemblages as Indicators of Water
     Quality in the Laurentian Great Lakes, Trans. Amer. Micros Soc.




     97:2-16.






Sullivan, R. A., Monteith, T. J. aqd Sonzogni, W. C. (1980).  Post-PLUARG




     Evaluation of Great Lakes Water Quality Management Studies and




     Programs, U.S. Environmental Protection Agency, Great Lakes National




     Program Office, Report No. EPA-905/9-80-C06-B, Chicago, Illinois.






Tiffany, M. A., Winchester, J. W. and R. H. Loucks (1969).  Natural and




     Pollution Sources of Iodine, Bromine, and Chlorine in the Great Lakes,




     JWPCF 41:1319 - 1329.






U.S. Environmental Protection Agency (1972).  Report to the Lake Michigan




     Enforcement Conference on Chloride, U.S. Environmental Protection




     Agency Region V, Chicago, Illinois






U.S. Army Corps of Engineers  (1979).  Lake Erie Wastewater Management Study




     Methodology Report, Buffalo District, N.Y., 146 p.
                                    41

-------
University of Michigan Research News (1978).  U.S.  Steel's New Wastewater




     Treatment Plant and the Impact of Increased Dissolved Solids  in Lake




     Michigan, XXIX, Nos 8/9, p. 14-19.






Upchurch, S. B. (1976).  Chemical Characteristics of the Great Lakes,  Great




     Lakes Basin Framework Study, Appendix 4,  Great Lakes Basin Commission,




     Ann Arbor, Michigan 151-238.






Upper Lakes Reference Group (1977).  The Waters of Lakes Huron and Superior,




     Report to the International Joint Commission,  Great Lakes Regional




     Office, Windsor, Ontario, Canada.






Weiler, R. R. and Chawla, V. K. (1969).  Dissolved Mineral Quality of Great




     Lakes Waters, Proc. 12th Conf."Great Lakes Res., 801-818.
                                    42

-------
                              List of Figures
Figure 1.  Annual mean chloride loads (and their standard deviation) to




           Lake Erie from the Detroit River.




Figure 2.  Projections of chloride concentrations over time in response




           to current external loads.




Figure 3.  Lake Michigan chloride load versus equilibrium chloride




           concentration for a range of possible hydraulic detention




           times.
                                    43

-------
   C  U
   O  C 03
•  -H    O   a >-i
   •o  a; 3
      Q co
   •o    to
   h  C ,
   tfl  JS r-l
   u  o js
   CO  -H U
      s c

   £~8
     > O
      1-1
   *O  OS T3
   C    0)
   (QUO]
   >—• >H CO
      o ua
   W  fci    >
   13  u «  -H
   ca  eu u  iJ
   O  Q to
   •J    -o
      0)
   0)  j: en
   -o  .u -l

                                                     HCH
                                               hCH
                                      h-OH
                                                                             0)
                                                                             oo
                                                                             co
                                                                                 CO
                                                                                 CD
                                                                             CD


                                                                             CO
                                                                             CO
                                                                                  CO
                  IO
                                              iq
                                              co
                                              q
                                              co
                                                                o
                                      eOl-

-------
Figure  2.  Projections  of'Chloride Concentration Over Time in
          Response to  Current External Loads.
    0    50    100   150   200   250   300   350   400  450  500
  (1975)       (2075)      (2175)      (2275)      (2375)      (2475)
                           Time (years)

-------
(SJE9A) 9uijj_ esuodsey onnaipAH
CO
CM
O  O
T-  O
  \    \ \
 O
 CO
    in
    CN
                            co
               o
               in
O
O
CN
                                                             O CO
                                                             S C
                                                             O O
                                                             ^^ •*-•

                                                                O
                                                             8
                                                             CO C
                                                                •Q
                                                                •C
                                                                CO
                                                                CO
                                                             O 13
                                                             o o
                                                                •a
                                                             o  S
                                                             o  Q
                                                             •^- —J
                                                                 CD
                                                                •JO

                                                             O  O
O

                                                                 iJ 60
                                                                   C
                                                                 « CO
                                                                 •o cs
                                                                 •H
                                                                 to (9
                                                                 O
        (-|/BUU) UOJ1B4U90UOQ
                                                                    O •
                                                                   K-l 0}
                                                                  00 -H H
                                                                 •H JJ
                                                                 J= «J C
                                                                  O h O
                                                                 •H ±J -H
                                                                 S C 4J
                                                                   0) C
                                                                  « u a>
                                                                 ^ c u
                                                                  a o a>
                                                                 J o Q
            en

            0)
            to
            3
            00
            •H

-------

-------